Compare commits

..

609 Commits

Author SHA1 Message Date
Chad Versace
5520221118 vk: Remove unneeded vulkan-138.h 2015-07-15 17:16:07 -07:00
Chad Versace
73a8f9543a vk: Bump vulkan.h version to 0.138 2015-07-15 17:16:07 -07:00
Chad Versace
55781f8d02 vk/0.138: Update VkResult values 2015-07-15 17:16:07 -07:00
Chad Versace
756d8064c1 vk/0.132: Do type-safety 2015-07-15 17:16:07 -07:00
Jason Ekstrand
927f54de68 vk/cmd_buffer: Move batch buffer padding to anv_batch_bo_finish() 2015-07-15 17:11:04 -07:00
Jason Ekstrand
9c0db9d349 vk/cmd_buffer: Rename bo_count to exec2_bo_count 2015-07-15 16:56:29 -07:00
Jason Ekstrand
6037b5d610 vk/cmd_buffer: Add a helper for allocating dynamic state
This matches what we do for surface state and makes the dynamic state pool
more opaque to things that need to get dynamic state.
2015-07-15 16:56:29 -07:00
Jason Ekstrand
7ccc8dd24a vk/private.h: Move cmd_buffer functions to near the cmd_buffer struct 2015-07-15 16:56:29 -07:00
Jason Ekstrand
d22d5f25fc vk: Split command buffer state into its own structure
Everything else in anv_cmd_buffer is the actual guts of the datastructure.
2015-07-15 16:56:29 -07:00
Jason Ekstrand
da4d9f6c7c vk: Move most of the anv_Cmd related stuff to its own file 2015-07-15 16:56:28 -07:00
Jason Ekstrand
d862099198 vk: Pull the guts of anv_cmd_buffer into its own file 2015-07-15 16:56:28 -07:00
Chad Versace
498ae009d3 vk/glsl: Replace raw casts
Needed for upcoming type-safety changes.
2015-07-15 15:51:37 -07:00
Chad Versace
6f140e8af1 vk/meta: Remove raw casts
Needed for upcoming type-safety changes.
2015-07-15 15:51:37 -07:00
Chad Versace
badbf0c94a vk/x11: Remove raw casts
The raw casts in the WSI functions will break the build when the
type-safety changes arrive.
2015-07-15 15:49:10 -07:00
Chad Versace
61a4bfe253 vk: Delete vkDbgSetObjectTag()
Because VkObject is going away.
2015-07-15 15:34:20 -07:00
Jason Ekstrand
e1c78ebe53 vk/device: Remove unneeded checks for NULL 2015-07-15 15:22:32 -07:00
Jason Ekstrand
f4748bff59 vk/device: Provide proper NULL handling in anv_device_free
The Vulkan spec does not specify that the free function provided to
CreateInstance must handle NULL properly so we do it in the wrapper.  If
this ever changes in the spec, we can delete the extra 2 lines.
2015-07-15 15:22:32 -07:00
Chad Versace
4c8e1e5888 vk: Stop internally calling anv_DestroyObject()
Replace each anv_DestroyObject() with anv_DestroyFoo().

Let vkDestroyObject() live for a while longer for Crucible's sake.
2015-07-15 15:11:16 -07:00
Chad Versace
f5ad06eb78 vk: Fix vkDestroyObject dispatch for VkRenderPass
It called anv_device_free() instead of anv_DestroyRenderPass().
2015-07-15 15:07:41 -07:00
Chad Versace
188f2328de vk: Fix vkCreate/DestroyRenderPass
While updating vkDestroyObject, I discovered that vkDestroyPass reliably
crashes. That hasn't been an issue yet, though, because it is never
called.

In vkCreateRenderPass:
    - Don't allocate empty attachment arrays.
    - Ensure that pointers to empty attachment arrays are NULL.
    - Store VkRenderPassCreateInfo::subpassCount as
      anv_render_pass::subpass_count.

In vkDestroyRenderPass:
    - Fix loop bounds: s/attachment_count/subpass_count/
    - Don't call anv_device_free on null pointers.
2015-07-15 15:07:41 -07:00
Chad Versace
c6270e8044 vk: Refactor create/destroy code for anv_descriptor_set
Define two new functions:
    anv_descriptor_set_create
    anv_descriptor_set_destroy
2015-07-15 14:31:22 -07:00
Chad Versace
365d80a91e vk: Replace some raw casts with safe casts
That is, replace some instances of
    (VkFoo) foo
with
    anv_foo_to_handle(foo)
2015-07-15 14:00:21 -07:00
Chad Versace
7529e7ce86 vk: Correct anv_CreateShaderModule's prototype
s/VkShader/VkShaderModule/

:sigh: I look forward to type-safety.
2015-07-15 13:59:47 -07:00
Chad Versace
8213be790e vk: Define struct anv_image_view, anv_buffer_view
Follow the pattern of anv_attachment_view. We need these structs to
implement the type-safety that arrived in the 0.132 header.
2015-07-15 12:19:29 -07:00
Chad Versace
43241a24bc vk/meta: Fix declared type of a shader module
s/VkShader/VkShaderModule/

I'm looking forward to a type-safe vulkan.h ;)
2015-07-15 11:49:37 -07:00
Chad Versace
94e473c993 vk: Remove struct anv_object
Trivial removal because vkDestroyObject() no longer uses it.
2015-07-15 11:29:43 -07:00
Jason Ekstrand
e375f722a6 vk/device: More documentation on surface state flushing 2015-07-15 11:09:02 -07:00
Connor Abbott
9aabe69028 vk/device: explain why a flush is necessary
Jason found this from experimenting, but the docs give a reasonable
explanation of why it's necessary.
2015-07-14 23:03:19 -07:00
Chad Versace
5f46c4608f vk: Fix indentation of anv_dynamic_cb_state 2015-07-14 18:19:10 -07:00
Chad Versace
0eeba6b80c vk: Add finishmes for VkDescriptorPool
VkDescriptorPool is a stub object. As a consequence, it's impossible to
free descriptor set memory.
2015-07-14 18:19:00 -07:00
Jason Ekstrand
2b5a4dc5f3 vk: Add vulkan-138 and remove vulkan-0.132
Now, 138 is the target and not 132.  Once object destruction is finished,
we can delete 138 as it will be identical to vulkan.h
2015-07-14 17:54:13 -07:00
Jason Ekstrand
1f658bed70 vk/device: Add stub support for command pools
Real support isn't really that far away.  We just need a data structure
with a linked list and a few tests.
2015-07-14 17:40:00 -07:00
Jason Ekstrand
ca7243b54e vk/vulkan.h: Add the stuff for cross-queue resource sharing
We only have one queue, so this is currently a no-op on our implementation.
2015-07-14 17:20:50 -07:00
Jason Ekstrand
553b4434ca vk/vulkan.h: Add a couple of size fields for specialization constants 2015-07-14 17:12:39 -07:00
Jason Ekstrand
e5db209d54 vk/vulkan.h: Move around buffer image granularities 2015-07-14 17:10:37 -07:00
Jason Ekstrand
c7fcfebd5b vk: Add stubs for all the sparse resource stuff 2015-07-14 17:06:11 -07:00
Jason Ekstrand
2a9136feb4 vk/image: Add a stub for the new ImageFormatProperties function
This lets the client query about things like multisample.  We don't do
multisample right now, so I'll let Chad deal with that when he gets to it.
2015-07-14 17:05:30 -07:00
Jason Ekstrand
2c4dc92f40 vk/vulkan.h: Rename FormatInfo to FormatProperties 2015-07-14 17:04:46 -07:00
Jason Ekstrand
d7f44852be vk/vulkan.h: Re-order some #define's 2015-07-14 16:41:39 -07:00
Jason Ekstrand
1fd3bc818a vk/vulkan.h: Rename a function parameter 2015-07-14 16:39:01 -07:00
Jason Ekstrand
2e2f48f840 vk: Remove abreviations 2015-07-14 16:34:31 -07:00
Jason Ekstrand
02db21ae11 vk: Add the new extension/layer enumeration entrypoints 2015-07-14 16:11:21 -07:00
Jason Ekstrand
a463eacb8f vk/vulkan.h: Change maxAnisotropy to a float 2015-07-14 15:04:11 -07:00
Jason Ekstrand
98957b18d2 vk/vulkan.h: Add the VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT flag 2015-07-14 15:03:39 -07:00
Jason Ekstrand
a35811d086 vk/vulkan.h: Rename a couple of function parameters
No functional change.
2015-07-14 15:03:01 -07:00
Jason Ekstrand
55723e97f1 vk: Split the memory requirements/binding functions 2015-07-14 14:59:39 -07:00
Jason Ekstrand
ccb2e5cd62 vk: Make barriers more precise (rev. 133) 2015-07-14 14:50:35 -07:00
Jason Ekstrand
30445f8f7a vk: Split the dynamic state binding function into one per state 2015-07-14 14:26:10 -07:00
Jason Ekstrand
d2c0870ff3 vk/vulkan.h: Rename a function parameter to match 132 2015-07-14 14:11:04 -07:00
Jason Ekstrand
8478350992 vk: Implement Multipass 2015-07-14 11:37:14 -07:00
Jason Ekstrand
68768c40be vk/vulkan.h: Re-arrange some enums and definitions in preparation for 131 2015-07-14 11:32:15 -07:00
Chad Versace
66cbb7f76d vk/0.132: Add vkDestroyRenderPass() 2015-07-14 11:21:31 -07:00
Chad Versace
6d0ed38db5 vk/0.132: Add vkDestroy*View()
vkDestroyColorAttachmentView
vkDestroyDepthStencilView

These functions are not in the 0.132 header, but adding them will help
us attain the type-safety API updates more quickly.
2015-07-14 11:19:22 -07:00
Chad Versace
1ca611cbad vk/0.132: Add vkDestroyCommandBuffer() 2015-07-14 11:11:41 -07:00
Chad Versace
6eec0b186c vk/0.132: Add vkDestroyImageView()
Just declare it in vulkan.h. Jason defined the function earlier
in image.c.
2015-07-14 11:09:14 -07:00
Chad Versace
4b2c5a98f0 vk/0.132: Add vkDestroyBufferView()
Just declare it in vulkan.h. Jason already defined the function
earlier in vulkan.c.
2015-07-14 11:06:57 -07:00
Chad Versace
08f7731f67 vk/0.132: Add vkDestroyFramebuffer() 2015-07-14 10:59:30 -07:00
Chad Versace
0c8456ef1e vk/0.132: Add vkDestroyDynamicDepthStencilState() 2015-07-14 10:54:51 -07:00
Chad Versace
b29c929e8e vk/0.132: Add vkDestroyDynamicColorBlendState() 2015-07-14 10:52:45 -07:00
Chad Versace
5e1737c42f vk/0.132: Add vkDestroyDynamicRasterState() 2015-07-14 10:51:08 -07:00
Chad Versace
d80fea1af6 vk/0.132: Add vkDestroyDynamicViewportState() 2015-07-14 10:42:45 -07:00
Chad Versace
9250e1e9e5 vk/0.132: Add vkDestroyDescriptorPool() 2015-07-14 10:38:22 -07:00
Chad Versace
f925ea31e7 vk/0.132: Add vkDestroyDescriptorSetLayout() 2015-07-14 10:36:49 -07:00
Chad Versace
ec5e2f4992 vk/0.132: Add vkDestroySampler() 2015-07-14 10:34:00 -07:00
Chad Versace
a684198935 vk/0.132: Add vkDestroyPipelineLayout() 2015-07-14 10:29:47 -07:00
Chad Versace
6e5ab5cf1b vk/0.132: Add vkDestroyPipeline() 2015-07-14 10:26:17 -07:00
Chad Versace
114015321e vk/0.132: Add vkDestroyPipelineCache() 2015-07-14 10:19:27 -07:00
Chad Versace
cb57bff36c vk/0.132: Add vkDestroyShader() 2015-07-14 10:16:22 -07:00
Chad Versace
8ae8e14ba7 vk/0.132: Add vkDestroyShaderModule() 2015-07-14 10:13:09 -07:00
Chad Versace
dd67c134ad vk/0.132: Add vkDestroyImage()
We only need to add it to vulkan.h because Jason defined the function
earlier in image.c.
2015-07-14 10:13:00 -07:00
Chad Versace
e18377f435 vk/0.132: Dispatch vkDestroyObject to new destructors
Oops. My recent commits added new destructors, but forgot to teach
vkDestroyObject about them. They are:
  vkDestroyFence
  vkDestroyEvent
  vkDestroySemaphore
  vkDestroyQueryPool
  vkDestroyBuffer
2015-07-14 09:58:22 -07:00
Chad Versace
e93b6d8eb1 vk/0.132: Add vkDestroyBuffer() 2015-07-14 09:47:45 -07:00
Chad Versace
584cb7a16f vk/0.132: Add vkDestroyQueryPool() 2015-07-14 09:44:58 -07:00
Chad Versace
68c7ef502d vk/0.132: Add vkDestroyEvent() 2015-07-14 09:33:47 -07:00
Chad Versace
549070b18c vk/0.132: Add vkDestroySemaphore() 2015-07-14 09:31:34 -07:00
Chad Versace
ebb191f145 vk/0.132: Add vkDestroyFence() 2015-07-14 09:29:35 -07:00
Chad Versace
435ccf4056 vk/0.132: Rename VkDynamic*State types
sed -i -e 's/VkDynamicVpState/VkDynamicViewportState/g' \
       -e 's/VkDynamicRsState/VkDynamicRasterState/g' \
       -e 's/VkDynamicCbState/VkDynamicColorBlendState/g' \
       -e 's/VkDynamicDsState/VkDynamicDepthStencilState/g' \
       $(git ls-files include/vulkan src/vulkan)
2015-07-13 16:19:28 -07:00
Connor Abbott
ffb51fd112 nir/spirv: update to SPIR-V revision 31
This means that now the internal version of glslangValidator is
required. This includes some changes due to the sampler/texture rework,
but doesn't actually enable anything more yet. We also don't yet handle
UBO's correctly, and don't handle matrix stride and row major/column
major yet.
2015-07-13 15:01:01 -07:00
Chad Versace
45f8723f44 vk/0.132: Move VkQueryControlFlags 2015-07-13 13:09:32 -07:00
Chad Versace
180c07ee50 vk/0.132: Move VkImageAspectFlags 2015-07-13 13:08:56 -07:00
Chad Versace
4b05a8cd31 vk/0.132: Move VkCmdBufferOptimizeFlags 2015-07-13 13:08:07 -07:00
Chad Versace
f1cf55fae6 vk/0.132: Move VkWaitEvent 2015-07-13 13:06:53 -07:00
Chad Versace
3112098776 vk/0.132: Move VkCmdBufferLevel 2015-07-13 13:06:33 -07:00
Chad Versace
c633ab5822 vk/0.132: Drop VK_ATTACHMENT_STORE_OP_RESOLVE_MSAA 2015-07-13 13:05:24 -07:00
Chad Versace
8f3b2187e1 vk/0.132: Rename bool32_t -> VkBool32
sed -i 's/bool32_t/VkBool32/g' \
  $(git ls-files src/vulkan include/vulkan)
2015-07-13 13:03:36 -07:00
Chad Versace
77dcfe3c70 vk/0.132: Remove stray typedef 2015-07-13 12:58:17 -07:00
Chad Versace
601d0891a6 vk/0.132: Move VKImageUsageFlags 2015-07-13 12:48:44 -07:00
Chad Versace
829810fa27 vk/0.132: Move VkImageType and VkImageTiling 2015-07-13 11:49:56 -07:00
Chad Versace
17c8232ecf vk/0.132: Import the 0.132 header
Import it as vulkan-0.132.h.
2015-07-13 11:47:12 -07:00
Chad Versace
a158ff55f0 vk/vulkan.h: Remove headers for old API versions
Remove the temporary headers for 0.90 and 0.130.
2015-07-13 11:46:30 -07:00
Chad Versace
1c4238a8e5 vk/0.130: Bump header version to 0.130
All APIs have been updated. This eliminates the diff between the
work-in-progress header and the 0.130 header.
2015-07-10 20:06:09 -07:00
Chad Versace
f43a304dc6 vk/0.130: Update vkAllocMemory to use VkMemoryType 2015-07-10 17:35:52 -07:00
Chad Versace
df2a013881 vk/0.130: Implement vkGetPhysicalDeviceMemoryProperties() 2015-07-10 17:35:52 -07:00
Chad Versace
c7f512721c vk/gem: Change signature of anv_gem_get_aperture()
Replace the anv_device parameter with anv_physical_device, because this needs
querying before vkCreateDevice.
2015-07-10 17:35:52 -07:00
Chad Versace
8cda3e9b1b vk/device: Add member anv_physical_device::fd
During anv_physical_device_init(), we opend the DRM device to do some
queries, then promptly closed it. Now we keep it open for the lifetime
of the anv_physical_device so that we can query it some more during
vkGetPhysicalDevice*Properties() [which will happen in follow-up
commits].
2015-07-10 17:35:52 -07:00
Chad Versace
4422bd4cf6 vk/device: Add func anv_physical_device_finish()
Because in a follow-up patch I need to do some non-trival teardown on
anv_physical_device. Currently, however, anv_physical_device_finish() is
currently a no-op that's just called in the right place.

Also, rename function fill_physical_device -> anv_physical_device_init
for symmetry.
2015-07-10 17:35:52 -07:00
Jason Ekstrand
7552e026da vk/device: Add an explicit destructor for RenderPass 2015-07-10 12:33:04 -07:00
Jason Ekstrand
8b342b39a3 vk/image: Add an explicit DestroyImage function 2015-07-10 12:30:58 -07:00
Jason Ekstrand
b94b8dfad5 vk/image: Add explicit constructors for buffer/image view types 2015-07-10 12:26:31 -07:00
Jason Ekstrand
18340883e3 nir: Add C++ versions of NIR_(SRC|DEST)_INIT 2015-07-10 11:57:33 -07:00
Chad Versace
9e64a2a8e4 mesa: Fix generation of git_sha1.h.tmp for gitlinks
Don't assume that $(top_srcdir)/.git is a directory. It may be a
gitlink file [1] if $(top_srcdir) is a submodule checkout or a linked
worktree [2].

[1] A "gitlink" is a text file that specifies the real location of
    the gitdir.
[2] Linked worktrees are a new feature in Git 2.5.

Cc: "10.6, 10.5" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
(cherry picked from commit 75784243df)
2015-07-10 11:24:25 -07:00
Jason Ekstrand
19f0a9b582 vk/query.c: Use the casting functions 2015-07-09 20:32:44 -07:00
Jason Ekstrand
6eb221c884 vk/pipeline.c: Use the casting functions 2015-07-09 20:28:08 -07:00
Jason Ekstrand
fb4e2195ec vk/formats.c: Use the casting functions 2015-07-09 20:24:17 -07:00
Jason Ekstrand
a52e208203 vk/image.c: Use the casting functions 2015-07-09 20:24:07 -07:00
Jason Ekstrand
b1de1d4f6e vk/device.c: One more use of a casting function 2015-07-09 20:23:46 -07:00
Jason Ekstrand
8739e8fbe2 vk/meta.c: Use the casting functions 2015-07-09 20:16:13 -07:00
Jason Ekstrand
92556c77f4 vk: Fix the build 2015-07-09 18:59:08 -07:00
Jason Ekstrand
098209eedf device.c: Use the cast helpers a bunch of places 2015-07-09 18:49:43 -07:00
Jason Ekstrand
73f9187e33 device.c: Use the cast helpers 2015-07-09 18:41:27 -07:00
Jason Ekstrand
7d24fab4ef vk/private.h: Add a bunch of static inline casting functions
We will need these as soon as we turn on type saftey.  We might as well
define and start using them now rather than later.
2015-07-09 18:40:54 -07:00
Jason Ekstrand
5c49730164 vk/device.c: Fix whitespace issues 2015-07-09 18:20:28 -07:00
Jason Ekstrand
c95f9b61f2 vk/device.c: Use ANV_FROM_HANDLE a bunch of places 2015-07-09 18:20:10 -07:00
Jason Ekstrand
335e88c8ee vk/vulkan.h: Add the pEnabledFeatures field to DeviceCreateInfo 2015-07-09 16:21:31 -07:00
Jason Ekstrand
34871cf7f3 vk/vulkan.h: Change the MsCreateInfo structure to the 130 version
We do nothing with it at the moment, so this is a no-op.
2015-07-09 16:19:54 -07:00
Jason Ekstrand
8c2c37fae7 vk: Remove the old GetPhysicalDeviceInfo call 2015-07-09 16:14:37 -07:00
Jason Ekstrand
1f907011a3 vk: Add the new PhysicalDeviceQueue queries 2015-07-09 16:14:37 -07:00
Jason Ekstrand
977a469bce vk: Support GetPhysicalDeviceProperties 2015-07-09 16:14:37 -07:00
Jason Ekstrand
65e0b304b6 vk: Add support for GetPhysicalDeviceLimits 2015-07-09 16:14:37 -07:00
Jason Ekstrand
f6d51f3fd3 vk: Add GetPhysicalDeviceFeatures 2015-07-09 16:14:37 -07:00
Chad Versace
5b75dffd04 vk/device: Fix vkEnumeratePhysicalDevices()
The Vulkan spec says that pPhysicalDeviceCount is an out parameter if
pPhysicalDevices is NULL; otherwise it's an inout parameter.

Mesa incorrectly treated it unconditionally as an inout parameter, which
could have lead to reading unitialized data.
2015-07-09 15:53:21 -07:00
Chad Versace
fa915b661d vk/device: Move device enumeration to vkEnumeratePhysicalDevices()
Don't enumerate devices in vkCreateInstance(). That's where global,
device-independent initialization should happen. Move device enumeration
to the more logical location, vkEnumeratePhysicalDevices().
2015-07-09 15:41:17 -07:00
Chad Versace
c34d314db3 vk/device: Be consistent about path to DRM device
Function fill_physical_device() has a 'path' parameter, and struct
anv_physical_device has a 'path' member. Sometimes these are used;
sometimes hardcoded "/dev/dri/renderD128" is used instead.

Be consistent. Hardcode "/dev/dri/renderD128" in exactly one location,
during initialization of the physical device.
2015-07-09 15:27:26 -07:00
Connor Abbott
cff06bbe7d vk/compiler: create an empty parameters list
Prevents problems when initializing the sanity_param_count.
2015-07-09 14:29:23 -04:00
Connor Abbott
3318a86d12 nir/spirv: fix wrong writemask for ALU operations 2015-07-09 14:28:39 -04:00
Connor Abbott
b8fedc19f5 nir/spirv: fix memory context for builtin variable
Fixes valgrind errors with func.depthstencil.basic.
2015-07-08 22:03:30 -04:00
Connor Abbott
e4292ac039 nir/spirv: zero out value array
Before values are pushed or annotated with a name, decoration, etc.,
they need to have an invalid type, NULL name, NULL decoration, etc.
ralloc zero's everything by accident, so this wasn't an issue in
practice, but we should be explicitly zero'ing it.
2015-07-08 22:03:30 -04:00
Connor Abbott
997831868f vk/compiler: create the right kind of program struct
This fixes Valgrind errors and gets all the tests to pass with
--use-spir-v.
2015-07-08 22:03:30 -04:00
Connor Abbott
a841e2c747 vk/compiler: mark inputs/outputs as read/written
This doesn't handle inputs and outputs larger than a vec4, but we plan
to add a varyiing splitting/packing pass to handle those anyways.
2015-07-08 22:03:30 -04:00
Jason Ekstrand
8640dc12dc vk/vulkan.h: Copy the VkStructureType enum from version 130
We now have the exact same structs which require pType.
2015-07-08 17:45:52 -07:00
Jason Ekstrand
5a4ebf6bc1 vk: Move to the new pipeline creation API's 2015-07-08 17:30:18 -07:00
Chad Versace
4fcb32a17d vk/0.130: Remove VkImageViewCreateInfo::minLod
It's now set solely through VkSampler.
2015-07-08 14:48:22 -07:00
Jason Ekstrand
367b9ba78f vk/vulkan.h: Move renderPassContinue from GraphicsBeginInfo to BeginInfo 2015-07-08 14:37:30 -07:00
Jason Ekstrand
d29ec8fa36 vk/vulkan.h: Update to the new UpdateDescriptorSets api 2015-07-08 14:24:56 -07:00
Jason Ekstrand
c8577b5f52 vk: Add a macro for creating anv variables from vulkan handles
This is very helpful for doing the mass bunch of casts at the top of a
function.  It will also be invaluable when we get type saftey in the API.
2015-07-08 14:24:14 -07:00
Chad Versace
ccb27a002c vk/0.130 Update VkObjectType values
Don't import any new enum tokens from the 0.130 header. Just update the
values of existing enums. This reduces the diff by about 16 lines.
2015-07-08 12:53:49 -07:00
Chad Versace
8985dd15a1 vk/0.130: Remove VkDescriptorUpdateMode
Nowhere used.
2015-07-08 12:51:46 -07:00
Chad Versace
e02dfa309a vk/0.130: Remove VK_DEVICE_CREATE_MULTI_DEVICE_IQ_MATCH_BIT 2015-07-08 12:49:48 -07:00
Chad Versace
e9034ed875 vk/0.130: Update vkCmdBlitImage signature
Add VkTexFilter param. Ignored for now.
2015-07-08 12:47:48 -07:00
Jason Ekstrand
aae45ab583 vk/vulkan.h: Add packing parameters to BufferImageCopy 2015-07-08 11:51:34 -07:00
Chad Versace
b4ef7f354b vk/0.130: Remove msaa members of VkDepthStencilViewCreateInfo 2015-07-08 11:50:51 -07:00
Jason Ekstrand
522ab835d6 vk/vulkan.h: Move over to the new border color enums 2015-07-08 11:44:52 -07:00
Jason Ekstrand
7598329774 vk/vulkan.h: Move VkFormatProperties 2015-07-08 11:16:45 -07:00
Jason Ekstrand
52940e8fcf vk/vulkan.h: Add RenderPassBeginContents 2015-07-08 10:57:13 -07:00
Jason Ekstrand
e19d6be2a9 vk/vulkan.h: Add command buffer levels 2015-07-08 10:53:32 -07:00
Jason Ekstrand
c84f2d3b8c vk/vulkan.h: Import the VkPipeEvent enum from 130
Now, VkPipeEventFlags is back in sync with VkPipeEvent
2015-07-08 10:49:46 -07:00
Jason Ekstrand
b20cc72603 vk/vulkan.h: Remove VkFormatInfoType 2015-07-08 10:39:31 -07:00
Jason Ekstrand
8e05bbeee9 vk/vulkan.h: Update extension handling to rev 130 2015-07-08 10:38:07 -07:00
Jason Ekstrand
cc29a5f4be vk/vulkan.h: Move format quering to the physical device 2015-07-08 09:34:47 -07:00
Jason Ekstrand
719fa8ac74 vk/vulkan.h: Remove some peer opening structs and STRUCTURE_TYPE enums 2015-07-08 09:25:13 -07:00
Jason Ekstrand
fc6dcc6227 vk: Add a copy of the v90 header. 2015-07-08 09:23:29 -07:00
Jason Ekstrand
12119282e6 vk/vulkan.h: Remove an unneeded comment 2015-07-08 09:18:09 -07:00
Jason Ekstrand
3c65a1ac14 vk/vulkan.h: Remove the MemoryRange stubs and add sparse stubs 2015-07-08 09:16:48 -07:00
Jason Ekstrand
bb6567f5d1 vk/vulkan.h: Switch BindObjectMemory to a device function and remove the index 2015-07-08 09:04:16 -07:00
Jason Ekstrand
e7acdda184 vk/vulkan.h: Switch to the split ProcAddr functions in 130 2015-07-07 18:51:53 -07:00
Jason Ekstrand
db24afee2f vk/vulkan.h: Switch from GetImageSubresourceInfo to GetImageSubresourceLayout 2015-07-07 18:20:18 -07:00
Jason Ekstrand
ef8980e256 vk/vulkan.h: Switch from GetObjectInfo to GetMemoryRequirements 2015-07-07 18:16:42 -07:00
Jason Ekstrand
d9c2caea6a vk: Update memory flushing functions to 130
This involves updating the prototype for FlushMappedMemory, adding
InvalidateMappedMemoryRanges, and removing PinSystemMemory.
2015-07-07 17:22:31 -07:00
Jason Ekstrand
d5349b1b18 vk/vulkan.h: Constify the pFences parameter to ResetFences 2015-07-07 17:18:00 -07:00
Jason Ekstrand
6aa1b89457 vk/vulkan.h: Move the definitions of Create(Framebuffer|RenderPass)
This better matches the 130 header.
2015-07-07 17:13:10 -07:00
Jason Ekstrand
0ff06540ae vk: Implement the GetRenderAreaGranularity function
At the moment, we're just going to scissor clears so a granularity of 1x1
is all we need.
2015-07-07 17:11:37 -07:00
Jason Ekstrand
435b062b26 vk/vulkan.h: Add a PipelineLayout parameter to BindDescriptorSets 2015-07-07 17:06:10 -07:00
Jason Ekstrand
518ca9e254 vk/vulkan.h: Add a compareEnable parameter to SamplerCreateInfo
Our hardware doesn't actually need this, so adding it is a no-op.
2015-07-07 16:49:04 -07:00
Jason Ekstrand
672590710b vk/vulkan.h: Remove initialCount from SemaphoreCreateInfo 2015-07-07 16:42:42 -07:00
Jason Ekstrand
80046a7d54 vk/vulkan.h: Update clear color handling to 130 2015-07-07 16:37:43 -07:00
Jason Ekstrand
3e4b00d283 meta: Use the VkClearColorValue structure for the color attribute 2015-07-07 16:27:06 -07:00
Jason Ekstrand
a35fef1ab2 vk/vulkan.h: Remove the pass argument from EndRenderPass 2015-07-07 16:22:23 -07:00
Jason Ekstrand
d2ca7e24b4 vk/vulkan.h: Rename VertexInputStateInfo to VertexInputStateCreateInfo 2015-07-07 16:15:55 -07:00
Jason Ekstrand
abbb776bbe vk/vulkan.h: Remove programPointSize
Instead, we auto-detect whether or not your shader writes gl_PointSize.  If
it does, we use 1.0, otherwise we take it from the shader.
2015-07-07 16:00:46 -07:00
Chad Versace
e7ddfe03ab vk/0.130: Stub vkCmdClear*Attachment() funcs
vkCmdClearColorAttachment
vkCmdClearDepthStencilAttachment
2015-07-07 15:57:37 -07:00
Chad Versace
f89e2e6304 vk/0.130: Define enum VkImageAspectFlagBits 2015-07-07 15:57:37 -07:00
Chad Versace
55ab1737d3 vk/0.130: Define VkRect3D 2015-07-07 15:55:53 -07:00
Chad Versace
11901a9100 vk/0.130: Update name of vkCmdClearDepthStencilImage() 2015-07-07 15:53:35 -07:00
Chad Versace
dff32238c7 vk/0.130: Stub vkCmdExecuteCommands() 2015-07-07 15:51:55 -07:00
Chad Versace
85c0d69be9 vk/0.130: Update vkCmdWaitEvents() signature 2015-07-07 15:49:57 -07:00
Chad Versace
0ecb789b71 vk: Remove unused 'v' param from stub() macro 2015-07-07 15:47:24 -07:00
Chad Versace
f78d684772 vk: Stub vkCmdPushConstants() from 0.130 header 2015-07-07 15:46:19 -07:00
Chad Versace
18ee32ef9d vk: Update vkCmdPipelineBarrier to 0.130 header 2015-07-07 15:43:41 -07:00
Chad Versace
4af79ab076 vk: Add func anv_clear_mask()
A little helper func for inspecting and clearing bitmasks.
2015-07-07 15:43:41 -07:00
Jason Ekstrand
788a8352b9 vk/vulkan.h: Remove some unused fields.
In particular, the following are removed:

 - disableVertexReuse
 - clipOrigin
 - depthMode
 - pointOrigin
 - provokingVertex
2015-07-07 15:33:00 -07:00
Jason Ekstrand
7fbed521bb vk/vulkan.h: Remove the explicit primitive restart index
Unfortunately, this requires some non-trivial changes to the driver.  Now
that the primitive restart index isn't given explicitly by the client, we
always use ~0 for everything like D3D does.  Unfortunately, our hardware is
awesome and a 32-bit version of ~0 doesn't match any 16-bit values.  This
means, we have to set it to either UINT16_MAX or UINT32_MAX depending on
the size of the index type.  Since we get the index type from
CmdBindIndexBuffer and the rest of the VF packet from the pipeline, we need
to lazy-emit the VF packet.
2015-07-07 15:33:00 -07:00
Chad Versace
d6b840beff vk: Delete some comments not present in 0.130 header
Deleting the comments reduces diff noise.
2015-07-07 15:16:13 -07:00
Chad Versace
84a5bc25e3 vk: Pull in remaining 0.130 handle types
This pulls in the definition of VkShaderModule and VkPipelineCache,
which nowhere used yet.
2015-07-07 15:13:01 -07:00
Chad Versace
f2899b1af2 vk: Pull in #defines from 0.130 header
Despite not being used yet, pulling in the macros does diminish the
header diff.
2015-07-07 15:11:30 -07:00
Jason Ekstrand
962d6932fa vk/vulkan.h: Rename (min|max)Depth to (min|max)DepthBounds 2015-07-07 12:37:54 -07:00
Jason Ekstrand
1fb859e4b2 vk/vulkan.h: Remove client-settable pointSize from DynamicRsState 2015-07-07 12:35:32 -07:00
Jason Ekstrand
245583075c vk/vulkan.h: Remove UINT8 index buffers 2015-07-07 11:26:49 -07:00
Jason Ekstrand
0a42332904 vk/vulkan.h: Re-order the object declarations 2015-07-07 11:26:49 -07:00
Kristian Høgsberg Kristensen
a1eea996d4 vk: Emit 3DSTATE_SAMPLE_MASK
This was missing and was causing the driver to not work with
execlists. Presumably we get a different initial hw context with
execlists enabled, that has sample mask 0 initially.

Set this to 0xffff for now.  When we add MS support, we need to take the
value from VkPipelineMsStateCreateInfo::sampleMask.
2015-07-06 23:54:12 -07:00
Kristian Høgsberg Kristensen
c325bb24b5 vk: Pull in new generated headers
The new headers use stdbool for enable/disable fields which
implicitly converts expressions like (flags & 8) to 0 or 1.
Also handles MBO (must-be-one) fields by setting them to one,
corrects a bspec typo (_3DPRIM_LISTSTRIP_ADJ -> LINESTRIP) and
makes a few enum values less clashy.
2015-07-06 22:12:26 -07:00
Chad Versace
23075bccb3 vk/image: Validate vkCreateImageView more
Exhaustively validate the function input.  If it's not validated and
doesn't have an anv_finishme(), then I overlooked it.
2015-07-06 18:28:26 -07:00
Chad Versace
69e11adecc vk/image: Add more info to VkImageViewType table
Convert the table from the direct mapping
  VkImageViewType -> SurfaceType

into a mapping to an info struct
  VkImageViewType -> struct anv_image_view_info
2015-07-06 18:28:26 -07:00
Chad Versace
b844f542e0 vk: Update VkImageViewType to 0.130.0
This splits 1D and 1D_ARRAY, 2D and 2D_ARRAY, CUBE and CUBE_ARRAY.

The new tokens are unused. This is just a header update.
2015-07-06 18:28:26 -07:00
Chad Versace
5b04db71ff vk/image: Move validation for vkCreateImageView
Move the validation from anv_CreateImageView() and anv_image_view_init()
to anv_validate_CreateImageView(). No new validation is added.
2015-07-06 18:27:14 -07:00
Jason Ekstrand
1f1b26bceb vk/vulkan.h: Rename VkRect to VkRect2D 2015-07-06 17:47:18 -07:00
Jason Ekstrand
63c1190e47 vk/vulkan.h: Rename count to arraySize in VkDescriptorSetLayoutBinding 2015-07-06 17:43:58 -07:00
Jason Ekstrand
d84f3155b1 vk/vulkan.h: Remove the Vk(Memory|Semaphor|Image)OpenInfo structs
We already deleted the functions that need them.  The structs are just
dangling uselessly.
2015-07-06 17:37:13 -07:00
Jason Ekstrand
65f9ccb4e7 vk/vulkan.h: Remove VK_MEMORY_PROPERTY_PREFER_HOST_LOCAL_BIT
We weren't doing anything with it, so this is a no-op
2015-07-06 17:33:45 -07:00
Jason Ekstrand
68fa750f2e vk/vulkan.h: Replace DEVICE_COHERENT_BIT with DEVICE_NON_COHERENT_BIT 2015-07-06 17:32:28 -07:00
Jason Ekstrand
d5b5bd67f6 vk/vulkan.h: Use the query result bits from revision 130
None of the important bits or names actually changed.  It just
added/removed some no-op names.

No functional change.
2015-07-06 17:27:11 -07:00
Jason Ekstrand
d843418c2e vk/vulkan.h: One more quick enum refactor clean-up 2015-07-06 17:26:29 -07:00
Jason Ekstrand
2b37fc28d1 vk/vulkan.h: Get rid of VERTEX_INPUT_STEP_RATE_DRAW
We never supported it, so no functional change.
2015-07-06 17:24:26 -07:00
Jason Ekstrand
a75967b1bb vk/vulkan.h: Remove the CLEAR_OPTIMAL image layout 2015-07-06 17:21:19 -07:00
Jason Ekstrand
2b404e5d00 vk: Rename CPU_READ/WRITE_BIT to HOST_READ/WRITE_BIT 2015-07-06 17:18:25 -07:00
Jason Ekstrand
c57ca3f16f vk/vulkan.h: Remove VK_IMAGE_CREATE_CLONEABLE_BIT 2015-07-06 17:14:30 -07:00
Jason Ekstrand
2de388c49c vk: Remove SHAREABLE bits
They were removed from the Vulkan API and we don't really use them because
there are no multi-GPU i965 systems.
2015-07-06 17:12:51 -07:00
Jason Ekstrand
1b0c47bba6 vk/vulkan.h: Re-order the logic op enums 2015-07-06 17:08:11 -07:00
Jason Ekstrand
c7cef662d0 vk/vulkan.h: Reformat a bunch of enums to match revision 130
In theory, no functional change.
2015-07-06 17:06:02 -07:00
Jason Ekstrand
8c5e48f307 vk: Rename NUM_SHADER_STAGE to SHADER_STAGE_NUM
This is a refactor of more than just the header but it lets us finish
reformating the shader stage enum.
2015-07-06 16:43:28 -07:00
Jason Ekstrand
d9176f2ec7 vk: Reformat a bunch of enums
This accounts for a number differences between the generated headers and
the hand-written header.  Not all reformatting is done in this commit but
it does make the headers much more diffable.

In theory, no functional change.
2015-07-06 16:41:31 -07:00
Jason Ekstrand
e95bf93e5a vk: Pull the VkResult enum from revision 130 2015-07-06 16:15:12 -07:00
Jason Ekstrand
1b7b580756 vk: re-arrange enums to match the order in revision 130 2015-07-06 16:11:05 -07:00
Jason Ekstrand
2fb524b369 vk: Rename a parameter in CmdBindDynamicStateObject 2015-07-06 15:37:17 -07:00
Jason Ekstrand
c5ffcc9958 vk: Remove multi-device stuff 2015-07-06 15:34:55 -07:00
Jason Ekstrand
c5ab5925df vk: Remove ClearDescriptorSets 2015-07-06 15:32:40 -07:00
Jason Ekstrand
ea5fbe1957 vk: Remove begin/end descriptor pool update 2015-07-06 15:32:27 -07:00
Jason Ekstrand
9a798fa946 vk: Remove stub for CloneImageData 2015-07-06 15:30:05 -07:00
Jason Ekstrand
78a0d23d4e vk: Remove the stub support for memory priorities 2015-07-06 15:28:10 -07:00
Jason Ekstrand
11cf214578 vk: Remove the stub support for explicit memory references 2015-07-06 15:27:58 -07:00
Jason Ekstrand
0dc7d4ac8a vk/vulkan.h: Reformat structs to match revision 130
Structs in the old version were specified as

typedef struct VkSomeThing_
{
   type                                        field; // comment
} VkSomeThing;

However, in the generated headers, you have

typedef struct {
   type                                        field;
} VkSomeThing;

This commit also removes some unneeded whitespaces.
2015-07-06 15:19:12 -07:00
Jason Ekstrand
19aabb5730 vk/vulkah.h: Re-arrange structures to match the order in 130 2015-07-06 15:09:30 -07:00
Connor Abbott
f9dbc34a18 nir/spirv: fix some bugs 2015-07-06 15:00:37 -07:00
Connor Abbott
f3ea3b6e58 nir/spirv: add support for builtins inside structures
We may be able to revert this depending on the outcome of bug 14190, but
for now it gets vertex shaders working with SPIR-V.
2015-07-06 15:00:37 -07:00
Connor Abbott
15047514c9 nir/spirv: fix a bug with structure creation
We were creating 2 extra bogus fields.
2015-07-06 15:00:37 -07:00
Connor Abbott
73351c6a18 nir/spirv: fix a bad assertion in the decoration handling
We should be asserting that the parent decoration didn't hand us
a member if the child decoration did, but different child decorations
may obviously have different members.
2015-07-06 15:00:37 -07:00
Connor Abbott
70d2336e7e nir/spirv: pull out logic for getting builtin locations
Also add support for more builtins.
2015-07-06 15:00:37 -07:00
Connor Abbott
aca5fc6af1 nir/spirv: plumb through the type of dereferences
We need this to know if a deref is of a builtin.
2015-07-06 15:00:37 -07:00
Connor Abbott
66375e2852 nir/spirv: handle structure member builtin decorations 2015-07-06 15:00:37 -07:00
Connor Abbott
23c179be75 nir/spirv: add a vtn_type struct
This will handle decorations that aren't in the glsl_type.
2015-07-06 15:00:37 -07:00
Connor Abbott
f9bb95ad4a nir/spirv: move 'type' into the union
Since SSA values now have their own types, it's more convenient to make
'type' only used when we want to look up an actual SPIR-V type, since
we're going to change its type soon to support various decorations that
are handled at the SPIR-V -> NIR level.
2015-07-06 15:00:37 -07:00
Jason Ekstrand
d5dccc1e7a vk: Move CreateFramebuffer and CreateRenderPass higher in the header
This matches where they are in the 130 header.
2015-07-06 14:41:43 -07:00
Jason Ekstrand
4a42f45514 vk: Remove atomic counters stubs 2015-07-06 14:38:45 -07:00
Jason Ekstrand
630b19a1c8 vk: Make vulkan.h look more like vulkan-130.h
Most of these changes are insubstantial.  The only potentially substantial
cyhange is that we added a few new #defines for API maximums.
2015-07-06 14:32:52 -07:00
Jason Ekstrand
2f9180b1b2 vk: Add a revision 130 header along-side the current header 2015-07-06 14:16:51 -07:00
Jason Ekstrand
1f1465f077 vk/meta: Add an initial implementation of ClearColorImage 2015-07-02 18:15:06 -07:00
Jason Ekstrand
8a6c8177e0 vk/meta: Factor the guts out of cmd_buffer_clear 2015-07-02 18:13:59 -07:00
Jason Ekstrand
beb0e25327 vk: Roll back to API v90
This is what version 0.1 of the Vulkan SDK is built against.
2015-07-01 16:44:12 -07:00
Jason Ekstrand
fa663c27f5 nir/spirv: Add initial structure member decoration support 2015-07-01 15:38:26 -07:00
Jason Ekstrand
e3d60d479b nir/spirv: Make vtn_handle_type match the other handler functions
Previously, the caller of vtn_handle_type had to handle actually inserting
the type.  However, this didn't really work if the type was decorated in
any way.
2015-07-01 15:34:10 -07:00
Jason Ekstrand
7a749aa4ba nir/spirv: Add basic support for Op[Group]MemberDecorate 2015-07-01 14:18:07 -07:00
Jason Ekstrand
682eb9489d vk/x11: Allow for the client querying the size of the format properties 2015-07-01 14:18:07 -07:00
Chad Versace
bba767a9af vk/formats: Fix entry for S8_UINT
I forgot to update this when fixing the depth formats.
2015-06-30 09:41:44 -07:00
Chad Versace
6720b47717 vk/formats: Document new meaning of anv_format::cpp
The way the code currently works is that anv_format::cpp is the cpp of
anv_format::surface_format.

Me and Kristian disagree about how the code *should* work. Despite that,
I think it's in our discussion's best interest to document how the code
*currently* works. That should eliminate confusion.

If and when the code begins to work differently, then we'll update the
anv_format comments.
2015-06-30 09:41:41 -07:00
Chad Versace
709fa463ec vk/depth: Add a FIXME
3DSTATE_DEPTH_BUFFER.Width,Height are wrong.
2015-06-26 22:15:03 -07:00
Chad Versace
5b3a1ceb83 vk/image: Enable 2d single-sample color miptrees
What's been tested, for both image views and color attachment views:

    - VK_FORMAT_R8G8B8A8_UNORM
    - VK_IMAGE_VIEW_TYPE_2D
    - mipLevels: 1, 2
    - baseMipLevel: 0, 1
    - arraySize: 1, 2
    - baseArraySlice: 0, 1

What's known to be broken:

    - Depth and stencil miptrees. To fix this, anv_depth_stencil_view
      needs major rework.
    - VkImageViewType != 2D
    - MSAA

Fixes Crucible tests:

  func.miptree.view-2d.levels02.array01.*
  func.miptree.view-2d.levels01.array02.*
  func.miptree.view-2d.levels02.array02.*
2015-06-26 22:11:15 -07:00
Chad Versace
c6e76aed9d vk/image: Define anv_surface, refactor anv_image
This prepares for upcoming miptree support.

anv_surface is a proxy for color surfaces, depth surfaces, and stencil
surfaces.  Embed two instances of anv_surface into anv_image: the
primary surface (color or depth), and an optional stencil surface.
2015-06-26 21:45:53 -07:00
Chad Versace
127cb3f6c5 vk/image: Reformat function signatures
Reformat them to match Mesa code-style.
2015-06-26 20:12:42 -07:00
Chad Versace
fdcd71f71d vk/image: Embed VkImageCreateInfo* into anv_image_create_info
All function signatures that matched this pattern,
  old: f(const VkImageCreateInfo *, const struct anv_image_create_info *)

were rewritten as
  new: f(const struct anv_image_create_info *)
2015-06-26 20:06:08 -07:00
Chad Versace
ca6cef3302 vk/image: Drop some tmp vars in anv_image_view_init()
Variables 'tile_mode' and 'format' are unneeded.
2015-06-26 19:50:04 -07:00
Chad Versace
9c46ba9ca2 vk/image: Abort on stencil image views
The code doesn't work. Not even close.

Replace the broken code with a FINISHME and abort.
2015-06-26 19:23:21 -07:00
Chad Versace
667529fbaa vk: Reindent struct anv_image 2015-06-26 15:27:20 -07:00
Chad Versace
74e3eb304f vk: Define MIN(a, b) macro 2015-06-26 15:09:07 -07:00
Chad Versace
55752fe94a vk: Rename functions ALIGN_*32 -> align_*32
ALIGN_U32 and ALIGN_I32 are functions, not macros. So stop using
allcaps.
2015-06-26 15:07:59 -07:00
Connor Abbott
6ee082718f Merge branch 'wip/nir-vtn' into vulkan
Adds composites and matrix multiplication, plus some control flow fixes.
2015-06-26 12:14:05 -07:00
Chad Versace
37d6e04ba1 vk/formats: Remove the cpp=0 stencil hack
The format table defined cpp = 0 for stencil-only formats. The real cpp
is 1.

When code begins to lie, especially about stencil buffers, code becomes
increasingly fragile as time progresses, and the damage becomes
increasingly hard to undo. (For precedent, see the painful history of
stencil buffer cpp in the git log for gen6 and gen7 in the i965 driver).
Let's undo the stencil buffer cpp lie now to avoid future pain.

In the format table, set cpp = 1 for VK_FORMAT_S8; replace checks for
cpp == 0; and delete all comments about the hack.
2015-06-26 09:58:22 -07:00
Chad Versace
67a7659d69 vk/image: Refactor anv_image_create()
From my experience with intel_mipmap_tree.c, I learned that for struct's
like anv_image and intel_mipmap_tree, which have sprawling
multi-function construction codepaths, it's easy to mistakenly use
unitialized struct members during construction.

Let's eliminate the risk of using unitialized anv_image members during
construction.  Fill the struct at the function bottom instead of
piecemeal throughout the constructor.
2015-06-26 09:32:59 -07:00
Chad Versace
5d7103ee15 vk/image: Group some assertions closer together
In anv_image_create(), group together the assertions on
VkImageCreateInfo.
2015-06-26 09:05:46 -07:00
Chad Versace
0349e8d607 vk/formats: #undef fmt at end of format table 2015-06-26 07:38:02 -07:00
Chad Versace
068b8a41e2 vk: Fix comment for anv_depth_stencil_view::stencil_qpitch
s/DEPTH/STENCIL/
2015-06-26 07:31:57 -07:00
Chad Versace
7ea707a42a vk/image: Add qpitch fields to anv_depth_stencil_view
For now, hard-code them to 0.
2015-06-25 20:10:16 -07:00
Chad Versace
b91a76de98 vk: Reindent and document struct anv_depth_stencil_view 2015-06-25 20:10:16 -07:00
Chad Versace
ebe1e768b8 vk/formats: Fix incorrect depth formats
anv_format::surface_format was incorrect for Vulkan depth formats.
For example, the format table mapped

    VK_FORMAT_D24_UNORM -> .surface_format = D24_UNORM_X8_UINT
    VK_FORMAT_D32_FLOAT -> .surface_format = D32_FLOAT

but should have mapped

    VK_FORMAT_D24_UNORM -> .surface_format = R24_UNORM_X8_TYPELESS
    VK_FORMAT_D32_FLOAT -> .surface_format = R32_FLOAT

The Crucible test func.depthstencil.basic passed despite the bug, but
only because it did not attempt to texture from the depth surface.

The core problem is that RENDER_SURFACE_STATE.SurfaceFormat and
3DSTATE_DEPTH_BUFFER.SurfaceFormat are distinct types. Considering them
as enum spaces, the two enum spaces have incompatible collisions.

Fix this by adding a new field 'depth_format' to struct anv_format.

Refer to brw_surface_formats.c:translate_tex_format() for precedent.
2015-06-25 20:10:16 -07:00
Chad Versace
45b804a049 vk/image: Rename local variable in anv_image_create()
This function has many local variables for info structs. Having one
named simply 'info' is confusing.  Rename it to 'format_info'.
2015-06-25 20:10:16 -07:00
Chad Versace
528071f004 vk/formats: Fix table entry for R8G8B8_SNORM
Now that anv_formats[] is formatted like a table, buggy entries are
easier to see.
2015-06-25 20:10:16 -07:00
Chad Versace
4c8146313f vk/formats: Rename anv_format::format -> surface_format
I misinterpreted anv_format::format as a VkFormat. Instead, it is
a hardware surface format (RENDER_SURFACE_STATE.SurfaceFormat). Rename
the field to 'surface_format' to make it unambiguous.
2015-06-25 20:10:16 -07:00
Chad Versace
4b8b451a1d vk/formats: Rename anv_format::channels -> num_channels
I misinterpreted anv_format::channels as a bitmask of channels.
Renaming it to 'num_channels' makes it unambiguous.
2015-06-25 20:10:16 -07:00
Chad Versace
af0ade0d6c vk: Reindent struct anv_format 2015-06-25 20:10:16 -07:00
Chad Versace
ae29fd1b55 vk/formats: Don't abbreviate tokens in the format table
Abbreviating the VK_FORMAT_* tokens doesn't help much. To the contrary,
it means grep and ctags can't find them.
2015-06-25 20:10:16 -07:00
Jason Ekstrand
d5e41a3a99 vk/compiler: Add the initial hacks to get SPIR-V up and going 2015-06-25 17:36:35 -07:00
Jason Ekstrand
c4c1d96a01 HACK: Get rid of sanity_param_count for FS 2015-06-25 17:36:34 -07:00
Jason Ekstrand
4f5ef945e0 i965: Don't print the GLSL IR if it doesn't exist 2015-06-25 17:36:34 -07:00
Jason Ekstrand
588acdb431 nir/spirv: Set the right location for shader input/outputs
We need to add FRAG_RESULT_DATA0 etc. to the input/output location.
2015-06-25 17:36:34 -07:00
Jason Ekstrand
333b8ddd6b nir/spirv: Set the interface type on uniform blocks 2015-06-25 17:36:34 -07:00
Jason Ekstrand
7e1792b1b7 nir/spirv: Set the system value mode on builtins 2015-06-25 17:36:34 -07:00
Jason Ekstrand
b72936fdad nir/spirv: Actually put variables on the right linked list 2015-06-25 17:36:34 -07:00
Jason Ekstrand
ee0a8f23e4 glsl: Move vert_attrib varying_slot and frag_result enums to shader_enums.h 2015-06-25 17:36:34 -07:00
Chad Versace
fa352969a2 vk/image: Check extent does not exceed surface type limits 2015-06-25 16:53:24 -07:00
Chad Versace
99031aa0f3 vk/image: Stop hardcoding SurfaceType of VkImageView
Instead, translate VkImageViewType to a gen SurfaceType.
2015-06-25 16:53:22 -07:00
Chad Versace
7ea121687c vk/image: Add anv_image::surf_type
This the gen SurfaceType, such as SURFTYPE_2D.
2015-06-25 16:52:16 -07:00
Chad Versace
cb30acaced vk/image: Add tables for gen SurfaceType
Tables for mapping VkImageType and VkImageViewType to gen SurfaceType.
Tables are unused.
2015-06-25 16:52:16 -07:00
Chad Versace
1132080d5d vk/util: Add anv_loge() for logging error messages 2015-06-25 16:52:16 -07:00
Chad Versace
5f2d469e37 vk: Add func anv_is_aligned() 2015-06-25 16:52:16 -07:00
Chad Versace
f7fb7575ef vk: Add anv_minify() 2015-06-25 16:52:05 -07:00
Chad Versace
7cec6c5dfd vk: Define MAX(a, b) macro 2015-06-25 16:29:42 -07:00
Jason Ekstrand
d178e15567 nir/spirv: Fix up some dererf ralloc parenting 2015-06-24 21:39:07 -07:00
Jason Ekstrand
845002e163 i965/nir: Handle returns as long as they're at the end of a function 2015-06-24 21:38:49 -07:00
Jason Ekstrand
2ecac045a4 i965/nir: Split NIR shader handling into two functions
The brw_create_nir function takes a GLSL or ARB shader and turns it into a
NIR shader.  The guts of the optimization and lowering code is now split
into a new brw_process_shader function.
2015-06-24 21:22:07 -07:00
Jason Ekstrand
e369a0eb41 nir/spirv: Use vtn_ssa_value for texture coordinates 2015-06-24 20:39:37 -07:00
Jason Ekstrand
d0bd2bc604 nir/spirv: Add support for the Uniform storage class
This is kida sketchy.  I'm not really sure this is the way it's supposed to
be used.
2015-06-24 20:32:05 -07:00
Jason Ekstrand
ba0d9d33d4 nir/spirv: Add support for some more decorations including built-in 2015-06-24 20:30:32 -07:00
Jason Ekstrand
1bc0a1ad98 nir/spirv: Make the header file C++ safe 2015-06-24 19:01:10 -07:00
Jason Ekstrand
88d02a1b27 vk: Build xmlconfig stuff into libi965_compiler 2015-06-24 15:59:09 -07:00
Kristian Høgsberg Kristensen
24dff4f8fa vk/headers: Handle MBO fields
These must be set to one.
2015-06-24 09:37:50 -07:00
Jason Ekstrand
a62edcce4e Merge remote-tracking branch 'mesa-public/master' into vulkan 2015-06-23 18:05:25 -07:00
Connor Abbott
dee4a94e69 nir/vtn: add support for phi nodes 2015-06-23 10:34:55 -07:00
Connor Abbott
fe1269cf28 nir/builder: add support for inserting before/after blocks 2015-06-23 10:34:22 -07:00
Connor Abbott
9a3dda101e nir/vtn: fix emitting code after loops
When we're done emitting the code for a loop, we need to visit the new
break block, which is the merge block of the current loop, rather than
the old merge block, which is the merge block of the loop containing the
one we just emitted code for.
2015-06-22 13:53:08 -07:00
Connor Abbott
e9c21d0ca0 unbreak things 2015-06-22 11:59:55 -07:00
Kristian Høgsberg Kristensen
9b9f973ca6 vk: Implement scratch buffers to make spilling work 2015-06-19 15:42:15 -07:00
Kristian Høgsberg Kristensen
9e59003fb1 vk: Undo relocs for scratch bos 2015-06-19 15:42:15 -07:00
Kristian Høgsberg Kristensen
b20794cfa8 vk/allocator: Get rid of non-memfd path
We can just use modern valgrind now.
2015-06-19 15:42:15 -07:00
Kristian Høgsberg Kristensen
aba75d0546 vk/headers: Make General State offsets relocations 2015-06-19 15:42:15 -07:00
Connor Abbott
841aab6f50 matrices matrices matrices 2015-06-18 18:52:44 -07:00
Connor Abbott
d0fc04aacf nir/types: be less strict about constructing matrix types 2015-06-18 18:51:51 -07:00
Connor Abbott
22854a60ef nir/builder: add a nir_fdot() convenience function 2015-06-18 17:34:55 -07:00
Connor Abbott
0e86ab7c0a nir/types: add a helper to transpose a matrix type 2015-06-18 17:34:12 -07:00
Connor Abbott
de4c31a085 fix glsl450 for composites 2015-06-18 17:33:08 -07:00
Kristian Høgsberg Kristensen
aedd3c9579 vk: Add missing gen7 RENDER_SURFACE_STATE struct 2015-06-17 21:42:29 -07:00
Connor Abbott
bf5a615659 composites composites composites 2015-06-17 16:25:38 -07:00
Kristian Høgsberg Kristensen
fa8a07748d vk: Compute CS exec mask and thread width max in pipeline
We compute the right mask and thread width max parameters as part of
pipeline creation and set them accordingly at vkCmdDispatch() and
vkCmdDispatchIndirect() time. These parameters depend only on the local
group size and the dispatch width of the program so we can figure this
out at pipeline create time.
2015-06-12 18:21:50 -07:00
Kristian Høgsberg Kristensen
c103c4990c vk: Set binding table layout for CS
We weren't setting the binding table layout for the backend compiler.
2015-06-12 18:21:49 -07:00
Kristian Høgsberg Kristensen
2fdd17d259 vk: Generate CS prog_data into the pipeline instance
We were generating the prog_data into a local variable and never
initializing the pipeline->cs_prog_data one.
2015-06-12 18:21:49 -07:00
Kristian Høgsberg Kristensen
00494c6cb7 vk: Document how depth/stencil formats work in anv_image_create()
This reverts commits

  e17ed04 * vk/image: Don't double-allocate stencil buffers
  1ee2d1c * vk/image: Teach anv_image_choose_tile_mode about WMAJOR

and instead adds a comment to describe the subtlety of how we create
images for stencil only formats.
2015-06-11 22:07:16 -07:00
Kristian Høgsberg Kristensen
fbc9fe3c92 vk: Use compute pipeline layout when binding compute sets 2015-06-11 21:57:43 -07:00
Kristian Høgsberg Kristensen
765175f5d1 vk: Implement basic compute shader support 2015-06-11 15:31:42 -07:00
Kristian Høgsberg Kristensen
7637b02aaa vk: Emit PIPELINE_SELECT on demand 2015-06-11 15:21:49 -07:00
Kristian Høgsberg Kristensen
405697eb3d vk: Stop asserting we have a fragment shader
Even for graphics, this is not a requirement, we can have a depth-only output pipeline.
2015-06-11 15:07:38 -07:00
Kristian Høgsberg Kristensen
e7edde60ba vk: Defer setting viewport dynamic state
We can't emit this until we've done a 3D pipeline select.
2015-06-11 15:04:09 -07:00
Kristian Høgsberg Kristensen
f7fe06cf0a vk: Disable shader stages in the graphics pipeline batch
We need to move this into the graphics pipeline batch so we don't  emit it
for compute pipelines.
2015-06-11 14:58:31 -07:00
Kristian Høgsberg Kristensen
9aae480cc4 vk: Don't emit STATE_SIP
We don't have a SIP kernel and don't enable exceptions.
2015-06-11 14:56:29 -07:00
Kristian Høgsberg Kristensen
923e923bbc vk: Compile fragment shader after VS and GS
Just moving code around to do shader stages in the natual order.
2015-06-11 14:55:50 -07:00
Jason Ekstrand
1dd63fcbed vk/entrypoints: Don't print every single function call 2015-06-11 10:10:13 -07:00
Kristian Høgsberg Kristensen
b581e924b6 vk: Remove left-over trp call 2015-06-11 09:26:49 -07:00
Kristian Høgsberg Kristensen
d76ea7644a vk: Set maximum point size range
We set both minimum and maximum point size to 0 in 3DSTATE_CLIP, which
will clip away all points.
2015-06-11 09:25:04 -07:00
Kristian Høgsberg Kristensen
a5b49d2799 vk: Use generated headers with fixed point support
The generated headers now convert float in the template struct to the
correct fixed point format.
2015-06-11 09:25:04 -07:00
Kristian Høgsberg Kristensen
ea7ef46cf9 vk: Regenerate headers with __gen_validate_value() 2015-06-11 09:25:03 -07:00
Jason Ekstrand
a566b1e08a vk/formats: Refactor format properties code
Along with the refactor, we now do the right thing when we hit an
unsupported format: Set the flags to 0 and return VK_SUCCESS.
2015-06-11 09:11:16 -07:00
Jason Ekstrand
2a3c29698c vk/image: Add a bunch of asserts 2015-06-10 21:04:51 -07:00
Jason Ekstrand
c8b62d109b vk: Add a couple vk_error calls 2015-06-10 21:04:13 -07:00
Jason Ekstrand
7153b56abc vk/private: Add a non-fatal assert 2015-06-10 21:03:50 -07:00
Jason Ekstrand
29d2bbb2b5 vk/cmd: Add an initial implementation of PipelineBarrier
We may want to do something more inteligent here later such as actually
handling image layout transitions.  However, this should do for now.
2015-06-10 16:37:33 -07:00
Jason Ekstrand
047ed02723 vk/emit: Use valgrind to validate every packed field 2015-06-10 12:43:02 -07:00
Jason Ekstrand
9cae3d18ac vk: Add valgrind checks in various emit functions
The check in batch_bo_finish should catch any undefined values in the batch
but isn't that great for debugging.  The checks in the various emit
functions will help get better granularity.
2015-06-09 21:51:37 -07:00
Jason Ekstrand
d5ad24e39b vk: Move the valgrind include and VG() macro to private.h 2015-06-09 21:51:37 -07:00
Chad Versace
e17ed04b03 vk/image: Don't double-allocate stencil buffers
If the main surface has format S8_UINT, then don't allocate the
auxiliary stencil surface.
2015-06-09 16:39:28 -07:00
Chad Versace
1ee2d1c3fc vk/image: Teach anv_image_choose_tile_mode about WMAJOR 2015-06-09 16:38:55 -07:00
Chad Versace
2d2e148952 vk/util: Add anv_abortf(), anv_abortfv()
Convenience functions to print an error message then abort.
2015-06-09 16:38:50 -07:00
Chad Versace
ffb1ee5d20 vk: Define anv_noreturn macro 2015-06-09 16:38:46 -07:00
Chad Versace
f1db3b3869 vk/image: Factor tile mode selection into separate function
Because it will eventually need to get smarter.
2015-06-09 16:38:42 -07:00
Jason Ekstrand
11e941900a vk/device: Actually allow destruction 2015-06-09 16:28:46 -07:00
Jason Ekstrand
5d4b6a01af vk/cmd_buffer: Properly initialize/reset dynamic states 2015-06-09 16:27:55 -07:00
Jason Ekstrand
634a6150b9 vk/pipeline: Zero out the depth-stencil state when not in use 2015-06-09 16:26:55 -07:00
Jason Ekstrand
919e7b7551 vk/device: Use anv_CreateDynamicViewportState instead of the vk one 2015-06-09 16:01:56 -07:00
Jason Ekstrand
0599d39dd9 vk/device: Dedent the vkCreateDynamicViewportState call 2015-06-09 15:53:26 -07:00
Chad Versace
d57c4cf999 vk/util: Annotate anv_finishme() as printflike 2015-06-09 14:46:49 -07:00
Chad Versace
822cb16abe vk: Define anv_printflike() macro 2015-06-09 14:46:45 -07:00
Chad Versace
081f617b5a vk/image: Stop hardcoding alignment of stencil surfaces
Look up the alignment from anv_tile_info_table.
2015-06-09 14:16:56 -07:00
Chad Versace
e6bd568f36 vk/image: Rewrite tile info table
- Reduce the number of table lookups in anv_image_create from 4 to 1.
- Add field for surface alignment.
- Shorten field names tile_width, tile_height -> width, height.
2015-06-09 14:16:45 -07:00
Chad Versace
5b777e2bcf vk/image: Delete an old comment 2015-06-09 14:14:29 -07:00
Jason Ekstrand
d842a6965f vk/compiler: Free the GL errors data 2015-06-09 12:36:23 -07:00
Jason Ekstrand
9f292219bf vk/compiler: Free more of prog_data when tearing down a pipeline 2015-06-09 12:36:23 -07:00
Jason Ekstrand
66b00d5e5a vk/queue: Embed the queue in and allocate it with the device 2015-06-09 12:36:23 -07:00
Jason Ekstrand
38f5eef59d vk/device: Free border color states when we have valgrind 2015-06-09 12:36:23 -07:00
Jason Ekstrand
999b56c507 vk/device: Destroy all batch buffers
Due to a copy+paste error, we were destroying all but the first batch or
surface state buffer.  Now we destroy them all.
2015-06-09 12:36:23 -07:00
Jason Ekstrand
3a38b0db5f vk/meta: Clean up temporary objects 2015-06-09 12:36:23 -07:00
Jason Ekstrand
9d6f55dedf vk/surface_view: Add a destructor 2015-06-09 12:36:23 -07:00
Chad Versace
e6162c2fef vk/image: Add anv_image::h_align,v_align
Use the new fields to compute RENDER_SURFACE_STATE.Surface*Alignment.
We still hardcode them to 4, though.
2015-06-09 12:19:24 -07:00
Jason Ekstrand
58afc24e57 vk/allocator: Remove the concept of a slave block pool
This reverts commit d24f8245db.
2015-06-08 17:46:32 -07:00
Jason Ekstrand
b6363c3f12 vk/device: Remove the binding table pools/streams 2015-06-08 17:45:57 -07:00
Jason Ekstrand
531549d9fc vk/pipeline: Move freeing the program stream to pipeline.c
It's created in pipeline.c so we should free it there.
2015-06-08 14:27:04 -07:00
Jason Ekstrand
66a4dab89a vk/pipeline: Don't destroy the program stream
It's freed in compiler.cpp and we don't want to free it twice.
2015-06-08 13:53:19 -07:00
Jason Ekstrand
920fb771d4 vk/allocator: Make the use of NULL_BLOCK in state_stream_finish explicit 2015-06-08 13:53:19 -07:00
Kristian Høgsberg Kristensen
52637c0996 vk: Quiet a few warnings 2015-06-08 08:51:40 -07:00
Kristian Høgsberg Kristensen
9eab70e54f vk: Create a minimal context for the compiler
This avoids the full brw context initialization and just sets up context
constants, initializes extensions and sets a few driver vfuncs for the
front-end GLSL compiler.
2015-06-08 08:51:40 -07:00
Jason Ekstrand
ce00233c13 vk/cmd_buffer: Use the dynamic state stream in emit_dynamic and merge_dynamic 2015-06-05 17:26:41 -07:00
Jason Ekstrand
e69588b764 vk/device: Use a 64-byte alignment for CC state 2015-06-05 17:26:26 -07:00
Jason Ekstrand
c2eeab305b vk/pipeline: Actually free the program stream and dynamic pool 2015-06-05 17:26:26 -07:00
Jason Ekstrand
ed2ca020f8 vk/allocator: Avoid double-free in the bo pool 2015-06-05 17:12:28 -07:00
Jason Ekstrand
aa523d3c62 vk/gem: Call VALGRIND_FREELIKE_BLOCK before unmapping 2015-06-05 16:41:49 -07:00
Chad Versace
87d98e1935 vk: Fix 2 incorrect typecasts
The compiler didn't find the cast errors because all Vulkan types are
just integers.
2015-06-04 14:32:22 -07:00
Chad Versace
b981379bcf vk: Make make clean remove generated spirv headers 2015-06-04 14:26:46 -07:00
Jason Ekstrand
8d930da35d vk/allocator: Remove an unneeded VG() wrapper 2015-06-04 09:14:33 -07:00
Jason Ekstrand
7f90e56e42 vk/device: Dissalow device destruction 2015-06-04 09:14:33 -07:00
Chad Versace
9cd42b3dea vk: Fix build
Commit 1286bd, which deleted vk.c, broke the build. Update the Makefile
to fix it.
2015-06-04 09:01:30 -07:00
Jason Ekstrand
251aea80b0 vk/DS: Mask stencil masks to 8 bits 2015-06-03 16:59:13 -07:00
Connor Abbott
47bd462b0c awesome control flow bugfixes/clarifications 2015-06-03 14:10:28 -04:00
Kristian Høgsberg Kristensen
a37d122e88 vk: Set color/blend state in meta clear if not set yet 2015-06-02 23:08:05 -07:00
Kristian Høgsberg Kristensen
1286bd3160 vk: Delete vk.c test case
We now have crucible up and running and all vk sub-cases have been moved
over. Delete this crufty old hack of a test case.
2015-06-02 22:57:42 -07:00
Kristian Høgsberg Kristensen
2f6aa424e9 vk: Update generated headers with support for 64 bit fields 2015-06-02 22:57:42 -07:00
Kristian Høgsberg Kristensen
5744d1763c vk: Set cb_state to NULL at cmd buffer create time
Dynamic color/blend state can be NULL in case we're not rendering to
color targets (only output to depth and/or stencil). Initialize
cmd_buffer->cb_state to NULL so we can reliably detect whether it's been
set or not.
2015-06-02 22:57:42 -07:00
Kristian Høgsberg Kristensen
c8f078537e vk: Implement vertexOffset parameter of vkCmdDrawIndexed()
As exposed by the func.draw_indexed test, we were ignoring the argument
and hardcoding 0.
2015-06-02 22:57:42 -07:00
Jason Ekstrand
e702197e3f vk/formats: Add a name to the metadata and better logging 2015-06-02 11:30:39 -07:00
Jason Ekstrand
fbafc946c6 vk/formats: Rework the formats table 2015-06-02 11:30:39 -07:00
Kristian Høgsberg Kristensen
f98c89ef31 vk: Move query related functionality to new file query.c 2015-06-01 21:52:45 -07:00
Jason Ekstrand
08748e3a0c i965: Use NIR by default for vertex shaders on GEN8+
GLSL IR vs. NIR shader-db results for SIMD8 vertex shaders on Broadwell:

   total instructions in shared programs: 2742062 -> 2681339 (-2.21%)
   instructions in affected programs:     1514770 -> 1454047 (-4.01%)
   helped:                                5813
   HURT:                                  1120

The gained programs are ARB vertext programs that were previously going
through the vec4 backend.  Now that we have prog_to_nir, ARB vertex
programs can go through the scalar backend so they show up as "gained" in
the shader-db results.

Acked-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Matt Turner <mattst88@gmail.com>
2015-06-01 12:25:58 -07:00
Jason Ekstrand
d4cbf6a728 vk/compiler: Add an index_count to the bind map and check for OOB 2015-06-01 12:25:58 -07:00
Jason Ekstrand
510b5c3bed vk/HACK: Plumb real descriptor set/index into textures 2015-06-01 12:25:58 -07:00
Jason Ekstrand
aded32bf04 NIR: Add a helper for doing sampler lowering for vulkan 2015-06-01 12:25:58 -07:00
Kristian Høgsberg Kristensen
5caa408579 vk: Indent tables to align '=' at column 48 2015-05-31 22:36:26 -07:00
Kristian Høgsberg Kristensen
76bb658518 vk: Add support for anisotropic bits 2015-05-31 22:15:34 -07:00
Kristian Høgsberg Kristensen
dc56e4f7b8 vk: Implement support for sampler border colors
This supports the three Vulkan border color types for float color
formats. The support for integer formats is a little trickier, as we
don't know the format of the texture at this time.
2015-05-31 17:20:48 -07:00
Jason Ekstrand
e497ac2c62 vk/device: Only flush the texture cache when setting state base address
After further examination, it appears that the other flushes and stalls
weren't actually needed.
2015-05-30 18:04:50 -07:00
Jason Ekstrand
2251305e1a vk/cmd_buffer: Track descriptor set dirtying per-stage 2015-05-30 10:07:29 -07:00
Jason Ekstrand
33cccbbb73 vk/device: Emit PIPE_CONTROL flushes surrounding new STATE_BASE_ADDRESS
According to the bspec, you're supposed to emit a PIPE_CONTROL with a CS
stall and a render target flush prior to chainging STATE_BASE_ADDRESS.  A
little experimentation, however, shows that this is not enough.  It also
appears as if you have to flush the texture cache after chainging base
address or things won't propagate properly.
2015-05-30 08:08:07 -07:00
Jason Ekstrand
b2b9fc9fad vk/allocator: Don't call VALGRIND_MALLOCLIKE_BLOCK on fresh gem_mmap's 2015-05-29 21:15:47 -07:00
Jason Ekstrand
03ffa9ca31 vk: Don't crash on partial descriptor sets 2015-05-29 20:43:10 -07:00
Jason Ekstrand
4ffbab5ae0 vk/device: Allow for starting a new surface state buffer
This commit allows for us to create a whole new surface state buffer when
the old one runs out of room.  We simply re-emit the state base address for
the new state, re-emit binding tables, and keep going.
2015-05-29 17:49:41 -07:00
Jason Ekstrand
c4bd5f87a0 vk/device: Do lazy surface state emission for binding tables
Before, we were emitting surface states up-front when binding tables were
updated.  Now, we wait to emit the surface states until we emit the binding
table.  This makes meta simpler and should make it easier to deal with
swapping out the surface state buffer.
2015-05-29 16:51:11 -07:00
Kristian Høgsberg Kristensen
4aecec0bd6 vk: Store dynamic slot index with struct anv_descriptor_slot
We need to make sure we use the right index into dynamic offset
array. Dynamic descriptors can be present or not in different stages and
to get the right offset, we need to compute the index at
vkCreateDescriptorSetLayout time.
2015-05-29 11:32:53 -07:00
Kristian Høgsberg Kristensen
fad418ff47 vk: Implement dynamic buffer offsets
We do this by creating a surface state on the fly that incorporates the
dynamic offset. This patch also refactor the descriptor set layout
constructor a bit to be less clever with switch statement fall
through. Instead of duplicating the subtle code to update the sampler
and surface slot map, we just use two switch statements.
2015-05-28 22:41:20 -07:00
Jason Ekstrand
9ffc1bed15 vk/device: Split state base address emit into its own function 2015-05-28 15:34:08 -07:00
Jason Ekstrand
468c89a351 vk/device: Use anv_batch_emit for MI_BATCH_BUFFER_START 2015-05-28 15:25:02 -07:00
Jason Ekstrand
2dc0f7fe5b vk/device: Actually destroy batch buffers 2015-05-28 13:08:21 -07:00
Jason Ekstrand
8cf932fd25 vk/query: Don't emit a CS stall by itself
Both the bspec and the simulator don't like this.  I'm not sure if stalling
at the scoreboard is right but it at least shuts up the simulator.
2015-05-28 10:27:53 -07:00
Jason Ekstrand
730ca0efb1 vk/device: Fixups for batch buffer chaining
Some how these didn't get merged with the other batch buffer chaining
stuff.  Oh well, it's here now.
2015-05-28 10:26:11 -07:00
Jason Ekstrand
de221a672d meta: Add a default ds_state and use it when no ds state is set 2015-05-28 10:06:45 -07:00
Jason Ekstrand
6eefeb1f84 vk/meta: Share the dummy RS and CB state between clear and blit 2015-05-28 10:00:38 -07:00
Kristian Høgsberg Kristensen
5a317ef4cb vk: Initialize dynamic state binding points to NULL
We rely on these being initialized to NULL so meta can reliably detect
whether or not they've been set. ds_state is also allowed to not be
present so we need a well-defined value for that.
2015-05-27 22:13:48 -07:00
Chad Versace
1435bf4bc4 .gitignore: Ignore spirv2nir binary 2015-05-27 17:01:09 -07:00
Chad Versace
f559fe9134 .gitignore: Scope Vulkan's generated source files
Don't ignore any file named entrypoints.{c,h}. Ignore it only if it's in
src/vulkan.
2015-05-27 16:59:53 -07:00
Chad Versace
ca385dcf2a vk: gitignore generated source files 2015-05-27 16:57:31 -07:00
Chad Versace
466f61e9f6 vk/glsl_scraper: Replace adhoc arg parsing with argparse 2015-05-27 16:56:02 -07:00
Chad Versace
fab9011c44 vk/image: Assert that VkImageTiling is valid 2015-05-27 16:21:04 -07:00
Chad Versace
c0739043b3 vk/image: Remove trailing whitespace 2015-05-27 16:15:47 -07:00
Chad Versace
4514e63893 vk/glsl: Reject invalid options
The script incorrectly interpreted --blah as the input filename.
2015-05-27 16:14:26 -07:00
Chad Versace
fd8b5e0df2 vk/glsl_scraper: Indent large text blocks
Indent them to the same level as if the text was code.

No changes in entrypoints.{c,h} after a clean build.
2015-05-27 16:09:31 -07:00
Chad Versace
df4b02f4ed vk/glsl_scraper: Fix code style for imports
Python style is one module imported per line, and imports are at the top
of the file.
2015-05-27 16:04:12 -07:00
Jason Ekstrand
b23885857f vk/meta: Actually create the CB state for blits 2015-05-27 12:06:30 -07:00
Jason Ekstrand
da8f148203 vk: Rework anv_batch and use chaining batch buffers
This mega-commit primarily does two things.  First, is to turn anv_batch
into a better abstraction of a batch.  Instead of actually having a BO, it
now has a few pointers to some piece of memory that are used to add data to
the "batch".  If it gets to the end, there is a function pointer that it
can call to attempt to grow the batch.

The second change is to start using chained batch buffers.  When the end of
the current batch BO is reached, it automatically creates a new one and
ineserts an MI_BATCH_BUFFER_START command to chain to it.  In this way, our
batch buffers are effectively infinite in length.
2015-05-27 11:48:28 -07:00
Jason Ekstrand
59def43fc8 Fixup for growable reloc lists 2015-05-27 11:48:28 -07:00
Jason Ekstrand
1c63575de8 vk/cmd_buffer: Allocate the surface_bo from device->batch_bo_pool 2015-05-27 11:48:28 -07:00
Jason Ekstrand
403266be05 vk/device: Make reloc lists growable 2015-05-27 11:48:28 -07:00
Jason Ekstrand
5ef81f0a05 vk/device: Use a bo pool for batch buffers 2015-05-27 11:48:28 -07:00
Jason Ekstrand
6f3e3c715a vk/allocator: Add a BO pool 2015-05-27 11:48:28 -07:00
Jason Ekstrand
59328bac10 vk/allocator: Add a free list that acts on pointers instead of offsets 2015-05-27 11:48:28 -07:00
Kristian Høgsberg
a1d30f867d vk: Add support for dynamic and pipeline color blend state 2015-05-26 17:12:37 -07:00
Kristian Høgsberg
2514ac5547 vk/test: Create and use color/blend dynamic and pipeline state 2015-05-26 17:12:37 -07:00
Kristian Høgsberg
1cd8437b9d vk/meta: Allocate and set color/blend state
For color blend, we have to set our own state to avoid inheriting bogus
blend state.
2015-05-26 17:12:37 -07:00
Kristian Høgsberg
610e6291da vk: Allocate samplers from dynamic stream 2015-05-26 11:50:34 -07:00
Kristian Høgsberg
b29f44218d vk: Emit color calc state
This involves pulling stencil ref values out of DS dynamic state and the
blend constant out of CB dynamic state.
2015-05-26 11:27:31 -07:00
Kristian Høgsberg
5e637c5d5a vk/pack: Generate length macros for structs 2015-05-26 11:27:31 -07:00
Kristian Høgsberg
998837764f vk: Program depth bias
This makes 3DSTATE_RASTER a split state command.
2015-05-26 11:27:31 -07:00
Kristian Høgsberg
0dbed616af vk: Add support for texture component swizzle
This also drops the share create_surface_state helper and moves filling
out SURFACE_STATE directly into anv_image_view_init() and
anv_color_attachment_view_init().
2015-05-26 11:27:29 -07:00
Kristian Høgsberg
cbe7ed416e vk: Implement dynamic and pipeline ds state 2015-05-25 20:20:31 -07:00
Kristian Høgsberg
37743f90bc vk: Set up depth and stencil buffers 2015-05-25 20:20:31 -07:00
Kristian Høgsberg
7c0d0021eb vk/test: Add new depth-stencil test
Not yet a depth stencil test, but will become one.
2015-05-25 20:20:31 -07:00
Kristian Høgsberg
0997a7b2e3 vk: Add basic MOCS settings
This matches what we do for GL.
2015-05-25 20:20:31 -07:00
Kristian Høgsberg
c03314bdd3 vk: Update to header files with nested struct support
This will let us do MOCS settings right.
2015-05-25 20:20:31 -07:00
Jason Ekstrand
ae8c93e023 vk/cmd_buffer: Initialize the pipeline pointer to NULL
If a meta operation is called before the pipeline is set, this can cause
uses of undefined values.  They *should* be harmless, but we might as well
shut up valgrind on this one too.
2015-05-25 17:14:49 -07:00
Jason Ekstrand
912944e59d vk/device: Use the correct number of viewports when creating default VP state
Fixes valgrind uninitialized value errors
2015-05-25 17:14:49 -07:00
Jason Ekstrand
1b211feb6c vk/compiler: Zero out the vs_prog_data struct when VS is disabled
Prevents uninitialized value errors
2015-05-25 17:14:49 -07:00
Jason Ekstrand
903bd4b056 vk/compiler: Fix up the binding hack and make it work in NIR 2015-05-25 12:57:32 -07:00
Jason Ekstrand
57153da2d5 vk: Actually implement some sort of destructor for all object types 2015-05-22 15:15:08 -07:00
Jason Ekstrand
0f0b5aecb8 vk/pipeline: Track VB's that are actually used by the pipeline
Previously, we just blasted out whatever VB's we had marked as "dirty"
regardless of which ones were used by the pipeline.  Given that the stride
of the VB is embedded in the pipeline this can cause problems.  One problem
is if the pipeline doesn't use the given VB binding we emit a bogus stride.
Another problem is that we weren't properly resetting the dirty bits when
the pipeline changed.
2015-05-21 16:58:53 -07:00
Jason Ekstrand
0a54751910 vk/device: Memset descriptor sets to 0 and handle descriptor set holes 2015-05-21 16:33:04 -07:00
Jason Ekstrand
519fe765e2 vk: Do relocations in surface states when they are created
Previously, we waited until later and did a pass through the used surfaces
and did the relocations then.  This lead to doing double-relocations which
was causing us to get bogus surface offsets.
2015-05-21 15:55:29 -07:00
Jason Ekstrand
ccf2bf9b99 vk/test: Use the glsl_scraper for building shaders 2015-05-21 12:24:02 -07:00
Jason Ekstrand
f3d70e4165 vk/glsl_scraper: Use the LunarG back-door for GLSL source 2015-05-21 12:22:44 -07:00
Jason Ekstrand
cb56372eeb vk/glsl_scraper: Use a fake GLSL version that glslang will accept 2015-05-21 12:21:02 -07:00
Jason Ekstrand
0e441cde71 vk: Bake the GLSL_VK_SHADER macro into the scraper output file 2015-05-21 12:21:00 -07:00
Jason Ekstrand
f17e835c26 vk/meta: Use glsl_scraper for our GLSL source
We are not yet using SPIR-V for meta but this is a first step.
2015-05-21 11:39:54 -07:00
Jason Ekstrand
b13c0f469b vk: More out-of-tree build fixes 2015-05-21 11:32:59 -07:00
Jason Ekstrand
f294154e42 vk: Fix for out-of-tree builds 2015-05-21 10:23:18 -07:00
Kristian Høgsberg
f9e66ea621 vk: Remove render pass stub call
This isn't really a stub.
2015-05-20 20:34:52 -07:00
Kristian Høgsberg
a29df71dd2 vk: Add WSI implementation 2015-05-20 20:34:52 -07:00
Kristian Høgsberg
f886647b75 vk: Add debug stubs 2015-05-20 20:34:52 -07:00
Kristian Høgsberg
63da974529 vk: Mark remaining unsupported formats as such 2015-05-20 20:34:52 -07:00
Kristian Høgsberg
387a1bb58f vk: Mark VK_FORMAT_UNDEFINED as 1 cpp, 1 channel 2015-05-20 20:34:52 -07:00
Kristian Høgsberg
a1bd426393 vk: Stream surface state instead of using the surface pool
Since the binding table pointer is only 16 bits, we can only have 64kb
of binding table state allocated at any given time. With a block size of
1kb, that amounts to just 64 command buffers, which is not enough.
2015-05-20 20:34:52 -07:00
Kristian Høgsberg
01504057f5 vk: Use surface_format_info from dri driver for vkGetFormatInfo 2015-05-20 20:34:52 -07:00
Chad Versace
a61f307996 vk: Fix result of vkCreateInstance
When fill_physical_device() fails, don't return VK_SUCCESS.
2015-05-20 19:51:10 -07:00
Jason Ekstrand
14929046ba vk/compiler: Add shader language detection
This commit adds support for the LunarG GLSL back-door as well as detecting
regular GLSL and SPIR-V.  The SPIR-V path doesn't exist yet, so that will
cause an assert-fail.
2015-05-20 17:05:41 -07:00
Jason Ekstrand
47c1cf5ce6 vk/test: Add a test for testing buffer copies 2015-05-20 16:20:04 -07:00
Jason Ekstrand
bea66ac5ad vk/meta: Add support for copying arbitrary size buffers 2015-05-20 16:20:04 -07:00
Jason Ekstrand
9557b85e3d vk/meta: Use the biggest format possible for buffer copies
This should substantially improve throughput of buffer copies.
2015-05-20 16:20:04 -07:00
Jason Ekstrand
13719e9225 vk/meta: Fix buffer copy extents 2015-05-20 16:20:04 -07:00
Jason Ekstrand
d7044a19b1 vk/meta: Use texture() instead of texture2D() 2015-05-19 12:44:35 -07:00
Jason Ekstrand
edff076188 vk: Use binding instead of index in uniform layout qualifiers
This more closely matches what the Vulkan docs say to do.
2015-05-19 12:44:22 -07:00
Jason Ekstrand
e37a89136f vk/glsl_scraper: Add a --glsl-only option 2015-05-19 11:29:07 -07:00
Jason Ekstrand
4bcf58a192 vk/glsl_scraper: Use the line number from the end of the macro
We used to use the line number from the start of the macro but this doesn't
seem to match the c preprocessor
2015-05-19 11:29:07 -07:00
Jason Ekstrand
1573913194 vk/glsl_scraper: Don't open files until needed
This prevents us from writing an empty file when the compile failed.
2015-05-19 11:29:07 -07:00
Kristian Høgsberg
e4c11f50b5 vk: Call finish for binding table state stream 2015-05-18 21:12:13 -07:00
Jason Ekstrand
851495d344 vk/meta: Use the new *view_init functions and stack-allocated views
This should save us a good deal of the leakage that meta currently has.
2015-05-18 20:57:43 -07:00
Jason Ekstrand
4668bbb161 vk/image: Factor view creation out into separate *_init functions
The *_init functions work basically the same as the Vulkan entrypoints
except that they act on an already-created view and take an optional
command buffer option.  If a command buffer is given, the surface state is
allocated out of the command buffer's state stream.
2015-05-18 20:57:43 -07:00
Jason Ekstrand
7c9f209427 Revert "vk/allocator: Don't use memfd when valgrind is detected"
This reverts commit b6ab076d6b.

It turns out setting USE_MEMFD to 0 is really bad because it means we can't
resize the pool.  Besides, valgrind SVN handles memfd so we really don't
need this fallback for valgrind anymore.
2015-05-18 20:57:43 -07:00
Jason Ekstrand
923691c70d vk: Use a separate block pool and state stream for binding tables
The binding table pointers packet only allows for a 16-bit binding table
address so all binding tables have to be in the first 64 KB of the surface
state BO.  We solve this by adding a slave block pool that pulls off the
first 64 KB worth of blocks and reserves them for binding tables.
2015-05-18 20:57:43 -07:00
Jason Ekstrand
d24f8245db vk/allocator: Add a concept of a slave block pool
We probably need a better name but this will do for now.
2015-05-18 20:57:43 -07:00
Kristian Høgsberg
997596e4c4 vk/test: Add test that prints format features 2015-05-18 20:52:44 -07:00
Kristian Høgsberg
241b59cba0 vk/test: Test timestamps and occlusion queries 2015-05-18 20:52:44 -07:00
Kristian Høgsberg
ae9ac47c74 vk: Make timestamp command work correctly
This was using the wrong timestamp register and needs to write a 64 bit
value.
2015-05-18 20:52:43 -07:00
Kristian Høgsberg
82ddab4b18 vk: Make occlusion query work, both copy and get functions 2015-05-18 20:52:43 -07:00
Kristian Høgsberg
1d40e6ade8 vk: Update generated header files
This fixes a problem where register addresses where incorrectly shifted.
2015-05-18 20:52:43 -07:00
Kristian Høgsberg
f330bad545 vk: Only fill render targets for meta clear
Clear inherits the render targets from the current render pass. This
means we need to fill out the binding table after switching to meta
bindings. However, meta copies etc happen outside a render pass and
break when we try to fill in the render targets. This change fills the
render targets only for meta clear.
2015-05-18 20:52:43 -07:00
Jason Ekstrand
b6c7d8c911 vk/pipeline: Use a state_stream for storing programs
Previously, we were effectively using a state_stream, it was just
hand-rolled based on a block pool.  Now we actually use the data structure.
2015-05-18 15:58:20 -07:00
Jason Ekstrand
4063b7deb8 vk/allocator: Add support for valgrind tracking of state pools and streams
We leave the block pool untracked so that reads/writes to freed blocks
will get caught and do the tracking at the state pool/stream level.  We
have to do a few extra gymnastics for streams because valgrind works in
terms of poitners and we work in terms of separate map and offset.
Fortunately, the users of the state pool and stream should always be using
the map pointer provided in the anv_state structure.  We just have to
track, per block, the map that was used when we initially got the block.
Then we can make sure we always use that map and valgrind should stay
happy.
2015-05-18 15:58:20 -07:00
Jason Ekstrand
b6ab076d6b vk/allocator: Don't use memfd when valgrind is detected 2015-05-18 15:58:20 -07:00
Jason Ekstrand
682d11a6e8 vk/allocator: Assert that block_pool_grow succeeds 2015-05-18 15:48:19 -07:00
Jason Ekstrand
28804fb9e4 vk/gem: VG_CLEAR the padding for the gem_mmap struct 2015-05-18 12:05:17 -07:00
Jason Ekstrand
8440b13f55 vk/meta: Rework the indentation style
No functional change.
2015-05-18 10:43:51 -07:00
Kristian Høgsberg
5286ef7849 vk: Provide more realistic values for device info 2015-05-18 10:27:08 -07:00
Kristian Høgsberg
69fd473321 vk: Use a temporary buffer for formatting in finishme
This is more likely to avoid breaking up the message when racing with
other threads.
2015-05-18 10:27:08 -07:00
Jason Ekstrand
cd7ab6ba4e vk/meta: Add an initial implementation of vkCmdCopyBuffer
Compile-tested only
2015-05-18 10:27:08 -07:00
Jason Ekstrand
c25ce55fd3 vk/meta: Add an initial implementation of vkCmdCopyBufferToImage
Compile-tested only
2015-05-18 10:27:08 -07:00
Jason Ekstrand
08bd554cda vk/meta: Add an initial implementation of vkCmdBlitImage
Compile-tested only
2015-05-18 10:27:08 -07:00
Jason Ekstrand
fb27d80781 vk/meta: Add an initial implementation of vkCmdCopyImage
Compile-tested only
2015-05-18 10:27:08 -07:00
Jason Ekstrand
c15f3834e3 vk/gem: Set the gem_mmap.flags parameter to 0 if it exists 2015-05-18 10:27:08 -07:00
Jason Ekstrand
f7b0f922be vk/gem: Only VK_CLEAR the addr_ptr in gen_mmap 2015-05-18 10:27:07 -07:00
Kristian Høgsberg
ca7e62d421 vk: Add a logger wrapper for the generated entrypoint 2015-05-18 10:27:07 -07:00
Kristian Høgsberg
eb92745b2e vk/gem: Just return -1 from anv_gem_wait() on error
We were returning -errno, unlike all the other gem functions.
2015-05-18 10:27:07 -07:00
Kristian Høgsberg
05754549e8 vk: Fix vkGetOjectInfo return values
We weren't properly returning the allocation count.
2015-05-18 10:27:07 -07:00
Kristian Høgsberg
6afb26452b vk: Implement fences
This basic implementation uses a throw-away bo for synchronization.
2015-05-18 10:27:07 -07:00
Kristian Høgsberg
e26a7ffbd9 vk/meta: Use anv_* internal entrypoints 2015-05-18 10:27:07 -07:00
Kristian Høgsberg
b7fac7a7d1 vk: Implement allocation count query 2015-05-18 10:27:07 -07:00
Kristian Høgsberg
783e6217fc vk: Change pData/pDataSize semantics
We now always copy the entire struct unless pData is NULL and
unconditionally write back the struct size. It's not clear this is
useful if the structs may grow over time, but it seems to be the
expected behaviour for now.
2015-05-18 10:27:07 -07:00
Kristian Høgsberg
b4b3bd1c51 vk: Return VK_SUCCESS from vkAllocDescriptorSets
This should've been returning VK_SUCCESS all along.
2015-05-18 10:27:07 -07:00
Kristian Høgsberg
a9f2115486 vk: Return VK_SUCCESS for all descriptor pool entry points 2015-05-18 10:27:07 -07:00
Kristian Høgsberg
60ebcbed54 vk: Start Implementing vkGetFormatInfo()
We move the format table and vkGetFormatInfo to their own file in the
process.
2015-05-18 10:27:07 -07:00
Kristian Høgsberg
454345da1e vk: Add script for generating ifunc entry points
This lets us generate a hash table for vkGetProcAddress and lets us call
public functions internally without the public entrypoint overhead.
2015-05-18 10:27:02 -07:00
Kristian Høgsberg
333bcc2072 vk: Fix vulkan header inconsistency
The function pointer typedef and the function prototype for
vkCmdClearColorImage() didn't agree. Fix the typedef to match the
prototype.
2015-05-17 21:08:31 -07:00
Kristian Høgsberg
b9eb56a404 vk: Add function pointer typedef for intel extension
Also guard function prototype by VK_PROTOTYPES.
2015-05-17 21:08:30 -07:00
Kristian Høgsberg
75cb85c56a vk: Add missing VKAPI for vkQueueRemoveMemReferences 2015-05-17 21:08:30 -07:00
Jason Ekstrand
a924ea0c75 Merge remote-tracking branch 'fdo-personal/wip/nir-vtn' into vulkan
This adds the SPIR-V -> NIR translator.
2015-05-16 12:43:16 -07:00
Jason Ekstrand
a63952510d nir/spirv: Don't assert that the current block is empty
It's possible that someone will give us SPIR-V code in which someone
needlessly branches to new blocks.  We should handle that ok now.
2015-05-16 12:34:34 -07:00
Jason Ekstrand
4e44dcc312 nir/spirv: Add initial support for samplers 2015-05-16 12:34:15 -07:00
Jason Ekstrand
d6f52dfb3e nir/spirv: Move Exp and Log to the list of currently unhandled ALU ops
NIR doesn't have the native opcodes for them anymore
2015-05-16 12:33:32 -07:00
Jason Ekstrand
a53e795524 nir/types: Add support for sampler types 2015-05-16 12:32:58 -07:00
Jason Ekstrand
0fa9211d7f nir/spirv: Make the global constants in spirv.h static
I've been promissed in a bug that this will be fixed in a future version of
the header.  However, in the interest of my branch building, I'm adding
these changes in myself for the moment.
2015-05-16 11:16:34 -07:00
Jason Ekstrand
036a4b1855 nir/spirv: Handle jump-to-loop in a more general way 2015-05-16 11:16:34 -07:00
Jason Ekstrand
56f533b3a0 nir/spirv: Handle boolean uniforms correctly 2015-05-16 11:16:34 -07:00
Jason Ekstrand
64bc58a88e nir/spirv: Handle control-flow with loops 2015-05-16 11:16:34 -07:00
Jason Ekstrand
3a2db9207d nir/spirv: Set a name on temporary variables 2015-05-16 11:16:34 -07:00
Jason Ekstrand
a28f8ad9f1 nir/spirv: Use the correct length for copying string literals 2015-05-16 11:16:34 -07:00
Jason Ekstrand
7b9c29e440 nir/spirv: Make vtn_ssa_value handle constants as well as ssa values 2015-05-16 11:16:33 -07:00
Jason Ekstrand
b0d1854efc nir/spirv: Add initial support for GLSL 4.50 builtins 2015-05-16 11:16:33 -07:00
Jason Ekstrand
1da9876486 nir/spirv: Split the core datastructures into a header file 2015-05-16 11:16:33 -07:00
Jason Ekstrand
98d78856f6 nir/spirv: Use the builder for all instructions
We don't actually use it to create all the instructions but we do use it
for insertion always.  This should make things far more consistent for
implementing extended instructions.
2015-05-16 11:16:33 -07:00
Jason Ekstrand
ff828749ea nir/spirv: Add support for a bunch of ALU operations 2015-05-16 11:16:33 -07:00
Jason Ekstrand
d2a7972557 nir/spirv: Add support for indirect array accesses 2015-05-16 11:16:33 -07:00
Jason Ekstrand
683c99908a nir/spirv: Explicitly type constants and SSA values 2015-05-16 11:16:33 -07:00
Jason Ekstrand
c5650148a9 nir/spirv: Handle OpBranchConditional
We do control-flow handling as a two-step process.  The first step is to
walk the instructions list and record various information about blocks and
functions.  This is where the acutal nir_function_overload objects get
created.  We also record the start/stop instruction for each block.  Then
a second pass walks over each of the functions and over the blocks in each
function in a way that's NIR-friendly and actually parses the instructions.
2015-05-16 11:16:33 -07:00
Jason Ekstrand
ebc152e4c9 nir/spirv: Add a helper for getting a value as an SSA value 2015-05-16 11:16:33 -07:00
Jason Ekstrand
f23afc549b nir/spirv: Split instruction handling into preamble and body sections 2015-05-16 11:16:33 -07:00
Jason Ekstrand
ae6d32c635 nir/spirv: Implement load/store instructiosn 2015-05-16 11:16:33 -07:00
Jason Ekstrand
88f6fbc897 nir: Add a helper for getting the tail of a deref chain 2015-05-16 11:16:33 -07:00
Jason Ekstrand
06acd174f3 nir/spirv: Actaully add variables to the funciton or shader 2015-05-16 11:16:33 -07:00
Jason Ekstrand
5045efa4aa nir/spirv: Add a vtn_untyped_value helper 2015-05-16 11:16:33 -07:00
Jason Ekstrand
01f3aa9c51 nir/spirv: Use vtn_value in the types code and fix a off-by-one error 2015-05-16 11:16:33 -07:00
Jason Ekstrand
6ff0830d64 nir/types: Add an is_vector_or_scalar helper 2015-05-16 11:16:33 -07:00
Jason Ekstrand
5acd472271 nir/spirv: Add support for deref chains 2015-05-16 11:16:33 -07:00
Jason Ekstrand
7182597e50 nir/types: Add a scalar type constructor 2015-05-16 11:16:32 -07:00
Jason Ekstrand
eccd798cc2 nir/spirv: Add support for OpLabel 2015-05-16 11:16:32 -07:00
Jason Ekstrand
a6cb9d9222 nir/spirv: Add support for declaring functions 2015-05-16 11:16:32 -07:00
Jason Ekstrand
8ee23dab04 nir/types: Add accessors for function parameter/return types 2015-05-16 11:16:32 -07:00
Jason Ekstrand
707b706d18 nir/spirv: Add support for declaring variables
Deref chains and variable load/store operations are still missing.
2015-05-16 11:16:32 -07:00
Jason Ekstrand
b2db85d8e4 nir/spirv: Add support for constants 2015-05-16 11:16:32 -07:00
Jason Ekstrand
3f83579664 nir/spirv: Add basic support for types 2015-05-16 11:16:32 -07:00
Jason Ekstrand
e9d3b1e694 nir/types: Add more helpers for creating types 2015-05-16 11:16:32 -07:00
Jason Ekstrand
fe550f0738 glsl/types: Expose the function_param and struct_field structs to C
Previously, they were hidden behind a #ifdef __cplusplus so C wouldn't find
them.  This commit simpliy moves the ifdef.
2015-05-16 11:16:32 -07:00
Jason Ekstrand
053778c493 glsl/types: Add support for function types 2015-05-16 11:16:32 -07:00
Jason Ekstrand
7b63b3de93 glsl: Add GLSL_TYPE_FUNCTION to the base types enums 2015-05-16 11:16:32 -07:00
Jason Ekstrand
2b570a49a9 nir/spirv: Rework the way values are added
Instead of having functions to add values and set various things, we just
have a function that does a few asserts and then returns the value.  The
caller is then responsible for setting the various fields.
2015-05-16 11:16:32 -07:00
Jason Ekstrand
f9a31ba044 nir/spirv: Add stub support for extension instructions 2015-05-16 11:16:32 -07:00
Jason Ekstrand
4763a13b07 REVERT: Add a simple helper program for testing SPIR-V -> NIR translation 2015-05-16 11:16:32 -07:00
Jason Ekstrand
cae8db6b7e glsl/compiler: Move the error_no_memory stub to standalone_scaffolding.cpp 2015-05-16 11:16:32 -07:00
Jason Ekstrand
98452cd8ae nir: Add the start of a SPIR-V to NIR translator
At the moment, it can handle the very basics of strings and can ignore
debug instructions.  It also has basic support for decorations.
2015-05-16 11:16:32 -07:00
Jason Ekstrand
573ca4a4a7 nir: Import the revision 30 SPIR-V header from Khronos 2015-05-16 11:16:31 -07:00
Jason Ekstrand
057bef8a84 vk/device: Use bias rather than layers for computing binding table size
Because we statically use the first 8 binding table entries for render
targets, we need to create a table of size 8 + surfaces.
2015-05-16 10:42:53 -07:00
Jason Ekstrand
22e61c9da4 vk/meta: Make clear a no-op if no layers need clearing
Among other things, this prevents recursive meta.
2015-05-16 10:30:05 -07:00
Jason Ekstrand
120394ac92 vk/meta: Save and restore the old bindings pointer
If we don't do this then recursive meta is completely broken.  What happens
is that the outer meta call may change the bindings pointer and the inner
meta call will change it again and, when it exits set it back to the
default.  However, the outer meta call may be relying on it being left
alone so it uses the non-meta descriptor sets instead of its own.
2015-05-16 10:28:04 -07:00
Jason Ekstrand
4223de769e vk/device: Simplify surface_count calculation 2015-05-16 10:23:09 -07:00
Jason Ekstrand
eb1952592e vk/glsl_helpers: Fix GLSL_VK_SHADER with respect to commas
Previously, the GLSL_VK_SHADER macro didn't work if the shader contained
commas outside of parentheses due to the way the C preprocessor works.
This commit fixes this by making it variadic again and doing it correctly
this time.
2015-05-15 22:17:07 -07:00
Kristian Høgsberg
3b9f32e893 vk: Make cmd_buffer->bindings a pointer
This lets us save and restore efficiently by just moving the pointer to
a temporary bindings struct for meta.
2015-05-15 18:12:07 -07:00
Kristian Høgsberg
9540130c41 vk: Move vertex buffers into struct anv_bindings 2015-05-15 16:34:31 -07:00
Kristian Høgsberg
0cfc493775 vk: Fix GLSL_VK_SHADER macro
Stringify doesn't work with __ARGV__. The last macro argument swallows
up excess arguments and as such we can just stringify that.
2015-05-15 16:15:04 -07:00
Kristian Høgsberg
af45f4a558 vk: Fix warning from missing initializer
Struct initializers need to be { 0, } to zero out the variable they're
initializing.
2015-05-15 16:07:17 -07:00
Kristian Høgsberg
bf096c9ec3 vk: Build binding tables at bind descriptor time
This changes the way descriptor sets and layouts work so that we fill
out binding table contents at the time we bind descriptor sets. We
manipulate the binding table contents and sampler state in a shadow-copy
in anv_cmd_buffer. At draw time, we allocate the actual binding table
and sampler state and flush the anv_cmd_buffer copies.
2015-05-15 16:05:31 -07:00
Kristian Høgsberg
1f6c220b45 vk: Update the bind map length to reflect MAX_SETS 2015-05-15 15:22:29 -07:00
Kristian Høgsberg
b806e80e66 vk: Flip back to using memfd for the allocators 2015-05-15 15:22:29 -07:00
Kristian Høgsberg
0a775e1eab vk: Rename dyn_state_pool to dynamic_state_pool
Given that we already tolerate surface_state_pool and the even longer
instruction_state_pool, there's no reason to arbitrarily abbreviate
dynamic.
2015-05-15 15:22:29 -07:00
Kristian Høgsberg
f5b0f1351f vk: Consolidate image, buffer and color attachment views
These are all just surface state, offset and a bo.
2015-05-15 15:22:29 -07:00
Jason Ekstrand
41db8db0f2 vk: Add a GLSL scraper utility
This new utility, glsl_scraper.py scrapes C files for instances of the
GLSL_VK_SHADER macro, pulls out the shader source, and compiles it to
SPIR-V.  The compilation is done using glslValidator.  The result is then
placed into another C file as arrays of dwords that can be easiliy handed
to a Vulkan driver.
2015-05-14 19:18:57 -07:00
Jason Ekstrand
79ace6def6 vk/meta: Add a magic GLSL shader source macro 2015-05-14 19:07:34 -07:00
Jason Ekstrand
018a0c1741 vk/meta: Add a better comment about the VS for blits 2015-05-14 11:39:32 -07:00
Jason Ekstrand
8c92701a69 vk/test: Use VK_IMAGE_TILING_OPTIMAL for the render target 2015-05-13 22:27:38 -07:00
Jason Ekstrand
4fb8bddc58 vk/test: Do a copy of the RT into a linear buffer and write that to a PNG 2015-05-13 22:23:30 -07:00
Jason Ekstrand
bd5b76d6d0 vk/meta: Add the start of a blit implementation
Currently, we only implement CopyImageToBuffer
2015-05-13 22:23:30 -07:00
Jason Ekstrand
94b8c0b810 vk/pipeline: Default to a SamplerCount of 1 for PS 2015-05-13 22:23:30 -07:00
Jason Ekstrand
d3d4776202 vk/pipeline: Add an extra flag for force-disabling the vertex shader
This way we can pass in a vertex shader and yet have the pipeline emit an
empty 3DSTATE_VS packet.  We need this for meta because we need to trick
the compiler into not deleting our inputs but at the same time disable the
VS so that we can use a rectlist.  This should go away once we actually get
SPIR-V.
2015-05-13 22:23:30 -07:00
Jason Ekstrand
a1309c5255 vk/pass: Emit a flushing pipe control at the end of the pass
This is rather crude but it at least makes sure that all the render targets
get flushed at the end of the pass.  We probably actually want to do
somthing based on image layout traansitions, but this will work for now.
2015-05-13 22:23:30 -07:00
Jason Ekstrand
07943656a7 vk/compiler: Set the binding table texture_start
This is by no means a complete solution to the binding table problems.
However, it does make texturing actually work.  Before, we were texturing
from the render target since they were both starting at 0.
2015-05-13 22:23:30 -07:00
Jason Ekstrand
cd197181f2 vk/compiler: Zero the prog data
We use prog_data[stage] != NULL to determine whether or not we need to
clean up that stage.  Make sure it default to NULL.
2015-05-13 22:22:59 -07:00
Jason Ekstrand
1f7dcf9d75 vk/image: Stash more information in images and views 2015-05-13 22:22:59 -07:00
Jason Ekstrand
43126388cd vk/meta: Save/restore more stuff in cmd_buffer_restore 2015-05-13 22:22:59 -07:00
Chad Versace
50806e8dec vk: Install headers
I need this for building a testsuite.
2015-05-13 17:49:26 -07:00
Kristian Høgsberg
83c7e1f1db vk: Add support for sampler descriptors 2015-05-13 14:47:11 -07:00
Kristian Høgsberg
4f9eaf77a5 vk: Use a typesafe anv_descriptor struct 2015-05-13 14:47:11 -07:00
Kristian Høgsberg
5c9d77600b vk: Create and bind a sampler in vk.c 2015-05-13 14:47:11 -07:00
Kristian Høgsberg
18acfa7301 vk: Fix copy-n-paste sType in vkCreateSampler 2015-05-13 14:47:11 -07:00
Kristian Høgsberg
a1ec789b0b vk: Add a dynamic state stream to anv_cmd_buffer
We'll need this for sampler state.
2015-05-13 14:47:11 -07:00
Kristian Høgsberg
3f52c016fa vk: Move struct anv_sampler to private.h 2015-05-13 14:47:11 -07:00
Kristian Høgsberg
a77229c979 vk: Allocate layout->count number of descriptors
layout->count is the number of descriptors the application
requested. layout->total is the number of entries we need across all
stages.
2015-05-13 14:47:11 -07:00
Kristian Høgsberg
a3fd136509 vk: Fill out sampler state from API values 2015-05-13 14:47:11 -07:00
Chad Versace
828817b88f vk: Ignore vk executable 2015-05-13 12:05:38 -07:00
Kristian Høgsberg
2b7a060178 vk: Fix stale error handling in vkQueueSubmit 2015-05-12 14:38:58 -07:00
Kristian Høgsberg
cb986ef597 vk: Submit all cmd buffers passed to vkQueueSubmit 2015-05-12 14:38:12 -07:00
Kristian Høgsberg
9905481552 vk: Add generated header for HSW and IVB (GEN75 and GEN7) 2015-05-12 14:29:04 -07:00
Jason Ekstrand
ffe9f60358 vk: Add stub() and stub_return() macros and mark piles of functions as stubs 2015-05-12 13:45:02 -07:00
Jason Ekstrand
d3b374ce59 vk/util: Add a anv_finishme function/macro 2015-05-12 13:43:36 -07:00
Jason Ekstrand
7727720585 vk/meta: Break setting up meta clear state into it's own functin 2015-05-12 13:03:50 -07:00
Jason Ekstrand
4336a1bc00 vk/pipeline: Add support for disabling the scissor in "extra" 2015-05-12 12:53:01 -07:00
Kristian Høgsberg
d77c34d1d2 vk: Add clear load-op for render passes 2015-05-11 23:25:29 -07:00
Kristian Høgsberg
b734e0bcc5 vk: Add support for driver-internal custom pipelines
This lets us disable the viewport, use rect lists and repclear.
2015-05-11 23:25:29 -07:00
Kristian Høgsberg
ad132bbe48 vk: Fix 3DSTATE_VERTEX_BUFFER emission
Set VertexBufferIndex to the attribute binding, not the location.
2015-05-11 23:25:29 -07:00
Kristian Høgsberg
6a895c6681 vk: Add 32 bpc signed and unsigned integer formats 2015-05-11 23:25:29 -07:00
Kristian Høgsberg
55b9b703ea vk: Add anv_batch_emit_merge() helper macro
This lets us emit a state packet by merging to half-backed versions,
typically one from the pipeline object and one from a dynamic state
objects.
2015-05-11 23:25:28 -07:00
Kristian Høgsberg
099faa1a2b vk: Store bo pointer in anv_image and anv_buffer
We don't need to point back to the memory object the bo came from.
Pointing directly to a bo lets us bind images and buffers to other
bos - like our allocator bos.
2015-05-11 23:25:28 -07:00
Kristian Høgsberg
4f25f5d86c vk: Support not having a vertex shader
This lets us bypass the vertex shader and pass data straight into
the rasterizer part of the pipeline.
2015-05-11 23:25:28 -07:00
Kristian Høgsberg
20ad071190 vk: Allow NULL as a valid pipeline layout
Vertex buffers and render targets aren't part of the layout so having
an empty layout is pretty common.
2015-05-11 22:12:56 -07:00
Kristian Høgsberg
769785c497 Add vulkan driver for BDW 2015-05-09 11:38:32 -07:00
2754 changed files with 131718 additions and 278523 deletions

View File

@@ -5,7 +5,6 @@
(c-file-style . "stroustrup")
(fill-column . 78)
(eval . (progn
(c-set-offset 'case-label '0)
(c-set-offset 'innamespace '0)
(c-set-offset 'inline-open '0)))
)

View File

@@ -1,101 +0,0 @@
language: c
sudo: false
cache:
directories:
- $HOME/.ccache
addons:
apt:
packages:
- libdrm-dev
- libudev-dev
- x11proto-xf86vidmode-dev
- libexpat1-dev
- libxcb-dri2-0-dev
- libx11-xcb-dev
- llvm-3.4-dev
- scons
env:
global:
- XORG_RELEASES=http://xorg.freedesktop.org/releases/individual
- XCB_RELEASES=http://xcb.freedesktop.org/dist
- XORGMACROS_VERSION=util-macros-1.19.0
- GLPROTO_VERSION=glproto-1.4.17
- DRI2PROTO_VERSION=dri2proto-2.8
- DRI3PROTO_VERSION=dri3proto-1.0
- PRESENTPROTO_VERSION=presentproto-1.0
- LIBPCIACCESS_VERSION=libpciaccess-0.13.4
- LIBDRM_VERSION=libdrm-2.4.65
- XCBPROTO_VERSION=xcb-proto-1.11
- LIBXCB_VERSION=libxcb-1.11
- LIBXSHMFENCE_VERSION=libxshmfence-1.2
- PKG_CONFIG_PATH=$HOME/prefix/lib/pkgconfig
matrix:
- BUILD=make
- BUILD=scons
install:
- export PATH="/usr/lib/ccache:$PATH"
- pip install --user mako
# Install dependencies where we require specific versions (or where
# disallowed by Travis CI's package whitelisting).
- wget $XORG_RELEASES/util/$XORGMACROS_VERSION.tar.bz2
- tar -jxvf $XORGMACROS_VERSION.tar.bz2
- (cd $XORGMACROS_VERSION && ./configure --prefix=$HOME/prefix && make install)
- wget $XORG_RELEASES/proto/$GLPROTO_VERSION.tar.bz2
- tar -jxvf $GLPROTO_VERSION.tar.bz2
- (cd $GLPROTO_VERSION && ./configure --prefix=$HOME/prefix && make install)
- wget $XORG_RELEASES/proto/$DRI2PROTO_VERSION.tar.bz2
- tar -jxvf $DRI2PROTO_VERSION.tar.bz2
- (cd $DRI2PROTO_VERSION && ./configure --prefix=$HOME/prefix && make install)
- wget $XORG_RELEASES/proto/$DRI3PROTO_VERSION.tar.bz2
- tar -jxvf $DRI3PROTO_VERSION.tar.bz2
- (cd $DRI3PROTO_VERSION && ./configure --prefix=$HOME/prefix && make install)
- wget $XORG_RELEASES/proto/$PRESENTPROTO_VERSION.tar.bz2
- tar -jxvf $PRESENTPROTO_VERSION.tar.bz2
- (cd $PRESENTPROTO_VERSION && ./configure --prefix=$HOME/prefix && make install)
- wget $XCB_RELEASES/$XCBPROTO_VERSION.tar.bz2
- tar -jxvf $XCBPROTO_VERSION.tar.bz2
- (cd $XCBPROTO_VERSION && ./configure --prefix=$HOME/prefix && make install)
- wget $XCB_RELEASES/$LIBXCB_VERSION.tar.bz2
- tar -jxvf $LIBXCB_VERSION.tar.bz2
- (cd $LIBXCB_VERSION && ./configure --prefix=$HOME/prefix && make install)
- wget $XORG_RELEASES/lib/$LIBPCIACCESS_VERSION.tar.bz2
- tar -jxvf $LIBPCIACCESS_VERSION.tar.bz2
- (cd $LIBPCIACCESS_VERSION && ./configure --prefix=$HOME/prefix && make install)
- wget http://dri.freedesktop.org/libdrm/$LIBDRM_VERSION.tar.bz2
- tar -jxvf $LIBDRM_VERSION.tar.bz2
- (cd $LIBDRM_VERSION && ./configure --prefix=$HOME/prefix && make install)
- wget $XORG_RELEASES/lib/$LIBXSHMFENCE_VERSION.tar.bz2
- tar -jxvf $LIBXSHMFENCE_VERSION.tar.bz2
- (cd $LIBXSHMFENCE_VERSION && ./configure --prefix=$HOME/prefix && make install)
# Disabled LLVM (and therefore r300 and r600) because the build fails
# with "undefined reference to `clock_gettime'" and "undefined
# reference to `setupterm'" in llvmpipe.
script:
- if test "x$BUILD" = xmake; then
./autogen.sh --enable-debug
--disable-gallium-llvm
--with-egl-platforms=x11,drm
--with-dri-drivers=i915,i965,radeon,r200,swrast,nouveau
--with-gallium-drivers=svga,swrast,vc4,virgl
;
make && make check;
elif test x$BUILD = xscons; then
scons;
fi

View File

@@ -21,8 +21,13 @@
# FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
# DEALINGS IN THE SOFTWARE.
# use c99 compiler by default
ifeq ($(LOCAL_CC),)
ifeq ($(LOCAL_IS_HOST_MODULE),true)
LOCAL_CFLAGS += -D_GNU_SOURCE
LOCAL_CC := $(HOST_CC) -std=c99 -D_GNU_SOURCE
else
LOCAL_CC := $(TARGET_CC) -std=c99
endif
endif
LOCAL_C_INCLUDES += \
@@ -32,14 +37,11 @@ LOCAL_C_INCLUDES += \
MESA_VERSION := $(shell cat $(MESA_TOP)/VERSION)
# define ANDROID_VERSION (e.g., 4.0.x => 0x0400)
LOCAL_CFLAGS += \
-Wno-unused-parameter \
-Wno-date-time \
-DPACKAGE_VERSION=\"$(MESA_VERSION)\" \
-DPACKAGE_BUGREPORT=\"https://bugs.freedesktop.org/enter_bug.cgi?product=Mesa\" \
-DANDROID_VERSION=0x0$(MESA_ANDROID_MAJOR_VERSION)0$(MESA_ANDROID_MINOR_VERSION)
LOCAL_CFLAGS += \
-D__STDC_LIMIT_MACROS \
-DHAVE___BUILTIN_EXPECT \
-DHAVE___BUILTIN_FFS \
-DHAVE___BUILTIN_FFSLL \
@@ -57,10 +59,6 @@ LOCAL_CFLAGS += \
-fvisibility=hidden \
-Wno-sign-compare
# mesa requires at least c99 compiler
LOCAL_CONLYFLAGS += \
-std=c99
ifeq ($(strip $(MESA_ENABLE_ASM)),true)
ifeq ($(TARGET_ARCH),x86)
LOCAL_CFLAGS += \
@@ -72,7 +70,7 @@ endif
ifeq ($(MESA_ENABLE_LLVM),true)
LOCAL_CFLAGS += \
-DHAVE_LLVM=0x0305 -DMESA_LLVM_VERSION_PATCH=2 \
-DHAVE_LLVM=0x0305 -DLLVM_VERSION_PATCH=2 \
-D__STDC_CONSTANT_MACROS \
-D__STDC_FORMAT_MACROS \
-D__STDC_LIMIT_MACROS
@@ -89,6 +87,3 @@ LOCAL_CPPFLAGS += \
ifeq ($(strip $(LOCAL_MODULE_TAGS)),)
LOCAL_MODULE_TAGS := optional
endif
# Quiet down the build system and remove any .h files from the sources
LOCAL_SRC_FILES := $(patsubst %.h, , $(LOCAL_SRC_FILES))

View File

@@ -24,7 +24,7 @@
# BOARD_GPU_DRIVERS should be defined. The valid values are
#
# classic drivers: i915 i965
# gallium drivers: swrast freedreno i915g ilo nouveau r300g r600g radeonsi vc4 virgl vmwgfx
# gallium drivers: swrast freedreno i915g ilo nouveau r300g r600g radeonsi vc4 vmwgfx
#
# The main target is libGLES_mesa. For each classic driver enabled, a DRI
# module will also be built. DRI modules will be loaded by libGLES_mesa.
@@ -42,15 +42,13 @@ $(call local-intermediates-dir)
endef
endif
MESA_DRI_MODULE_REL_PATH := dri
MESA_DRI_MODULE_PATH := $(TARGET_OUT_SHARED_LIBRARIES)/$(MESA_DRI_MODULE_REL_PATH)
MESA_DRI_MODULE_UNSTRIPPED_PATH := $(TARGET_OUT_SHARED_LIBRARIES_UNSTRIPPED)/$(MESA_DRI_MODULE_REL_PATH)
MESA_COMMON_MK := $(MESA_TOP)/Android.common.mk
MESA_PYTHON2 := python
DRM_GRALLOC_TOP := hardware/drm_gralloc
classic_drivers := i915 i965
gallium_drivers := swrast freedreno i915g ilo nouveau r300g r600g radeonsi vmwgfx vc4 virgl
gallium_drivers := swrast freedreno i915g ilo nouveau r300g r600g radeonsi vmwgfx vc4
MESA_GPU_DRIVERS := $(strip $(BOARD_GPU_DRIVERS))
@@ -90,11 +88,11 @@ ifneq ($(strip $(MESA_GPU_DRIVERS)),)
SUBDIRS := \
src/loader \
src/mapi \
src/compiler \
src/compiler/glsl \
src/glsl \
src/mesa \
src/util \
src/egl \
src/egl/main \
src/egl/drivers/dri2 \
src/mesa/drivers/dri
ifeq ($(strip $(MESA_BUILD_GALLIUM)),true)

View File

@@ -24,21 +24,15 @@ SUBDIRS = src
AM_DISTCHECK_CONFIGURE_FLAGS = \
--enable-dri3 \
--enable-gallium-tests \
--enable-gallium-osmesa \
--enable-gbm \
--enable-gles1 \
--enable-gles2 \
--enable-glx-tls \
--enable-nine \
--enable-opencl \
--enable-va \
--enable-vdpau \
--enable-xa \
--enable-xvmc \
--disable-llvm-shared-libs \
--with-egl-platforms=x11,wayland,drm,surfaceless \
--with-dri-drivers=i915,i965,nouveau,radeon,r200,swrast \
--with-gallium-drivers=i915,ilo,nouveau,r300,r600,radeonsi,freedreno,svga,swrast,vc4,virgl
--with-egl-platforms=x11,wayland,drm
ACLOCAL_AMFLAGS = -I m4
@@ -54,6 +48,7 @@ noinst_HEADERS = \
include/c99_alloca.h \
include/c99_compat.h \
include/c99_math.h \
include/c99 \
include/c11 \
include/D3D9 \
include/HaikuGL \

View File

@@ -1 +1 @@
11.2.1
10.7.0-devel

View File

@@ -1,73 +0,0 @@
# http://www.appveyor.com/docs/appveyor-yml
#
# To setup AppVeyor for your own personal repositories do the following:
# - Sign up
# - Add a new project
# - Select Git and fill in the Git clone URL
# - Setup a Git hook as explained in
# https://github.com/appveyor/webhooks#installing-git-hook
# - Check 'Settings > General > Skip branches without appveyor.yml'
# - Check 'Settings > General > Rolling builds'
# - Setup the global or project notifications to your liking
#
# Note that kicking (or restarting) a build via the web UI will not work, as it
# will fail to find appveyor.yml . The Git hook is the most practical way to
# kick a build.
#
# See also:
# - http://help.appveyor.com/discussions/problems/2209-node-grunt-build-specify-a-project-or-solution-file-the-directory-does-not-contain-a-project-or-solution-file
# - http://help.appveyor.com/discussions/questions/1184-build-config-vs-appveyoryaml
version: '{build}'
branches:
except:
- /^travis.*$/
# Don't download the full Mesa history to speed up cloning. However the clone
# depth must not be too small, otherwise builds might fail when lots of patches
# are committed in succession, because the desired commit is not found on the
# truncated history.
#
# See also:
# - https://www.appveyor.com/blog/2014/06/04/shallow-clone-for-git-repositories
clone_depth: 100
cache:
- win_flex_bison-2.4.5.zip
- llvm-3.3.1-msvc2013-mtd.7z
environment:
WINFLEXBISON_ARCHIVE: win_flex_bison-2.4.5.zip
LLVM_ARCHIVE: llvm-3.3.1-msvc2013-mtd.7z
install:
# Check pip
- python --version
- python -m pip --version
# Install Mako
- python -m pip install --egg Mako
# Install SCons
- python -m pip install --egg scons==2.4.1
- scons --version
# Install flex/bison
- if not exist "%WINFLEXBISON_ARCHIVE%" appveyor DownloadFile "http://downloads.sourceforge.net/project/winflexbison/%WINFLEXBISON_ARCHIVE%"
- 7z x -y -owinflexbison\ "%WINFLEXBISON_ARCHIVE%" > nul
- set Path=%CD%\winflexbison;%Path%
- win_flex --version
- win_bison --version
# Download and extract LLVM
- if not exist "%LLVM_ARCHIVE%" appveyor DownloadFile "https://people.freedesktop.org/~jrfonseca/llvm/%LLVM_ARCHIVE%"
- 7z x -y "%LLVM_ARCHIVE%" > nul
- mkdir llvm\bin
- set LLVM=%CD%\llvm
build_script:
- scons -j%NUMBER_OF_PROCESSORS% MSVC_VERSION=12.0 llvm=1
# It's possible to setup notification here, as described in
# http://www.appveyor.com/docs/notifications#appveyor-yml-configuration , but
# doing so would cause the notification settings to be replicated across all
# repos, which is most likely undesired. So it's better to rely on the
# Appveyor global/project notification settings.

View File

@@ -15,14 +15,17 @@
# $ DRYRUN=yes bin/bugzilla_mesa.sh mesa-9.0.2..mesa-9.0.3 | wc -l
# regex pattern: trim before bug number
trim_before='s/.*show_bug.cgi?id=\([0-9]*\).*/\1/'
# regex pattern: trim before url
trim_before='s/.*\(http\)/\1/'
# regex pattern: reconstruct the url
use_after='s,^,https://bugs.freedesktop.org/show_bug.cgi?id=,'
# regex pattern: trim after url
trim_after='s/\(show_bug.cgi?id=[0-9]*\).*/\1/'
# regex pattern: always use https
use_https='s/http:/https:/'
# extract fdo urls from commit log
urls=$(git log $* | grep 'bugs.freedesktop.org/show_bug' | sed -e $trim_before | sort -n -u | sed -e $use_after)
urls=$(git log $* | grep 'bugs.freedesktop.org/show_bug' | sed -e $trim_before -e $trim_after -e $use_https | sort | uniq)
# if DRYRUN is set to "yes", simply print the URLs and don't fetch the
# details from fdo bugzilla.

View File

@@ -1,35 +0,0 @@
#!/bin/sh
# Script for generating a list of candidates which fix commits that have been
# previously cherry-picked to a stable branch.
#
# Usage examples:
#
# $ bin/get-extra-pick-list.sh
# $ bin/get-extra-pick-list.sh > picklist
# $ bin/get-extra-pick-list.sh | tee picklist
# Use the last branchpoint as our limit for the search
# XXX: there should be a better way for this
latest_branchpoint=`git branch | grep \* | cut -c 3-`-branchpoint
# Grep for commits with "cherry picked from commit" in the commit message.
git log --reverse --grep="cherry picked from commit" $latest_branchpoint..HEAD |\
grep "cherry picked from commit" |\
sed -e 's/^[[:space:]]*(cherry picked from commit[[:space:]]*//' -e 's/)//' |\
cut -c -8 |\
while read sha
do
# Check if the original commit is referenced in master
git log -n1 --pretty=oneline --grep=$sha $latest_branchpoint..origin/master |\
cut -c -8 |\
while read candidate
do
# Check if the potential fix, hasn't landed in branch yet.
found=`git log -n1 --pretty=oneline --reverse --grep=$candidate $latest_branchpoint..HEAD |wc -l`
if test $found = 0
then
echo Commit $candidate might need to be picked, as it references $sha
fi
done
done

View File

@@ -44,7 +44,7 @@ AC_INIT([Mesa], [MESA_VERSION],
AC_CONFIG_AUX_DIR([bin])
AC_CONFIG_MACRO_DIR([m4])
AC_CANONICAL_SYSTEM
AM_INIT_AUTOMAKE([foreign tar-ustar dist-xz subdir-objects])
AM_INIT_AUTOMAKE([foreign tar-ustar dist-xz])
dnl We only support native Windows builds (MinGW/MSVC) through SCons.
case "$host_os" in
@@ -64,25 +64,22 @@ m4_ifdef([AM_PROG_AR], [AM_PROG_AR])
dnl Set internal versions
OSMESA_VERSION=8
AC_SUBST([OSMESA_VERSION])
OPENCL_VERSION=1
AC_SUBST([OPENCL_VERSION])
dnl Versions for external dependencies
LIBDRM_REQUIRED=2.4.60
LIBDRM_REQUIRED=2.4.38
LIBDRM_RADEON_REQUIRED=2.4.56
LIBDRM_AMDGPU_REQUIRED=2.4.63
LIBDRM_INTEL_REQUIRED=2.4.61
LIBDRM_NVVIEUX_REQUIRED=2.4.66
LIBDRM_NOUVEAU_REQUIRED=2.4.66
LIBDRM_FREEDRENO_REQUIRED=2.4.67
LIBDRM_INTEL_REQUIRED=2.4.60
LIBDRM_NVVIEUX_REQUIRED=2.4.33
LIBDRM_NOUVEAU_REQUIRED="2.4.33 libdrm >= 2.4.41"
LIBDRM_FREEDRENO_REQUIRED=2.4.57
DRI2PROTO_REQUIRED=2.6
DRI3PROTO_REQUIRED=1.0
PRESENTPROTO_REQUIRED=1.0
LIBUDEV_REQUIRED=151
GLPROTO_REQUIRED=1.4.14
LIBOMXIL_BELLAGIO_REQUIRED=0.0
LIBVA_REQUIRED=0.38.0
VDPAU_REQUIRED=1.1
LIBVA_REQUIRED=0.35.0
VDPAU_REQUIRED=0.4.1
WAYLAND_REQUIRED=1.2.0
XCB_REQUIRED=1.9.3
XCBDRI2_REQUIRED=1.8
@@ -98,7 +95,7 @@ AC_PROG_CXX
AM_PROG_CC_C_O
AM_PROG_AS
AX_CHECK_GNU_MAKE
AC_CHECK_PROGS([PYTHON2], [python2.7 python2 python])
AC_CHECK_PROGS([PYTHON2], [python2 python])
AC_PROG_SED
AC_PROG_MKDIR_P
@@ -107,13 +104,11 @@ AC_SYS_LARGEFILE
LT_PREREQ([2.2])
LT_INIT([disable-static])
AC_CHECK_PROG(RM, rm, [rm -f])
AX_PROG_BISON([],
AS_IF([test ! -f "$srcdir/src/compiler/glsl/glcpp/glcpp-parse.c"],
AS_IF([test ! -f "$srcdir/src/glsl/glcpp/glcpp-parse.c"],
[AC_MSG_ERROR([bison not found - unable to compile glcpp-parse.y])]))
AX_PROG_FLEX([],
AS_IF([test ! -f "$srcdir/src/compiler/glsl/glcpp/glcpp-lex.c"],
AS_IF([test ! -f "$srcdir/src/glsl/glcpp/glcpp-lex.c"],
[AC_MSG_ERROR([flex not found - unable to compile glcpp-lex.l])]))
AC_CHECK_PROG(INDENT, indent, indent, cat)
@@ -197,13 +192,6 @@ if test "x$GCC" = xyes -a "x$acv_mesa_CLANG" = xno; then
fi
fi
dnl We don't support building Mesa with Sun C compiler
dnl https://bugs.freedesktop.org/show_bug.cgi?id=93189
AC_CHECK_DECL([__SUNPRO_C], [SUNCC=yes], [SUNCC=no])
if test "x$SUNCC" = xyes; then
AC_MSG_ERROR([Building with Sun C compiler is not supported, use GCC instead.])
fi
dnl Check for compiler builtins
AX_GCC_BUILTIN([__builtin_bswap32])
AX_GCC_BUILTIN([__builtin_bswap64])
@@ -217,14 +205,11 @@ AX_GCC_BUILTIN([__builtin_popcount])
AX_GCC_BUILTIN([__builtin_popcountll])
AX_GCC_BUILTIN([__builtin_unreachable])
AX_GCC_FUNC_ATTRIBUTE([const])
AX_GCC_FUNC_ATTRIBUTE([flatten])
AX_GCC_FUNC_ATTRIBUTE([format])
AX_GCC_FUNC_ATTRIBUTE([malloc])
AX_GCC_FUNC_ATTRIBUTE([packed])
AX_GCC_FUNC_ATTRIBUTE([pure])
AX_GCC_FUNC_ATTRIBUTE([unused])
AX_GCC_FUNC_ATTRIBUTE([warn_unused_result])
AM_CONDITIONAL([GEN_ASM_OFFSETS], test "x$GEN_ASM_OFFSETS" = xyes)
@@ -245,7 +230,7 @@ _SAVE_LDFLAGS="$LDFLAGS"
_SAVE_CPPFLAGS="$CPPFLAGS"
dnl Compiler macros
DEFINES="-D__STDC_LIMIT_MACROS -D__STDC_CONSTANT_MACROS"
DEFINES=""
AC_SUBST([DEFINES])
case "$host_os" in
linux*|*-gnu*|gnu*)
@@ -296,16 +281,14 @@ if test "x$GCC" = xyes; then
# Work around aliasing bugs - developers should comment this out
CFLAGS="$CFLAGS -fno-strict-aliasing"
# We don't want floating-point math functions to set errno or trap
CFLAGS="$CFLAGS -fno-math-errno -fno-trapping-math"
# gcc's builtin memcmp is slower than glibc's
# http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43052
CFLAGS="$CFLAGS -fno-builtin-memcmp"
# Flags to help ensure that certain portions of the code -- and only those
# portions -- can be built with MSVC:
# - src/util, src/gallium/auxiliary, rc/gallium/drivers/llvmpipe, and
# - src/util, src/gallium/auxiliary, and src/gallium/drivers/llvmpipe needs
# to build with Windows SDK 7.0.7600, which bundles MSVC 2008
# - non-Linux/Posix OpenGL portions needs to build on MSVC 2013 (which
# supports most of C99)
# - the rest has no compiler compiler restrictions
@@ -322,6 +305,9 @@ if test "x$GCC" = xyes; then
AC_MSG_RESULT([yes])],
AC_MSG_RESULT([no]));
CFLAGS="$save_CFLAGS"
MSVC2008_COMPAT_CFLAGS="$MSVC2013_COMPAT_CFLAGS -Werror=declaration-after-statement"
MSVC2008_COMPAT_CXXFLAGS="$MSVC2013_COMPAT_CXXFLAGS"
fi
if test "x$GXX" = xyes; then
CXXFLAGS="$CXXFLAGS -Wall"
@@ -349,6 +335,8 @@ fi
AC_SUBST([MSVC2013_COMPAT_CFLAGS])
AC_SUBST([MSVC2013_COMPAT_CXXFLAGS])
AC_SUBST([MSVC2008_COMPAT_CFLAGS])
AC_SUBST([MSVC2008_COMPAT_CXXFLAGS])
dnl even if the compiler appears to support it, using visibility attributes isn't
dnl going to do anything useful currently on cygwin apart from emit lots of warnings
@@ -377,11 +365,10 @@ save_CFLAGS="$CFLAGS"
CFLAGS="$SSE41_CFLAGS $CFLAGS"
AC_COMPILE_IFELSE([AC_LANG_SOURCE([[
#include <smmintrin.h>
int param;
int main () {
__m128i a = _mm_set1_epi32 (param), b = _mm_set1_epi32 (param + 1), c;
__m128i a = _mm_set1_epi32 (0), b = _mm_set1_epi32 (0), c;
c = _mm_max_epu32(a, b);
return _mm_cvtsi128_si32(c);
return 0;
}]])], SSE41_SUPPORTED=1)
CFLAGS="$save_CFLAGS"
if test "x$SSE41_SUPPORTED" = x1; then
@@ -390,61 +377,6 @@ fi
AM_CONDITIONAL([SSE41_SUPPORTED], [test x$SSE41_SUPPORTED = x1])
AC_SUBST([SSE41_CFLAGS], $SSE41_CFLAGS)
dnl Check for Endianness
AC_C_BIGENDIAN(
little_endian=no,
little_endian=yes,
little_endian=no,
little_endian=no
)
dnl Check for POWER8 Architecture
PWR8_CFLAGS="-mpower8-vector"
have_pwr8_intrinsics=no
AC_MSG_CHECKING(whether gcc supports -mpower8-vector)
save_CFLAGS=$CFLAGS
CFLAGS="$PWR8_CFLAGS $CFLAGS"
AC_COMPILE_IFELSE([AC_LANG_SOURCE([[
#if defined(__GNUC__) && (__GNUC__ < 4 || (__GNUC__ == 4 && __GNUC_MINOR__ < 8))
#error "Need GCC >= 4.8 for sane POWER8 support"
#endif
#include <altivec.h>
int main () {
vector unsigned char r;
vector unsigned int v = vec_splat_u32 (1);
r = __builtin_vec_vgbbd ((vector unsigned char) v);
return 0;
}]])], have_pwr8_intrinsics=yes)
CFLAGS=$save_CFLAGS
AC_ARG_ENABLE(pwr8,
[AC_HELP_STRING([--disable-pwr8-inst],
[disable POWER8-specific instructions])],
[enable_pwr8=$enableval], [enable_pwr8=auto])
if test "x$enable_pwr8" = xno ; then
have_pwr8_intrinsics=disabled
fi
if test $have_pwr8_intrinsics = yes && test $little_endian = yes ; then
DEFINES="$DEFINES -D_ARCH_PWR8"
CXXFLAGS="$CXXFLAGS $PWR8_CFLAGS"
CFLAGS="$CFLAGS $PWR8_CFLAGS"
else
PWR8_CFLAGS=
fi
AC_MSG_RESULT($have_pwr8_intrinsics)
if test "x$enable_pwr8" = xyes && test $have_pwr8_intrinsics = no ; then
AC_MSG_ERROR([POWER8 compiler support not detected])
fi
if test $have_pwr8_intrinsics = yes && test $little_endian = no ; then
AC_MSG_WARN([POWER8 optimization is enabled only on POWER8 Little-Endian])
fi
AC_SUBST([PWR8_CFLAGS], $PWR8_CFLAGS)
dnl Can't have static and shared libraries, default to static if user
dnl explicitly requested. If both disabled, set to static since shared
dnl was explicitly requested.
@@ -470,29 +402,8 @@ AC_ARG_ENABLE([debug],
[enable_debug="$enableval"],
[enable_debug=no]
)
AC_ARG_ENABLE([profile],
[AS_HELP_STRING([--enable-profile],
[enable profiling of code @<:@default=disabled@:>@])],
[enable_profile="$enableval"],
[enable_profile=no]
)
if test "x$enable_profile" = xyes; then
DEFINES="$DEFINES -DPROFILE"
if test "x$GCC" = xyes; then
CFLAGS="$CFLAGS -fno-omit-frame-pointer"
fi
if test "x$GXX" = xyes; then
CXXFLAGS="$CXXFLAGS -fno-omit-frame-pointer"
fi
fi
if test "x$enable_debug" = xyes; then
DEFINES="$DEFINES -DDEBUG"
if test "x$enable_profile" = xyes; then
AC_MSG_WARN([Debug and Profile are enabled at the same time])
fi
if test "x$GCC" = xyes; then
if ! echo "$CFLAGS" | grep -q -e '-g'; then
CFLAGS="$CFLAGS -g"
@@ -613,32 +524,15 @@ AM_CONDITIONAL(HAVE_COMPAT_SYMLINKS, test "x$HAVE_COMPAT_SYMLINKS" = xyes)
dnl
dnl library names
dnl
dnl Unfortunately we need to do a few things that libtool can't help us with,
dnl so we need some knowledge of shared library filenames:
dnl
dnl LIB_EXT is the extension used when creating symlinks for alternate
dnl filenames for a shared library which will be dynamically loaded
dnl
dnl IMP_LIB_EXT is the extension used when checking for the presence of a
dnl the file for a shared library we wish to link with
dnl
case "$host_os" in
darwin* )
LIB_EXT='dylib'
IMP_LIB_EXT=$LIB_EXT
;;
LIB_EXT='dylib' ;;
cygwin* )
LIB_EXT='dll'
IMP_LIB_EXT='dll.a'
;;
LIB_EXT='dll' ;;
aix* )
LIB_EXT='a'
IMP_LIB_EXT=$LIB_EXT
;;
LIB_EXT='a' ;;
* )
LIB_EXT='so'
IMP_LIB_EXT=$LIB_EXT
;;
LIB_EXT='so' ;;
esac
AC_SUBST([LIB_EXT])
@@ -757,7 +651,6 @@ fi
AC_CHECK_HEADER([xlocale.h], [DEFINES="$DEFINES -DHAVE_XLOCALE_H"])
AC_CHECK_HEADER([sys/sysctl.h], [DEFINES="$DEFINES -DHAVE_SYS_SYSCTL_H"])
AC_CHECK_FUNC([strtof], [DEFINES="$DEFINES -DHAVE_STRTOF"])
AC_CHECK_FUNC([mkostemp], [DEFINES="$DEFINES -DHAVE_MKOSTEMP"])
dnl Check to see if dlopen is in default libraries (like Solaris, which
dnl has it in libc), or if libdl is needed to get it.
@@ -845,11 +738,6 @@ linux*)
dri3_default=no
;;
esac
if test "x$enable_dri" = xno; then
dri3_default=no
fi
AC_ARG_ENABLE([dri3],
[AS_HELP_STRING([--enable-dri3],
[enable DRI3 @<:@default=auto@:>@])],
@@ -949,7 +837,7 @@ GALLIUM_DRIVERS_DEFAULT="r300,r600,svga,swrast"
AC_ARG_WITH([gallium-drivers],
[AS_HELP_STRING([--with-gallium-drivers@<:@=DIRS...@:>@],
[comma delimited Gallium drivers list, e.g.
"i915,ilo,nouveau,r300,r600,radeonsi,freedreno,svga,swrast,vc4,virgl"
"i915,ilo,nouveau,r300,r600,radeonsi,freedreno,svga,swrast,vc4"
@<:@default=r300,r600,svga,swrast@:>@])],
[with_gallium_drivers="$withval"],
[with_gallium_drivers="$GALLIUM_DRIVERS_DEFAULT"])
@@ -1022,13 +910,6 @@ fi
AM_CONDITIONAL(HAVE_DRI_GLX, test "x$enable_glx" = xyes -a \
"x$enable_dri" = xyes)
# Check for libdrm
PKG_CHECK_MODULES([LIBDRM], [libdrm >= $LIBDRM_REQUIRED],
[have_libdrm=yes], [have_libdrm=no])
if test "x$have_libdrm" = xyes; then
DEFINES="$DEFINES -DHAVE_LIBDRM"
fi
# Select which platform-dependent DRI code gets built
case "$host_os" in
darwin*)
@@ -1039,15 +920,10 @@ gnu*|cygwin*)
dri_platform='drm' ;;
esac
if test "x$enable_dri" = xyes -a "x$dri_platform" = xdrm -a "x$have_libdrm" = xyes; then
have_drisw_kms='yes'
fi
AM_CONDITIONAL(HAVE_DRICOMMON, test "x$enable_dri" = xyes )
AM_CONDITIONAL(HAVE_DRISW, test "x$enable_dri" = xyes )
AM_CONDITIONAL(HAVE_DRISW_KMS, test "x$have_drisw_kms" = xyes )
AM_CONDITIONAL(HAVE_DRI2, test "x$enable_dri" = xyes -a "x$dri_platform" = xdrm -a "x$have_libdrm" = xyes )
AM_CONDITIONAL(HAVE_DRI3, test "x$enable_dri3" = xyes -a "x$dri_platform" = xdrm -a "x$have_libdrm" = xyes )
AM_CONDITIONAL(HAVE_DRI2, test "x$enable_dri" = xyes -a "x$dri_platform" = xdrm )
AM_CONDITIONAL(HAVE_DRI3, test "x$enable_dri3" = xyes -a "x$dri_platform" = xdrm )
AM_CONDITIONAL(HAVE_APPLEDRI, test "x$enable_dri" = xyes -a "x$dri_platform" = xapple )
AC_ARG_ENABLE([shared-glapi],
@@ -1076,8 +952,14 @@ dnl
dnl Driver specific build directories
dnl
if test -n "$with_gallium_drivers" -a "x$enable_glx$enable_xlib_glx" = xyesyes; then
case "x$enable_glx$enable_xlib_glx" in
xyesyes)
NEED_WINSYS_XLIB="yes"
;;
esac
if test "x$enable_dri" = xyes; then
enable_gallium_loader="$enable_shared_pipe_drivers"
fi
if test "x$enable_gallium_osmesa" = xyes; then
@@ -1213,11 +1095,6 @@ AC_MSG_RESULT([$with_sha1])
AC_SUBST(SHA1_LIBS)
AC_SUBST(SHA1_CFLAGS)
# Enable a define for SHA1
if test "x$with_sha1" != "x"; then
DEFINES="$DEFINES -DHAVE_SHA1"
fi
# Allow user to configure out the shader-cache feature
AC_ARG_ENABLE([shader-cache],
AS_HELP_STRING([--disable-shader-cache], [Disable binary shader cache]),
@@ -1234,6 +1111,13 @@ if test "x$with_sha1" = "x"; then
fi
AM_CONDITIONAL([ENABLE_SHADER_CACHE], [test x$enable_shader_cache = xyes])
# Check for libdrm
PKG_CHECK_MODULES([LIBDRM], [libdrm >= $LIBDRM_REQUIRED],
[have_libdrm=yes], [have_libdrm=no])
if test "x$have_libdrm" = xyes; then
DEFINES="$DEFINES -DHAVE_LIBDRM"
fi
case "$host_os" in
linux*)
need_pci_id=yes ;;
@@ -1269,6 +1153,10 @@ AC_ARG_ENABLE([driglx-direct],
[driglx_direct="$enableval"],
[driglx_direct="yes"])
# Check for libcaca
PKG_CHECK_EXISTS([caca], [have_libcaca=yes], [have_libcaca=no])
AM_CONDITIONAL([HAVE_LIBCACA], [test x$have_libcaca = xyes])
dnl
dnl libGL configuration per driver
dnl
@@ -1310,8 +1198,7 @@ xyesno)
if test x"$enable_dri3" = xyes; then
PKG_CHECK_EXISTS([xcb >= $XCB_REQUIRED], [], AC_MSG_ERROR([DRI3 requires xcb >= $XCB_REQUIRED]))
dri3_modules="xcb-dri3 xcb-present xcb-sync xshmfence >= $XSHMFENCE_REQUIRED"
PKG_CHECK_MODULES([XCB_DRI3], [$dri3_modules])
dri_modules="$dri_modules xcb-dri3 xcb-present xcb-sync xshmfence >= $XSHMFENCE_REQUIRED"
fi
fi
if test x"$dri_platform" = xapple ; then
@@ -1398,16 +1285,6 @@ AC_SUBST(GLX_TLS, ${GLX_USE_TLS})
AS_IF([test "x$GLX_USE_TLS" = xyes -a "x$ax_pthread_ok" = xyes],
[DEFINES="${DEFINES} -DGLX_USE_TLS"])
dnl Read-only text section on x86 hardened platforms
AC_ARG_ENABLE([glx-read-only-text],
[AS_HELP_STRING([--enable-glx-read-only-text],
[Disable writable .text section on x86 (decreases performance) @<:@default=disabled@:>@])],
[enable_glx_read_only_text="$enableval"],
[enable_glx_read_only_text=no])
if test "x$enable_glx_read_only_text" = xyes; then
DEFINES="$DEFINES -DGLX_X86_READONLY_TEXT"
fi
dnl
dnl More DRI setup
dnl
@@ -1480,7 +1357,7 @@ if test "x$enable_dri" = xyes; then
fi
;;
darwin*)
DEFINES="$DEFINES -DGLX_ALIAS_UNSUPPORTED -DBUILDING_MESA"
DEFINES="$DEFINES -DGLX_ALIAS_UNSUPPORTED"
if test "x$with_dri_drivers" = "xyes"; then
with_dri_drivers="swrast"
fi
@@ -1501,6 +1378,26 @@ if test "x$enable_dri" = xyes; then
[AC_MSG_ERROR([Expat library required for DRI not found])])
EXPAT_LIBS="-lexpat"])
DRICOMMON_NEED_LIBDRM=no
# If we are building any DRI driver other than swrast.
if test -n "$with_dri_drivers"; then
if test "x$with_dri_drivers" != xswrast; then
# ... libdrm is required
if test "x$have_libdrm" != xyes; then
AC_MSG_ERROR([DRI drivers requires libdrm >= $LIBDRM_REQUIRED])
fi
DRICOMMON_NEED_LIBDRM=yes
fi
fi
# If we're building any gallium DRI driver other than swrast
if test -n "$with_gallium_drivers" -a "x$DRICOMMON_NEED_LIBDRM" = xno; then
if test "x$with_gallium_drivers" != xswrast; then
# ... build a libdrm aware dricommon
DRICOMMON_NEED_LIBDRM=yes
fi
fi
# put all the necessary libs together
DRI_LIB_DEPS="$DRI_LIB_DEPS $SELINUX_LIBS $LIBDRM_LIBS $EXPAT_LIBS -lm $PTHREAD_LIBS $DLOPEN_LIBS"
fi
@@ -1528,7 +1425,7 @@ if test -n "$with_dri_drivers"; then
;;
xnouveau)
HAVE_NOUVEAU_DRI=yes;
PKG_CHECK_MODULES([NVVIEUX], [libdrm_nouveau >= $LIBDRM_NVVIEUX_REQUIRED])
PKG_CHECK_MODULES([NOUVEAU], [libdrm_nouveau >= $LIBDRM_NVVIEUX_REQUIRED])
;;
xradeon)
HAVE_RADEON_DRI=yes;
@@ -1636,6 +1533,8 @@ GBM_PC_LIB_PRIV="$DLOPEN_LIBS"
AC_SUBST([GBM_PC_REQ_PRIV])
AC_SUBST([GBM_PC_LIB_PRIV])
AM_CONDITIONAL(HAVE_VULKAN, true)
dnl
dnl EGL configuration
dnl
@@ -1652,12 +1551,6 @@ if test "x$enable_egl" = xyes; then
if test "x$enable_shared_glapi" = xno; then
AC_MSG_ERROR([egl_dri2 requires --enable-shared-glapi])
fi
if test "x$enable_dri3" = xyes; then
HAVE_EGL_DRIVER_DRI3=1
if test "x$enable_shared_glapi" = xno; then
AC_MSG_ERROR([egl_dri3 requires --enable-shared-glapi])
fi
fi
else
# Avoid building an "empty" libEGL. Drop/update this
# when other backends (haiku?) come along.
@@ -1669,8 +1562,6 @@ fi
AM_CONDITIONAL(HAVE_EGL, test "x$enable_egl" = xyes)
AC_SUBST([EGL_LIB_DEPS])
gallium_st="mesa"
dnl
dnl XA configuration
dnl
@@ -1683,7 +1574,7 @@ if test "x$enable_xa" = xyes; then
enabling XA.
Example: ./configure --enable-xa --with-gallium-drivers=svga...])
fi
gallium_st="$gallium_st xa"
enable_gallium_loader=$enable_shared_pipe_drivers
fi
AM_CONDITIONAL(HAVE_ST_XA, test "x$enable_xa" = xyes)
@@ -1728,25 +1619,25 @@ AM_CONDITIONAL(NEED_GALLIUM_VL_WINSYS, test "x$need_gallium_vl_winsys" = xyes)
if test "x$enable_xvmc" = xyes; then
PKG_CHECK_MODULES([XVMC], [xvmc >= $XVMC_REQUIRED])
gallium_st="$gallium_st xvmc"
enable_gallium_loader=$enable_shared_pipe_drivers
fi
AM_CONDITIONAL(HAVE_ST_XVMC, test "x$enable_xvmc" = xyes)
if test "x$enable_vdpau" = xyes; then
PKG_CHECK_MODULES([VDPAU], [vdpau >= $VDPAU_REQUIRED])
gallium_st="$gallium_st vdpau"
enable_gallium_loader=$enable_shared_pipe_drivers
fi
AM_CONDITIONAL(HAVE_ST_VDPAU, test "x$enable_vdpau" = xyes)
if test "x$enable_omx" = xyes; then
PKG_CHECK_MODULES([OMX], [libomxil-bellagio >= $LIBOMXIL_BELLAGIO_REQUIRED])
gallium_st="$gallium_st omx"
enable_gallium_loader=$enable_shared_pipe_drivers
fi
AM_CONDITIONAL(HAVE_ST_OMX, test "x$enable_omx" = xyes)
if test "x$enable_va" = xyes; then
PKG_CHECK_MODULES([VA], [libva >= $LIBVA_REQUIRED])
gallium_st="$gallium_st va"
enable_gallium_loader=$enable_shared_pipe_drivers
fi
AM_CONDITIONAL(HAVE_ST_VA, test "x$enable_va" = xyes)
@@ -1760,15 +1651,11 @@ if test "x$enable_nine" = xyes; then
if test "x$with_gallium_drivers" = xswrast; then
AC_MSG_ERROR([nine requires at least one non-swrast gallium driver])
fi
if test $GCC_VERSION_MAJOR -lt 4 -o $GCC_VERSION_MAJOR -eq 4 -a $GCC_VERSION_MINOR -lt 6; then
AC_MSG_ERROR([gcc >= 4.6 is required to build nine])
fi
if test "x$enable_dri3" = xno; then
AC_MSG_WARN([using nine together with wine requires DRI3 enabled system])
fi
gallium_st="$gallium_st nine"
enable_gallium_loader=$enable_shared_pipe_drivers
fi
AM_CONDITIONAL(HAVE_ST_NINE, test "x$enable_nine" = xyes)
@@ -1783,15 +1670,7 @@ AC_ARG_WITH([clang-libdir],
[CLANG_LIBDIR=''])
PKG_CHECK_EXISTS([libclc], [have_libclc=yes], [have_libclc=no])
PKG_CHECK_MODULES([LIBELF], [libelf], [have_libelf=yes], [have_libelf=no])
if test "x$have_libelf" = xno; then
LIBELF_LIBS=''
LIBELF_CFLAGS=''
AC_CHECK_LIB([elf], [elf_memory], [have_libelf=yes;LIBELF_LIBS=-lelf], [have_libelf=no])
AC_SUBST([LIBELF_LIBS])
AC_SUBST([LIBELF_CFLAGS])
fi
AC_CHECK_LIB([elf], [elf_memory], [have_libelf=yes;ELF_LIB=-lelf])
if test "x$enable_opencl" = xyes; then
if test -z "$with_gallium_drivers"; then
@@ -1814,7 +1693,8 @@ if test "x$enable_opencl" = xyes; then
AC_SUBST([LIBCLC_LIBEXECDIR])
fi
gallium_st="$gallium_st clover"
# XXX: Use $enable_shared_pipe_drivers once converted to use static/shared pipe-drivers
enable_gallium_loader=yes
if test "x$enable_opencl_icd" = xyes; then
OPENCL_LIBNAME="MesaOpenCL"
@@ -1885,9 +1765,6 @@ egl_platforms=`IFS=', '; echo $with_egl_platforms`
for plat in $egl_platforms; do
case "$plat" in
wayland)
test "x$have_libdrm" != xyes &&
AC_MSG_ERROR([EGL platform wayland requires libdrm >= $LIBDRM_REQUIRED])
PKG_CHECK_MODULES([WAYLAND], [wayland-client >= $WAYLAND_REQUIRED wayland-server >= $WAYLAND_REQUIRED])
if test "x$WAYLAND_SCANNER" = x; then
@@ -1911,6 +1788,9 @@ for plat in $egl_platforms; do
AC_MSG_ERROR([EGL platform surfaceless requires libdrm >= $LIBDRM_REQUIRED])
;;
android|gdi|null)
;;
*)
AC_MSG_ERROR([EGL platform '$plat' does not exist])
;;
@@ -1931,6 +1811,9 @@ else
EGL_NATIVE_PLATFORM="_EGL_INVALID_PLATFORM"
fi
if echo "$egl_platforms" | grep -q 'x11'; then
NEED_WINSYS_XLIB=yes
fi
AM_CONDITIONAL(HAVE_EGL_PLATFORM_X11, echo "$egl_platforms" | grep -q 'x11')
AM_CONDITIONAL(HAVE_EGL_PLATFORM_WAYLAND, echo "$egl_platforms" | grep -q 'wayland')
AM_CONDITIONAL(HAVE_EGL_PLATFORM_DRM, echo "$egl_platforms" | grep -q 'drm')
@@ -2094,6 +1977,10 @@ AC_SUBST([XVMC_LIB_INSTALL_DIR])
dnl
dnl Gallium Tests
dnl
if test "x$enable_gallium_tests" = xyes; then
# XXX: Use $enable_shared_pipe_drivers once converted to use static/shared pipe-drivers
enable_gallium_loader=yes
fi
AM_CONDITIONAL(HAVE_GALLIUM_TESTS, test "x$enable_gallium_tests" = xyes)
dnl Directory for VDPAU libs
@@ -2148,17 +2035,18 @@ gallium_require_llvm() {
}
gallium_require_drm_loader() {
if test "x$need_pci_id$have_pci_id" = xyesno; then
AC_MSG_ERROR([Gallium drm loader requires libudev >= $LIBUDEV_REQUIRED or sysfs])
if test "x$enable_gallium_loader" = xyes; then
if test "x$need_pci_id$have_pci_id" = xyesno; then
AC_MSG_ERROR([Gallium drm loader requires libudev >= $LIBUDEV_REQUIRED or sysfs])
fi
enable_gallium_drm_loader=yes
fi
if test "x$enable_va" = xyes && test "x$7" != x; then
GALLIUM_TARGET_DIRS="$GALLIUM_TARGET_DIRS $7"
fi
}
dnl This is for Glamor. Skip this if OpenGL is disabled.
require_egl_drm() {
if test "x$enable_opengl" = xno; then
return 0
fi
case "$with_egl_platforms" in
*drm*)
;;
@@ -2180,7 +2068,7 @@ radeon_llvm_check() {
if test "x$enable_gallium_llvm" != "xyes"; then
AC_MSG_ERROR([--enable-gallium-llvm is required when building $1])
fi
llvm_check_version_for "3" "6" "0" $1
llvm_check_version_for "3" "4" "2" $1
if test true && $LLVM_CONFIG --targets-built | grep -iqvw $amdgpu_llvm_target_name ; then
AC_MSG_ERROR([LLVM $amdgpu_llvm_target_name not enabled in your LLVM build.])
fi
@@ -2239,7 +2127,6 @@ if test -n "$with_gallium_drivers"; then
xradeonsi)
HAVE_GALLIUM_RADEONSI=yes
PKG_CHECK_MODULES([RADEON], [libdrm_radeon >= $LIBDRM_RADEON_REQUIRED])
PKG_CHECK_MODULES([AMDGPU], [libdrm_amdgpu >= $LIBDRM_AMDGPU_REQUIRED])
gallium_require_drm "radeonsi"
gallium_require_drm_loader
radeon_llvm_check "radeonsi"
@@ -2268,16 +2155,11 @@ if test -n "$with_gallium_drivers"; then
gallium_require_drm "vc4"
gallium_require_drm_loader
PKG_CHECK_MODULES([SIMPENROSE], [simpenrose],
[USE_VC4_SIMULATOR=yes;
DEFINES="$DEFINES -DUSE_VC4_SIMULATOR"],
[USE_VC4_SIMULATOR=no])
;;
xvirgl)
HAVE_GALLIUM_VIRGL=yes
gallium_require_drm "virgl"
gallium_require_drm_loader
require_egl_drm "virgl"
case "$host_cpu" in
i?86 | x86_64 | amd64)
USE_VC4_SIMULATOR=yes
;;
esac
;;
*)
AC_MSG_ERROR([Unknown Gallium driver: $driver])
@@ -2295,19 +2177,12 @@ dnl in LLVM_LIBS.
if test "x$MESA_LLVM" != x0; then
if ! $LLVM_CONFIG --libs ${LLVM_COMPONENTS} >/dev/null; then
AC_MSG_ERROR([Calling ${LLVM_CONFIG} failed])
fi
LLVM_LIBS="`$LLVM_CONFIG --libs ${LLVM_COMPONENTS}`"
dnl llvm-config may not give the right answer when llvm is a built as a
dnl single shared library, so we must work the library name out for
dnl ourselves.
dnl (See https://llvm.org/bugs/show_bug.cgi?id=6823)
if test "x$enable_llvm_shared_libs" = xyes; then
dnl We can't use $LLVM_VERSION because it has 'svn' stripped out,
LLVM_SO_NAME=LLVM-`$LLVM_CONFIG --version`
AS_IF([test -f "$LLVM_LIBDIR/lib$LLVM_SO_NAME.$IMP_LIB_EXT"], [llvm_have_one_so=yes])
AS_IF([test -f "$LLVM_LIBDIR/lib$LLVM_SO_NAME.so"], [llvm_have_one_so=yes])
if test "x$llvm_have_one_so" = xyes; then
dnl LLVM was built using auto*, so there is only one shared object.
@@ -2315,7 +2190,7 @@ if test "x$MESA_LLVM" != x0; then
else
dnl If LLVM was built with CMake, there will be one shared object per
dnl component.
AS_IF([test ! -f "$LLVM_LIBDIR/libLLVMTarget.$IMP_LIB_EXT"],
AS_IF([test ! -f "$LLVM_LIBDIR/libLLVMTarget.so"],
[AC_MSG_ERROR([Could not find llvm shared libraries:
Please make sure you have built llvm with the --enable-shared option
and that your llvm libraries are installed in $LLVM_LIBDIR
@@ -2353,18 +2228,41 @@ AM_CONDITIONAL(HAVE_GALLIUM_FREEDRENO, test "x$HAVE_GALLIUM_FREEDRENO" = xyes)
AM_CONDITIONAL(HAVE_GALLIUM_SOFTPIPE, test "x$HAVE_GALLIUM_SOFTPIPE" = xyes)
AM_CONDITIONAL(HAVE_GALLIUM_LLVMPIPE, test "x$HAVE_GALLIUM_LLVMPIPE" = xyes)
AM_CONDITIONAL(HAVE_GALLIUM_VC4, test "x$HAVE_GALLIUM_VC4" = xyes)
AM_CONDITIONAL(HAVE_GALLIUM_VIRGL, test "x$HAVE_GALLIUM_VIRGL" = xyes)
AM_CONDITIONAL(HAVE_GALLIUM_STATIC_TARGETS, test "x$enable_shared_pipe_drivers" = xno)
if test "x$enable_dri" = xyes; then
GALLIUM_PIPE_LOADER_DEFINES="$GALLIUM_PIPE_LOADER_DEFINES -DHAVE_PIPE_LOADER_DRI"
fi
# NOTE: anything using xcb or other client side libs ends up in separate
# _CLIENT variables. The pipe loader is built in two variants,
# one that is standalone and does not link any x client libs (for
# use by XA tracker in particular, but could be used in any case
# where communication with xserver is not desired).
if test "x$enable_gallium_loader" = xyes; then
if test "x$NEED_WINSYS_XLIB" = xyes; then
GALLIUM_PIPE_LOADER_DEFINES="$GALLIUM_PIPE_LOADER_DEFINES -DHAVE_PIPE_LOADER_XLIB"
fi
if test "x$have_drisw_kms" = xyes; then
GALLIUM_PIPE_LOADER_DEFINES="$GALLIUM_PIPE_LOADER_DEFINES -DHAVE_PIPE_LOADER_KMS"
if test "x$enable_dri" = xyes; then
GALLIUM_PIPE_LOADER_DEFINES="$GALLIUM_PIPE_LOADER_DEFINES -DHAVE_PIPE_LOADER_DRI"
fi
if test "x$enable_gallium_drm_loader" = xyes; then
GALLIUM_PIPE_LOADER_DEFINES="$GALLIUM_PIPE_LOADER_DEFINES -DHAVE_PIPE_LOADER_DRM"
PKG_CHECK_MODULES([GALLIUM_PIPE_LOADER_XCB], [xcb xcb-dri2],
pipe_loader_have_xcb=yes, pipe_loader_have_xcb=no)
if test "x$pipe_loader_have_xcb" = xyes; then
GALLIUM_PIPE_LOADER_CLIENT_DEFINES="$GALLIUM_PIPE_LOADER_CLIENT_DEFINES -DHAVE_PIPE_LOADER_XCB"
GALLIUM_PIPE_LOADER_CLIENT_LIBS="$GALLIUM_PIPE_LOADER_CLIENT_LIBS $GALLIUM_PIPE_LOADER_XCB_LIBS $LIBDRM_LIBS"
fi
fi
GALLIUM_PIPE_LOADER_CLIENT_DEFINES="$GALLIUM_PIPE_LOADER_CLIENT_DEFINES $GALLIUM_PIPE_LOADER_DEFINES"
GALLIUM_PIPE_LOADER_CLIENT_LIBS="$GALLIUM_PIPE_LOADER_CLIENT_LIBS $GALLIUM_PIPE_LOADER_LIBS"
AC_SUBST([GALLIUM_PIPE_LOADER_DEFINES])
AC_SUBST([GALLIUM_PIPE_LOADER_LIBS])
AC_SUBST([GALLIUM_PIPE_LOADER_CLIENT_DEFINES])
AC_SUBST([GALLIUM_PIPE_LOADER_CLIENT_LIBS])
fi
AC_SUBST([GALLIUM_PIPE_LOADER_DEFINES])
AM_CONDITIONAL(HAVE_I915_DRI, test x$HAVE_I915_DRI = xyes)
AM_CONDITIONAL(HAVE_I965_DRI, test x$HAVE_I965_DRI = xyes)
@@ -2379,6 +2277,8 @@ AM_CONDITIONAL(NEED_RADEON_DRM_WINSYS, test "x$HAVE_GALLIUM_R300" = xyes -o \
AM_CONDITIONAL(NEED_WINSYS_XLIB, test "x$NEED_WINSYS_XLIB" = xyes)
AM_CONDITIONAL(NEED_RADEON_LLVM, test x$NEED_RADEON_LLVM = xyes)
AM_CONDITIONAL(USE_R600_LLVM_COMPILER, test x$USE_R600_LLVM_COMPILER = xyes)
AM_CONDITIONAL(HAVE_LOADER_GALLIUM, test x$enable_gallium_loader = xyes)
AM_CONDITIONAL(HAVE_DRM_LOADER_GALLIUM, test x$enable_gallium_drm_loader = xyes)
AM_CONDITIONAL(HAVE_GALLIUM_COMPUTE, test x$enable_opencl = xyes)
AM_CONDITIONAL(HAVE_MESA_LLVM, test x$MESA_LLVM = x1)
AM_CONDITIONAL(USE_VC4_SIMULATOR, test x$USE_VC4_SIMULATOR = xyes)
@@ -2386,6 +2286,9 @@ if test "x$USE_VC4_SIMULATOR" = xyes -a "x$HAVE_GALLIUM_ILO" = xyes; then
AC_MSG_ERROR([VC4 simulator on x86 replaces i965 driver build, so ilo must be disabled.])
fi
AC_SUBST([ELF_LIB])
AM_CONDITIONAL(DRICOMMON_NEED_LIBDRM, test "x$DRICOMMON_NEED_LIBDRM" = xyes)
AM_CONDITIONAL(HAVE_LIBDRM, test "x$have_libdrm" = xyes)
AM_CONDITIONAL(HAVE_X11_DRIVER, test "x$enable_xlib_glx" = xyes)
AM_CONDITIONAL(HAVE_OSMESA, test "x$enable_osmesa" = xyes)
@@ -2421,6 +2324,13 @@ AC_SUBST([XA_MINOR], $XA_MINOR)
AC_SUBST([XA_TINY], $XA_TINY)
AC_SUBST([XA_VERSION], "$XA_MAJOR.$XA_MINOR.$XA_TINY")
PKG_CHECK_MODULES(VALGRIND, [valgrind],
[have_valgrind=yes], [have_valgrind=no])
if test "x$have_valgrind" = "xyes"; then
AC_DEFINE([HAVE_VALGRIND], 1,
[Use valgrind intrinsics to suppress false warnings])
fi
dnl Restore LDFLAGS and CPPFLAGS
LDFLAGS="$_SAVE_LDFLAGS"
CPPFLAGS="$_SAVE_CPPFLAGS"
@@ -2438,8 +2348,8 @@ CXXFLAGS="$CXXFLAGS $USER_CXXFLAGS"
dnl Substitute the config
AC_CONFIG_FILES([Makefile
src/Makefile
src/compiler/Makefile
src/egl/Makefile
src/egl/drivers/dri2/Makefile
src/egl/main/Makefile
src/egl/main/egl.pc
src/egl/wayland/wayland-drm/Makefile
src/egl/wayland/wayland-egl/Makefile
@@ -2448,7 +2358,6 @@ AC_CONFIG_FILES([Makefile
src/gallium/auxiliary/Makefile
src/gallium/auxiliary/pipe-loader/Makefile
src/gallium/drivers/freedreno/Makefile
src/gallium/drivers/ddebug/Makefile
src/gallium/drivers/i915/Makefile
src/gallium/drivers/ilo/Makefile
src/gallium/drivers/llvmpipe/Makefile
@@ -2463,7 +2372,6 @@ AC_CONFIG_FILES([Makefile
src/gallium/drivers/svga/Makefile
src/gallium/drivers/trace/Makefile
src/gallium/drivers/vc4/Makefile
src/gallium/drivers/virgl/Makefile
src/gallium/state_trackers/clover/Makefile
src/gallium/state_trackers/dri/Makefile
src/gallium/state_trackers/glx/xlib/Makefile
@@ -2480,7 +2388,6 @@ AC_CONFIG_FILES([Makefile
src/gallium/targets/libgl-xlib/Makefile
src/gallium/targets/omx/Makefile
src/gallium/targets/opencl/Makefile
src/gallium/targets/opencl/mesa.icd
src/gallium/targets/osmesa/Makefile
src/gallium/targets/osmesa/osmesa.pc
src/gallium/targets/pipe-loader/Makefile
@@ -2496,7 +2403,6 @@ AC_CONFIG_FILES([Makefile
src/gallium/winsys/intel/drm/Makefile
src/gallium/winsys/nouveau/drm/Makefile
src/gallium/winsys/radeon/drm/Makefile
src/gallium/winsys/amdgpu/drm/Makefile
src/gallium/winsys/svga/drm/Makefile
src/gallium/winsys/sw/dri/Makefile
src/gallium/winsys/sw/kms-dri/Makefile
@@ -2504,10 +2410,9 @@ AC_CONFIG_FILES([Makefile
src/gallium/winsys/sw/wrapper/Makefile
src/gallium/winsys/sw/xlib/Makefile
src/gallium/winsys/vc4/drm/Makefile
src/gallium/winsys/virgl/drm/Makefile
src/gallium/winsys/virgl/vtest/Makefile
src/gbm/Makefile
src/gbm/main/gbm.pc
src/glsl/Makefile
src/glx/Makefile
src/glx/apple/Makefile
src/glx/tests/Makefile
@@ -2533,6 +2438,7 @@ AC_CONFIG_FILES([Makefile
src/mesa/drivers/osmesa/osmesa.pc
src/mesa/drivers/x11/Makefile
src/mesa/main/tests/Makefile
src/vulkan/Makefile
src/util/Makefile
src/util/tests/hash_table/Makefile])
@@ -2598,9 +2504,6 @@ if test "$enable_egl" = yes; then
if test "x$HAVE_EGL_DRIVER_DRI2" != "x"; then
egl_drivers="$egl_drivers builtin:egl_dri2"
fi
if test "x$HAVE_EGL_DRIVER_DRI3" != "x"; then
egl_drivers="$egl_drivers builtin:egl_dri3"
fi
echo " EGL drivers: $egl_drivers"
fi
@@ -2616,8 +2519,7 @@ fi
echo ""
if test -n "$with_gallium_drivers"; then
echo " Gallium drivers: $gallium_drivers"
echo " Gallium st: $gallium_st"
echo " Gallium: yes"
else
echo " Gallium: no"
fi

View File

@@ -92,54 +92,54 @@ GL 3.3, GLSL 3.30 --- all DONE: i965, nv50, nvc0, r600, radeonsi, llvmpipe, soft
GL_ARB_vertex_type_2_10_10_10_rev DONE ()
GL 4.0, GLSL 4.00 --- all DONE: nvc0, r600, radeonsi
GL 4.0, GLSL 4.00:
GL_ARB_draw_buffers_blend DONE (i965, nv50, llvmpipe, softpipe)
GL_ARB_draw_indirect DONE (i965, llvmpipe, softpipe)
GL_ARB_gpu_shader5 DONE (i965)
GL_ARB_draw_buffers_blend DONE (i965, nv50, nvc0, r600, radeonsi, llvmpipe, softpipe)
GL_ARB_draw_indirect DONE (i965, nvc0, r600, radeonsi, llvmpipe, softpipe)
GL_ARB_gpu_shader5 DONE (i965, nvc0)
- 'precise' qualifier DONE
- Dynamically uniform sampler array indices DONE (softpipe)
- Dynamically uniform UBO array indices DONE ()
- Dynamically uniform sampler array indices DONE (r600, softpipe)
- Dynamically uniform UBO array indices DONE (r600)
- Implicit signed -> unsigned conversions DONE
- Fused multiply-add DONE ()
- Packing/bitfield/conversion functions DONE (softpipe)
- Enhanced textureGather DONE (softpipe)
- Geometry shader instancing DONE (llvmpipe, softpipe)
- Packing/bitfield/conversion functions DONE (r600, radeonsi, softpipe)
- Enhanced textureGather DONE (r600, radeonsi, softpipe)
- Geometry shader instancing DONE (r600, llvmpipe, softpipe)
- Geometry shader multiple streams DONE ()
- Enhanced per-sample shading DONE ()
- Interpolation functions DONE ()
- Enhanced per-sample shading DONE (r600, radeonsi)
- Interpolation functions DONE (r600)
- New overload resolution rules DONE
GL_ARB_gpu_shader_fp64 DONE (llvmpipe, softpipe)
GL_ARB_sample_shading DONE (i965, nv50)
GL_ARB_shader_subroutine DONE (i965, nv50, llvmpipe, softpipe)
GL_ARB_tessellation_shader DONE (i965)
GL_ARB_texture_buffer_object_rgb32 DONE (i965, llvmpipe, softpipe)
GL_ARB_texture_cube_map_array DONE (i965, nv50, llvmpipe, softpipe)
GL_ARB_texture_gather DONE (i965, nv50, llvmpipe, softpipe)
GL_ARB_texture_query_lod DONE (i965, nv50, softpipe)
GL_ARB_transform_feedback2 DONE (i965, nv50, llvmpipe, softpipe)
GL_ARB_transform_feedback3 DONE (i965, nv50, llvmpipe, softpipe)
GL_ARB_gpu_shader_fp64 DONE (nvc0, softpipe)
GL_ARB_sample_shading DONE (i965, nv50, nvc0, r600, radeonsi)
GL_ARB_shader_subroutine started (Dave)
GL_ARB_tessellation_shader started (Chris, Ilia)
GL_ARB_texture_buffer_object_rgb32 DONE (i965, nvc0, r600, radeonsi, llvmpipe, softpipe)
GL_ARB_texture_cube_map_array DONE (i965, nv50, nvc0, r600, radeonsi, llvmpipe, softpipe)
GL_ARB_texture_gather DONE (i965, nv50, nvc0, r600, radeonsi, llvmpipe, softpipe)
GL_ARB_texture_query_lod DONE (i965, nv50, nvc0, r600, radeonsi)
GL_ARB_transform_feedback2 DONE (i965, nv50, nvc0, r600, radeonsi, llvmpipe, softpipe)
GL_ARB_transform_feedback3 DONE (i965, nv50, nvc0, r600, radeonsi, llvmpipe, softpipe)
GL 4.1, GLSL 4.10 --- all DONE: nvc0, r600, radeonsi
GL 4.1, GLSL 4.10:
GL_ARB_ES2_compatibility DONE (i965, nv50, llvmpipe, softpipe)
GL_ARB_ES2_compatibility DONE (i965, nv50, nvc0, r600, radeonsi, llvmpipe, softpipe)
GL_ARB_get_program_binary DONE (0 binary formats)
GL_ARB_separate_shader_objects DONE (all drivers)
GL_ARB_shader_precision DONE (all drivers that support GLSL 4.10)
GL_ARB_vertex_attrib_64bit DONE (llvmpipe, softpipe)
GL_ARB_viewport_array DONE (i965, nv50, llvmpipe, softpipe)
GL_ARB_shader_precision started (Micah)
GL_ARB_vertex_attrib_64bit DONE (nvc0, softpipe)
GL_ARB_viewport_array DONE (i965, nv50, nvc0, r600, llvmpipe)
GL 4.2, GLSL 4.20:
GL_ARB_texture_compression_bptc DONE (i965, nvc0, r600, radeonsi)
GL_ARB_compressed_texture_pixel_storage DONE (all drivers)
GL_ARB_shader_atomic_counters DONE (i965, nvc0)
GL_ARB_shader_atomic_counters DONE (i965)
GL_ARB_texture_storage DONE (all drivers)
GL_ARB_transform_feedback_instanced DONE (i965, nv50, nvc0, r600, radeonsi, llvmpipe, softpipe)
GL_ARB_base_instance DONE (i965, nv50, nvc0, r600, radeonsi, llvmpipe, softpipe)
GL_ARB_shader_image_load_store DONE (i965)
GL_ARB_shader_image_load_store in progress (curro)
GL_ARB_conservative_depth DONE (all drivers that support GLSL 1.30)
GL_ARB_shading_language_420pack DONE (all drivers that support GLSL 1.30)
GL_ARB_shading_language_packing DONE (all drivers)
@@ -149,27 +149,27 @@ GL 4.2, GLSL 4.20:
GL 4.3, GLSL 4.30:
GL_ARB_arrays_of_arrays DONE (all drivers that support GLSL 1.30)
GL_ARB_arrays_of_arrays started (Timothy)
GL_ARB_ES3_compatibility DONE (all drivers that support GLSL 3.30)
GL_ARB_clear_buffer_object DONE (all drivers)
GL_ARB_compute_shader DONE (i965)
GL_ARB_copy_image DONE (i965, nv50, nvc0, r600, radeonsi)
GL_ARB_compute_shader in progress (jljusten)
GL_ARB_copy_image DONE (i965) (gallium - in progress, VMware)
GL_KHR_debug DONE (all drivers)
GL_ARB_explicit_uniform_location DONE (all drivers that support GLSL)
GL_ARB_fragment_layer_viewport DONE (i965, nv50, nvc0, r600, radeonsi, llvmpipe)
GL_ARB_fragment_layer_viewport DONE (nv50, nvc0, r600, llvmpipe)
GL_ARB_framebuffer_no_attachments DONE (i965)
GL_ARB_internalformat_query2 in progress (elima)
GL_ARB_internalformat_query2 not started
GL_ARB_invalidate_subdata DONE (all drivers)
GL_ARB_multi_draw_indirect DONE (i965, nvc0, r600, radeonsi, llvmpipe, softpipe)
GL_ARB_program_interface_query DONE (all drivers)
GL_ARB_robust_buffer_access_behavior not started
GL_ARB_shader_image_size DONE (i965)
GL_ARB_shader_storage_buffer_object DONE (i965, nvc0)
GL_ARB_shader_image_size in progress (Martin Peres)
GL_ARB_shader_storage_buffer_object in progress (Iago Toral, Samuel Iglesias)
GL_ARB_stencil_texturing DONE (i965/gen8+, nv50, nvc0, r600, radeonsi, llvmpipe, softpipe)
GL_ARB_texture_buffer_range DONE (nv50, nvc0, i965, r600, radeonsi, llvmpipe)
GL_ARB_texture_query_levels DONE (all drivers that support GLSL 1.30)
GL_ARB_texture_storage_multisample DONE (all drivers that support GL_ARB_texture_multisample)
GL_ARB_texture_view DONE (i965, nv50, nvc0, r600, radeonsi, llvmpipe, softpipe)
GL_ARB_texture_view DONE (i965, nv50, nvc0, llvmpipe, softpipe)
GL_ARB_vertex_attrib_binding DONE (all drivers)
@@ -177,16 +177,10 @@ GL 4.4, GLSL 4.40:
GL_MAX_VERTEX_ATTRIB_STRIDE DONE (all drivers)
GL_ARB_buffer_storage DONE (i965, nv50, nvc0, r600, radeonsi)
GL_ARB_clear_texture DONE (i965, nv50, nvc0)
GL_ARB_enhanced_layouts in progress (Timothy)
- compile-time constant expressions DONE
- explicit byte offsets for blocks in progress
- forced alignment within blocks in progress
- specified vec4-slot component numbers in progress
- specified transform/feedback layout in progress
- input/output block locations DONE
GL_ARB_clear_texture DONE (i965) (gallium - in progress, VMware)
GL_ARB_enhanced_layouts not started
GL_ARB_multi_bind DONE (all drivers)
GL_ARB_query_buffer_object DONE (nvc0)
GL_ARB_query_buffer_object not started
GL_ARB_texture_mirror_clamp_to_edge DONE (i965, nv50, nvc0, r600, radeonsi, llvmpipe, softpipe)
GL_ARB_texture_stencil8 DONE (nv50, nvc0, r600, radeonsi, llvmpipe, softpipe)
GL_ARB_vertex_type_10f_11f_11f_rev DONE (i965, nv50, nvc0, r600, radeonsi, llvmpipe, softpipe)
@@ -195,30 +189,39 @@ GL 4.5, GLSL 4.50:
GL_ARB_ES3_1_compatibility not started
GL_ARB_clip_control DONE (i965, nv50, nvc0, r600, radeonsi, llvmpipe, softpipe)
GL_ARB_conditional_render_inverted DONE (i965, nv50, nvc0, r600, radeonsi, llvmpipe, softpipe)
GL_ARB_conditional_render_inverted DONE (i965, nv50, nvc0, llvmpipe, softpipe)
GL_ARB_cull_distance in progress (Tobias)
GL_ARB_derivative_control DONE (i965, nv50, nvc0, r600, radeonsi)
GL_ARB_derivative_control DONE (i965, nv50, nvc0, r600)
GL_ARB_direct_state_access DONE (all drivers)
GL_ARB_get_texture_sub_image DONE (all drivers)
GL_ARB_shader_texture_image_samples DONE (i965, nv50, nvc0, r600, radeonsi)
GL_ARB_texture_barrier DONE (i965, nv50, nvc0, r600, radeonsi)
GL_KHR_context_flush_control DONE (all - but needs GLX/EGL extension to be useful)
- Transform Feedback object DONE
- Buffer object DONE
- Framebuffer object DONE
- Renderbuffer object DONE
- Texture object DONE
- Vertex array object DONE
- Sampler object DONE
- Program Pipeline object DONE
- Query object DONE (will require changes when GL_ARB_query_buffer_object lands)
GL_ARB_get_texture_sub_image started (Brian Paul)
GL_ARB_shader_texture_image_samples not started
GL_ARB_texture_barrier DONE (nv50, nvc0, r600, radeonsi)
GL_KHR_context_flush_control DONE (all - but needs GLX/EXT extension to be useful)
GL_KHR_robust_buffer_access_behavior not started
GL_KHR_robustness 90% done (the ARB variant)
GL_EXT_shader_integer_mix DONE (all drivers that support GLSL)
These are the extensions cherry-picked to make GLES 3.1
GLES3.1, GLSL ES 3.1
GL_ARB_arrays_of_arrays DONE (all drivers that support GLSL 1.30)
GL_ARB_compute_shader DONE (i965)
GL_ARB_arrays_of_arrays started (Timothy)
GL_ARB_compute_shader in progress (jljusten)
GL_ARB_draw_indirect DONE (i965, nvc0, r600, radeonsi, llvmpipe, softpipe)
GL_ARB_explicit_uniform_location DONE (all drivers that support GLSL)
GL_ARB_framebuffer_no_attachments DONE (i965)
GL_ARB_program_interface_query DONE (all drivers)
GL_ARB_shader_atomic_counters DONE (i965, nvc0)
GL_ARB_shader_image_load_store DONE (i965)
GL_ARB_shader_image_size DONE (i965)
GL_ARB_shader_storage_buffer_object DONE (i965, nvc0)
GL_ARB_shader_atomic_counters DONE (i965)
GL_ARB_shader_image_load_store in progress (curro)
GL_ARB_shader_image_size in progress (Martin Peres)
GL_ARB_shader_storage_buffer_object in progress (Iago Toral, Samuel Iglesias)
GL_ARB_shading_language_packing DONE (all drivers)
GL_ARB_separate_shader_objects DONE (all drivers)
GL_ARB_stencil_texturing DONE (i965/gen8+, nv50, nvc0, r600, radeonsi, llvmpipe, softpipe)
@@ -229,35 +232,10 @@ GLES3.1, GLSL ES 3.1
GS5 Packing/bitfield/conversion functions DONE (i965, nvc0, r600, radeonsi)
GL_EXT_shader_integer_mix DONE (all drivers that support GLSL)
Additional functionality not covered above:
glMemoryBarrierByRegion DONE
glGetTexLevelParameter[fi]v - needs updates DONE
glGetBooleani_v - restrict to GLES enums
gl_HelperInvocation support DONE (i965, nvc0, r600)
GLES3.2, GLSL ES 3.2
GL_EXT_color_buffer_float DONE (all drivers)
GL_KHR_blend_equation_advanced not started
GL_KHR_debug DONE (all drivers)
GL_KHR_robustness 90% done (the ARB variant)
GL_KHR_texture_compression_astc_ldr DONE (i965/gen9+)
GL_OES_copy_image not started (based on GL_ARB_copy_image, which is done for some drivers)
GL_OES_draw_buffers_indexed not started
GL_OES_draw_elements_base_vertex DONE (all drivers)
GL_OES_geometry_shader started (Marta)
GL_OES_gpu_shader5 not started (based on parts of GL_ARB_gpu_shader5, which is done for some drivers)
GL_OES_primitive_bounding box not started
GL_OES_sample_shading not started (based on parts of GL_ARB_sample_shading, which is done for some drivers)
GL_OES_sample_variables not started (based on parts of GL_ARB_sample_shading, which is done for some drivers)
GL_OES_shader_image_atomic not started (based on parts of GL_ARB_shader_image_load_store, which is done for some drivers)
GL_OES_shader_io_blocks not started (based on parts of GLSL 1.50, which is done)
GL_OES_shader_multisample_interpolation not started (based on parts of GL_ARB_gpu_shader5, which is done)
GL_OES_tessellation_shader not started (based on GL_ARB_tessellation_shader, which is done for some drivers)
GL_OES_texture_border_clamp not started (based on GL_ARB_texture_border_clamp, which is done)
GL_OES_texture_buffer not started (based on GL_ARB_texture_buffer_object, GL_ARB_texture_buffer_range, and GL_ARB_texture_buffer_object_rgb32 that are all done)
GL_OES_texture_cube_map_array not started (based on GL_ARB_texture_cube_map_array, which is done for all drivers)
GL_OES_texture_stencil8 DONE (all drivers that support GL_ARB_texture_stencil8)
GL_OES_texture_storage_multisample_2d_array DONE (all drivers that support GL_ARB_texture_multisample)
Additional functions not covered above:
glMemoryBarrierByRegion
glGetTexLevelParameter[fi]v - needs updates to restrict to GLES enums
glGetBooleani_v - needs updates to restrict to GLES enums
More info about these features and the work involved can be found at
http://dri.freedesktop.org/wiki/MissingFunctionality

View File

@@ -2,8 +2,8 @@ The software may implement third party technologies (e.g. third party
libraries) that are not licensed to you by AMD and for which you may need
to obtain licenses from other parties. Unless explicitly stated otherwise,
these third party technologies are not licensed hereunder. Such third
party technologies include, but are not limited, to H.264, H.265, HEVC, MPEG-2,
MPEG-4, AVC, and VC-1.
party technologies include, but are not limited, to H.264, MPEG-2, MPEG-4,
AVC, and VC-1.
For MPEG-2 Encoding Products ANY USE OF THIS PRODUCT IN ANY MANNER OTHER
THAN PERSONAL USE THAT COMPLIES WITH THE MPEG-2 STANDARD FOR ENCODING VIDEO

View File

@@ -87,13 +87,6 @@ created in a <code>lib64</code> directory at the top of the Mesa source
tree.</p>
</dd>
<dt><code>--sysconfdir=DIR</code></dt>
<dd><p>This option specifies the directory where the configuration
files will be installed. The default is <code>${prefix}/etc</code>.
Currently there's only one config file provided when dri drivers are
enabled - it's <code>drirc</code>.</p>
</dd>
<dt><code>--enable-static, --disable-shared</code></dt>
<dd><p>By default, Mesa
will build shared libraries. Either of these options will force static
@@ -224,7 +217,7 @@ GLX.
<dt><code>--with-expat=DIR</code>
<dd><p><strong>DEPRECATED</strong>, use <code>PKG_CONFIG_PATH</code> instead.</p>
<p>The DRI-enabled libGL uses expat to
parse the DRI configuration files in <code>${sysconfdir}/drirc</code> and
parse the DRI configuration files in <code>/etc/drirc</code> and
<code>~/.drirc</code>. This option allows a specific expat installation
to be used. For example, <code>--with-expat=/usr/local</code> will
search for expat headers and libraries in <code>/usr/local/include</code>

View File

@@ -90,14 +90,14 @@
<li><a href="http://www.opengl.org" target="_parent">OpenGL website</a>
<li><a href="http://dri.freedesktop.org" target="_parent">DRI website</a>
<li><a href="http://www.freedesktop.org" target="_parent">freedesktop.org</a>
<li><a href="http://planet.freedesktop.org" target="_parent">Developer blogs</a>
</ul>
<b>Hosted by:</b>
<br>
<blockquote>
<a href="http://sourceforge.net"
target="_parent">sourceforge.net</a>
target="_parent"><img src="http://sourceforge.net/sflogo.php?group_id=3&amp;type=1"
width="88" height="31" align="bottom" alt="Sourceforge.net" border="0"></a>
</blockquote>
</body>

View File

@@ -88,10 +88,10 @@ types such as <code>EGLNativeDisplayType</code> or
<code>EGLNativeWindowType</code> defined for.</p>
<p>The available platforms are <code>x11</code>, <code>drm</code>,
<code>wayland</code>, <code>surfaceless</code>, <code>android</code>,
and <code>haiku</code>. The <code>android</code> platform
<code>wayland</code>, <code>null</code>, <code>android</code>,
<code>haiku</code>, and <code>gdi</code>. The <code>android</code> platform
can only be built as a system component, part of AOSP, while the
<code>haiku</code> platform can only be built with SCons.
<code>haiku</code> and <code>gdi</code> platforms can only be built with SCons.
Unless for special needs, the build system should
select the right platforms automatically.</p>

View File

@@ -91,20 +91,11 @@ This is only valid for versions &gt;= 3.0.
<li> Mesa may not really implement all the features of the given version.
(for developers only)
</ul>
<li>MESA_GLES_VERSION_OVERRIDE - changes the value returned by
glGetString(GL_VERSION) for OpenGL ES.
<ul>
<li> The format should be MAJOR.MINOR
<li> Examples: 2.0, 3.0, 3.1
<li> Mesa may not really implement all the features of the given version.
(for developers only)
</ul>
<li>MESA_GLSL_VERSION_OVERRIDE - changes the value returned by
glGetString(GL_SHADING_LANGUAGE_VERSION). Valid values are integers, such as
"130". Mesa will not really implement all the features of the given language version
if it's higher than what's normally reported. (for developers only)
<li>MESA_GLSL - <a href="shading.html#envvars">shading language compiler options</a>
<li>MESA_NO_MINMAX_CACHE - when set, the minmax index cache is globally disabled.
</ul>
@@ -162,7 +153,6 @@ See the <a href="xlibdriver.html">Xlib software driver page</a> for details.
<li>no16 - suppress generation of 16-wide fragment shaders. useful for debugging broken shaders</li>
<li>blorp - emit messages about the blorp operations (blits &amp; clears)</li>
<li>nodualobj - suppress generation of dual-object geometry shader code</li>
<li>optimizer - dump shader assembly to files at each optimization pass and iteration that make progress</li>
</ul>
</ul>
@@ -188,14 +178,6 @@ Mesa EGL supports different sets of environment variables. See the
<li>GALLIUM_HUD - draws various information on the screen, like framerate,
cpu load, driver statistics, performance counters, etc.
Set GALLIUM_HUD=help and run e.g. glxgears for more info.
<li>GALLIUM_HUD_PERIOD - sets the hud update rate in seconds (float). Use zero
to update every frame. The default period is 1/2 second.
<li>GALLIUM_HUD_VISIBLE - control default visibility, defaults to true.
<li>GALLIUM_HUD_TOGGLE_SIGNAL - toggle visibility via user specified signal.
Especially useful to toggle hud at specific points of application and
disable for unencumbered viewing the rest of the time. For example, set
GALLIUM_HUD_VISIBLE to false and GALLIUM_HUD_SIGNAL_TOGGLE to 10 (SIGUSR1).
Use kill -10 <pid> to toggle the hud as desired.
<li>GALLIUM_LOG_FILE - specifies a file for logging all errors, warnings, etc.
rather than stderr.
<li>GALLIUM_PRINT_OPTIONS - if non-zero, print all the Gallium environment
@@ -232,7 +214,7 @@ See src/mesa/state_tracker/st_debug.c for other options.
<li>LP_PERF - a comma-separated list of options to selectively no-op various
parts of the driver. See the source code for details.
<li>LP_NUM_THREADS - an integer indicating how many threads to use for rendering.
Zero turns off threading completely. The default value is the number of CPU
Zero turns of threading completely. The default value is the number of CPU
cores present.
</ul>
@@ -247,31 +229,6 @@ for details.
</ul>
<h3>VA-API state tracker environment variables</h3>
<ul>
<li>VAAPI_MPEG4_ENABLED - enable MPEG4 for VA-API, disabled by default.
</ul>
<h3>VC4 driver environment variables</h3>
<ul>
<li>VC4_DEBUG - a comma-separated list of named flags, which do various things:
<ul>
<li>cl - dump command list during creation</li>
<li>qpu - dump generated QPU instructions</li>
<li>qir - dump QPU IR during program compile</li>
<li>nir - dump NIR during program compile</li>
<li>tgsi - dump TGSI during program compile</li>
<li>shaderdb - dump program compile information for shader-db analysis</li>
<li>perf - print during performance-related events</li>
<li>norast - skip actual hardware execution of commands</li>
<li>always_flush - flush after each draw call</li>
<li>always_sync - wait for finish after each flush</li>
<li>dump - write a GPU command stream trace file (VC4 simulator only)</li>
</ul>
</ul>
<p>
Other Gallium drivers have their own environment variables. These may change
frequently so the source code should be consulted for details.

View File

@@ -16,163 +16,6 @@
<h1>News</h1>
<h2>February 10, 2016</h2>
<p>
<a href="relnotes/11.1.2.html">Mesa 11.1.2</a> is released.
This is a bug-fix release.
</p>
<h2>January 22, 2016</h2>
<p>
<a href="relnotes/11.0.9.html">Mesa 11.0.9</a> is released.
This is a bug-fix release.
<br>
NOTE: It is anticipated that 11.0.9 will be the final release in the 11.0
series. Users of 11.0 are encouraged to migrate to the 11.1 series in order
to obtain future fixes.
</p>
<h2>January 13, 2016</h2>
<p>
<a href="relnotes/11.1.1.html">Mesa 11.1.1</a> is released.
This is a bug-fix release.
</p>
<h2>December 21, 2015</h2>
<p>
<a href="relnotes/11.0.8.html">Mesa 11.0.8</a> is released.
This is a bug-fix release.
</p>
<h2>December 15, 2015</h2>
<p>
<a href="relnotes/11.1.0.html">Mesa 11.1.0</a> is released. This is a new
development release. See the release notes for more information about
the release.
</p>
<h2>December 9, 2015</h2>
<p>
<a href="relnotes/11.0.7.html">Mesa 11.0.7</a> is released.
This is a bug-fix release.
</p>
<p>
Mesa demos 8.3.0 is also released.
See the <a href="http://lists.freedesktop.org/archives/mesa-announce/2015-December/000191.html">announcement</a> for more information about the release.
You can download it from <a href="ftp://ftp.freedesktop.org/pub/mesa/demos/8.3.0/">ftp.freedesktop.org/pub/mesa/demos/8.3.0/</a>.
</p>
<h2>November 21, 2015</h2>
<p>
<a href="relnotes/11.0.6.html">Mesa 11.0.6</a> is released.
This is a bug-fix release.
</p>
<h2>November 11, 2015</h2>
<p>
<a href="relnotes/11.0.5.html">Mesa 11.0.5</a> is released.
This is a bug-fix release.
</p>
<h2>October 24, 2015</h2>
<p>
<a href="relnotes/11.0.4.html">Mesa 11.0.4</a> is released.
This is a bug-fix release.
</p>
<h2>October 10, 2015</h2>
<p>
<a href="relnotes/11.0.3.html">Mesa 11.0.3</a> is released.
This is a bug-fix release.
</p>
<h2>October 3, 2015</h2>
<p>
<a href="relnotes/10.6.9.html">Mesa 10.6.9</a> is released.
This is a bug-fix release.
<br>
NOTE: It is anticipated that 10.6.9 will be the final release in the 10.6
series. Users of 10.6 are encouraged to migrate to the 11.0 series in order
to obtain future fixes.
</p>
<h2>September 28, 2015</h2>
<p>
<a href="relnotes/11.0.2.html">Mesa 11.0.2</a> is released.
This is a bug-fix release.
</p>
<h2>September 26, 2015</h2>
<p>
<a href="relnotes/11.0.1.html">Mesa 11.0.1</a> is released.
This is a bug-fix release.
</p>
<h2>September 20, 2015</h2>
<p>
<a href="relnotes/10.6.8.html">Mesa 10.6.8</a> is released.
This is a bug-fix release.
</p>
<h2>September 12, 2015</h2>
<p>
<a href="relnotes/11.0.0.html">Mesa 11.0.0</a> is released. This is a new
development release. See the release notes for more information about
the release.
</p>
<h2>September 10, 2015</h2>
<p>
<a href="relnotes/10.6.7.html">Mesa 10.6.7</a> is released.
This is a bug-fix release.
</p>
<h2>September 4, 2015</h2>
<p>
<a href="relnotes/10.6.6.html">Mesa 10.6.6</a> is released.
This is a bug-fix release.
</p>
<h2>August 22, 2015</h2>
<p>
<a href="relnotes/10.6.5.html">Mesa 10.6.5</a> is released.
This is a bug-fix release.
</p>
<h2>August 11, 2015</h2>
<p>
<a href="relnotes/10.6.4.html">Mesa 10.6.4</a> is released.
This is a bug-fix release.
</p>
<h2>July 26, 2015</h2>
<p>
<a href="relnotes/10.6.3.html">Mesa 10.6.3</a> is released.
This is a bug-fix release.
</p>
<h2>July 11, 2015</h2>
<p>
<a href="relnotes/10.6.2.html">Mesa 10.6.2</a> is released.
This is a bug-fix release.
</p>
<h2>July 04, 2015</h2>
<p>
<a href="relnotes/10.5.9.html">Mesa 10.5.9</a> is released.
This is a bug-fix release.
<br>
NOTE: It is anticipated that 10.5.9 will be the final release in the 10.5
series. Users of 10.5 are encouraged to migrate to the 10.6 series in order
to obtain future fixes.
</p>
<h2>June 29, 2015</h2>
<p>
<a href="relnotes/10.6.1.html">Mesa 10.6.1</a> is released.
This is a bug-fix release.
</p>
<h2>June 20, 2015</h2>
<p>
<a href="relnotes/10.5.8.html">Mesa 10.5.8</a> is released.

View File

@@ -39,7 +39,7 @@ Version 2.6.4 or later should work.
</li>
<br>
<li><a href="http://www.makotemplates.org/">Python Mako module</a> -
Python Mako module is required. Version 0.3.4 or later should work.
Python Mako module is required. Version 0.7.3 or later should work.
</li>
</br>
<li><a href="http://www.scons.org/">SCons</a> is required for building on
@@ -58,9 +58,6 @@ On Windows with MinGW, install flex and bison with:
For MSVC on Windows, install
<a href="http://winflexbison.sourceforge.net/">Win flex-bison</a>.
</li>
<br>
<li>For building on Windows, Microsoft Visual Studio 2013 or later is required.
</li>
</ul>

View File

@@ -21,29 +21,6 @@ The release notes summarize what's new or changed in each Mesa release.
</p>
<ul>
<li><a href="relnotes/11.1.2.html">11.1.2 release notes</a>
<li><a href="relnotes/11.0.9.html">11.0.9 release notes</a>
<li><a href="relnotes/11.1.1.html">11.1.1 release notes</a>
<li><a href="relnotes/11.0.8.html">11.0.8 release notes</a>
<li><a href="relnotes/11.1.0.html">11.1.0 release notes</a>
<li><a href="relnotes/11.0.7.html">11.0.7 release notes</a>
<li><a href="relnotes/11.0.6.html">11.0.6 release notes</a>
<li><a href="relnotes/11.0.5.html">11.0.5 release notes</a>
<li><a href="relnotes/11.0.4.html">11.0.4 release notes</a>
<li><a href="relnotes/11.0.3.html">11.0.3 release notes</a>
<li><a href="relnotes/10.6.9.html">10.6.9 release notes</a>
<li><a href="relnotes/11.0.2.html">11.0.2 release notes</a>
<li><a href="relnotes/11.0.1.html">11.0.1 release notes</a>
<li><a href="relnotes/10.6.8.html">10.6.8 release notes</a>
<li><a href="relnotes/11.0.0.html">11.0.0 release notes</a>
<li><a href="relnotes/10.6.7.html">10.6.7 release notes</a>
<li><a href="relnotes/10.6.6.html">10.6.6 release notes</a>
<li><a href="relnotes/10.6.5.html">10.6.5 release notes</a>
<li><a href="relnotes/10.6.4.html">10.6.4 release notes</a>
<li><a href="relnotes/10.6.3.html">10.6.3 release notes</a>
<li><a href="relnotes/10.6.2.html">10.6.2 release notes</a>
<li><a href="relnotes/10.5.9.html">10.5.9 release notes</a>
<li><a href="relnotes/10.6.1.html">10.6.1 release notes</a>
<li><a href="relnotes/10.5.8.html">10.5.8 release notes</a>
<li><a href="relnotes/10.6.0.html">10.6.0 release notes</a>
<li><a href="relnotes/10.5.7.html">10.5.7 release notes</a>

View File

@@ -1,140 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 10.5.9 Release Notes / July 04, 2015</h1>
<p>
Mesa 10.5.9 is a bug fix release which fixes bugs found since the 10.5.8 release.
</p>
<p>
Mesa 10.5.9 implements the OpenGL 3.3 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 3.3. OpenGL
3.3 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
0c081b59572ee9732e7438d34adc3817fe8cc8d4b58abc0e71fd4b4c904945cb mesa-10.5.9.tar.gz
71c69f31d3dbc35cfa79950e58a01d27030378d8c7ef1259a0b31d4d0487f4ec mesa-10.5.9.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=84225">Bug 84225</a> - Allow constant-index-expression sampler array indexing with GLSL-ES &lt; 300</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88999">Bug 88999</a> - [SKL] Compiz crashes after opening unity dash</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=89118">Bug 89118</a> - [SKL Bisected]many Ogles3conform cases core dumped</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90537">Bug 90537</a> - radeonsi bo/va conflict on RADEON_GEM_VA (rscreen-&gt;ws-&gt;buffer_from_handle returns NULL)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90839">Bug 90839</a> - [10.5.5/10.6 regression, bisected] PBO glDrawPixels no longer using blit fastpath</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90873">Bug 90873</a> - Kernel hang, TearFree On, Mate desktop environment</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91056">Bug 91056</a> - The Bard's Tale (2005, native) has rendering issues</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91117">Bug 91117</a> - Nimbus (running in wine) has rendering issues, objects are semi-transparent</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91124">Bug 91124</a> - Civilization V (in Wine) has rendering issues: text missing, menu bar corrupted</li>
</ul>
<h2>Changes</h2>
<p>Ben Widawsky (2):</p>
<ul>
<li>i965/gen9: Implement Push Constant Buffer workaround</li>
<li>i965/skl: Use 1 register for uniform pull constant payload</li>
</ul>
<p>Boyan Ding (1):</p>
<ul>
<li>egl/x11: Remove duplicate call to dri2_x11_add_configs_for_visuals</li>
</ul>
<p>Chris Wilson (3):</p>
<ul>
<li>i965: Fix HW blitter pitch limits</li>
<li>i915: Blit RGBX&lt;-&gt;RGBA drawpixels</li>
<li>i965: Export format comparison for blitting between miptrees</li>
</ul>
<p>Emil Velikov (6):</p>
<ul>
<li>docs: Add sha256sums for the 10.5.8 release</li>
<li>configure: warn about shared_glapi &amp; xlib-glx only when both are set</li>
<li>configure: error out when building backend-less libEGL</li>
<li>configure: error out when building libEGL without shared-glapi</li>
<li>gbm: do not (over)link against libglapi.so</li>
<li>Update version to 10.5.9</li>
</ul>
<p>Frank Henigman (1):</p>
<ul>
<li>gbm: dlopen libglapi so gbm_create_device works</li>
</ul>
<p>Ilia Mirkin (8):</p>
<ul>
<li>glsl: add version checks to conditionals for builtin variable enablement</li>
<li>mesa: add GL_PROGRAM_PIPELINE support in KHR_debug calls</li>
<li>glsl: binding point is a texture unit, which is a combined space</li>
<li>nvc0: always put all tfb bufs into bufctx</li>
<li>nv50,nvc0: make sure to pushbuf_refn before putting bo into pushbuf_data</li>
<li>nv50/ir: propagate modifier to right arg when const-folding mad</li>
<li>nv50/ir: fix emission of address reg in 3rd source</li>
<li>nv50/ir: copy joinAt when splitting both before and after</li>
</ul>
<p>Mario Kleiner (2):</p>
<ul>
<li>nouveau: Use dup fd as key in drm-winsys hash table to fix ZaphodHeads.</li>
<li>winsys/radeon: Use dup fd as key in drm-winsys hash table to fix ZaphodHeads.</li>
</ul>
<p>Michel Dänzer (1):</p>
<ul>
<li>winsys/radeon: Unmap GPU VM address range when destroying BO</li>
</ul>
<p>Tapani Pälli (6):</p>
<ul>
<li>glsl: Allow dynamic sampler array indexing with GLSL ES &lt; 3.00</li>
<li>mesa/glsl: new compiler option EmitNoIndirectSampler</li>
<li>i915: use EmitNoIndirectSampler</li>
<li>mesa/st: use EmitNoIndirectSampler if !ARB_gpu_shader5</li>
<li>i965: use EmitNoIndirectSampler for gen &lt; 7</li>
<li>glsl: validate sampler array indexing for 'constant-index-expression'</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,104 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 10.6.1 Release Notes / June 29, 2015</h1>
<p>
Mesa 10.6.1 is a bug fix release which fixes bugs found since the 10.6.0 release.
</p>
<p>
Mesa 10.6.1 implements the OpenGL 3.3 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 3.3. OpenGL
3.3 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
b4cccd4d0eabcc2bca00c3175d3ad88fdda57ffdb883a7998525b873a21fe607 mesa-10.6.1.tar.gz
6c80a2b647e57c85dc36e609d9aed17f878f0d8e0cf9ace86d14cf604101e1eb mesa-10.6.1.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90347">Bug 90347</a> - [NVE0+] Failure to insert texbar under some circumstances (causing bad colors in Terasology)</li>
</ul>
<h2>Changes</h2>
<p>Anuj Phogat (4):</p>
<ul>
<li>mesa: Handle integer formats in need_rgb_to_luminance_conversion()</li>
<li>mesa: Use helper function need_rgb_to_luminance_conversion()</li>
<li>mesa: Turn need_rgb_to_luminance_conversion() in to a global function</li>
<li>meta: Abort meta path if ReadPixels need rgb to luminance conversion</li>
</ul>
<p>Ben Widawsky (1):</p>
<ul>
<li>i965/gen9: Implement Push Constant Buffer workaround</li>
</ul>
<p>Boyan Ding (2):</p>
<ul>
<li>egl/x11: Set version of swrastLoader to 2</li>
<li>egl/x11: Remove duplicate call to dri2_x11_add_configs_for_visuals</li>
</ul>
<p>Emil Velikov (6):</p>
<ul>
<li>docs: Add sha256sums for the 10.6.0 release</li>
<li>configure: warn about shared_glapi &amp; xlib-glx only when both are set</li>
<li>configure: error out when building backend-less libEGL</li>
<li>configure: error out when building libEGL without shared-glapi</li>
<li>gbm: do not (over)link against libglapi.so</li>
<li>Update version to 10.6.1</li>
</ul>
<p>Frank Henigman (1):</p>
<ul>
<li>gbm: dlopen libglapi so gbm_create_device works</li>
</ul>
<p>Ilia Mirkin (9):</p>
<ul>
<li>nvc0/ir: fix collection of first uses for texture barrier insertion</li>
<li>nv50,nvc0: clamp uniform size to 64k</li>
<li>nvc0/ir: can't have a join on a load with an indirect source</li>
<li>glsl: handle conversions to double when comparing param matches</li>
<li>glsl: add version checks to conditionals for builtin variable enablement</li>
<li>mesa: add GL_PROGRAM_PIPELINE support in KHR_debug calls</li>
<li>glsl: binding point is a texture unit, which is a combined space</li>
<li>nvc0: always put all tfb bufs into bufctx</li>
<li>nv50,nvc0: make sure to pushbuf_refn before putting bo into pushbuf_data</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,165 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 10.6.2 Release Notes / July 11, 2015</h1>
<p>
Mesa 10.6.2 is a bug fix release which fixes bugs found since the 10.6.1 release.
</p>
<p>
Mesa 10.6.2 implements the OpenGL 3.3 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 3.3. OpenGL
3.3 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
9c7ab9300dda6c912faaaff97995ec1820ba21d114d9cf555f145cbad90995f4 mesa-10.6.2.tar.gz
05753d3db4212900927b9894221a1669a10f56786e86a7e818b6e18a0817dca9 mesa-10.6.2.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=73528">Bug 73528</a> - Deferred lighting in Second Life causes system hiccups and screen flickering</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=80500">Bug 80500</a> - Flickering shadows in unreleased title trace</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=82186">Bug 82186</a> - [r600g] BARTS GPU lockup with minecraft shaders</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=84225">Bug 84225</a> - Allow constant-index-expression sampler array indexing with GLSL-ES &lt; 300</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90537">Bug 90537</a> - radeonsi bo/va conflict on RADEON_GEM_VA (rscreen-&gt;ws-&gt;buffer_from_handle returns NULL)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90873">Bug 90873</a> - Kernel hang, TearFree On, Mate desktop environment</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91022">Bug 91022</a> - [g45 g965 bisected] assertions generated from textureGrad cube samplers fix</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91047">Bug 91047</a> - [SNB Bisected] Messed up Fog in Super Smash Bros. Melee in Dolphin</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91056">Bug 91056</a> - The Bard's Tale (2005, native) has rendering issues</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91117">Bug 91117</a> - Nimbus (running in wine) has rendering issues, objects are semi-transparent</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91124">Bug 91124</a> - Civilization V (in Wine) has rendering issues: text missing, menu bar corrupted</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91173">Bug 91173</a> - Oddworld: Stranger's Wrath HD: disfigured models in wrong colors</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91226">Bug 91226</a> - Crash in glLinkProgram (NEW)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91231">Bug 91231</a> - [NV92] Psychonauts (native) segfaults on start when DRI3 enabled</li>
</ul>
<h2>Changes</h2>
<p>Chris Wilson (1):</p>
<ul>
<li>loader: Look for any version of currently linked libudev.so</li>
</ul>
<p>Emil Velikov (2):</p>
<ul>
<li>docs: Add sha256 checksums for the 10.6.1 release</li>
<li>Update version to 10.6.2</li>
</ul>
<p>Ilia Mirkin (8):</p>
<ul>
<li>nv50/ir: propagate modifier to right arg when const-folding mad</li>
<li>nv50/ir: fix emission of address reg in 3rd source</li>
<li>nv50/ir: copy joinAt when splitting both before and after</li>
<li>mesa: reset the source packing when creating temp transfer image</li>
<li>nv50/ir: don't emit src2 in immediate form</li>
<li>mesa/prog: relative offsets into constbufs are not constant</li>
<li>nv50/ir: UCMP arguments are float, so make sure modifiers are applied</li>
<li>nvc0: turn sample counts off during blit</li>
</ul>
<p>Kenneth Graunke (5):</p>
<ul>
<li>i965/fs: Fix ir_txs in emit_texture_gen4_simd16().</li>
<li>i965: Reserve more batch space to accomodate Gen6 perfmonitors.</li>
<li>i965/vs: Fix matNxM vertex attributes where M != 4.</li>
<li>Revert "glsl: clone inputs and outputs during linking"</li>
<li>Revert "i965: Delete linked GLSL IR when using NIR."</li>
</ul>
<p>Marek Olšák (3):</p>
<ul>
<li>r600g: disable single-sample fast color clear due to hangs</li>
<li>radeonsi: fix a hang with DrawTransformFeedback on 4 SE chips</li>
<li>st/dri: don't set PIPE_BIND_SCANOUT for MSAA surfaces</li>
</ul>
<p>Mario Kleiner (2):</p>
<ul>
<li>nouveau: Use dup fd as key in drm-winsys hash table to fix ZaphodHeads.</li>
<li>winsys/radeon: Use dup fd as key in drm-winsys hash table to fix ZaphodHeads.</li>
</ul>
<p>Matt Turner (2):</p>
<ul>
<li>i965/fs: Don't mess up stride for uniform integer multiplication.</li>
<li>Revert SHA1 additions.</li>
</ul>
<p>Michel Dänzer (1):</p>
<ul>
<li>winsys/radeon: Unmap GPU VM address range when destroying BO</li>
</ul>
<p>Mike Stroyan (2):</p>
<ul>
<li>meta: Only change and restore viewport 0 in mesa meta mode</li>
<li>i965: allocate at least 1 BLEND_STATE element</li>
</ul>
<p>Neil Roberts (4):</p>
<ul>
<li>i965/skl: Set the pulls bary bit in 3DSTATE_PS_EXTRA</li>
<li>glsl: Add missing check for whether an expression is an add operation</li>
<li>glsl: Make sure not to dereference NULL</li>
<li>i965: Don't try to print the GLSL IR if it has been freed</li>
</ul>
<p>Tapani Pälli (8):</p>
<ul>
<li>glsl: clone inputs and outputs during linking</li>
<li>i965: Delete linked GLSL IR when using NIR.</li>
<li>glsl: Allow dynamic sampler array indexing with GLSL ES &lt; 3.00</li>
<li>mesa/glsl: new compiler option EmitNoIndirectSampler</li>
<li>i965: use EmitNoIndirectSampler for gen &lt; 7</li>
<li>i915: use EmitNoIndirectSampler</li>
<li>mesa/st: use EmitNoIndirectSampler if !ARB_gpu_shader5</li>
<li>glsl: validate sampler array indexing for 'constant-index-expression'</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,106 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 10.6.3 Release Notes / July 26, 2015</h1>
<p>
Mesa 10.6.3 is a bug fix release which fixes bugs found since the 10.6.2 release.
</p>
<p>
Mesa 10.6.3 implements the OpenGL 3.3 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 3.3. OpenGL
3.3 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
c27e1e33798e69a6d2d2425aee8ac7b4c0b243066a65dd76cbb182ea31b1c7f2 mesa-10.6.3.tar.gz
58592e07c350cd2e8969b73fa83048c657a39fe2f13f3b88f5e5818fe2e4676d mesa-10.6.3.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90728">Bug 90728</a> - dvd playback with vlc and vdpau causes segmentation fault</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91337">Bug 91337</a> - OSMesaGetProcAdress(&quot;OSMesaPixelStore&quot;) returns nil</li>
</ul>
<h2>Changes</h2>
<p>Brian Paul (1):</p>
<ul>
<li>osmesa: fix OSMesaPixelsStore typo</li>
</ul>
<p>Chad Versace (1):</p>
<ul>
<li>mesa: Fix generation of git_sha1.h.tmp for gitlinks</li>
</ul>
<p>Christian König (2):</p>
<ul>
<li>vl: cleanup video buffer private when the decoder is destroyed</li>
<li>st/vdpau: fix mixer size checks</li>
</ul>
<p>Emil Velikov (3):</p>
<ul>
<li>docs: Add sha256 checksums for the 10.6.2 release</li>
<li>auxiliary/vl: use the correct screen index</li>
<li>Update version to 10.6.3</li>
</ul>
<p>Francisco Jerez (1):</p>
<ul>
<li>i965/gen9: Use custom MOCS entries set up by the kernel.</li>
</ul>
<p>Ilia Mirkin (5):</p>
<ul>
<li>nv50, nvc0: enable at least one color RT if alphatest is enabled</li>
<li>nvc0/ir: fix txq on indirect samplers</li>
<li>nvc0/ir: don't worry about sampler in txq handling</li>
<li>gm107/ir: fix indirect txq emission</li>
<li>nv50: fix max level clamping on G80</li>
</ul>
<p>Kenneth Graunke (1):</p>
<ul>
<li>program: Allow redundant OPTION ARB_fog_* directives.</li>
</ul>
<p>Rob Clark (1):</p>
<ul>
<li>xa: don't leak fences</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,137 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 10.6.4 Release Notes / August 11, 2015</h1>
<p>
Mesa 10.6.4 is a bug fix release which fixes bugs found since the 10.6.3 release.
</p>
<p>
Mesa 10.6.4 implements the OpenGL 3.3 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 3.3. OpenGL
3.3 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
4960bf17d8b5d6a6503c6954ec6cf480b5cd930797bac901c60bea192675f85e mesa-10.6.4.tar.gz
8f5ac103f0f503de2f7a985b0df349bd4ecdfe7f51c714be146fa5a9a3c07b77 mesa-10.6.4.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=73512">Bug 73512</a> - [clover] mesa.icd. should contain full path</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91290">Bug 91290</a> - SIGSEGV glcpp/glcpp-parse.y:1077</li>
</ul>
<h2>Changes</h2>
<p>Anuj Phogat (6):</p>
<ul>
<li>mesa: Turn get_readpixels_transfer_ops() in to a global function</li>
<li>meta: Fix transfer operations check in meta pbo path for readpixels</li>
<li>meta: Abort meta pbo path if readpixels need signed-unsigned conversion</li>
<li>meta: Don't do fragment color clamping in _mesa_meta_pbo_GetTexSubImage</li>
<li>mesa: Add a helper function _mesa_need_luminance_to_rgb_conversion()</li>
<li>meta: Fix reading luminance texture as rgba in _mesa_meta_pbo_GetTexSubImage()</li>
</ul>
<p>Ben Widawsky (1):</p>
<ul>
<li>i965/skl: Add production thread counts and URB size</li>
</ul>
<p>Eduardo Lima Mitev (3):</p>
<ul>
<li>mesa: Fix errors values returned by glShaderBinary()</li>
<li>mesa: Validate target before resolving tex obj in glTex(ture)SubImageXD</li>
<li>mesa: Fix error returned by glCopyTexImage2D() upon an invalid internal format</li>
</ul>
<p>Emil Velikov (6):</p>
<ul>
<li>docs: Add checksums for mesa 10.6.3 tarballs</li>
<li>configure.ac: do not set HAVE_DRI(23) when libdrm is missing</li>
<li>egl/wayland: libdrm is a hard requirement, treat it as such</li>
<li>winsys/radeon: don't leak the fd when it is 0</li>
<li>bugzilla_mesa.sh: sort the bugs list by number</li>
<li>Update version to 10.6.4</li>
</ul>
<p>Francisco Jerez (1):</p>
<ul>
<li>i965/fs: Fix fs_inst::regs_read() for sources in the ATTR file.</li>
</ul>
<p>Frank Binns (2):</p>
<ul>
<li>egl/dri: Add error info needed for EGL_EXT_image_dma_buf_import extension</li>
<li>egl: Add eglQuerySurface surface type check for EGL_LARGEST_PBUFFER attrib</li>
</ul>
<p>Igor Gnatenko (1):</p>
<ul>
<li>opencl: use versioned .so in mesa.icd</li>
</ul>
<p>Ilia Mirkin (1):</p>
<ul>
<li>nvc0: fix geometry program revalidation of clipping params</li>
</ul>
<p>Kenneth Graunke (1):</p>
<ul>
<li>glsl: Fix a bug where LHS swizzles of swizzles were too small.</li>
</ul>
<p>Marek Olšák (6):</p>
<ul>
<li>st/mesa: don't call st_validate_state in BlitFramebuffer</li>
<li>radeonsi: upload shader rodata after updating scratch relocations</li>
<li>st/mesa: don't ignore texture buffer state changes</li>
<li>radeonsi: rework how shader pointers to descriptors are set</li>
<li>radeonsi: completely rework updating descriptors without CP DMA</li>
<li>r600g: fix the CB_SHADER_MASK setup</li>
</ul>
<p>Samuel Iglesias Gonsalvez (1):</p>
<ul>
<li>glsl/glcpp: fix SIGSEGV when checking error condition for macro redefinition</li>
</ul>
<p>Samuel Pitoiset (1):</p>
<ul>
<li>nv50: avoid segfault with enabled but unbound vertex attrib</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,124 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 10.6.5 Release Notes / August 22, 2015</h1>
<p>
Mesa 10.6.5 is a bug fix release which fixes bugs found since the 10.6.4 release.
</p>
<p>
Mesa 10.6.5 implements the OpenGL 3.3 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 3.3. OpenGL
3.3 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
afe290fc7af75a25df5ee52396a9f09e5dba85fb3e159304bdda265b8564b0d4 mesa-10.6.5.tar.gz
fb6fac3c85bcfa9d06b8dd439169f23f0c0924a88e44362e738b99b1feff762f mesa-10.6.5.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=85252">Bug 85252</a> - Segfault in compiler while processing ternary operator with void arguments</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91570">Bug 91570</a> - Upgrading mesa to 10.6 causes segfault in OpenGL applications with GeForce4 MX 440 / AGP 8X</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91610">Bug 91610</a> - [BSW] GPU hang for spec.shaders.point-vertex-id gl_instanceid divisor</li>
</ul>
<h2>Changes</h2>
<p>Adam Jackson (1):</p>
<ul>
<li>glx: Fix __glXWireToEvent for BufferSwapComplete</li>
</ul>
<p>Alex Deucher (2):</p>
<ul>
<li>radeonsi: add new OLAND pci id</li>
<li>radeonsi: properly set the raster_config for KV</li>
</ul>
<p>Emil Velikov (4):</p>
<ul>
<li>docs: add sha256 checksums for 10.6.4</li>
<li>vc4: add missing nir include, to fix the build</li>
<li>Revert "radeonsi: properly set the raster_config for KV"</li>
<li>Update version to 10.6.5</li>
</ul>
<p>Frank Binns (1):</p>
<ul>
<li>egl/x11: don't abort when creating a DRI2 drawable fails</li>
</ul>
<p>Ilia Mirkin (3):</p>
<ul>
<li>nouveau: no need to do tnl wakeup, state updates are always hooked up</li>
<li>gm107/ir: indirect handle goes first on maxwell also</li>
<li>nv50,nvc0: take level into account when doing eng2d multi-layer blits</li>
</ul>
<p>Jason Ekstrand (4):</p>
<ul>
<li>meta/copy_image: Stash off the scissor</li>
<li>mesa/formats: Only do byteswapping for packed formats</li>
<li>mesa/formats: Fix swizzle flipping for big-endian targets</li>
<li>mesa/formats: Don't flip channels of null array formats</li>
</ul>
<p>Marek Olšák (3):</p>
<ul>
<li>radeonsi: fix polygon offset scale</li>
<li>r600g: fix polygon offset scale</li>
<li>r600g: allow setting geometry shader sampler states</li>
</ul>
<p>Neil Roberts (1):</p>
<ul>
<li>i965/bdw: Fix setting the instancing state for the SGVS element</li>
</ul>
<p>Oded Gabbay (2):</p>
<ul>
<li>mesa: clear existing swizzle info before bitwise-OR</li>
<li>mesa/formats: don't byteswap when building array formats</li>
</ul>
<p>Renaud Gaubert (1):</p>
<ul>
<li>glsl: avoid compiler's segfault when processing operators with void arguments</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,164 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 10.6.6 Release Notes / September 04, 2015</h1>
<p>
Mesa 10.6.6 is a bug fix release which fixes bugs found since the 10.6.5 release.
</p>
<p>
Mesa 10.6.6 implements the OpenGL 3.3 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 3.3. OpenGL
3.3 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
416517aa9df4791f97d34451a9e4da33c966afcd18c115c5769b92b15b018ef5 mesa-10.6.6.tar.gz
570f2154b7340ff5db61ff103bc6e85165b8958798b78a50fa2df488e98e5778 mesa-10.6.6.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=84677">Bug 84677</a> - Triangle disappears with glPolygonMode GL_LINE</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90734">Bug 90734</a> - glBufferSubData is corrupting data when buffer is &gt; 32k</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90748">Bug 90748</a> - [BDW Bisected]dEQP-GLES3.functional.fbo.completeness.renderable.texture.depth.rg_half_float_oes fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90902">Bug 90902</a> - [bsw][regression] dEQP: &quot;Found invalid pixel values&quot;</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90925">Bug 90925</a> - &quot;high fidelity&quot;: Segfault in _mesa_program_resource_find_name</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91254">Bug 91254</a> - (regresion) video using VA-API on Intel slow and freeze system with mesa 10.6 or 10.6.1</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91292">Bug 91292</a> - [BDW+] glVertexAttribDivisor not working in combination with glPolygonMode</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91673">Bug 91673</a> - Segfault when calling glTexSubImage2D on storage texture to bound FBO</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91726">Bug 91726</a> - R600 asserts in tgsi_cmp/make_src_for_op3</li>
</ul>
<h2>Changes</h2>
<p>Chris Wilson (2):</p>
<ul>
<li>i965: Prevent coordinate overflow in intel_emit_linear_blit</li>
<li>i965: Always re-emit the pipeline select during invariant state emission</li>
</ul>
<p>Daniel Scharrer (1):</p>
<ul>
<li>mesa: add missing queries for ARB_direct_state_access</li>
</ul>
<p>Dave Airlie (8):</p>
<ul>
<li>mesa/arb_gpu_shader_fp64: add support for glGetUniformdv</li>
<li>mesa/texgetimage: fix missing stencil check</li>
<li>st/readpixels: fix accel path for skipimages.</li>
<li>texcompress_s3tc/fxt1: fix stride checks (v1.1)</li>
<li>mesa/readpixels: check strides are equal before skipping conversion</li>
<li>mesa: enable texture stencil8 for multisample</li>
<li>r600/sb: update last_cf for finalize if.</li>
<li>r600g: fix calculation for gpr allocation</li>
</ul>
<p>David Heidelberg (1):</p>
<ul>
<li>st/nine: Require gcc &gt;= 4.6</li>
</ul>
<p>Emil Velikov (2):</p>
<ul>
<li>docs: add sha256 checksums for 10.6.5</li>
<li>get-pick-list.sh: Require explicit "10.6" for nominating stable patches</li>
</ul>
<p>Glenn Kennard (4):</p>
<ul>
<li>r600g: Fix assert in tgsi_cmp</li>
<li>r600g/sb: Handle undef in read port tracker</li>
<li>r600g/sb: Don't read junk after EOP</li>
<li>r600g/sb: Don't crash on empty if jump target</li>
</ul>
<p>Ilia Mirkin (5):</p>
<ul>
<li>st/mesa: fix assignments with 4-operand arguments (i.e. BFI)</li>
<li>st/mesa: pass through 4th opcode argument in bitmap/pixel visitors</li>
<li>nv50,nvc0: disable depth bounds test on blit</li>
<li>nv50: fix 2d engine blits for 64- and 128-bit formats</li>
<li>mesa: only copy the requested teximage faces</li>
</ul>
<p>Jason Ekstrand (1):</p>
<ul>
<li>i965/fs: Split VGRFs after lowering pull constants</li>
</ul>
<p>Kenneth Graunke (3):</p>
<ul>
<li>i965: Fix copy propagation type changes.</li>
<li>Revert "i965: Advertise a line width of 40.0 on Cherryview and Skylake."</li>
<li>i965: Momentarily pretend to support ARB_texture_stencil8 for blits.</li>
</ul>
<p>Marek Olšák (3):</p>
<ul>
<li>gallium/radeon: fix the ADDRESS_HI mask for EVENT_WRITE CIK packets</li>
<li>mesa: create multisample fallback textures like normal textures</li>
<li>radeonsi: fix a Unigine Heaven hang when drirc is missing</li>
</ul>
<p>Matt Turner (1):</p>
<ul>
<li>i965/fs: Handle MRF destinations in lower_integer_multiplication().</li>
</ul>
<p>Neil Roberts (2):</p>
<ul>
<li>i965: Swap the order of the vertex ID and edge flag attributes</li>
<li>i965/bdw: Fix 3DSTATE_VF_INSTANCING when the edge flag is used</li>
</ul>
<p>Tapani Pälli (5):</p>
<ul>
<li>mesa: update fbo state in glTexStorage</li>
<li>glsl: build stageref mask using IR, not symbol table</li>
<li>glsl: expose build_program_resource_list function</li>
<li>glsl: create program resource list after LinkShader</li>
<li>mesa: add GL_RED, GL_RG support for floating point textures</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,75 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 10.6.7 Release Notes / September 10, 2015</h1>
<p>
Mesa 10.6.7 is a bug fix release which fixes bugs found since the 10.6.6 release.
</p>
<p>
Mesa 10.6.7 implements the OpenGL 3.3 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 3.3. OpenGL
3.3 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
4ba10c59abee30d72476543a57afd2f33803dabf4620dc333b335d47966ff842 mesa-10.6.7.tar.gz
feb1f640b915dada88a7c793dfaff0ae23580f8903f87a6b76469253de0d28d8 mesa-10.6.7.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90751">Bug 90751</a> - [BDW Bisected]dEQP-GLES3.functional.fbo.completeness.renderable.texture.stencil.stencil_index8 fails</li>
</ul>
<h2>Changes</h2>
<p>Dave Airlie (1):</p>
<ul>
<li>mesa/teximage: use correct extension for accept stencil texture.</li>
</ul>
<p>Emil Velikov (3):</p>
<ul>
<li>docs: add sha256 checksums for 10.6.6</li>
<li>Revert "i965: Momentarily pretend to support ARB_texture_stencil8 for blits."</li>
<li>Update version to 10.6.7</li>
</ul>
<p>Kenneth Graunke (1):</p>
<ul>
<li>glsl: Handle attribute aliasing in attribute storage limit check.</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,136 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 10.6.8 Release Notes / September 20, 2015</h1>
<p>
Mesa 10.6.8 is a bug fix release which fixes bugs found since the 10.6.7 release.
</p>
<p>
Mesa 10.6.8 implements the OpenGL 3.3 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 3.3. OpenGL
3.3 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
1f34dba2a8059782e3e4e0f18b9628004e253b2c69085f735b846d2e63c9e250 mesa-10.6.8.tar.gz
e36ee5ceeadb3966fb5ce5b4cf18322dbb76a4f075558ae49c3bba94f57d58fd mesa-10.6.8.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90621">Bug 90621</a> - Mesa fail to build from git</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91526">Bug 91526</a> - World of Warcraft (on Wine) has UI corruption with nouveau</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91719">Bug 91719</a> - [SNB,HSW,BYT] dEQP regressions associated with using NIR for vertex shaders</li>
</ul>
<h2>Changes</h2>
<p>Alejandro Piñeiro (1):</p>
<ul>
<li>i965/vec4: fill src_reg type using the constructor type parameter</li>
</ul>
<p>Antia Puentes (1):</p>
<ul>
<li>i965/vec4: Fix saturation errors when coalescing registers</li>
</ul>
<p>Emil Velikov (2):</p>
<ul>
<li>docs: add sha256 checksums for 10.6.7</li>
<li>cherry-ignore: add commit non applicable for 10.6</li>
</ul>
<p>Hans de Goede (4):</p>
<ul>
<li>nv30: Fix creation of scanout buffers</li>
<li>nv30: Implement color resolve for msaa</li>
<li>nv30: Fix max width / height checks in nv30 sifm code</li>
<li>nv30: Disable msaa unless requested from the env by NV30_MAX_MSAA</li>
</ul>
<p>Ian Romanick (2):</p>
<ul>
<li>mesa: Pass the type to _mesa_uniform_matrix as a glsl_base_type</li>
<li>mesa: Don't allow wrong type setters for matrix uniforms</li>
</ul>
<p>Ilia Mirkin (5):</p>
<ul>
<li>st/mesa: don't fall back to 16F when 32F is requested</li>
<li>nvc0: always emit a full shader colormask</li>
<li>nvc0: remove BGRA4 format support</li>
<li>st/mesa: avoid integer overflows with buffers &gt;= 512MB</li>
<li>nv50, nvc0: fix max texture buffer size to 128M elements</li>
</ul>
<p>Jason Ekstrand (1):</p>
<ul>
<li>i965/vec4: Don't reswizzle hardware registers</li>
</ul>
<p>Jose Fonseca (1):</p>
<ul>
<li>gallivm: Workaround LLVM PR23628.</li>
</ul>
<p>Kenneth Graunke (1):</p>
<ul>
<li>i965: Momentarily pretend to support ARB_texture_stencil8 for blits.</li>
</ul>
<p>Oded Gabbay (1):</p>
<ul>
<li>llvmpipe: convert double to long long instead of unsigned long long</li>
</ul>
<p>Ray Strode (1):</p>
<ul>
<li>gbm: convert gbm bo format to fourcc format on dma-buf import</li>
</ul>
<p>Ulrich Weigand (1):</p>
<ul>
<li>mesa: Fix texture compression on big-endian systems</li>
</ul>
<p>Vinson Lee (1):</p>
<ul>
<li>gallivm: Do not use NoFramePointerElim with LLVM 3.7.</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,130 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 10.6.9 Release Notes / Octover 03, 2015</h1>
<p>
Mesa 10.6.9 is a bug fix release which fixes bugs found since the 10.6.8 release.
</p>
<p>
Mesa 10.6.9 implements the OpenGL 3.3 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 3.3. OpenGL
3.3 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
3406876aac67546d0c3e2cb97da330b62644c313e7992b95618662e13c54296a mesa-10.6.9.tar.gz
b04c4de6280b863babc2929573da17218d92e9e4ba6272d548d135415723e8c3 mesa-10.6.9.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=38109">Bug 38109</a> - i915 driver crashes if too few vertices are submitted (Mesa 7.10.2)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=55552">Bug 55552</a> - Compile errors with --enable-mangling</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86281">Bug 86281</a> - brw_meta_fast_clear (brw=brw&#64;entry=0x7fffd4097a08, fb=fb&#64;entry=0x7fffd40fa900, buffers=buffers&#64;entry=2, partial_clear=partial_clear&#64;entry=false)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91970">Bug 91970</a> - [BSW regression] dEQP-GLES3.functional.shaders.precision.int.highp_mul_vertex</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92072">Bug 92072</a> - Wine breakage since d082c5324 (st/mesa: don't call st_validate_state in BlitFramebuffer)</li>
</ul>
<h2>Changes</h2>
<p>Brian Paul (1):</p>
<ul>
<li>st/mesa: try PIPE_BIND_RENDER_TARGET when choosing float texture formats</li>
</ul>
<p>Chris Wilson (1):</p>
<ul>
<li>i965: Remove early release of DRI2 miptree</li>
</ul>
<p>Emil Velikov (4):</p>
<ul>
<li>docs: add sha256 checksums for 10.6.8</li>
<li>cherry-ignore: add commit non applicable for 10.6</li>
<li>cherry-ignore: add commit non applicable for 10.6</li>
<li>Update version to 10.6.9</li>
</ul>
<p>Iago Toral Quiroga (1):</p>
<ul>
<li>mesa: Fix GL_FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE for default framebuffer.</li>
</ul>
<p>Ian Romanick (5):</p>
<ul>
<li>t_dd_dmatmp: Make "count" actually be the count</li>
<li>t_dd_dmatmp: Clean up improper code formatting from previous patch</li>
<li>t_dd_dmatmp: Use '&amp; 3' instead of '% 4' everywhere</li>
<li>t_dd_dmatmp: Pull out common 'count -= count &amp; 3' code</li>
<li>t_dd_dmatmp: Use addition instead of subtraction in loop bounds</li>
</ul>
<p>Jeremy Huddleston (1):</p>
<ul>
<li>configure.ac: Add support to enable read-only text segment on x86.</li>
</ul>
<p>Kristian Høgsberg Kristensen (1):</p>
<ul>
<li>i965: Respect stride and subreg_offset for ATTR registers</li>
</ul>
<p>Kyle Brenneman (3):</p>
<ul>
<li>glx: Fix build errors with --enable-mangling (v2)</li>
<li>mapi: Make _glapi_get_stub work with "gl" or "mgl" prefix.</li>
<li>glx: Don't hard-code the name "libGL.so.1" in driOpenDriver (v3)</li>
</ul>
<p>Leo Liu (1):</p>
<ul>
<li>radeon/vce: fix vui time_scale zero error</li>
</ul>
<p>Marek Olšák (1):</p>
<ul>
<li>st/mesa: fix front buffer regression after dropping st_validate_state in Blit</li>
</ul>
<p>Roland Scheidegger (1):</p>
<ul>
<li>mesa: fix mipmap generation for immutable, compressed textures</li>
</ul>
</div>
</body>
</html>

61
docs/relnotes/10.7.0.html Normal file
View File

@@ -0,0 +1,61 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 10.7.0 Release Notes / TBD</h1>
<p>
Mesa 10.7.0 is a new development release.
People who are concerned with stability and reliability should stick
with a previous release or wait for Mesa 10.7.1.
</p>
<p>
Mesa 10.7.0 implements the OpenGL 3.3 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 3.3. OpenGL
3.3 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
TBD.
</pre>
<h2>New features</h2>
<p>
Note: some of the new features are only available with certain drivers.
</p>
<ul>
<li>GL_ARB_framebuffer_no_attachments on i965</li>
<li>GL_ARB_shader_stencil_export on llvmpipe</li>
</ul>
<h2>Bug fixes</h2>
TBD.
<h2>Changes</h2>
TBD.
</div>
</body>
</html>

View File

@@ -1,259 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 11.0.0 Release Notes / September 12, 2015</h1>
<p>
Mesa 11.0.0 is a new development release.
People who are concerned with stability and reliability should stick
with a previous release or wait for Mesa 11.0.1.
</p>
<p>
Mesa 11.0.0 implements the OpenGL 4.1 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 4.1. OpenGL
4.1 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
7d7e4ddffa3b162506efa01e2cc41e329caa4995336b92e5cc21f2e1fb36c1b3 mesa-11.0.0.tar.gz
e095a3eb2eca9dfde7efca8946527c8ae20a0cc938a8c78debc7f158ad44af32 mesa-11.0.0.tar.xz
</pre>
<h2>New features</h2>
<p>
Note: some of the new features are only available with certain drivers.
</p>
<ul>
<li>New hardware support for AMD GCN 1.2 GPUs: Tonga, Iceland, Carrizo, Fiji</li>
<li>OpenGL 4.1 on radeonsi, nvc0</li>
<li>OpenGL ES 3.0 on freedreno (a3xx, a4xx)
<li>GL_AMD_vertex_shader_viewport_index on radeonsi</li>
<li>GL_ARB_conditional_render_inverted on r600, radeonsi</li>
<li>GL_ARB_depth_buffer_float on a4xx</li>
<li>GL_ARB_derivative_control on radeonsi</li>
<li>GL_ARB_draw_buffers, GL_ARB_draw_buffers_blend on a4xx</li>
<li>GL_ARB_fragment_layer_viewport on radeonsi</li>
<li>GL_ARB_framebuffer_no_attachments on i965</li>
<li>GL_ARB_get_texture_sub_image for all drivers</li>
<li>GL_ARB_gpu_shader5 on radeonsi</li>
<li>GL_ARB_gpu_shader_fp64 on llvmpipe, radeonsi</li>
<li>GL_ARB_shader_image_load_store on i965</li>
<li>GL_ARB_shader_precision on radeonsi, nvc0</li>
<li>GL_ARB_shader_image_size on i965</li>
<li>GL_ARB_shader_stencil_export on llvmpipe</li>
<li>GL_ARB_shader_subroutine on core profile all drivers</li>
<li>GL_ARB_tessellation_shader on nvc0, radeonsi</li>
<li>GL_ARB_transform_feedback2, GL_ARB_transform_feedback_instanced, GL_EXT_transform_feedback on a3xx, a4xx</li>
<li>GL_ARB_vertex_attrib_64bit on llvmpipe, radeonsi</li>
<li>GL_ARB_viewport_array on radeonsi</li>
<li>GL_EXT_depth_bounds_test on radeonsi, nv30, nv50, nvc0</li>
<li>GL_EXT_texture_compression_s3tc on freedreno (a3xx)</li>
<li>GL_NV_read_depth (GLES) on all drivers</li>
<li>GL_NV_read_depth_stencil (GLES) on all drivers</li>
<li>GL_NV_read_stencil (GLES) on all drivers</li>
<li>GL_OES_texture_float on all r300, r600, radeonsi, nv30, nv50, nvc0, softpipe, llvmpipe</li>
<li>GL_OES_texture_half_float on all r300, r600, radeonsi, nv30, nv50, nvc0, softpipe, llvmpipe</li>
<li>GL_OES_texture_float_linear on all r300, r600, radeonsi, nv30, nv50, nvc0, softpipe, llvmpipe</li>
<li>GL_OES_texture_half_float_linear on all r300, r600, radeonsi, nv30, nv50, nvc0, softpipe, llvmpipe</li>
<li>GL_EXT_draw_buffers2 on a4xx</li>
<li>GLX_ARB_create_context_robustness on r600, radeonsi</li>
<li>EGL_EXT_create_context_robustness on r600, radeonsi</li>
<li>EGL_KHR_gl_colorspace on r600, radeonsi, nv50, nvc0</li>
<li>EGL_KHR_gl_texture_3D_image on r600, radeonsi, nv50, nvc0</li>
<li>EGL 1.5 on r600, radeonsi, nv50, nvc0</li>
</ul>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=51658">Bug 51658</a> - r200 (&amp; possibly radeon) DRI fixes for gnome shell on Mesa 8.0.3</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=65525">Bug 65525</a> - [llvmpipe] lp_scene.h:210:lp_scene_alloc: Assertion `size &lt;= (64 * 1024)' failed.</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=66346">Bug 66346</a> - shader_query.cpp:49: error: invalid conversion from 'void*' to 'GLuint'</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=73512">Bug 73512</a> - [clover] mesa.icd. should contain full path</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=73528">Bug 73528</a> - Deferred lighting in Second Life causes system hiccups and screen flickering</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=74329">Bug 74329</a> - Please expose OES_texture_float and OES_texture_half_float on the ES3 context</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=80500">Bug 80500</a> - Flickering shadows in unreleased title trace</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=82186">Bug 82186</a> - [r600g] BARTS GPU lockup with minecraft shaders</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=84225">Bug 84225</a> - Allow constant-index-expression sampler array indexing with GLSL-ES &lt; 300</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=84677">Bug 84677</a> - Triangle disappears with glPolygonMode GL_LINE</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=85252">Bug 85252</a> - Segfault in compiler while processing ternary operator with void arguments</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=89131">Bug 89131</a> - [Bisected] Graphical corruption in Weston, shows old framebuffer pieces</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90000">Bug 90000</a> - [i965 Bisected NIR] Piglit/gglean_fragprog1-z-write_test fail</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90073">Bug 90073</a> - Leaks in xcb_dri3_open_reply_fds() and get_render_node_from_id_path_tag</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90249">Bug 90249</a> - Fails to build egl_dri2 on osx</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90310">Bug 90310</a> - Fails to build gallium_dri.so at linking stage with clang because of multiple redefinitions</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90347">Bug 90347</a> - [NVE0+] Failure to insert texbar under some circumstances (causing bad colors in Terasology)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90466">Bug 90466</a> - arm: linker error ndefined reference to `nir_metadata_preserve'</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90520">Bug 90520</a> - Register spilling clobbers registers used elsewhere in the shader</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90537">Bug 90537</a> - radeonsi bo/va conflict on RADEON_GEM_VA (rscreen-&gt;ws-&gt;buffer_from_handle returns NULL)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90547">Bug 90547</a> - [BDW/BSW/SKL Bisected]Piglit/glean&#64;vertprog1-rsq_test_2_(reciprocal_square_root_of_negative_value) fais</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90580">Bug 90580</a> - [HSW bisected] integer multiplication bug</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90600">Bug 90600</a> - IOError: [Errno 2] No such file or directory: 'gl_API.xml'</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90621">Bug 90621</a> - Mesa fail to build from git</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90629">Bug 90629</a> - [i965] SIMD16 dual_source_blend assertion `src[i].file != GRF || src[i].width == dst.width' failed</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90691">Bug 90691</a> - [BSW]Piglit/spec/nv_conditional_render/dlist fails intermittently</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90728">Bug 90728</a> - dvd playback with vlc and vdpau causes segmentation fault</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90734">Bug 90734</a> - glBufferSubData is corrupting data when buffer is &gt; 32k</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90748">Bug 90748</a> - [BDW Bisected]dEQP-GLES3.functional.fbo.completeness.renderable.texture.depth.rg_half_float_oes fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90749">Bug 90749</a> - [BDW Bisected]dEQP-GLES3.functional.rasterization.fbo.rbo_multisample_max.primitives.lines_wide fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90751">Bug 90751</a> - [BDW Bisected]dEQP-GLES3.functional.fbo.completeness.renderable.texture.stencil.stencil_index8 fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90797">Bug 90797</a> - [ALL bisected] Mesa change cause performance case manhattan fail.</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90817">Bug 90817</a> - swrast fails to load with certain remote X servers</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90830">Bug 90830</a> - [bsw bisected regression] GPU hang for spec.arb_gpu_shader5.execution.sampler_array_indexing.vs-nonzero-base</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90839">Bug 90839</a> - [10.5.5/10.6 regression, bisected] PBO glDrawPixels no longer using blit fastpath</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90873">Bug 90873</a> - Kernel hang, TearFree On, Mate desktop environment</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90887">Bug 90887</a> - PhiMovesPass in register allocator broken</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90895">Bug 90895</a> - [IVB/HSW/BDW/BSW Bisected] GLB2.7 Egypt, GfxBench3.0 T-Rex &amp; ALU and many SynMark cases performance reduced by 10-23%</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90902">Bug 90902</a> - [bsw][regression] dEQP: &quot;Found invalid pixel values&quot;</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90903">Bug 90903</a> - egl_dri2.c:dri2_load fails to load libglapi on osx</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90904">Bug 90904</a> - OSX: EXC_BAD_ACCESS when using translate_sse + gallium + softpipe/llvmpipe</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90905">Bug 90905</a> - mesa: Finish subdir-objects transition</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90925">Bug 90925</a> - &quot;high fidelity&quot;: Segfault in _mesa_program_resource_find_name</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91022">Bug 91022</a> - [g45 g965 bisected] assertions generated from textureGrad cube samplers fix</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91047">Bug 91047</a> - [SNB Bisected] Messed up Fog in Super Smash Bros. Melee in Dolphin</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91056">Bug 91056</a> - The Bard's Tale (2005, native) has rendering issues</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91077">Bug 91077</a> - dri2_glx.c:1186: undefined reference to `loader_open_device'</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91099">Bug 91099</a> - [llvmpipe] piglit glsl-max-varyings &gt;max_varying_components regression</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91101">Bug 91101</a> - [softpipe] piglit glsl-1.50&#64;execution&#64;geometry&#64;max-input-components regression</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91117">Bug 91117</a> - Nimbus (running in wine) has rendering issues, objects are semi-transparent</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91124">Bug 91124</a> - Civilization V (in Wine) has rendering issues: text missing, menu bar corrupted</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91173">Bug 91173</a> - Oddworld: Stranger's Wrath HD: disfigured models in wrong colors</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91193">Bug 91193</a> - [290x] Dota2 reborn ingame rendering breaks with git-af4b9c7</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91222">Bug 91222</a> - lp_test_format regression on CentOS 7</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91226">Bug 91226</a> - Crash in glLinkProgram (NEW)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91231">Bug 91231</a> - [NV92] Psychonauts (native) segfaults on start when DRI3 enabled</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91254">Bug 91254</a> - (regresion) video using VA-API on Intel slow and freeze system with mesa 10.6 or 10.6.1</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91290">Bug 91290</a> - SIGSEGV glcpp/glcpp-parse.y:1077</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91292">Bug 91292</a> - [BDW+] glVertexAttribDivisor not working in combination with glPolygonMode</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91337">Bug 91337</a> - OSMesaGetProcAdress(&quot;OSMesaPixelStore&quot;) returns nil</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91418">Bug 91418</a> - Visual Studio 2015 vsnprintf build error</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91425">Bug 91425</a> - [regression, bisected] Piglit spec/ext_packed_float/ getteximage-invalid-format-for-packed-type fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91441">Bug 91441</a> - make check DispatchSanity_test.GL30 regression</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91444">Bug 91444</a> - regression bisected radeonsi: don't change pipe_resource in resource_copy_region</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91461">Bug 91461</a> - gl_TessLevel* writes have no effect for all but the last TCS invocation</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91513">Bug 91513</a> - [IVB/HSW/BDW/SKL Bisected] Lightsmark performance reduced by 7%-10%</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91526">Bug 91526</a> - World of Warcraft (on Wine) has UI corruption with nouveau</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91544">Bug 91544</a> - [i965, regression, bisected] regression of several tests in 93977d3a151675946c03e</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91551">Bug 91551</a> - DXTn compressed normal maps produce severe artifacts on all NV5x and NVDx chipsets</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91570">Bug 91570</a> - Upgrading mesa to 10.6 causes segfault in OpenGL applications with GeForce4 MX 440 / AGP 8X</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91591">Bug 91591</a> - rounding.h:102:2: error: #error &quot;Unsupported or undefined LONG_BIT&quot;</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91610">Bug 91610</a> - [BSW] GPU hang for spec.shaders.point-vertex-id gl_instanceid divisor</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91673">Bug 91673</a> - Segfault when calling glTexSubImage2D on storage texture to bound FBO</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91726">Bug 91726</a> - R600 asserts in tgsi_cmp/make_src_for_op3</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91847">Bug 91847</a> - glGenerateTextureMipmap not working (no errors) unless glActiveTexture(GL_TEXTURE1) is called before</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91857">Bug 91857</a> - Mesa 10.6.3 linker is slow</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91881">Bug 91881</a> - regression: GPU lockups since mesa-11.0.0_rc1 on RV620 (r600) driver</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91890">Bug 91890</a> - [nve7] witcher2: blurry image &amp; DATA_ERRORs (class 0xa097 mthd 0x2380/0x238c)</li>
</ul>
<h2>Changes</h2>
<li>Removed the EGL loader from the Linux SCons build.</li>
</div>
</body>
</html>

View File

@@ -1,134 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 11.0.1 Release Notes / September 26, 2015</h1>
<p>
Mesa 11.0.1 is a bug fix release which fixes bugs found since the 11.0.0 release.
</p>
<p>
Mesa 11.0.1 implements the OpenGL 4.1 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 4.1. OpenGL
4.1 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
6dab262877e12c0546a0e2970c6835a0f217e6d4026ccecb3cd5dd733d1ce867 mesa-11.0.1.tar.gz
43d0dfcd1f1e36f07f8228cd76d90175d3fc74c1ed25d7071794a100a98ef2a6 mesa-11.0.1.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=38109">Bug 38109</a> - i915 driver crashes if too few vertices are submitted (Mesa 7.10.2)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91114">Bug 91114</a> - ES3-CTS.gtf.GL3Tests.shadow.shadow_execution_vert fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91716">Bug 91716</a> - [bisected] piglit.shaders.glsl-vs-int-attrib regresses on 32 bit BYT, HSW, IVB, SNB</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91719">Bug 91719</a> - [SNB,HSW,BYT] dEQP regressions associated with using NIR for vertex shaders</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92009">Bug 92009</a> - ES3-CTS.gtf.GL3Tests.packed_pixels.packed_pixels fails</li>
</ul>
<h2>Changes</h2>
<p>Antia Puentes (2):</p>
<ul>
<li>i965/vec4: Fix saturation errors when coalescing registers</li>
<li>i965/vec4_nir: Load constants as integers</li>
</ul>
<p>Anuj Phogat (1):</p>
<ul>
<li>meta: Abort meta pbo path if TexSubImage need signed unsigned conversion</li>
</ul>
<p>Emil Velikov (2):</p>
<ul>
<li>docs: add sha256 checksums for 11.0.0</li>
<li>Update version to 11.0.1</li>
</ul>
<p>Iago Toral Quiroga (1):</p>
<ul>
<li>mesa: Fix GL_FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE for default framebuffer.</li>
</ul>
<p>Ian Romanick (5):</p>
<ul>
<li>t_dd_dmatmp: Make "count" actually be the count</li>
<li>t_dd_dmatmp: Clean up improper code formatting from previous patch</li>
<li>t_dd_dmatmp: Use '&amp; 3' instead of '% 4' everywhere</li>
<li>t_dd_dmatmp: Pull out common 'count -= count &amp; 3' code</li>
<li>t_dd_dmatmp: Use addition instead of subtraction in loop bounds</li>
</ul>
<p>Ilia Mirkin (6):</p>
<ul>
<li>st/mesa: avoid integer overflows with buffers &gt;= 512MB</li>
<li>nv50, nvc0: fix max texture buffer size to 128M elements</li>
<li>freedreno/a3xx: fix blending of L8 format</li>
<li>nv50,nvc0: detect underlying resource changes and update tic</li>
<li>nv50,nvc0: flush texture cache in presence of coherent bufs</li>
<li>radeonsi: load fmask ptr relative to the resources array</li>
</ul>
<p>Jason Ekstrand (2):</p>
<ul>
<li>nir: Fix a bunch of ralloc parenting errors</li>
<li>i965/vec4: Don't reswizzle hardware registers</li>
</ul>
<p>Jeremy Huddleston (1):</p>
<ul>
<li>configure.ac: Add support to enable read-only text segment on x86.</li>
</ul>
<p>Ray Strode (1):</p>
<ul>
<li>gbm: convert gbm bo format to fourcc format on dma-buf import</li>
</ul>
<p>Tapani Pälli (2):</p>
<ul>
<li>mesa: fix errors when reading depth with glReadPixels</li>
<li>i965: fix textureGrad for cubemaps</li>
</ul>
<p>Ulrich Weigand (1):</p>
<ul>
<li>mesa: Fix texture compression on big-endian systems</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,85 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 11.0.2 Release Notes / September 28, 2015</h1>
<p>
Mesa 11.0.2 is a bug fix release which fixes bugs found since the 11.0.1 release.
</p>
<p>
Mesa 11.0.2 implements the OpenGL 4.1 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 4.1. OpenGL
4.1 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
45170773500d6ae2f9eb93fc85efee69f7c97084411ada4eddf92f78bca56d20 mesa-11.0.2.tar.gz
fce11fb27eb87adf1e620a76455d635c6136dfa49ae58c53b34ef8d0c7b7eae4 mesa-11.0.2.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91582">Bug 91582</a> - [bisected] Regression in DEQP gles2.functional.negative_api.texture.texsubimage2d_neg_offset</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91970">Bug 91970</a> - [BSW regression] dEQP-GLES3.functional.shaders.precision.int.highp_mul_vertex</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92095">Bug 92095</a> - [Regression, bisected] arb_shader_atomic_counters.compiler.builtins.frag</li>
</ul>
<h2>Changes</h2>
<p>Eduardo Lima Mitev (3):</p>
<ul>
<li>mesa: Fix order of format+type and internal format checks for glTexImageXD ops</li>
<li>mesa: Move _mesa_base_tex_format() from teximage to glformats files</li>
<li>mesa: Use the effective internal format instead for validation</li>
</ul>
<p>Emil Velikov (2):</p>
<ul>
<li>docs: add sha256 checksums for 11.0.1</li>
<li>Update version to 11.0.2</li>
</ul>
<p>Kristian Høgsberg Kristensen (1):</p>
<ul>
<li>i965: Respect stride and subreg_offset for ATTR registers</li>
</ul>
<p>Matt Turner (1):</p>
<ul>
<li>glsl: Expose gl_MaxTess{Control,Evaluation}AtomicCounters.</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,185 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 11.0.3 Release Notes / October 10, 2015</h1>
<p>
Mesa 11.0.3 is a bug fix release which fixes bugs found since the 11.0.2 release.
</p>
<p>
Mesa 11.0.3 implements the OpenGL 4.1 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 4.1. OpenGL
4.1 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
c2210e3daecc10ed9fdcea500327652ed6effc2f47c4b9cee63fb08f560d7117 mesa-11.0.3.tar.gz
ab2992eece21adc23c398720ef8c6933cb69ea42e1b2611dc09d031e17e033d6 mesa-11.0.3.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=55552">Bug 55552</a> - Compile errors with --enable-mangling</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=71789">Bug 71789</a> - [r300g] Visuals not found in (default) depth = 24</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91044">Bug 91044</a> - piglit spec/egl_khr_create_context/valid debug flag gles* fail</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91342">Bug 91342</a> - Very dark textures on some objects in indoors environments in Postal 2</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91596">Bug 91596</a> - EGL_KHR_gl_colorspace (v2) causes problem with Android-x86 GUI</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91718">Bug 91718</a> - piglit.spec.arb_shader_image_load_store.invalid causes intermittent GPU HANG</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92072">Bug 92072</a> - Wine breakage since d082c5324 (st/mesa: don't call st_validate_state in BlitFramebuffer)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92265">Bug 92265</a> - Black windows in weston after update mesa to 11.0.2-1</li>
</ul>
<h2>Changes</h2>
<p>Brian Paul (1):</p>
<ul>
<li>st/mesa: try PIPE_BIND_RENDER_TARGET when choosing float texture formats</li>
</ul>
<p>Daniel Scharrer (1):</p>
<ul>
<li>mesa: Add abs input modifier to base for POW in ffvertex_prog</li>
</ul>
<p>Emil Velikov (3):</p>
<ul>
<li>docs: add sha256 checksums for 11.0.2</li>
<li>Revert "nouveau: make sure there's always room to emit a fence"</li>
<li>Update version to 11.0.3</li>
</ul>
<p>Francisco Jerez (1):</p>
<ul>
<li>i965/fs: Fix hang on IVB and VLV with image format mismatch.</li>
</ul>
<p>Ian Romanick (1):</p>
<ul>
<li>meta: Handle array textures in scaled MSAA blits</li>
</ul>
<p>Ilia Mirkin (6):</p>
<ul>
<li>nouveau: be more careful about freeing temporary transfer buffers</li>
<li>nouveau: delay deleting buffer with unflushed fence</li>
<li>nouveau: wait to unref the transfer's bo until it's no longer used</li>
<li>nv30: pretend to have packed texture/surface formats</li>
<li>nv30: always go through translate module on big-endian</li>
<li>nouveau: make sure there's always room to emit a fence</li>
</ul>
<p>Jason Ekstrand (1):</p>
<ul>
<li>mesa: Correctly handle GL_BGRA_EXT in ES3 format_and_type checks</li>
</ul>
<p>Kyle Brenneman (3):</p>
<ul>
<li>glx: Fix build errors with --enable-mangling (v2)</li>
<li>mapi: Make _glapi_get_stub work with "gl" or "mgl" prefix.</li>
<li>glx: Don't hard-code the name "libGL.so.1" in driOpenDriver (v3)</li>
</ul>
<p>Leo Liu (1):</p>
<ul>
<li>radeon/vce: fix vui time_scale zero error</li>
</ul>
<p>Marek Olšák (21):</p>
<ul>
<li>st/mesa: fix front buffer regression after dropping st_validate_state in Blit</li>
<li>radeonsi: handle index buffer alloc failures</li>
<li>radeonsi: handle constant buffer alloc failures</li>
<li>gallium/radeon: handle buffer_map staging buffer failures better</li>
<li>gallium/radeon: handle buffer alloc failures in r600_draw_rectangle</li>
<li>gallium/radeon: add a fail path for depth MSAA texture readback</li>
<li>radeonsi: report alloc failure from si_shader_binary_read</li>
<li>radeonsi: add malloc fail paths to si_create_shader_state</li>
<li>radeonsi: skip drawing if the tess factor ring allocation fails</li>
<li>radeonsi: skip drawing if GS ring allocations fail</li>
<li>radeonsi: handle shader precompile failures</li>
<li>radeonsi: handle fixed-func TCS shader create failure</li>
<li>radeonsi: skip drawing if VS, TCS, TES, GS fail to compile or upload</li>
<li>radeonsi: skip drawing if PS fails to compile or upload</li>
<li>radeonsi: skip drawing if updating the scratch buffer fails</li>
<li>radeonsi: don't forget to update scratch relocations for LS, HS, ES shaders</li>
<li>radeonsi: handle dummy constant buffer allocation failure</li>
<li>gallium/u_blitter: handle allocation failures</li>
<li>radeonsi: add scratch buffer to the buffer list when it's re-allocated</li>
<li>st/dri: don't use _ctx in client_wait_sync</li>
<li>egl/dri2: don't require a context for ClientWaitSync (v2)</li>
</ul>
<p>Matthew Waters (1):</p>
<ul>
<li>egl: rework handling EGL_CONTEXT_FLAGS</li>
</ul>
<p>Michel Dänzer (1):</p>
<ul>
<li>st/dri: Use packed RGB formats</li>
</ul>
<p>Roland Scheidegger (1):</p>
<ul>
<li>mesa: fix mipmap generation for immutable, compressed textures</li>
</ul>
<p>Tom Stellard (3):</p>
<ul>
<li>gallium/radeon: Use call_once() when initailizing LLVM targets</li>
<li>gallivm: Allow drivers and state trackers to initialize gallivm LLVM targets v2</li>
<li>radeon/llvm: Initialize gallivm targets when initializing the AMDGPU target v2</li>
</ul>
<p>Varad Gautam (1):</p>
<ul>
<li>egl: restore surface type before linking config to its display</li>
</ul>
<p>Ville Syrjälä (3):</p>
<ul>
<li>i830: Fix collision between I830_UPLOAD_RASTER_RULES and I830_UPLOAD_TEX(0)</li>
<li>i915: Fix texcoord vs. varying collision in fragment programs</li>
<li>i915: Remember to call intel_prepare_render() before blitting</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,168 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 11.0.4 Release Notes / October 24, 2015</h1>
<p>
Mesa 11.0.4 is a bug fix release which fixes bugs found since the 11.0.3 release.
</p>
<p>
Mesa 11.0.4 implements the OpenGL 4.1 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 4.1. OpenGL
4.1 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
ed412ca6a46d1bd055120e5c12806c15419ae8c4dd6d3f6ea20a83091d5c78bf mesa-11.0.4.tar.gz
40201bf7fc6fa12a6d9edfe870b41eb4dd6669154e3c42c48a96f70805f5483d mesa-11.0.4.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86281">Bug 86281</a> - brw_meta_fast_clear (brw=brw&#64;entry=0x7fffd4097a08, fb=fb&#64;entry=0x7fffd40fa900, buffers=buffers&#64;entry=2, partial_clear=partial_clear&#64;entry=false)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86720">Bug 86720</a> - [radeon] Europa Universalis 4 freezing during game start (10.3.3+, still broken on 11.0.2)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91788">Bug 91788</a> - [HSW Regression] Synmark2_v6 Multithread performance case FPS reduced by 36%</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92304">Bug 92304</a> - [cts] cts.shaders.negative conformance tests fail</li>
</ul>
<h2>Changes</h2>
<p>Alejandro Piñeiro (2):</p>
<ul>
<li>i965/vec4: check writemask when bailing out at register coalesce</li>
<li>i965/vec4: fill src_reg type using the constructor type parameter</li>
</ul>
<p>Brian Paul (2):</p>
<ul>
<li>vbo: fix incorrect switch statement in init_mat_currval()</li>
<li>mesa: fix incorrect opcode in save_BlendFunci()</li>
</ul>
<p>Chih-Wei Huang (3):</p>
<ul>
<li>mesa: android: Fix the incorrect path of sse_minmax.c</li>
<li>nv50/ir: use C++11 standard std::unordered_map if possible</li>
<li>nv30: include the header of ffs prototype</li>
</ul>
<p>Chris Wilson (1):</p>
<ul>
<li>i965: Remove early release of DRI2 miptree</li>
</ul>
<p>Dave Airlie (1):</p>
<ul>
<li>mesa/uniforms: fix get_uniform for doubles (v2)</li>
</ul>
<p>Emil Velikov (1):</p>
<ul>
<li>docs: add sha256 checksums for 11.0.3</li>
</ul>
<p>Francisco Jerez (5):</p>
<ul>
<li>i965: Don't tell the hardware about our UAV access.</li>
<li>mesa: Expose function to calculate whether a shader image unit is valid.</li>
<li>mesa: Skip redundant texture completeness checking during image validation.</li>
<li>i965: Use _mesa_is_image_unit_valid() instead of gl_image_unit::_Valid.</li>
<li>mesa: Get rid of texture-dependent image unit derived state.</li>
</ul>
<p>Ian Romanick (8):</p>
<ul>
<li>glsl: Allow built-in functions as constant expressions in OpenGL ES 1.00</li>
<li>ff_fragment_shader: Use binding to set the sampler unit</li>
<li>glsl/linker: Use constant_initializer instead of constant_value to initialize uniforms</li>
<li>glsl: Use constant_initializer instead of constant_value to determine whether to keep an unused uniform</li>
<li>glsl: Only set ir_variable::constant_value for const-decorated variables</li>
<li>glsl: Restrict initializers for global variables to constant expression in ES</li>
<li>glsl: Add method to determine whether an expression contains the sequence operator</li>
<li>glsl: In later GLSL versions, sequence operator is cannot be a constant expression</li>
</ul>
<p>Ilia Mirkin (1):</p>
<ul>
<li>nouveau: make sure there's always room to emit a fence</li>
</ul>
<p>Indrajit Das (1):</p>
<ul>
<li>st/va: Used correct parameter to derive the value of the "h" variable in vlVaCreateImage</li>
</ul>
<p>Jonathan Gray (1):</p>
<ul>
<li>configure.ac: ensure RM is set</li>
</ul>
<p>Krzysztof Sobiecki (1):</p>
<ul>
<li>st/fbo: use pipe_surface_release instead of pipe_surface_reference</li>
</ul>
<p>Leo Liu (1):</p>
<ul>
<li>st/omx/dec/h264: fix field picture type 0 poc disorder</li>
</ul>
<p>Marek Olšák (3):</p>
<ul>
<li>st/mesa: fix clip state dependencies</li>
<li>radeonsi: fix a GS copy shader leak</li>
<li>gallium: add PIPE_SHADER_CAP_MAX_UNROLL_ITERATIONS_HINT</li>
</ul>
<p>Nicolai Hähnle (1):</p>
<ul>
<li>u_vbuf: fix vb slot assignment for translated buffers</li>
</ul>
<p>Rob Clark (1):</p>
<ul>
<li>freedreno/a3xx: cache-flush is needed after MEM_WRITE</li>
</ul>
<p>Tapani Pälli (3):</p>
<ul>
<li>mesa: add GL_UNSIGNED_INT_24_8 to _mesa_pack_depth_span</li>
<li>mesa: Set api prefix to version string when overriding version</li>
<li>mesa: fix ARRAY_SIZE query for GetProgramResourceiv</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,172 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 11.0.5 Release Notes / November 11, 2015</h1>
<p>
Mesa 11.0.5 is a bug fix release which fixes bugs found since the 11.0.4 release.
</p>
<p>
Mesa 11.0.5 implements the OpenGL 4.1 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 4.1. OpenGL
4.1 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
8495ef5c06f7f726452462b7d408a5b40048373ff908f2283a3b4d1f49b45ee6 mesa-11.0.5.tar.gz
9c255a2a6695fcc6ef4a279e1df0aeaf417dc142f39ee59dfb533d80494bb67a mesa-11.0.5.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91993">Bug 91993</a> - Graphical glitch in Astromenace (open-source game).</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92214">Bug 92214</a> - Flightgear crashes during splashboot with R600 driver, LLVM 3.7.0 and mesa 11.0.2</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92437">Bug 92437</a> - osmesa: Expose GL entry points for Windows build, via .def file</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92476">Bug 92476</a> - [cts] ES2-CTS.gtf.GL2ExtensionTests.egl_image.egl_image fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92623">Bug 92623</a> - Differences in prog_data ignored when caching fragment programs (causes hangs)</li>
</ul>
<h2>Changes</h2>
<p>Alex Deucher (1):</p>
<ul>
<li>radeon/uvd: don't expose HEVC on old UVD hw (v3)</li>
</ul>
<p>Ben Widawsky (1):</p>
<ul>
<li>i965/skl: Add GT4 PCI IDs</li>
</ul>
<p>Emil Velikov (4):</p>
<ul>
<li>docs: add sha256 checksums for 11.0.4</li>
<li>cherry-ignore: ignore a possible wrong nomination</li>
<li>Revert "mesa/glformats: Undo code changes from _mesa_base_tex_format() move"</li>
<li>Update version to 11.0.5</li>
</ul>
<p>Emmanuel Gil Peyrot (1):</p>
<ul>
<li>gbm.h: Add a missing stddef.h include for size_t.</li>
</ul>
<p>Eric Anholt (1):</p>
<ul>
<li>vc4: When the create ioctl fails, free our cache and try again.</li>
</ul>
<p>Ian Romanick (1):</p>
<ul>
<li>i965: Fix is-renderable check in intel_image_target_renderbuffer_storage</li>
</ul>
<p>Ilia Mirkin (3):</p>
<ul>
<li>nvc0: respect edgeflag attribute width</li>
<li>nouveau: set MaxDrawBuffers to the same value as MaxColorAttachments</li>
<li>nouveau: relax fence emit space assert</li>
</ul>
<p>Ivan Kalvachev (1):</p>
<ul>
<li>r600g: Fix special negative immediate constants when using ABS modifier.</li>
</ul>
<p>Jason Ekstrand (2):</p>
<ul>
<li>nir/lower_vec_to_movs: Pass the shader around directly</li>
<li>nir: Report progress from lower_vec_to_movs().</li>
</ul>
<p>Jose Fonseca (2):</p>
<ul>
<li>gallivm: Translate all util_cpu_caps bits to LLVM attributes.</li>
<li>gallivm: Explicitly disable unsupported CPU features.</li>
</ul>
<p>Julien Isorce (4):</p>
<ul>
<li>st/va: pass picture desc to begin and decode</li>
<li>nvc0: fix crash when nv50_miptree_from_handle fails</li>
<li>st/va: do not destroy old buffer when new one failed</li>
<li>st/va: add more errors checks in vlVaBufferSetNumElements and vlVaMapBuffer</li>
</ul>
<p>Kenneth Graunke (6):</p>
<ul>
<li>i965: Fix missing BRW_NEW_*_PROG_DATA flagging caused by cache reuse.</li>
<li>nir: Report progress from nir_split_var_copies().</li>
<li>nir: Properly invalidate metadata in nir_split_var_copies().</li>
<li>nir: Properly invalidate metadata in nir_opt_copy_prop().</li>
<li>nir: Properly invalidate metadata in nir_lower_vec_to_movs().</li>
<li>nir: Properly invalidate metadata in nir_opt_remove_phis().</li>
</ul>
<p>Marek Olšák (1):</p>
<ul>
<li>radeonsi: add register definitions for Stoney</li>
</ul>
<p>Nanley Chery (1):</p>
<ul>
<li>mesa/glformats: Undo code changes from _mesa_base_tex_format() move</li>
</ul>
<p>Nicolai Hähnle (1):</p>
<ul>
<li>st/mesa: fix mipmap generation for immutable textures with incomplete pyramids</li>
</ul>
<p>Nigel Stewart (1):</p>
<ul>
<li>osmesa: Expose GL entry points for Windows build via DEF file.</li>
</ul>
<p>Roland Scheidegger (1):</p>
<ul>
<li>gallivm: disable f16c when not using AVX</li>
</ul>
<p>Samuel Li (2):</p>
<ul>
<li>radeonsi: add support for Stoney asics (v3)</li>
<li>radeonsi: add Stoney pci ids</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,145 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 11.0.6 Release Notes / November 21, 2015</h1>
<p>
Mesa 11.0.6 is a bug fix release which fixes bugs found since the 11.0.5 release.
</p>
<p>
Mesa 11.0.6 implements the OpenGL 4.1 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 4.1. OpenGL
4.1 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
4bdf054af66ebabf3eca0616f9f5e44c2f234695661b570261c391bc2f4f7482 mesa-11.0.6.tar.gz
8340e64cdc91999840404c211496f3de38e7b4cb38db34e2f72f1642c5134760 mesa-11.0.6.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91780">Bug 91780</a> - Rendering issues with geometry shader</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92588">Bug 92588</a> - [HSW,BDW,BSW,SKL-Y][GLES 3.1 CTS] ES31-CTS.arrays_of_arrays.InteractionFunctionCalls2 - assert</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92738">Bug 92738</a> - Randon R7 240 doesn't work on 16KiB page size platform</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92860">Bug 92860</a> - [radeonsi][bisected] st/mesa: implement ARB_copy_image - Corruption in ARK Survival Evolved</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92900">Bug 92900</a> - [regression bisected] About 700 piglit regressions is what could go wrong</li>
</ul>
<h2>Changes</h2>
<p>Alex Deucher (1):</p>
<ul>
<li>radeonsi: enable optimal raster config setting for fiji (v2)</li>
</ul>
<p>Ben Widawsky (1):</p>
<ul>
<li>i965/skl/gt4: Fix URB programming restriction.</li>
</ul>
<p>Boyuan Zhang (2):</p>
<ul>
<li>st/vaapi: fix vaapi VC-1 simple/main corruption v2</li>
<li>radeon/uvd: fix VC-1 simple/main profile decode v2</li>
</ul>
<p>Dave Airlie (1):</p>
<ul>
<li>r600: initialised PGM_RESOURCES_2 for ES/GS</li>
</ul>
<p>Emil Velikov (4):</p>
<ul>
<li>docs: add sha256 checksums for 11.0.5</li>
<li>cherry-ignore: add the swrast front buffer support</li>
<li>automake: use static llvm for make distcheck</li>
<li>Update version to 11.0.6</li>
</ul>
<p>Eric Anholt (3):</p>
<ul>
<li>vc4: Return GL_OUT_OF_MEMORY when buffer allocation fails.</li>
<li>vc4: Return NULL when we can't make our shadow for a sampler view.</li>
<li>vc4: Add support for nir_op_uge, using the carry bit on QPU_A_SUB.</li>
</ul>
<p>Ian Romanick (2):</p>
<ul>
<li>meta/generate_mipmap: Don't leak the sampler object</li>
<li>meta/generate_mipmap: Only modify the draw framebuffer binding in fallback_required</li>
</ul>
<p>Ilia Mirkin (2):</p>
<ul>
<li>mesa/copyimage: allow width/height to not be multiples of block</li>
<li>nouveau: don't expose HEVC decoding support</li>
</ul>
<p>Jason Ekstrand (1):</p>
<ul>
<li>nir/vars_to_ssa: Rework copy set handling in lower_copies_to_load_store</li>
</ul>
<p>Kenneth Graunke (1):</p>
<ul>
<li>glsl: Allow implicit int -&gt; uint conversions for the % operator.</li>
</ul>
<p>Marek Olšák (1):</p>
<ul>
<li>radeonsi: initialize SX_PS_DOWNCONVERT to 0 on Stoney</li>
</ul>
<p>Michel Dänzer (1):</p>
<ul>
<li>winsys/radeon: Use CPU page size instead of hardcoding 4096 bytes v3</li>
</ul>
<p>Oded Gabbay (1):</p>
<ul>
<li>llvmpipe: use simple coeffs calc for 128bit vectors</li>
</ul>
<p>Roland Scheidegger (2):</p>
<ul>
<li>radeon: fix bgrx8/xrgb8 blits</li>
<li>r200: fix bgrx8/xrgb8 blits</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,154 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 11.0.7 Release Notes / December 9, 2015</h1>
<p>
Mesa 11.0.7 is a bug fix release which fixes bugs found since the 11.0.6 release.
</p>
<p>
Mesa 11.0.7 implements the OpenGL 4.1 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 4.1. OpenGL
4.1 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
07c27004ff68b288097d17b2faa7bdf15ec73c96b7e6c9835266e544adf0a62f mesa-11.0.7.tar.gz
e7e90a332ede6c8fd08eff90786a3fd1605a4e62ebf3a9b514047838194538cb mesa-11.0.7.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90348">Bug 90348</a> - Spilling failure of b96 merged value</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92363">Bug 92363</a> - [BSW/BDW] ogles1conform Gets test fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92438">Bug 92438</a> - Segfault in pushbuf_kref when running the android emulator (qemu) on nv50</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93110">Bug 93110</a> - [NVE4] textureSize() and textureQueryLevels() uses a texture bound during the previous draw call</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93126">Bug 93126</a> - wrongly claim supporting GL_EXT_texture_rg</li>
</ul>
<h2>Changes</h2>
<p>Chris Wilson (1):</p>
<ul>
<li>meta: Compute correct buffer size with SkipRows/SkipPixels</li>
</ul>
<p>Daniel Stone (1):</p>
<ul>
<li>egl/wayland: Ignore rects from SwapBuffersWithDamage</li>
</ul>
<p>Dave Airlie (4):</p>
<ul>
<li>texgetimage: consolidate 1D array handling code.</li>
<li>r600: geometry shader gsvs itemsize workaround</li>
<li>r600: rv670 use at least 16es/gs threads</li>
<li>r600: workaround empty geom shader.</li>
</ul>
<p>Emil Velikov (4):</p>
<ul>
<li>docs: add sha256 checksums for 11.0.6</li>
<li>get-pick-list.sh: Require explicit "11.0" for nominating stable patches</li>
<li>mesa; add get-extra-pick-list.sh script into bin/</li>
<li>Update version to 11.0.7</li>
</ul>
<p>François Tigeot (1):</p>
<ul>
<li>xmlconfig: Add support for DragonFly</li>
</ul>
<p>Ian Romanick (22):</p>
<ul>
<li>mesa: Make bind_vertex_buffer avilable outside varray.c</li>
<li>mesa: Refactor update_array_format to make _mesa_update_array_format_public</li>
<li>mesa: Refactor enable_vertex_array_attrib to make _mesa_enable_vertex_array_attrib</li>
<li>i965: Pass brw_context instead of gl_context to brw_draw_rectlist</li>
<li>i965: Use DSA functions for VBOs in brw_meta_fast_clear</li>
<li>i965: Use internal functions for buffer object access</li>
<li>i965: Don't pollute the buffer object namespace in brw_meta_fast_clear</li>
<li>meta: Use DSA functions for PBO in create_texture_for_pbo</li>
<li>meta: Use _mesa_NamedBufferData and _mesa_NamedBufferSubData for users of _mesa_meta_setup_vertex_objects</li>
<li>i965: Use _mesa_NamedBufferSubData for users of _mesa_meta_setup_vertex_objects</li>
<li>meta: Don't leave the VBO bound after _mesa_meta_setup_vertex_objects</li>
<li>meta: Track VBO using gl_buffer_object instead of GL API object handle</li>
<li>meta: Use DSA functions for VBOs in _mesa_meta_setup_vertex_objects</li>
<li>meta: Use internal functions for buffer object and VAO access</li>
<li>meta: Don't pollute the buffer object namespace in _mesa_meta_setup_vertex_objects</li>
<li>meta: Partially convert _mesa_meta_DrawTex to DSA</li>
<li>meta: Track VBO using gl_buffer_object instead of GL API object handle in _mesa_meta_DrawTex</li>
<li>meta: Use internal functions for buffer object and VAO access in _mesa_meta_DrawTex</li>
<li>meta: Don't pollute the buffer object namespace in _mesa_meta_DrawTex</li>
<li>meta/TexSubImage: Don't pollute the buffer object namespace</li>
<li>meta/generate_mipmap: Don't leak the framebuffer object</li>
<li>glsl: Fix off-by-one error in array size check assertion</li>
</ul>
<p>Ilia Mirkin (7):</p>
<ul>
<li>nvc0/ir: actually emit AFETCH on kepler</li>
<li>nir: fix typo in idiv lowering, causing large-udiv-udiv failures</li>
<li>nouveau: use the buffer usage to determine placement when no binding</li>
<li>nv50,nvc0: properly handle buffer storage invalidation on dsa buffer</li>
<li>nv50/ir: fix (un)spilling of 3-wide results</li>
<li>mesa: support GL_RED/GL_RG in ES2 contexts when driver support exists</li>
<li>nvc0/ir: start offset at texBindBase for txq, like regular texturing</li>
</ul>
<p>Jonathan Gray (1):</p>
<ul>
<li>automake: fix some occurrences of hardcoded -ldl and -lpthread</li>
</ul>
<p>Leo Liu (1):</p>
<ul>
<li>radeon/vce: disable Stoney VCE for 11.0</li>
</ul>
<p>Marta Lofstedt (1):</p>
<ul>
<li>gles2: Update gl2ext.h to revision: 32120</li>
</ul>
<p>Oded Gabbay (1):</p>
<ul>
<li>llvmpipe: disable VSX in ppc due to LLVM PPC bug</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,200 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 11.0.8 Release Notes / December 9, 2015</h1>
<p>
Mesa 11.0.8 is a bug fix release which fixes bugs found since the 11.0.7 release.
</p>
<p>
Mesa 11.0.8 implements the OpenGL 4.1 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 4.1. OpenGL
4.1 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
ab9db87b54d7525e4b611b82577ea9a9eae55927558df57b190059d5ecd9406f mesa-11.0.8.tar.gz
5696e4730518b6805d2ed5def393c4293f425a2c2c01bd5ed4bdd7ad62f7ad75 mesa-11.0.8.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91806">Bug 91806</a> - configure does not test whether assembler supports sse4.1</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92849">Bug 92849</a> - [IVB HSW BDW] piglit image load/store load-from-cleared-image.shader_test fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92909">Bug 92909</a> - Offset/alignment issue with layout std140 and vec3</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93004">Bug 93004</a> - Guild Wars 2 crash on nouveau DX11 cards</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93215">Bug 93215</a> - [Regression bisected] Ogles1conform Automatic mipmap generation test is fail</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93266">Bug 93266</a> - gl_arb_shading_language_420pack does not allow binding of image variables</li>
</ul>
<h2>Changes</h2>
<p>Boyuan Zhang (1):</p>
<ul>
<li>radeon/uvd: uv pitch separation for stoney</li>
</ul>
<p>Dave Airlie (9):</p>
<ul>
<li>r600: do SQ flush ES ring rolling workaround</li>
<li>r600: SMX returns CONTEXT_DONE early workaround</li>
<li>r600/shader: split address get out to a function.</li>
<li>r600/shader: add utility functions to do single slot arithmatic</li>
<li>r600g: fix geom shader input indirect indexing.</li>
<li>r600: handle geometry dynamic input array index</li>
<li>radeonsi: handle doubles in lds load path.</li>
<li>mesa/varray: set double arrays to non-normalised.</li>
<li>mesa/shader: return correct attribute location for double matrix arrays</li>
</ul>
<p>Emil Velikov (8):</p>
<ul>
<li>docs: add sha256 checksums for 11.0.7</li>
<li>cherry-ignore: don't pick a specific i965 formats patch</li>
<li>Revert "i965/nir: Remove unused indirect handling"</li>
<li>Revert "i965/state: Get rid of dword_pitch arguments to buffer functions"</li>
<li>Revert "i965/vec4: Use a stride of 1 and byte offsets for UBOs"</li>
<li>Revert "i965/fs: Use a stride of 1 and byte offsets for UBOs"</li>
<li>Revert "i965/vec4: Use byte offsets for UBO pulls on Sandy Bridge"</li>
<li>Update version to 11.0.8</li>
</ul>
<p>Francisco Jerez (1):</p>
<ul>
<li>i965: Resolve color and flush for all active shader images in intel_update_state().</li>
</ul>
<p>Ian Romanick (1):</p>
<ul>
<li>meta/generate_mipmap: Work-around GLES 1.x problem with GL_DRAW_FRAMEBUFFER</li>
</ul>
<p>Ilia Mirkin (17):</p>
<ul>
<li>freedreno/a4xx: support lod_bias</li>
<li>freedreno/a4xx: fix 5_5_5_1 texture sampler format</li>
<li>freedreno/a4xx: point regid to "red" even for alpha-only rb formats</li>
<li>nvc0/ir: fold postfactor into immediate</li>
<li>nv50/ir: deal with loops with no breaks</li>
<li>nv50/ir: the mad source might not have a defining instruction</li>
<li>nv50/ir: fix instruction permutation logic</li>
<li>nv50/ir: don't forget to mark flagsDef on cvt in txb lowering</li>
<li>nv50/ir: fix DCE to not generate 96-bit loads</li>
<li>nv50/ir: avoid looking at uninitialized srcMods entries</li>
<li>gk110/ir: fix imul hi emission with limm arg</li>
<li>gk104/ir: sampler doesn't matter for txf</li>
<li>gk110/ir: fix imad sat/hi flag emission for immediate args</li>
<li>nv50/ir: fix cutoff for using r63 vs r127 when replacing zero</li>
<li>nv50/ir: can't have predication and immediates</li>
<li>glsl: assign varying locations to tess shaders when doing SSO</li>
<li>ttn: add TEX2 support</li>
</ul>
<p>Jason Ekstrand (5):</p>
<ul>
<li>i965/vec4: Use byte offsets for UBO pulls on Sandy Bridge</li>
<li>i965/fs: Use a stride of 1 and byte offsets for UBOs</li>
<li>i965/vec4: Use a stride of 1 and byte offsets for UBOs</li>
<li>i965/state: Get rid of dword_pitch arguments to buffer functions</li>
<li>i965/nir: Remove unused indirect handling</li>
</ul>
<p>Jonathan Gray (2):</p>
<ul>
<li>configure.ac: use pkg-config for libelf</li>
<li>configure: check for python2.7 for PYTHON2</li>
</ul>
<p>Kenneth Graunke (2):</p>
<ul>
<li>i965: Fix fragment shader struct inputs.</li>
<li>i965: Fix scalar vertex shader struct outputs.</li>
</ul>
<p>Marek Olšák (8):</p>
<ul>
<li>radeonsi: fix occlusion queries on Fiji</li>
<li>radeonsi: fix a hang due to uninitialized border color registers</li>
<li>radeonsi: fix Fiji for LLVM &lt;= 3.7</li>
<li>radeonsi: don't call of u_prims_for_vertices for patches and rectangles</li>
<li>radeonsi: apply the streamout workaround to Fiji as well</li>
<li>gallium/radeon: fix Hyper-Z hangs by programming PA_SC_MODE_CNTL_1 correctly</li>
<li>tgsi/scan: add flag colors_written</li>
<li>r600g: write all MRTs only if there is exactly one output (fixes a hang)</li>
</ul>
<p>Matt Turner (1):</p>
<ul>
<li>glsl: Allow binding of image variables with 420pack.</li>
</ul>
<p>Neil Roberts (2):</p>
<ul>
<li>i965: Add MESA_FORMAT_B8G8R8X8_SRGB to brw_format_for_mesa_format</li>
<li>i965: Add B8G8R8X8_SRGB to the alpha format override</li>
</ul>
<p>Oded Gabbay (1):</p>
<ul>
<li>configura.ac: fix test for SSE4.1 assembler support</li>
</ul>
<p>Patrick Rudolph (2):</p>
<ul>
<li>nv50,nvc0: fix use-after-free when vertex buffers are unbound</li>
<li>gallium/util: return correct number of bound vertex buffers</li>
</ul>
<p>Samuel Pitoiset (1):</p>
<ul>
<li>nvc0: free memory allocated by the prog which reads MP perf counters</li>
</ul>
<p>Tapani Pälli (1):</p>
<ul>
<li>i965: use _Shader to get fragment program when updating surface state</li>
</ul>
<p>Tom Stellard (2):</p>
<ul>
<li>radeonsi: Rename si_shader::ls_rsrc{1,2} to si_shader::rsrc{1,2}</li>
<li>radeonsi/compute: Use the compiler's COMPUTE_PGM_RSRC* register values</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,127 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 11.0.9 Release Notes / January 22, 2016</h1>
<p>
Mesa 11.0.9 is a bug fix release which fixes bugs found since the 11.0.8 release.
</p>
<p>
Mesa 11.0.9 implements the OpenGL 4.1 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 4.1. OpenGL
4.1 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
1597c2e983f476f98efdd6cd58b5298896d18479ff542bdeff28b98b129ede05 mesa-11.0.9.tar.gz
a1262ff1c66a16ccf341186cf0e57b306b8589eb2cc5ce92ffb6788ab01d2b01 mesa-11.0.9.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91596">Bug 91596</a> - EGL_KHR_gl_colorspace (v2) causes problem with Android-x86 GUI</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92229">Bug 92229</a> - [APITRACE] SOMA have serious graphical errors</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93257">Bug 93257</a> - [SKL, bisected] ASTC dEQP tests segfault</li>
</ul>
<h2>Changes</h2>
<p>Emil Velikov (6):</p>
<ul>
<li>docs: add sha256 checksums for 11.0.8</li>
<li>cherry-ignore: add patch already in branch</li>
<li>cherry-ignore: add the dri3 glx null check patch</li>
<li>i915: correctly parse/set the context flags</li>
<li>egl/dri2: expose srgb configs when KHR_gl_colorspace is available</li>
<li>Update version to 11.0.9</li>
</ul>
<p>Grazvydas Ignotas (1):</p>
<ul>
<li>r600: fix constant buffer size programming</li>
</ul>
<p>Ilia Mirkin (5):</p>
<ul>
<li>nvc0: don't forget to reset VTX_TMP bufctx slot after blit completion</li>
<li>nv50/ir: float(s32 &amp; 0xff) = float(u8), not s8</li>
<li>nv50,nvc0: make sure there's pushbuf space and that we ref the bo early</li>
<li>nv50,nvc0: fix crash when increasing bsp bo size for h264</li>
<li>nvc0: scale up inter_bo size so that it's 16M for a 4K video</li>
</ul>
<p>Kenneth Graunke (2):</p>
<ul>
<li>ralloc: Fix ralloc_adopt() to the old context's last child's parent.</li>
<li>nvc0: Set winding order regardless of domain.</li>
</ul>
<p>Marek Olšák (1):</p>
<ul>
<li>radeonsi: don't miss changes to SPI_TMPRING_SIZE</li>
</ul>
<p>Miklós Máté (1):</p>
<ul>
<li>mesa: Don't leak ATIfs instructions in DeleteFragmentShader</li>
</ul>
<p>Neil Roberts (1):</p>
<ul>
<li>i965: Fix crash when calling glViewport with no surface bound</li>
</ul>
<p>Nicolai Hähnle (6):</p>
<ul>
<li>gallium/radeon: only dispose locally created target machine in radeon_llvm_compile</li>
<li>mesa/bufferobj: make _mesa_delete_buffer_object externally accessible</li>
<li>st/mesa: use _mesa_delete_buffer_object</li>
<li>radeon: use _mesa_delete_buffer_object</li>
<li>i915: use _mesa_delete_buffer_object</li>
<li>i965: use _mesa_delete_buffer_object</li>
</ul>
<p>Oded Gabbay (1):</p>
<ul>
<li>llvmpipe: use vpkswss when dst is signed</li>
</ul>
<p>Rob Herring (1):</p>
<ul>
<li>freedreno/ir3: fix 32-bit builds with pointer-to-int-cast error enabled</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,281 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 11.1.0 Release Notes / 15 December 2015</h1>
<p>
Mesa 11.1.0 is a new development release.
People who are concerned with stability and reliability should stick
with a previous release or wait for Mesa 11.1.1.
</p>
<p>
Mesa 11.1.0 implements the OpenGL 4.1 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 4.1. OpenGL
4.1 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
e3bc44be4df5e4dc728dfda7b55b1aaeadfce36eca6a367b76cc07598070cb2d mesa-11.1.0.tar.gz
9befe03b04223eb1ede177fa8cac001e2850292c8c12a3ec9929106afad9cf1f mesa-11.1.0.tar.xz
</pre>
<h2>New features</h2>
<p>
Note: some of the new features are only available with certain drivers.
</p>
<ul>
<li>OpenGL 3.1 support on freedreno (a3xx, a4xx)</li>
<li>OpenGL 3.3 support for VMware guest VM driver (supported by Workstation 12
and Fusion 8).
<li>GL_AMD_performance_monitor on nv50</li>
<li>GL_ARB_arrays_of_arrays on i965</li>
<li>GL_ARB_blend_func_extended on freedreno (a3xx)</li>
<li>GL_ARB_clear_texture on nv50, nvc0</li>
<li>GL_ARB_clip_control on freedreno/a4xx</li>
<li>GL_ARB_copy_image on nv50, nvc0, radeonsi</li>
<li>GL_ARB_depth_clamp on freedreno/a4xx</li>
<li>GL_ARB_fragment_layer_viewport on i965 (gen6+)</li>
<li>GL_ARB_gpu_shader_fp64 on r600 for Cypress/Cayman/Aruba chips</li>
<li>GL_ARB_gpu_shader5 on r600 for Evergreen and later chips</li>
<li>GL_ARB_seamless_cubemap_per_texture on freedreno/a4xx</li>
<li>GL_ARB_shader_clock on i965 (gen7+)</li>
<li>GL_ARB_shader_stencil_export on i965 (gen9+)</li>
<li>GL_ARB_shader_storage_buffer_object on i965</li>
<li>GL_ARB_shader_texture_image_samples on i965, nv50, nvc0, r600, radeonsi</li>
<li>GL_ARB_texture_barrier / GL_NV_texture_barrier on i965</li>
<li>GL_ARB_texture_buffer_range on freedreno/a3xx</li>
<li>GL_ARB_texture_compression_bptc on freedreno/a4xx</li>
<li>GL_ARB_texture_query_lod on softpipe</li>
<li>GL_ARB_texture_view on radeonsi and r600 (for evergeen and newer)</li>
<li>GL_ARB_vertex_type_2_10_10_10_rev on freedreno (a3xx, a4xx)</li>
<li>GL_EXT_blend_func_extended on all drivers that support the ARB version</li>
<li>GL_EXT_buffer_storage implemented for when ES 3.1 support is gained</li>
<li>GL_EXT_draw_elements_base_vertex on all drivers</li>
<li>GL_EXT_texture_compression_rgtc / latc on freedreno (a3xx & a4xx)</li>
<li>GL_KHR_debug (GLES)</li>
<li>GL_NV_conditional_render on freedreno</li>
<li>GL_OES_draw_elements_base_vertex on all drivers</li>
<li>EGL_KHR_create_context on softpipe, llvmpipe</li>
<li>EGL_KHR_gl_colorspace on softpipe, llvmpipe</li>
<li>new virgl gallium driver for qemu virtio-gpu</li>
<li>16x multisampling on i965 (gen9+)</li>
<li>GL_EXT_shader_samples_identical on i965.</li>
</ul>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=28130">Bug 28130</a> - vbo: premature flushing breaks GL_LINE_LOOP</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=38109">Bug 38109</a> - i915 driver crashes if too few vertices are submitted (Mesa 7.10.2)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=49779">Bug 49779</a> - Extra line segments in GL_LINE_LOOP</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=55552">Bug 55552</a> - Compile errors with --enable-mangling</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=71789">Bug 71789</a> - [r300g] Visuals not found in (default) depth = 24</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=79783">Bug 79783</a> - Distorted output in obs-studio where other vendors &quot;work&quot;</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=80821">Bug 80821</a> - When LIBGL_ALWAYS_SOFTWARE is set, KHR_create_context is not supported</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=81174">Bug 81174</a> - Gallium: GL_LINE_LOOP broken with more than 512 points</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=83508">Bug 83508</a> - [UBO] Assertion for array of blocks</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=84677">Bug 84677</a> - Triangle disappears with glPolygonMode GL_LINE</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86281">Bug 86281</a> - brw_meta_fast_clear (brw=brw&#64;entry=0x7fffd4097a08, fb=fb&#64;entry=0x7fffd40fa900, buffers=buffers&#64;entry=2, partial_clear=partial_clear&#64;entry=false)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86469">Bug 86469</a> - Unreal Engine demo doesn't run</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86720">Bug 86720</a> - [radeon] Europa Universalis 4 freezing during game start (10.3.3+, still broken on 11.0.2)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=89014">Bug 89014</a> - PIPE_QUERY_GPU_FINISHED is not acting as expected on SI</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90175">Bug 90175</a> - [hsw bisected][PATCH] atomic counters doesn't work for a binding point different to zero</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90348">Bug 90348</a> - Spilling failure of b96 merged value</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90631">Bug 90631</a> - Compilation failure for fragment shader with many branches on Sandy Bridge</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90734">Bug 90734</a> - glBufferSubData is corrupting data when buffer is &gt; 32k</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90887">Bug 90887</a> - PhiMovesPass in register allocator broken</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91044">Bug 91044</a> - piglit spec/egl_khr_create_context/valid debug flag gles* fail</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91114">Bug 91114</a> - ES3-CTS.gtf.GL3Tests.shadow.shadow_execution_vert fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91254">Bug 91254</a> - (regresion) video using VA-API on Intel slow and freeze system with mesa 10.6 or 10.6.1</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91292">Bug 91292</a> - [BDW+] glVertexAttribDivisor not working in combination with glPolygonMode</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91342">Bug 91342</a> - Very dark textures on some objects in indoors environments in Postal 2</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91526">Bug 91526</a> - World of Warcraft (on Wine) has UI corruption with nouveau</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91551">Bug 91551</a> - DXTn compressed normal maps produce severe artifacts on all NV5x and NVDx chipsets</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91596">Bug 91596</a> - EGL_KHR_gl_colorspace (v2) causes problem with Android-x86 GUI</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91716">Bug 91716</a> - [bisected] piglit.shaders.glsl-vs-int-attrib regresses on 32 bit BYT, HSW, IVB, SNB</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91718">Bug 91718</a> - piglit.spec.arb_shader_image_load_store.invalid causes intermittent GPU HANG</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91719">Bug 91719</a> - [SNB,HSW,BYT] dEQP regressions associated with using NIR for vertex shaders</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91726">Bug 91726</a> - R600 asserts in tgsi_cmp/make_src_for_op3</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91780">Bug 91780</a> - Rendering issues with geometry shader</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91785">Bug 91785</a> - make check DispatchSanity_test.GLES31 regression</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91788">Bug 91788</a> - [HSW Regression] Synmark2_v6 Multithread performance case FPS reduced by 36%</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91847">Bug 91847</a> - glGenerateTextureMipmap not working (no errors) unless glActiveTexture(GL_TEXTURE1) is called before</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91857">Bug 91857</a> - Mesa 10.6.3 linker is slow</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91881">Bug 91881</a> - regression: GPU lockups since mesa-11.0.0_rc1 on RV620 (r600) driver</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91890">Bug 91890</a> - [nve7] witcher2: blurry image &amp; DATA_ERRORs (class 0xa097 mthd 0x2380/0x238c)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91898">Bug 91898</a> - src/util/mesa-sha1.c:250:25: fatal error: openssl/sha.h: No such file or directory</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91927">Bug 91927</a> - [SKL] [regression] piglit compressed textures tests fail with kernel upgrade</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91930">Bug 91930</a> - Program with GtkGLArea widget does not redraw</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91970">Bug 91970</a> - [BSW regression] dEQP-GLES3.functional.shaders.precision.int.highp_mul_vertex</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91985">Bug 91985</a> - [regression, bisected] FTBFS with commit f9caabe8f1: R600_UCP_CONST_BUFFER is undefined</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91993">Bug 91993</a> - Graphical glitch in Astromenace (open-source game).</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92009">Bug 92009</a> - ES3-CTS.gtf.GL3Tests.packed_pixels.packed_pixels fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92033">Bug 92033</a> - [SNB,regression,dEQP,bisected] functional.shaders.random tests regressed</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92052">Bug 92052</a> - nir/nir_builder.h:79: error: expected primary-expression before . token</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92054">Bug 92054</a> - make check gbm-symbols-check regression</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92066">Bug 92066</a> - [ILK,G45,regression] New assertion on BRW_MAX_MRF breaks ilk and g45</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92072">Bug 92072</a> - Wine breakage since d082c5324 (st/mesa: don't call st_validate_state in BlitFramebuffer)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92095">Bug 92095</a> - [Regression, bisected] arb_shader_atomic_counters.compiler.builtins.frag</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92122">Bug 92122</a> - [bisected, cts] Regression with Assault Android Cactus</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92124">Bug 92124</a> - shader_query.cpp:841:34: error: strndup was not declared in this scope</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92183">Bug 92183</a> - linker.cpp:3187:46: error: strtok_r was not declared in this scope</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92193">Bug 92193</a> - [SKL] ES2-CTS.gtf.GL2ExtensionTests.compressed_astc_texture.compressed_astc_texture fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92214">Bug 92214</a> - Flightgear crashes during splashboot with R600 driver, LLVM 3.7.0 and mesa 11.0.2</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92221">Bug 92221</a> - Unintended code changes in _mesa_base_tex_format commit</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92265">Bug 92265</a> - Black windows in weston after update mesa to 11.0.2-1</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92304">Bug 92304</a> - [cts] cts.shaders.negative conformance tests fail</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92363">Bug 92363</a> - [BSW/BDW] ogles1conform Gets test fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92437">Bug 92437</a> - osmesa: Expose GL entry points for Windows build, via .def file</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92438">Bug 92438</a> - Segfault in pushbuf_kref when running the android emulator (qemu) on nv50</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92476">Bug 92476</a> - [cts] ES2-CTS.gtf.GL2ExtensionTests.egl_image.egl_image fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92588">Bug 92588</a> - [HSW,BDW,BSW,SKL-Y][GLES 3.1 CTS] ES31-CTS.arrays_of_arrays.InteractionFunctionCalls2 - assert</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92621">Bug 92621</a> - [G965 ILK G45] Regression: 24 piglit regressions in glsl-1.10</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92623">Bug 92623</a> - Differences in prog_data ignored when caching fragment programs (causes hangs)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92634">Bug 92634</a> - gallium's vl_mpeg12_decoder does not work with st/va</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92639">Bug 92639</a> - [Regression bisected] Ogles1conform mustpass.c fail</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92641">Bug 92641</a> - [SKL BSW] [Regression] Ogles1conform userclip.c fail</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92645">Bug 92645</a> - kodi vdpau interop fails since mesa,meta: move gl_texture_object::TargetIndex initializations</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92705">Bug 92705</a> - [clover] fail to build with llvm-svn/clang-svn 3.8</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92709">Bug 92709</a> - &quot;LLVM triggered Diagnostic Handler: unsupported call to function ldexpf in main&quot; when starting race in stuntrally</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92738">Bug 92738</a> - Randon R7 240 doesn't work on 16KiB page size platform</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92744">Bug 92744</a> - [g965 Regression bisected] Performance regression and piglit assertions due to liveness analysis</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92770">Bug 92770</a> - [SNB, regression, dEQP] deqp-gles3.functional.shaders.discard.dynamic_loop_texture</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92824">Bug 92824</a> - [regression, bisected] `make check` dispatch-sanity broken by GL_EXT_buffer_storage</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92849">Bug 92849</a> - [IVB HSW BDW] piglit image load/store load-from-cleared-image.shader_test fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92859">Bug 92859</a> - [regression, bisected] validate_intrinsic_instr: Assertion triggered</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92860">Bug 92860</a> - [radeonsi][bisected] st/mesa: implement ARB_copy_image - Corruption in ARK Survival Evolved</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92900">Bug 92900</a> - [regression bisected] About 700 piglit regressions is what could go wrong</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92909">Bug 92909</a> - Offset/alignment issue with layout std140 and vec3</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92985">Bug 92985</a> - Mac OS X build error &quot;ar: no archive members specified&quot;</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93015">Bug 93015</a> - Tonga Elemental segfault + VM faults since radeon: implement r600_query_hw_get_result via function pointers</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93048">Bug 93048</a> - [CTS regression] mesa af2723 breaks GL Conformance for debug extension</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93063">Bug 93063</a> - drm_helper.h:227:1: error: static declaration of pipe_virgl_create_screen follows non-static declaration</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93091">Bug 93091</a> - [opencl] segfault when running any opencl programs (like clinfo)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93126">Bug 93126</a> - wrongly claim supporting GL_EXT_texture_rg</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93180">Bug 93180</a> - [regression] arb_separate_shader_objects.active sampler conflict fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93235">Bug 93235</a> - [regression] dispatch sanity broken by GetPointerv</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93266">Bug 93266</a> - gl_arb_shading_language_420pack does not allow binding of image variables</li>
</ul>
<h2>Changes</h2>
<li>MPEG4 decoding has been disabled by default in the VAAPI driver</li>
</div>
</body>
</html>

View File

@@ -1,197 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 11.1.1 Release Notes / January 13, 2016</h1>
<p>
Mesa 11.1.1 is a bug fix release which fixes bugs found since the 11.1.0 release.
</p>
<p>
Mesa 11.1.1 implements the OpenGL 4.1 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 4.1. OpenGL
4.1 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
b15089817540ba0bffd0aad323ecf3a8ff6779568451827c7274890b4a269d58 mesa-11.1.1.tar.gz
64db074fc514136b5fb3890111f0d50604db52f0b1e94ba3fcb0fe8668a7fd20 mesa-11.1.1.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91806">Bug 91806</a> - configure does not test whether assembler supports sse4.1</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92229">Bug 92229</a> - [APITRACE] SOMA have serious graphical errors</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92233">Bug 92233</a> - Unigine Heaven 4.0 silhuette run</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93004">Bug 93004</a> - Guild Wars 2 crash on nouveau DX11 cards</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93215">Bug 93215</a> - [Regression bisected] Ogles1conform Automatic mipmap generation test is fail</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93257">Bug 93257</a> - [SKL, bisected] ASTC dEQP tests segfault</li>
</ul>
<h2>Changes</h2>
<p>Brian Paul (1):</p>
<ul>
<li>st/mesa: check state-&gt;mesa in early return check in st_validate_state()</li>
</ul>
<p>Dave Airlie (6):</p>
<ul>
<li>mesa/varray: set double arrays to non-normalised.</li>
<li>mesa/shader: return correct attribute location for double matrix arrays</li>
<li>glsl: pass stage into mark function</li>
<li>glsl/fp64: add helper for dual slot double detection.</li>
<li>glsl: fix count_attribute_slots to allow for different 64-bit handling</li>
<li>glsl: only update doubles inputs for vertex inputs.</li>
</ul>
<p>Emil Velikov (4):</p>
<ul>
<li>docs: add sha256 checksums for 11.0.1</li>
<li>cherry-ignore: drop the "re-enable" DCC on Stoney</li>
<li>cherry-ignore: don't pick a specific i965 formats patch</li>
<li>Update version to 11.1.1</li>
</ul>
<p>Eric Anholt (2):</p>
<ul>
<li>vc4: Warn instead of abort()ing on exec ioctl failures.</li>
<li>vc4: Keep sample mask writes from being reordered after TLB writes</li>
</ul>
<p>Grazvydas Ignotas (1):</p>
<ul>
<li>r600: fix constant buffer size programming</li>
</ul>
<p>Ian Romanick (1):</p>
<ul>
<li>meta/generate_mipmap: Work-around GLES 1.x problem with GL_DRAW_FRAMEBUFFER</li>
</ul>
<p>Ilia Mirkin (9):</p>
<ul>
<li>nv50/ir: can't have predication and immediates</li>
<li>gk104/ir: simplify and fool-proof texbar algorithm</li>
<li>glsl: assign varying locations to tess shaders when doing SSO</li>
<li>glx/dri3: a drawable might not be bound at wait time</li>
<li>nvc0: don't forget to reset VTX_TMP bufctx slot after blit completion</li>
<li>nv50/ir: float(s32 &amp; 0xff) = float(u8), not s8</li>
<li>nv50,nvc0: make sure there's pushbuf space and that we ref the bo early</li>
<li>nv50,nvc0: fix crash when increasing bsp bo size for h264</li>
<li>nvc0: scale up inter_bo size so that it's 16M for a 4K video</li>
</ul>
<p>Jonathan Gray (2):</p>
<ul>
<li>configure.ac: use pkg-config for libelf</li>
<li>configure: check for python2.7 for PYTHON2</li>
</ul>
<p>Kenneth Graunke (5):</p>
<ul>
<li>ralloc: Fix ralloc_adopt() to the old context's last child's parent.</li>
<li>drirc: Disable ARB_blend_func_extended for Heaven 4.0/Valley 1.0.</li>
<li>glsl: Fix varying struct locations when varying packing is disabled.</li>
<li>nvc0: Set winding order regardless of domain.</li>
<li>nir: Add a lower_fdiv option, turn fdiv into fmul/frcp.</li>
</ul>
<p>Marek Olšák (7):</p>
<ul>
<li>tgsi/scan: add flag colors_written</li>
<li>r600g: write all MRTs only if there is exactly one output (fixes a hang)</li>
<li>radeonsi: don't call of u_prims_for_vertices for patches and rectangles</li>
<li>radeonsi: apply the streamout workaround to Fiji as well</li>
<li>gallium/radeon: fix Hyper-Z hangs by programming PA_SC_MODE_CNTL_1 correctly</li>
<li>program: add _mesa_reserve_parameter_storage</li>
<li>st/mesa: fix GLSL uniform updates for glBitmap &amp; glDrawPixels (v2)</li>
</ul>
<p>Mark Janes (1):</p>
<ul>
<li>Add missing platform information for KBL</li>
</ul>
<p>Miklós Máté (1):</p>
<ul>
<li>mesa: Don't leak ATIfs instructions in DeleteFragmentShader</li>
</ul>
<p>Neil Roberts (3):</p>
<ul>
<li>i965: Add MESA_FORMAT_B8G8R8X8_SRGB to brw_format_for_mesa_format</li>
<li>i965: Add B8G8R8X8_SRGB to the alpha format override</li>
<li>i965: Fix crash when calling glViewport with no surface bound</li>
</ul>
<p>Nicolai Hähnle (2):</p>
<ul>
<li>gallium/radeon: only dispose locally created target machine in radeon_llvm_compile</li>
<li>gallium/radeon: fix regression in a number of driver queries</li>
</ul>
<p>Oded Gabbay (1):</p>
<ul>
<li>configura.ac: fix test for SSE4.1 assembler support</li>
</ul>
<p>Patrick Rudolph (2):</p>
<ul>
<li>nv50,nvc0: fix use-after-free when vertex buffers are unbound</li>
<li>gallium/util: return correct number of bound vertex buffers</li>
</ul>
<p>Rob Herring (1):</p>
<ul>
<li>freedreno/ir3: fix 32-bit builds with pointer-to-int-cast error enabled</li>
</ul>
<p>Samuel Pitoiset (3):</p>
<ul>
<li>nvc0: free memory allocated by the prog which reads MP perf counters</li>
<li>nv50,nvc0: free memory allocated by performance metrics</li>
<li>nv50: free memory allocated by the prog which reads MP perf counters</li>
</ul>
<p>Sarah Sharp (1):</p>
<ul>
<li>mesa: Add KBL PCI IDs and platform information.</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,182 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 11.1.2 Release Notes / February 10, 2016</h1>
<p>
Mesa 11.1.2 is a bug fix release which fixes bugs found since the 11.1.1 release.
</p>
<p>
Mesa 11.1.2 implements the OpenGL 4.1 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 4.1. OpenGL
4.1 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
ba0e7462b2936b86e6684c26fbb55519f8d9ad31d13a1c1e1afbe41e73466eea mesa-11.1.2.tar.gz
8f72aead896b340ba0f7a4a474bfaf71681f5d675592aec1cb7ba698e319148b mesa-11.1.2.tar.xz
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91596">Bug 91596</a> - EGL_KHR_gl_colorspace (v2) causes problem with Android-x86 GUI</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93628">Bug 93628</a> - Exception: attempt to use unavailable module DRM when building MesaGL 11.1.0 on windows</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93648">Bug 93648</a> - Random lines being rendered when playing Dolphin (geometry shaders related, w/ apitrace)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93650">Bug 93650</a> - GL_ARB_separate_shader_objects is buggy (PCSX2)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93717">Bug 93717</a> - Meta mipmap generation can corrupt texture state</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93722">Bug 93722</a> - Segfault when compiling shader with a subroutine that takes a parameter</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93731">Bug 93731</a> - glUniformSubroutinesuiv segfaults when subroutine uniform is bound to a specific location</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93761">Bug 93761</a> - A conditional discard in a fragment shader causes no depth writing at all</li>
</ul>
<h2>Changes</h2>
<p>Ben Widawsky (1):</p>
<ul>
<li>i965/bxt: Fix conservative wm thread counts.</li>
</ul>
<p>Dave Airlie (1):</p>
<ul>
<li>glsl: fix subroutine lowering reusing actual parmaters</li>
</ul>
<p>Emil Velikov (6):</p>
<ul>
<li>docs: add sha256 checksums for 11.1.1</li>
<li>cherry-ignore: drop the i965/kbl .num_slices patch</li>
<li>i915: correctly parse/set the context flags</li>
<li>targets/dri: android: use WHOLE static libraries</li>
<li>egl/dri2: expose srgb configs when KHR_gl_colorspace is available</li>
<li>Update version to 11.1.2</li>
</ul>
<p>Eric Anholt (2):</p>
<ul>
<li>vc4: Don't record the seqno of a failed job submit.</li>
<li>vc4: Throttle outstanding rendering after submission.</li>
</ul>
<p>François Tigeot (1):</p>
<ul>
<li>gallium: Add DragonFly support</li>
</ul>
<p>Grazvydas Ignotas (1):</p>
<ul>
<li>r600g: don't leak driver const buffers</li>
</ul>
<p>Ian Romanick (2):</p>
<ul>
<li>meta/blit: Restore GL_DEPTH_STENCIL_TEXTURE_MODE state for GL_TEXTURE_RECTANGLE</li>
<li>meta: Use internal functions to set texture parameters</li>
</ul>
<p>Ilia Mirkin (6):</p>
<ul>
<li>st/mesa: use surface format to generate mipmaps when available</li>
<li>glsl: always compute proper varying type, irrespective of varying packing</li>
<li>nvc0: avoid crashing when there are holes in vertex array bindings</li>
<li>nv50,nvc0: fix buffer clearing to respect engine alignment requirements</li>
<li>nv50/ir: fix false global CSE on instructions with multiple defs</li>
<li>st/mesa: treat a write as a read for range purposes</li>
</ul>
<p>Jason Ekstrand (3):</p>
<ul>
<li>i965/vec4: Use UW type for multiply into accumulator on GEN8+</li>
<li>i965/fs/generator: Take an actual shader stage rather than a string</li>
<li>i965/fs: Always set channel 2 of texture headers in some stages</li>
</ul>
<p>Jose Fonseca (2):</p>
<ul>
<li>scons: Conditionally use DRM module on pipe-loader.</li>
<li>pipe-loader: Fix PATH_MAX define on MSVC.</li>
</ul>
<p>Karol Herbst (1):</p>
<ul>
<li>nv50/ir: fix memory corruption when spilling and redoing RA</li>
</ul>
<p>Kenneth Graunke (2):</p>
<ul>
<li>glsl: Make bitfield_insert/extract and bfi/bfm non-vectorizable.</li>
<li>glsl: Allow implicit int -&gt; uint conversions for bitwise operators (&amp;, ^, |).</li>
</ul>
<p>Leo Liu (2):</p>
<ul>
<li>vl: add zig zag scan for list 4x4</li>
<li>st/omx/dec/h264: fix corruption when scaling matrix present flag set</li>
</ul>
<p>Marek Olšák (1):</p>
<ul>
<li>radeonsi: don't miss changes to SPI_TMPRING_SIZE</li>
</ul>
<p>Nicolai Hähnle (11):</p>
<ul>
<li>mesa/bufferobj: make _mesa_delete_buffer_object externally accessible</li>
<li>st/mesa: use _mesa_delete_buffer_object</li>
<li>radeon: use _mesa_delete_buffer_object</li>
<li>i915: use _mesa_delete_buffer_object</li>
<li>i965: use _mesa_delete_buffer_object</li>
<li>util/u_pstipple.c: copy immediates during transformation</li>
<li>radeonsi: extract the VGT_GS_MODE calculation into its own function</li>
<li>radeonsi: ensure that VGT_GS_MODE is sent when necessary</li>
<li>radeonsi: add DCC buffer for sampler views on new CS</li>
<li>st/mesa: use the correct address generation functions in st_TexSubImage blit</li>
<li>radeonsi: fix discard-only fragment shaders (11.1 version)</li>
</ul>
<p>Timothy Arceri (4):</p>
<ul>
<li>glsl: fix segfault linking subroutine uniform with explicit location</li>
<li>mesa: fix segfault in glUniformSubroutinesuiv()</li>
<li>glsl: fix interface block error message</li>
<li>glsl: create helper to remove outer vertex index array used by some stages</li>
</ul>
</div>
</body>
</html>

View File

@@ -1,296 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 11.2.0 Release Notes / 4 April 2016</h1>
<p>
Mesa 11.2.0 is a new development release.
People who are concerned with stability and reliability should stick
with a previous release or wait for Mesa 11.2.1.
</p>
<p>
Mesa 11.2.0 implements the OpenGL 4.1 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 4.1. OpenGL
4.1 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
dea3d8143929aad5c24ef0993ddb05807b30c284b488fc62903adfcc1c127887 mesa-11.2.0.tar.gz
1c1fed2674abf3f16ed2623e9a5694d6752c293194e18462ebc644a19cfaafb2 mesa-11.2.0.tar.xz
</pre>
<h2>New features</h2>
<p>
Note: some of the new features are only available with certain drivers.
</p>
<ul>
<li>GL_ARB_arrays_of_arrays on all gallium drivers that provide GLSL 1.30</li>
<li>GL_ARB_base_instance on freedreno/a4xx</li>
<li>GL_ARB_compute_shader on i965</li>
<li>GL_ARB_copy_image on r600</li>
<li>GL_ARB_indirect_parameters on nvc0</li>
<li>GL_ARB_query_buffer_object on nvc0</li>
<li>GL_ARB_shader_atomic_counters on nvc0</li>
<li>GL_ARB_shader_draw_parameters on i965, nvc0</li>
<li>GL_ARB_shader_storage_buffer_object on nvc0</li>
<li>GL_ARB_tessellation_shader on i965 and r600 (evergreen/cayman only)</li>
<li>GL_ARB_texture_buffer_object_rgb32 on freedreno/a4xx</li>
<li>GL_ARB_texture_buffer_range on freedreno/a4xx</li>
<li>GL_ARB_texture_query_lod on freedreno/a4xx</li>
<li>GL_ARB_texture_rgb10_a2ui on freedreno/a4xx</li>
<li>GL_ARB_texture_view on freedreno/a4xx</li>
<li>GL_ARB_vertex_type_10f_11f_11f_rev on freedreno/a4xx</li>
<li>GL_KHR_texture_compression_astc_ldr on freedreno/a4xx</li>
<li>GL_AMD_performance_monitor on radeonsi (CIK+ only)</li>
<li>GL_ATI_meminfo on r600, radeonsi</li>
<li>GL_NVX_gpu_memory_info on r600, radeonsi</li>
<li>New OSMesaCreateContextAttribs() function (for creating core profile
contexts)</li>
</ul>
<h2>Bug fixes</h2>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=27512">Bug 27512</a> - Illegal instruction _mesa_x86_64_transform_points4_general</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=75165">Bug 75165</a> - compute.c:464:49: error: function definition is not allowed here</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=79783">Bug 79783</a> - Distorted output in obs-studio where other vendors &quot;work&quot;</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=89330">Bug 89330</a> - piglit glsl-1.50 invariant-qualifier-in-out-block-01 regression</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=89969">Bug 89969</a> - nouveau: add support for chunk decoding in order to support vaapi (st/va)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90348">Bug 90348</a> - Spilling failure of b96 merged value</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91526">Bug 91526</a> - World of Warcraft (on Wine) has UI corruption with nouveau</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91596">Bug 91596</a> - EGL_KHR_gl_colorspace (v2) causes problem with Android-x86 GUI</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91806">Bug 91806</a> - configure does not test whether assembler supports sse4.1</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91927">Bug 91927</a> - [SKL] [regression] piglit compressed textures tests fail with kernel upgrade</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92193">Bug 92193</a> - [SKL] ES2-CTS.gtf.GL2ExtensionTests.compressed_astc_texture.compressed_astc_texture fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92229">Bug 92229</a> - [APITRACE] SOMA have serious graphical errors</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92233">Bug 92233</a> - Unigine Heaven 4.0 silhuette run</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92363">Bug 92363</a> - [BSW/BDW] ogles1conform Gets test fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92438">Bug 92438</a> - Segfault in pushbuf_kref when running the android emulator (qemu) on nv50</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92589">Bug 92589</a> - [BDW BSW SKL CTS] ES31-CTS.texture_gather.* GPU_HANG</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92595">Bug 92595</a> - [HSW,BDW,SKL][GLES 3.1 CTS] Big difference in the results for the ES31-CTS.shader_bitfield_operation.* tests</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92609">Bug 92609</a> - [BDW, BSW] piglit sampling-2d-array-as-2d-layer fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92687">Bug 92687</a> - Add support for ARB_internalformat_query2</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92706">Bug 92706</a> - glBlitFramebuffer refuses to blit RGBA to RGB with MSAA</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92709">Bug 92709</a> - &quot;LLVM triggered Diagnostic Handler: unsupported call to function ldexpf in main&quot; when starting race in stuntrally</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92743">Bug 92743</a> - Centroid shouldn't have to match between the FS and the VS</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92759">Bug 92759</a> - [Regression, bisected] Visuals without alpha bits are not sRGB-capable</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92849">Bug 92849</a> - [IVB HSW BDW] piglit image load/store load-from-cleared-image.shader_test fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92909">Bug 92909</a> - Offset/alignment issue with layout std140 and vec3</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93004">Bug 93004</a> - Guild Wars 2 crash on nouveau DX11 cards</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93048">Bug 93048</a> - [CTS regression] mesa af2723 breaks GL Conformance for debug extension</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93063">Bug 93063</a> - drm_helper.h:227:1: error: static declaration of pipe_virgl_create_screen follows non-static declaration</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93091">Bug 93091</a> - [opencl] segfault when running any opencl programs (like clinfo)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93092">Bug 93092</a> - lp_test_format regression</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93126">Bug 93126</a> - wrongly claim supporting GL_EXT_texture_rg</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93180">Bug 93180</a> - [regression] arb_separate_shader_objects.active sampler conflict fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93189">Bug 93189</a> - &quot;./util/u_inlines.h&quot;, line 83: operands have incompatible types: void &quot;:&quot; int</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93215">Bug 93215</a> - [Regression bisected] Ogles1conform Automatic mipmap generation test is fail</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93235">Bug 93235</a> - [regression] dispatch sanity broken by GetPointerv</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93257">Bug 93257</a> - [SKL, bisected] ASTC dEQP tests segfault</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93264">Bug 93264</a> - Tonga VM Faults since llvm ScheduleDAGInstrs: Rework schedule graph builder.</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93266">Bug 93266</a> - gl_arb_shading_language_420pack does not allow binding of image variables</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93300">Bug 93300</a> - Two Worlds 2 renders water incorrectly</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93312">Bug 93312</a> - [SKL][GLES 3.1 CTS] ES31-CTS.layout_binding* GPU_HANG</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93320">Bug 93320</a> - [HSW,BDW,SKL][GLES 3.1 CTS] ES31-CTS.vertex_attrib_binding.advanced-bindingUpdate fail</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93322">Bug 93322</a> - [HSW,BDW,SKL][GLES 3.1 CTS] ES31-CTS.compute_shader.resource-ubo fail</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93323">Bug 93323</a> - [HSW,BDW,SKL][GLES 3.1 CTS]ES31-CTS.shader_image_load_store.basic-allTargets-store-fs fail</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93325">Bug 93325</a> - [HSW,BDW,SKL]ES31-CTS.explicit_uniform_location.uniform-loc-* 2 tests fail</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93339">Bug 93339</a> - glLinkProgram() should fail when a varying is never written to in a previous stage</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93348">Bug 93348</a> - [HSW,BDW,SKL][GLES 3.1 CTS] ES31-CTS.compute_shader.* segfault</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93358">Bug 93358</a> - [HSW] Unreal Elemental demo - assertion error in copy_image_with_blitter</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93387">Bug 93387</a> - inverse() shouldnt be exposed in GLSL 1.20 and 1.30</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93388">Bug 93388</a> - [i965, regression, bisection] MESA_FORMAT_B8G8R8X8_SRGB changes break kwin</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93407">Bug 93407</a> - [SKL][GLES 3.1 CTS]ES31-CTS.compute_shader.resources-texture fail</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93410">Bug 93410</a> - [BDW,SKL][GLES 3.1 CTS]ES31-CTS.shader_image_load_store.negative-linkErrors fail</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93418">Bug 93418</a> - Geometry Shaders output wrong vertices on Sandy Bridge</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93426">Bug 93426</a> - [SKL,BDW,BSW,BXT] CTS regression: es2-cts.gtf.gl2fixedtests.buffer_objects.buffer_object,s</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93524">Bug 93524</a> - Clover doesn't build</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93526">Bug 93526</a> - GfxBench 4 tessellation demos misrender</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93532">Bug 93532</a> - [HSW,BDW,SKL][GLES 3.1 CTS] ES31-CTS.compute_shader.*. Regression, bisected.</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93540">Bug 93540</a> - [BISECTED, HSW] Rendering issue in Heaven (and other benchmarks)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93560">Bug 93560</a> - opt_combine_constants failing fabsf(reg-&gt;f) == table.imm[i].val assertion</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93599">Bug 93599</a> - Strange green flashes with &quot;Metro: Last Light Redux&quot; + &quot;Metro 2033 Redux&quot; with Intel Mesa driver</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93648">Bug 93648</a> - Random lines being rendered when playing Dolphin (geometry shaders related, w/ apitrace)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93650">Bug 93650</a> - GL_ARB_separate_shader_objects is buggy (PCSX2)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93667">Bug 93667</a> - Crash in eglCreateImageKHR with huge texture size</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93696">Bug 93696</a> - [HSW,BDW;SKL][GLES 3.1 CTS]ES31-CTS.explicit_uniform_location.uniform-loc-mix-with-implicit-max-* fail</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93700">Bug 93700</a> - [SKL, regression] deqp-gles2.functional.texture.completeness</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93717">Bug 93717</a> - Meta mipmap generation can corrupt texture state</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93722">Bug 93722</a> - Segfault when compiling shader with a subroutine that takes a parameter</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93725">Bug 93725</a> - [HSW, regression, bisected] ES31-CTS.texture_gather.*depth*</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93731">Bug 93731</a> - glUniformSubroutinesuiv segfaults when subroutine uniform is bound to a specific location</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93761">Bug 93761</a> - A conditional discard in a fragment shader causes no depth writing at all</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93790">Bug 93790</a> - [HSW] Use after free with compute programs</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93792">Bug 93792</a> - [HSW] intel_mipmap_tree.c:1325: intel_miptree_copy_slice: Assertion `src_mt-&gt;format == dst_mt-&gt;format</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93813">Bug 93813</a> - Incorrect viewport range when GL_CLIP_ORIGIN is GL_UPPER_LEFT</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93840">Bug 93840</a> - [i965] Alien: Isolation fails with GL_ARB_compute_shader enabled</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93862">Bug 93862</a> - [Bisected] &quot;drm/amdgpu: fix amdgpu_bo_pin_restricted VRAM placing v2&quot; is bad</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93878">Bug 93878</a> - [llvmpipe][softpipe] piglit arb_gpu_shader_fp64-double-gettransformfeedbackvarying regression</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93957">Bug 93957</a> - [HSW] Mishandling of sample count when using an attachment-less framebuffer (assertion error)</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93961">Bug 93961</a> - virgl build failure after 2016-02-01 changes - no previous prototype for 'virgl_drm_winsys_create'</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93962">Bug 93962</a> - [HSW, regression, bisected, CTS] ES2-CTS.gtf.GL2FixedTests.scissor.scissor - segfault/asserts</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93989">Bug 93989</a> - build: flex-2.5.39 seems to be failing for glsl_lexer.ll</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94016">Bug 94016</a> - make check MesaExtensionsTest.AlphabeticallySorted regression</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94019">Bug 94019</a> - [bisected] 3D acceleration broken with gallium/radeon: just get num_tile_pipes from the winsys</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94050">Bug 94050</a> - test_vec4_register_coalesce regression</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94073">Bug 94073</a> - Miscompilation of abs_vec3_vert_xvary_ref.vert in WebGL conformance</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94081">Bug 94081</a> - [HSW] compute shader shared var + atomic op = fail</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94088">Bug 94088</a> - [llvmpipe] SIGFPE pthread_barrier_destroy.c:40</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94091">Bug 94091</a> - Tonga unreal elemental segfault since radeonsi: put image, fmask, and sampler descriptors into one array</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94100">Bug 94100</a> - [HSW] compute indirect dispatch with 0 work groups causes gpu hang</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94134">Bug 94134</a> - [regression] piglit.spec.arb_texture_view.sampling-2d-array-as-2d-layer assertion</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94139">Bug 94139</a> - [regression, HSW, IVB] piglit.spec.arb_compute_shader.minmax</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94150">Bug 94150</a> - UE4 Suntemple rendering errors</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94186">Bug 94186</a> - Crash when launching glxinfo and World of Warcraft with RV790</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94188">Bug 94188</a> - define (or undef) defined behaves stupidly</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94193">Bug 94193</a> - [llvmpipe] Line antialiasing looks different when GL_LINE_STIPPLE is enabled with pattern 0xffff</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94199">Bug 94199</a> - Shader abort/crash</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94253">Bug 94253</a> - [llvmpipe] piglit gl-1.0-swapbuffers-behavior regression</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94254">Bug 94254</a> - [llvmpipe] [softpipe] piglit read-front regression</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94257">Bug 94257</a> - [softpipe] piglit glx-copy-sub-buffer regression</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94274">Bug 94274</a> - [swrast] piglit arb_occlusion_query2-render regression</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94284">Bug 94284</a> - [radeonsi] outlast segfault on start</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94388">Bug 94388</a> - r600_blit.c:281: r600_decompress_depth_textures: Assertion `tex-&gt;is_depth &amp;&amp; !tex-&gt;is_flushing_texture' failed.</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94412">Bug 94412</a> - Trine 3 misrender</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94481">Bug 94481</a> - softpipe - access violation in img_filter_2d_nearest</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94524">Bug 94524</a> - Wrong gl_TessLevelOuter interpretation for isolines</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94595">Bug 94595</a> - [Mesa AMD&amp;swrast] Texture views attached as framebuffers return their viewed tecture's color encoding and render incorrectly</li>
</ul>
<h2>Changes</h2>
Microsoft Visual Studio 2013 or later is now required for building
on Windows.
Previously, Visual Studio 2008 and later were supported.
TBD.
</div>
</body>
</html>

View File

@@ -1,118 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<title>Mesa Release Notes</title>
<link rel="stylesheet" type="text/css" href="../mesa.css">
</head>
<body>
<div class="header">
<h1>The Mesa 3D Graphics Library</h1>
</div>
<iframe src="../contents.html"></iframe>
<div class="content">
<h1>Mesa 11.2.1 Release Notes / April 17, 2016</h1>
<p>
Mesa 11.2.1 is a bug fix release which fixes bugs found since the 11.2.0 release.
</p>
<p>
Mesa 11.2.1 implements the OpenGL 4.1 API, but the version reported by
glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 4.1. OpenGL
4.1 is <strong>only</strong> available if requested at context creation
because compatibility contexts are not supported.
</p>
<h2>SHA256 checksums</h2>
<pre>
TBD
</pre>
<h2>New features</h2>
<p>None</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=93962">Bug 93962</a> - [HSW, regression, bisected, CTS] ES2-CTS.gtf.GL2FixedTests.scissor.scissor - segfault/asserts</li>
</ul>
<h2>Changes</h2>
<p>Brian Paul (2):</p>
<ul>
<li>st/mesa: fix glReadBuffer() assertion failure</li>
<li>st/mesa: fix memleak in glDrawPixels cache code</li>
</ul>
<p>Christian Schmidbauer (1):</p>
<ul>
<li>st/nine: specify WINAPI only for i386 and amd64</li>
</ul>
<p>Emil Velikov (3):</p>
<ul>
<li>docs: add sha256 checksums for 11.2.0</li>
<li>configure.ac: update the path of the generated files</li>
<li>Update version to 11.2.1</li>
</ul>
<p>Ilia Mirkin (1):</p>
<ul>
<li>glsl: allow usage of the keyword buffer before GLSL 430 / ESSL 310</li>
</ul>
<p>Iurie Salomov (1):</p>
<ul>
<li>va: check null context in vlVaDestroyContext</li>
</ul>
<p>Jason Ekstrand (2):</p>
<ul>
<li>i965/tiled_memcopy: Add aligned mem_copy parameters to the [de]tiling functions</li>
<li>i965/tiled_memcpy: Rework the RGBA -&gt; BGRA mem_copy functions</li>
</ul>
<p>Kenneth Graunke (3):</p>
<ul>
<li>i965: Fix textureSize() depth value for 1 layer surfaces on Gen4-6.</li>
<li>i965: Use brw-&gt;urb.min_vs_urb_entries instead of 32 for BLORP.</li>
<li>glsl: Lower variable indexing of system value arrays unconditionally.</li>
</ul>
<p>Marek Olšák (1):</p>
<ul>
<li>drirc: add a workaround for blackness in Warsow</li>
</ul>
<p>Nicolai Hähnle (1):</p>
<ul>
<li>radeonsi: fix bounds check in si_create_vertex_elements</li>
</ul>
<p>Samuel Pitoiset (1):</p>
<ul>
<li>nv50/ir: do not try to attach JOIN ops to ATOM</li>
</ul>
<p>Thomas Hindoe Paaboel Andersen (1):</p>
<ul>
<li>st/va: avoid dereference after free in vlVaDestroyImage</li>
</ul>
</div>
</body>
</html>

View File

@@ -63,20 +63,6 @@ execution. These are generally used for debugging.
Example: export MESA_GLSL=dump,nopt
</p>
<p>
Shaders can be dumped and replaced on runtime for debugging purposes. Mesa
needs to be configured with '--with-sha1' to enable this functionality. This
feature is not currently supported by SCons build.
This is controlled via following environment variables:
<ul>
<li><b>MESA_SHADER_DUMP_PATH</b> - path where shader sources are dumped
<li><b>MESA_SHADER_READ_PATH</b> - path where replacement shaders are read
</ul>
Note, path set must exist before running for dumping or replacing to work.
When both are set, these paths should be different so the dumped shaders do
not clobber the replacement shaders.
</p>
<h2 id="support">GLSL Version</h2>

View File

@@ -1,176 +0,0 @@
Name
EXT_shader_samples_identical
Name Strings
GL_EXT_shader_samples_identical
Contact
Ian Romanick, Intel (ian.d.romanick 'at' intel.com)
Contributors
Chris Forbes, Mesa
Magnus Wendt, Intel
Neil S. Roberts, Intel
Graham Sellers, AMD
Status
XXX - Not complete yet.
Version
Last Modified Date: November 19, 2015
Revision: 6
Number
TBD
Dependencies
OpenGL 3.2, or OpenGL ES 3.1, or ARB_texture_multisample is required.
This extension is written against the OpenGL 4.5 (Core Profile)
Specification
Overview
Multisampled antialiasing has become a common method for improving the
quality of rendered images. Multisampling differs from supersampling in
that the color of a primitive that covers all or part of a pixel is
resolved once, regardless of the number of samples covered. If a large
polygon is rendered, the colors of all samples in each interior pixel will
be the same. This suggests a simple compression scheme that can reduce
the necessary memory bandwidth requirements. In one such scheme, each
sample is stored in a separate slice of the multisample surface. An
additional multisample control surface (MCS) contains a mapping from pixel
samples to slices.
If all the values stored in the MCS for a particular pixel are the same,
then all the samples have the same value. Applications can take advantage
of this information to reduce the bandwidth of reading multisample
textures. A custom multisample resolve filter could optimize resolving
pixels where every sample is identical by reading the color once.
color = texelFetch(sampler, coordinate, 0);
if (!textureSamplesIdenticalEXT(sampler, coordinate)) {
for (int i = 1; i < MAX_SAMPLES; i++) {
vec4 c = texelFetch(sampler, coordinate, i);
//... accumulate c into color
}
}
New Procedures and Functions
None.
New Tokens
None.
Additions to the OpenGL 4.5 (Core Profile) Specification
None.
Modifications to The OpenGL Shading Language Specification, Version 4.50.5
Including the following line in a shader can be used to control the
language features described in this extension:
#extension GL_EXT_shader_samples_identical
A new preprocessor #define is added to the OpenGL Shading Language:
#define GL_EXT_shader_samples_identical
Add to the table in section 8.7 "Texture Lookup Functions"
Syntax:
bool textureSamplesIdenticalEXT(gsampler2DMS sampler, ivec2 coord)
bool textureSamplesIdenticalEXT(gsampler2DMSArray sampler,
ivec3 coord)
Description:
Returns true if it can be determined that all samples within the texel
of the multisample texture bound to <sampler> at <coord> contain the
same values or false if this cannot be determined."
Additions to the AGL/EGL/GLX/WGL Specifications
None
Errors
None
New State
None
New Implementation Dependent State
None
Issues
1) What should the new functions be called?
RESOLVED: textureSamplesIdenticalEXT. Initially
textureAllSamplesIdenticalEXT was considered, but
textureSamplesIdenticalEXT is more similar to the existing textureSamples
function.
2) It seems like applications could implement additional optimization if
they were provided with raw MCS data. Should this extension also
provide that data?
There are a number of challenges in providing raw MCS data. The biggest
problem being that the amount of MCS data depends on the number of
samples, and that is not known at compile time. Additionally, without new
texelFetch functions, applications would have difficulty utilizing the
information.
Another option is to have a function that returns an array of tuples of
sample number and count. This also has difficulties with the maximum
array size not being known at compile time.
RESOLVED: Do not expose raw MCS data in this extension.
3) Should this extension also extend SPIR-V?
RESOLVED: Yes, but this has not yet been written.
4) Is it possible for textureSamplesIdenticalEXT to report false negatives?
RESOLVED: Yes. It is possible that the underlying hardware may not detect
that separate writes of the same color to different samples of a pixel are
the same. The shader function is at the whim of the underlying hardware
implementation. It is also possible that a compressed multisample surface
is not used. In that case the function will likely always return false.
Revision History
Rev Date Author Changes
--- ---------- -------- ---------------------------------------------
1 2014/08/20 cforbes Initial version
2 2015/10/23 idr Change from MESA to EXT. Rebase on OpenGL 4.5,
and add dependency on OpenGL ES 3.1. Initial
draft of overview section and issues 1 through
3.
3 2015/10/27 idr Typo fixes.
4 2015/11/10 idr Rename extension from EXT_shader_multisample_compression
to EXT_shader_samples_identical.
Add issue #4.
5 2015/11/18 idr Fix some typos spotted by gsellers. Change the
name of the name of the function to
textureSamplesIdenticalEXT.
6 2015/11/19 idr Fix more typos spotted by Nicolai Hähnle.

View File

@@ -42,7 +42,9 @@ Tungsten Graphics, Inc. have supported the ongoing development of Mesa.
<li>The
<a href="http://www.mesa3d.org">Mesa</a>
website is hosted by
<a href="http://sourceforge.net">sourceforge.net</a>.
<a href="http://sourceforge.net">
<img src="http://sourceforge.net/sflogo.php?group_id=3&amp;type=1"
width="88" height="31" align="bottom" alt="Sourceforge.net" border="0"></a>
<br>
<br>

View File

@@ -30,10 +30,6 @@
<dt><a href="http://www.valgrind.org">Valgrind</a></dt>
<dd>is a very useful tool for tracking down
memory-related problems in your code.</dd>
<dt><a href="http:scan.coverity.com/projects/mesa">Coverity</a><dt>
<dd>provides static code analysis of Mesa. If you create an account
you can see the results and try to fix outstanding issues.</dd>
</dl>
</div>

View File

@@ -26,31 +26,6 @@ VMware Workstation running on Linux or Windows and VMware Fusion running on
MacOS are all supported.
</p>
<p>
With the August 2015 Workstation 12 / Fusion 8 releases, OpenGL 3.3
is supported in the guest.
This requires:
<ul>
<li>The VM is configured for virtual hardware version 12.
<li>The host OS, GPU and graphics driver supports DX11 (Windows) or
OpenGL 4.0 (Linux, Mac)
<li>On Linux, the vmwgfx kernel module must be version 2.9.0 or later.
<li>A recent version of Mesa with the updated svga gallium driver.
</ul>
</p>
<p>
Otherwise, OpenGL 2.1 is supported.
</p>
<p>
OpenGL 3.3 support can be disabled by setting the environment variable
SVGA_VGPU10=0.
You will then have OpenGL 2.1 support.
This may be useful to work around application bugs (such as incorrect use
of the OpenGL 3.x core profile).
</p>
<p>
Most modern Linux distros include the SVGA3D driver so end users shouldn't
be concerned with this information.
@@ -148,33 +123,10 @@ To get the latest code from git:
<h2>Building the Code</h2>
<ul>
<li>
Determine where the GL-related libraries reside on your system and set
the LIBDIR environment variable accordingly.
<br><br>
For 32-bit Ubuntu systems:
<pre>
export LIBDIR=/usr/lib/i386-linux-gnu
</pre>
For 64-bit Ubuntu systems:
<pre>
export LIBDIR=/usr/lib/x86_64-linux-gnu
</pre>
For 32-bit Fedora systems:
<pre>
export LIBDIR=/usr/lib
</pre>
For 64-bit Fedora systems:
<pre>
export LIBDIR=/usr/lib64
</pre>
</li>
<li>Build libdrm:
<li>Build libdrm: If you're on a 32-bit system, you should skip the --libdir configure option. Note also the comment about toolchain libdrm above.
<pre>
cd $TOP/drm
./autogen.sh --prefix=/usr --libdir=${LIBDIR}
./autogen.sh --prefix=/usr --libdir=/usr/lib64
make
sudo make install
</pre>
@@ -185,9 +137,12 @@ The libxatracker library is used exclusively by the X server to do render,
copy and video acceleration:
<br>
The following configure options doesn't build the EGL system.
<br>
As before, if you're on a 32-bit system, you should skip the --libdir
configure option.
<pre>
cd $TOP/mesa
./autogen.sh --prefix=/usr --libdir=${LIBDIR} --with-gallium-drivers=svga --with-dri-drivers=swrast --enable-xa --disable-dri3 --enable-glx-tls
./autogen.sh --prefix=/usr --libdir=/usr/lib64 --with-gallium-drivers=svga --with-dri-drivers= --enable-xa --disable-dri3
make
sudo make install
</pre>
@@ -197,39 +152,25 @@ if they're not installed in your system. You should be told what's missing.
<br>
<br>
<li>xf86-video-vmware: Now, once libxatracker is installed, we proceed with
building and replacing the current Xorg driver.
First check if your system is 32- or 64-bit.
<li>xf86-video-vmware: Now, once libxatracker is installed, we proceed with building and replacing the current Xorg driver. First check if your system is 32- or 64-bit. If you're building for a 32-bit system, you will not be needing the --libdir=/usr/lib64 option to autogen.
<pre>
cd $TOP/xf86-video-vmware
./autogen.sh --prefix=/usr --libdir=${LIBDIR}
./autogen.sh --prefix=/usr --libdir=/usr/lib64
make
sudo make install
</pre>
<li>vmwgfx kernel module. First make sure that any old version of this kernel module is removed from the system by issuing
<pre>
<pre>
sudo rm /lib/modules/`uname -r`/kernel/drivers/gpu/drm/vmwgfx.ko*
</pre>
Build and install:
<pre>
</pre>
Then
<pre>
cd $TOP/vmwgfx
make
sudo make install
sudo depmod -a
</pre>
If you're using a Ubuntu OS:
<pre>
sudo update-initramfs -u
</pre>
If you're using a Fedora OS:
<pre>
sudo dracut --force
</pre>
Add 'vmwgfx' to the /etc/modules file:
<pre>
echo vmwgfx | sudo tee -a /etc/modules
</pre>
sudo cp 00-vmwgfx.rules /etc/udev/rules.d
sudo depmod -ae
</pre>
Note: some distros put DRM kernel drivers in different directories.
For example, sometimes vmwgfx.ko might be found in
@@ -286,16 +227,6 @@ If you don't see this, try setting this environment variable:
then rerun glxinfo and examine the output for error messages.
</p>
<p>
If OpenGL 3.3 is not working (you only get OpenGL 2.1):
</p>
<ul>
<li>Make sure the VM uses hardware version 12.
<li>Make sure the vmwgfx kernel module is version 2.9.0 or later.
<li>Check the vmware.log file for errors.
<li>Run 'dmesg | grep vmwgfx' and look for "DX: yes".
</div>
</body>
</html>

1
doxygen/.gitignore vendored
View File

@@ -1,4 +1,3 @@
*.db
*.tag
*.tmp
agpgart

View File

@@ -33,4 +33,3 @@ subset: $(SUBSET:.doxy=.tag)
clean:
-rm -rf $(FULL:.doxy=) $(SUBSET:.doxy=)
-rm -rf *.tag
-rm -rf *.db

View File

@@ -178,17 +178,11 @@ typedef struct _RGNDATA {
#undef WINAPI
#endif /* WINAPI*/
#ifdef __GNUC__
#if (defined(__x86_64__) && !defined(__ILP32__)) || defined(_M_X64)
#define WINAPI __attribute__((ms_abi))
#elif defined(__i386) || defined(_M_IX86)
#define WINAPI __attribute__((__stdcall__))
#else /* neither amd64 nor i386 */
#define WINAPI
#endif
#else /* __GNUC__ */
#define WINAPI
#endif
#if defined(__x86_64__) || defined(_M_X64)
#define WINAPI __attribute__((ms_abi))
#else /* x86_64 */
#define WINAPI __attribute__((__stdcall__))
#endif /* x86_64 */
/* Implementation caps */
#define D3DPRESENT_BACK_BUFFERS_MAX 3
@@ -233,7 +227,6 @@ typedef struct _RGNDATA {
#define D3DERR_DRIVERINVALIDCALL MAKE_D3DHRESULT(2157)
#define D3DERR_DEVICEREMOVED MAKE_D3DHRESULT(2160)
#define D3DERR_DEVICEHUNG MAKE_D3DHRESULT(2164)
#define S_PRESENT_OCCLUDED MAKE_D3DSTATUS(2168)
/********************************************************
* Bitmasks *

View File

@@ -77,7 +77,7 @@ typedef HDC EGLNativeDisplayType;
typedef HBITMAP EGLNativePixmapType;
typedef HWND EGLNativeWindowType;
#elif defined(__WINSCW__) || defined(__SYMBIAN32__) /* Symbian */
#elif defined(__APPLE__) || defined(__WINSCW__) || defined(__SYMBIAN32__) /* Symbian */
typedef int EGLNativeDisplayType;
typedef void *EGLNativeWindowType;
@@ -105,7 +105,7 @@ typedef struct ANativeWindow* EGLNativeWindowType;
typedef struct egl_native_pixmap_t* EGLNativePixmapType;
typedef void* EGLNativeDisplayType;
#elif defined(__unix__) || defined(__APPLE__)
#elif defined(__unix__)
#if defined(MESA_EGL_NO_X11_HEADERS)

View File

@@ -6,7 +6,7 @@ extern "C" {
#endif
/*
** Copyright (c) 2013-2015 The Khronos Group Inc.
** Copyright (c) 2013-2014 The Khronos Group Inc.
**
** Permission is hereby granted, free of charge, to any person obtaining a
** copy of this software and/or associated documentation files (the
@@ -33,7 +33,7 @@ extern "C" {
** used to make the header, and the header can be found at
** http://www.opengl.org/registry/
**
** Khronos $Revision: 31811 $ on $Date: 2015-08-10 17:01:11 +1000 (Mon, 10 Aug 2015) $
** Khronos $Revision: 29735 $ on $Date: 2015-02-02 19:00:01 -0800 (Mon, 02 Feb 2015) $
*/
#if defined(_WIN32) && !defined(APIENTRY) && !defined(__CYGWIN__) && !defined(__SCITECH_SNAP__)
@@ -53,7 +53,7 @@ extern "C" {
#define GLAPI extern
#endif
#define GL_GLEXT_VERSION 20150809
#define GL_GLEXT_VERSION 20150202
/* Generated C header for:
* API: gl
@@ -1041,22 +1041,6 @@ typedef unsigned short GLhalf;
#define GL_COLOR_ATTACHMENT13 0x8CED
#define GL_COLOR_ATTACHMENT14 0x8CEE
#define GL_COLOR_ATTACHMENT15 0x8CEF
#define GL_COLOR_ATTACHMENT16 0x8CF0
#define GL_COLOR_ATTACHMENT17 0x8CF1
#define GL_COLOR_ATTACHMENT18 0x8CF2
#define GL_COLOR_ATTACHMENT19 0x8CF3
#define GL_COLOR_ATTACHMENT20 0x8CF4
#define GL_COLOR_ATTACHMENT21 0x8CF5
#define GL_COLOR_ATTACHMENT22 0x8CF6
#define GL_COLOR_ATTACHMENT23 0x8CF7
#define GL_COLOR_ATTACHMENT24 0x8CF8
#define GL_COLOR_ATTACHMENT25 0x8CF9
#define GL_COLOR_ATTACHMENT26 0x8CFA
#define GL_COLOR_ATTACHMENT27 0x8CFB
#define GL_COLOR_ATTACHMENT28 0x8CFC
#define GL_COLOR_ATTACHMENT29 0x8CFD
#define GL_COLOR_ATTACHMENT30 0x8CFE
#define GL_COLOR_ATTACHMENT31 0x8CFF
#define GL_DEPTH_ATTACHMENT 0x8D00
#define GL_STENCIL_ATTACHMENT 0x8D20
#define GL_FRAMEBUFFER 0x8D40
@@ -2875,17 +2859,6 @@ GLAPI void APIENTRY glTextureBarrier (void);
#define GL_ARB_ES3_1_compatibility 1
#endif /* GL_ARB_ES3_1_compatibility */
#ifndef GL_ARB_ES3_2_compatibility
#define GL_ARB_ES3_2_compatibility 1
#define GL_PRIMITIVE_BOUNDING_BOX_ARB 0x92BE
#define GL_MULTISAMPLE_LINE_WIDTH_RANGE_ARB 0x9381
#define GL_MULTISAMPLE_LINE_WIDTH_GRANULARITY_ARB 0x9382
typedef void (APIENTRYP PFNGLPRIMITIVEBOUNDINGBOXARBPROC) (GLfloat minX, GLfloat minY, GLfloat minZ, GLfloat minW, GLfloat maxX, GLfloat maxY, GLfloat maxZ, GLfloat maxW);
#ifdef GL_GLEXT_PROTOTYPES
GLAPI void APIENTRY glPrimitiveBoundingBoxARB (GLfloat minX, GLfloat minY, GLfloat minZ, GLfloat minW, GLfloat maxX, GLfloat maxY, GLfloat maxZ, GLfloat maxW);
#endif
#endif /* GL_ARB_ES3_2_compatibility */
#ifndef GL_ARB_ES3_compatibility
#define GL_ARB_ES3_compatibility 1
#endif /* GL_ARB_ES3_compatibility */
@@ -3299,10 +3272,6 @@ GLAPI GLboolean APIENTRY glIsProgramARB (GLuint program);
#define GL_FRAGMENT_SHADER_DERIVATIVE_HINT_ARB 0x8B8B
#endif /* GL_ARB_fragment_shader */
#ifndef GL_ARB_fragment_shader_interlock
#define GL_ARB_fragment_shader_interlock 1
#endif /* GL_ARB_fragment_shader_interlock */
#ifndef GL_ARB_framebuffer_no_attachments
#define GL_ARB_framebuffer_no_attachments 1
#endif /* GL_ARB_framebuffer_no_attachments */
@@ -3363,91 +3332,6 @@ GLAPI void APIENTRY glFramebufferTextureFaceARB (GLenum target, GLenum attachmen
#define GL_ARB_gpu_shader_fp64 1
#endif /* GL_ARB_gpu_shader_fp64 */
#ifndef GL_ARB_gpu_shader_int64
#define GL_ARB_gpu_shader_int64 1
#define GL_INT64_ARB 0x140E
#define GL_INT64_VEC2_ARB 0x8FE9
#define GL_INT64_VEC3_ARB 0x8FEA
#define GL_INT64_VEC4_ARB 0x8FEB
#define GL_UNSIGNED_INT64_VEC2_ARB 0x8FF5
#define GL_UNSIGNED_INT64_VEC3_ARB 0x8FF6
#define GL_UNSIGNED_INT64_VEC4_ARB 0x8FF7
typedef void (APIENTRYP PFNGLUNIFORM1I64ARBPROC) (GLint location, GLint64 x);
typedef void (APIENTRYP PFNGLUNIFORM2I64ARBPROC) (GLint location, GLint64 x, GLint64 y);
typedef void (APIENTRYP PFNGLUNIFORM3I64ARBPROC) (GLint location, GLint64 x, GLint64 y, GLint64 z);
typedef void (APIENTRYP PFNGLUNIFORM4I64ARBPROC) (GLint location, GLint64 x, GLint64 y, GLint64 z, GLint64 w);
typedef void (APIENTRYP PFNGLUNIFORM1I64VARBPROC) (GLint location, GLsizei count, const GLint64 *value);
typedef void (APIENTRYP PFNGLUNIFORM2I64VARBPROC) (GLint location, GLsizei count, const GLint64 *value);
typedef void (APIENTRYP PFNGLUNIFORM3I64VARBPROC) (GLint location, GLsizei count, const GLint64 *value);
typedef void (APIENTRYP PFNGLUNIFORM4I64VARBPROC) (GLint location, GLsizei count, const GLint64 *value);
typedef void (APIENTRYP PFNGLUNIFORM1UI64ARBPROC) (GLint location, GLuint64 x);
typedef void (APIENTRYP PFNGLUNIFORM2UI64ARBPROC) (GLint location, GLuint64 x, GLuint64 y);
typedef void (APIENTRYP PFNGLUNIFORM3UI64ARBPROC) (GLint location, GLuint64 x, GLuint64 y, GLuint64 z);
typedef void (APIENTRYP PFNGLUNIFORM4UI64ARBPROC) (GLint location, GLuint64 x, GLuint64 y, GLuint64 z, GLuint64 w);
typedef void (APIENTRYP PFNGLUNIFORM1UI64VARBPROC) (GLint location, GLsizei count, const GLuint64 *value);
typedef void (APIENTRYP PFNGLUNIFORM2UI64VARBPROC) (GLint location, GLsizei count, const GLuint64 *value);
typedef void (APIENTRYP PFNGLUNIFORM3UI64VARBPROC) (GLint location, GLsizei count, const GLuint64 *value);
typedef void (APIENTRYP PFNGLUNIFORM4UI64VARBPROC) (GLint location, GLsizei count, const GLuint64 *value);
typedef void (APIENTRYP PFNGLGETUNIFORMI64VARBPROC) (GLuint program, GLint location, GLint64 *params);
typedef void (APIENTRYP PFNGLGETUNIFORMUI64VARBPROC) (GLuint program, GLint location, GLuint64 *params);
typedef void (APIENTRYP PFNGLGETNUNIFORMI64VARBPROC) (GLuint program, GLint location, GLsizei bufSize, GLint64 *params);
typedef void (APIENTRYP PFNGLGETNUNIFORMUI64VARBPROC) (GLuint program, GLint location, GLsizei bufSize, GLuint64 *params);
typedef void (APIENTRYP PFNGLPROGRAMUNIFORM1I64ARBPROC) (GLuint program, GLint location, GLint64 x);
typedef void (APIENTRYP PFNGLPROGRAMUNIFORM2I64ARBPROC) (GLuint program, GLint location, GLint64 x, GLint64 y);
typedef void (APIENTRYP PFNGLPROGRAMUNIFORM3I64ARBPROC) (GLuint program, GLint location, GLint64 x, GLint64 y, GLint64 z);
typedef void (APIENTRYP PFNGLPROGRAMUNIFORM4I64ARBPROC) (GLuint program, GLint location, GLint64 x, GLint64 y, GLint64 z, GLint64 w);
typedef void (APIENTRYP PFNGLPROGRAMUNIFORM1I64VARBPROC) (GLuint program, GLint location, GLsizei count, const GLint64 *value);
typedef void (APIENTRYP PFNGLPROGRAMUNIFORM2I64VARBPROC) (GLuint program, GLint location, GLsizei count, const GLint64 *value);
typedef void (APIENTRYP PFNGLPROGRAMUNIFORM3I64VARBPROC) (GLuint program, GLint location, GLsizei count, const GLint64 *value);
typedef void (APIENTRYP PFNGLPROGRAMUNIFORM4I64VARBPROC) (GLuint program, GLint location, GLsizei count, const GLint64 *value);
typedef void (APIENTRYP PFNGLPROGRAMUNIFORM1UI64ARBPROC) (GLuint program, GLint location, GLuint64 x);
typedef void (APIENTRYP PFNGLPROGRAMUNIFORM2UI64ARBPROC) (GLuint program, GLint location, GLuint64 x, GLuint64 y);
typedef void (APIENTRYP PFNGLPROGRAMUNIFORM3UI64ARBPROC) (GLuint program, GLint location, GLuint64 x, GLuint64 y, GLuint64 z);
typedef void (APIENTRYP PFNGLPROGRAMUNIFORM4UI64ARBPROC) (GLuint program, GLint location, GLuint64 x, GLuint64 y, GLuint64 z, GLuint64 w);
typedef void (APIENTRYP PFNGLPROGRAMUNIFORM1UI64VARBPROC) (GLuint program, GLint location, GLsizei count, const GLuint64 *value);
typedef void (APIENTRYP PFNGLPROGRAMUNIFORM2UI64VARBPROC) (GLuint program, GLint location, GLsizei count, const GLuint64 *value);
typedef void (APIENTRYP PFNGLPROGRAMUNIFORM3UI64VARBPROC) (GLuint program, GLint location, GLsizei count, const GLuint64 *value);
typedef void (APIENTRYP PFNGLPROGRAMUNIFORM4UI64VARBPROC) (GLuint program, GLint location, GLsizei count, const GLuint64 *value);
#ifdef GL_GLEXT_PROTOTYPES
GLAPI void APIENTRY glUniform1i64ARB (GLint location, GLint64 x);
GLAPI void APIENTRY glUniform2i64ARB (GLint location, GLint64 x, GLint64 y);
GLAPI void APIENTRY glUniform3i64ARB (GLint location, GLint64 x, GLint64 y, GLint64 z);
GLAPI void APIENTRY glUniform4i64ARB (GLint location, GLint64 x, GLint64 y, GLint64 z, GLint64 w);
GLAPI void APIENTRY glUniform1i64vARB (GLint location, GLsizei count, const GLint64 *value);
GLAPI void APIENTRY glUniform2i64vARB (GLint location, GLsizei count, const GLint64 *value);
GLAPI void APIENTRY glUniform3i64vARB (GLint location, GLsizei count, const GLint64 *value);
GLAPI void APIENTRY glUniform4i64vARB (GLint location, GLsizei count, const GLint64 *value);
GLAPI void APIENTRY glUniform1ui64ARB (GLint location, GLuint64 x);
GLAPI void APIENTRY glUniform2ui64ARB (GLint location, GLuint64 x, GLuint64 y);
GLAPI void APIENTRY glUniform3ui64ARB (GLint location, GLuint64 x, GLuint64 y, GLuint64 z);
GLAPI void APIENTRY glUniform4ui64ARB (GLint location, GLuint64 x, GLuint64 y, GLuint64 z, GLuint64 w);
GLAPI void APIENTRY glUniform1ui64vARB (GLint location, GLsizei count, const GLuint64 *value);
GLAPI void APIENTRY glUniform2ui64vARB (GLint location, GLsizei count, const GLuint64 *value);
GLAPI void APIENTRY glUniform3ui64vARB (GLint location, GLsizei count, const GLuint64 *value);
GLAPI void APIENTRY glUniform4ui64vARB (GLint location, GLsizei count, const GLuint64 *value);
GLAPI void APIENTRY glGetUniformi64vARB (GLuint program, GLint location, GLint64 *params);
GLAPI void APIENTRY glGetUniformui64vARB (GLuint program, GLint location, GLuint64 *params);
GLAPI void APIENTRY glGetnUniformi64vARB (GLuint program, GLint location, GLsizei bufSize, GLint64 *params);
GLAPI void APIENTRY glGetnUniformui64vARB (GLuint program, GLint location, GLsizei bufSize, GLuint64 *params);
GLAPI void APIENTRY glProgramUniform1i64ARB (GLuint program, GLint location, GLint64 x);
GLAPI void APIENTRY glProgramUniform2i64ARB (GLuint program, GLint location, GLint64 x, GLint64 y);
GLAPI void APIENTRY glProgramUniform3i64ARB (GLuint program, GLint location, GLint64 x, GLint64 y, GLint64 z);
GLAPI void APIENTRY glProgramUniform4i64ARB (GLuint program, GLint location, GLint64 x, GLint64 y, GLint64 z, GLint64 w);
GLAPI void APIENTRY glProgramUniform1i64vARB (GLuint program, GLint location, GLsizei count, const GLint64 *value);
GLAPI void APIENTRY glProgramUniform2i64vARB (GLuint program, GLint location, GLsizei count, const GLint64 *value);
GLAPI void APIENTRY glProgramUniform3i64vARB (GLuint program, GLint location, GLsizei count, const GLint64 *value);
GLAPI void APIENTRY glProgramUniform4i64vARB (GLuint program, GLint location, GLsizei count, const GLint64 *value);
GLAPI void APIENTRY glProgramUniform1ui64ARB (GLuint program, GLint location, GLuint64 x);
GLAPI void APIENTRY glProgramUniform2ui64ARB (GLuint program, GLint location, GLuint64 x, GLuint64 y);
GLAPI void APIENTRY glProgramUniform3ui64ARB (GLuint program, GLint location, GLuint64 x, GLuint64 y, GLuint64 z);
GLAPI void APIENTRY glProgramUniform4ui64ARB (GLuint program, GLint location, GLuint64 x, GLuint64 y, GLuint64 z, GLuint64 w);
GLAPI void APIENTRY glProgramUniform1ui64vARB (GLuint program, GLint location, GLsizei count, const GLuint64 *value);
GLAPI void APIENTRY glProgramUniform2ui64vARB (GLuint program, GLint location, GLsizei count, const GLuint64 *value);
GLAPI void APIENTRY glProgramUniform3ui64vARB (GLuint program, GLint location, GLsizei count, const GLuint64 *value);
GLAPI void APIENTRY glProgramUniform4ui64vARB (GLuint program, GLint location, GLsizei count, const GLuint64 *value);
#endif
#endif /* GL_ARB_gpu_shader_int64 */
#ifndef GL_ARB_half_float_pixel
#define GL_ARB_half_float_pixel 1
typedef unsigned short GLhalfARB;
@@ -3827,16 +3711,6 @@ GLAPI void APIENTRY glGetQueryObjectuivARB (GLuint id, GLenum pname, GLuint *par
#define GL_ARB_occlusion_query2 1
#endif /* GL_ARB_occlusion_query2 */
#ifndef GL_ARB_parallel_shader_compile
#define GL_ARB_parallel_shader_compile 1
#define GL_MAX_SHADER_COMPILER_THREADS_ARB 0x91B0
#define GL_COMPLETION_STATUS_ARB 0x91B1
typedef void (APIENTRYP PFNGLMAXSHADERCOMPILERTHREADSARBPROC) (GLuint count);
#ifdef GL_GLEXT_PROTOTYPES
GLAPI void APIENTRY glMaxShaderCompilerThreadsARB (GLuint count);
#endif
#endif /* GL_ARB_parallel_shader_compile */
#ifndef GL_ARB_pipeline_statistics_query
#define GL_ARB_pipeline_statistics_query 1
#define GL_VERTICES_SUBMITTED_ARB 0x82EE
@@ -3879,10 +3753,6 @@ GLAPI void APIENTRY glPointParameterfvARB (GLenum pname, const GLfloat *params);
#define GL_COORD_REPLACE_ARB 0x8862
#endif /* GL_ARB_point_sprite */
#ifndef GL_ARB_post_depth_coverage
#define GL_ARB_post_depth_coverage 1
#endif /* GL_ARB_post_depth_coverage */
#ifndef GL_ARB_program_interface_query
#define GL_ARB_program_interface_query 1
#endif /* GL_ARB_program_interface_query */
@@ -3956,26 +3826,6 @@ GLAPI void APIENTRY glGetnMinmaxARB (GLenum target, GLboolean reset, GLenum form
#define GL_ARB_robustness_isolation 1
#endif /* GL_ARB_robustness_isolation */
#ifndef GL_ARB_sample_locations
#define GL_ARB_sample_locations 1
#define GL_SAMPLE_LOCATION_SUBPIXEL_BITS_ARB 0x933D
#define GL_SAMPLE_LOCATION_PIXEL_GRID_WIDTH_ARB 0x933E
#define GL_SAMPLE_LOCATION_PIXEL_GRID_HEIGHT_ARB 0x933F
#define GL_PROGRAMMABLE_SAMPLE_LOCATION_TABLE_SIZE_ARB 0x9340
#define GL_SAMPLE_LOCATION_ARB 0x8E50
#define GL_PROGRAMMABLE_SAMPLE_LOCATION_ARB 0x9341
#define GL_FRAMEBUFFER_PROGRAMMABLE_SAMPLE_LOCATIONS_ARB 0x9342
#define GL_FRAMEBUFFER_SAMPLE_LOCATION_PIXEL_GRID_ARB 0x9343
typedef void (APIENTRYP PFNGLFRAMEBUFFERSAMPLELOCATIONSFVARBPROC) (GLenum target, GLuint start, GLsizei count, const GLfloat *v);
typedef void (APIENTRYP PFNGLNAMEDFRAMEBUFFERSAMPLELOCATIONSFVARBPROC) (GLuint framebuffer, GLuint start, GLsizei count, const GLfloat *v);
typedef void (APIENTRYP PFNGLEVALUATEDEPTHVALUESARBPROC) (void);
#ifdef GL_GLEXT_PROTOTYPES
GLAPI void APIENTRY glFramebufferSampleLocationsfvARB (GLenum target, GLuint start, GLsizei count, const GLfloat *v);
GLAPI void APIENTRY glNamedFramebufferSampleLocationsfvARB (GLuint framebuffer, GLuint start, GLsizei count, const GLfloat *v);
GLAPI void APIENTRY glEvaluateDepthValuesARB (void);
#endif
#endif /* GL_ARB_sample_locations */
#ifndef GL_ARB_sample_shading
#define GL_ARB_sample_shading 1
#define GL_SAMPLE_SHADING_ARB 0x8C36
@@ -4002,26 +3852,14 @@ GLAPI void APIENTRY glMinSampleShadingARB (GLfloat value);
#define GL_ARB_separate_shader_objects 1
#endif /* GL_ARB_separate_shader_objects */
#ifndef GL_ARB_shader_atomic_counter_ops
#define GL_ARB_shader_atomic_counter_ops 1
#endif /* GL_ARB_shader_atomic_counter_ops */
#ifndef GL_ARB_shader_atomic_counters
#define GL_ARB_shader_atomic_counters 1
#endif /* GL_ARB_shader_atomic_counters */
#ifndef GL_ARB_shader_ballot
#define GL_ARB_shader_ballot 1
#endif /* GL_ARB_shader_ballot */
#ifndef GL_ARB_shader_bit_encoding
#define GL_ARB_shader_bit_encoding 1
#endif /* GL_ARB_shader_bit_encoding */
#ifndef GL_ARB_shader_clock
#define GL_ARB_shader_clock 1
#endif /* GL_ARB_shader_clock */
#ifndef GL_ARB_shader_draw_parameters
#define GL_ARB_shader_draw_parameters 1
#endif /* GL_ARB_shader_draw_parameters */
@@ -4041,12 +3879,7 @@ GLAPI void APIENTRY glMinSampleShadingARB (GLfloat value);
#ifndef GL_ARB_shader_objects
#define GL_ARB_shader_objects 1
#ifdef __APPLE__
#ifdef BUILDING_MESA
/* Avoid uint <-> void* warnings */
typedef unsigned long GLhandleARB;
#else
typedef void *GLhandleARB;
#endif
#else
typedef unsigned int GLhandleARB;
#endif
@@ -4191,10 +4024,6 @@ GLAPI void APIENTRY glGetShaderSourceARB (GLhandleARB obj, GLsizei maxLength, GL
#define GL_ARB_shader_texture_lod 1
#endif /* GL_ARB_shader_texture_lod */
#ifndef GL_ARB_shader_viewport_layer_array
#define GL_ARB_shader_viewport_layer_array 1
#endif /* GL_ARB_shader_viewport_layer_array */
#ifndef GL_ARB_shading_language_100
#define GL_ARB_shading_language_100 1
#define GL_SHADING_LANGUAGE_VERSION_ARB 0x8B8C
@@ -4268,20 +4097,12 @@ GLAPI void APIENTRY glNamedBufferPageCommitmentARB (GLuint buffer, GLintptr offs
#define GL_MAX_SPARSE_3D_TEXTURE_SIZE_ARB 0x9199
#define GL_MAX_SPARSE_ARRAY_TEXTURE_LAYERS_ARB 0x919A
#define GL_SPARSE_TEXTURE_FULL_ARRAY_CUBE_MIPMAPS_ARB 0x91A9
typedef void (APIENTRYP PFNGLTEXPAGECOMMITMENTARBPROC) (GLenum target, GLint level, GLint xoffset, GLint yoffset, GLint zoffset, GLsizei width, GLsizei height, GLsizei depth, GLboolean commit);
typedef void (APIENTRYP PFNGLTEXPAGECOMMITMENTARBPROC) (GLenum target, GLint level, GLint xoffset, GLint yoffset, GLint zoffset, GLsizei width, GLsizei height, GLsizei depth, GLboolean resident);
#ifdef GL_GLEXT_PROTOTYPES
GLAPI void APIENTRY glTexPageCommitmentARB (GLenum target, GLint level, GLint xoffset, GLint yoffset, GLint zoffset, GLsizei width, GLsizei height, GLsizei depth, GLboolean commit);
GLAPI void APIENTRY glTexPageCommitmentARB (GLenum target, GLint level, GLint xoffset, GLint yoffset, GLint zoffset, GLsizei width, GLsizei height, GLsizei depth, GLboolean resident);
#endif
#endif /* GL_ARB_sparse_texture */
#ifndef GL_ARB_sparse_texture2
#define GL_ARB_sparse_texture2 1
#endif /* GL_ARB_sparse_texture2 */
#ifndef GL_ARB_sparse_texture_clamp
#define GL_ARB_sparse_texture_clamp 1
#endif /* GL_ARB_sparse_texture_clamp */
#ifndef GL_ARB_stencil_texturing
#define GL_ARB_stencil_texturing 1
#endif /* GL_ARB_stencil_texturing */
@@ -4434,12 +4255,6 @@ GLAPI void APIENTRY glGetCompressedTexImageARB (GLenum target, GLint level, void
#define GL_DOT3_RGBA_ARB 0x86AF
#endif /* GL_ARB_texture_env_dot3 */
#ifndef GL_ARB_texture_filter_minmax
#define GL_ARB_texture_filter_minmax 1
#define GL_TEXTURE_REDUCTION_MODE_ARB 0x9366
#define GL_WEIGHTED_AVERAGE_ARB 0x9367
#endif /* GL_ARB_texture_filter_minmax */
#ifndef GL_ARB_texture_float
#define GL_ARB_texture_float 1
#define GL_TEXTURE_RED_TYPE_ARB 0x8C10
@@ -4934,11 +4749,6 @@ GLAPI void APIENTRY glBlendBarrierKHR (void);
#define GL_KHR_debug 1
#endif /* GL_KHR_debug */
#ifndef GL_KHR_no_error
#define GL_KHR_no_error 1
#define GL_CONTEXT_FLAG_NO_ERROR_BIT_KHR 0x00000008
#endif /* GL_KHR_no_error */
#ifndef GL_KHR_robust_buffer_access_behavior
#define GL_KHR_robust_buffer_access_behavior 1
#endif /* GL_KHR_robust_buffer_access_behavior */
@@ -5081,6 +4891,7 @@ typedef void (APIENTRYP PFNGLPOINTPARAMETERXVOESPROC) (GLenum pname, const GLfix
typedef void (APIENTRYP PFNGLPOINTSIZEXOESPROC) (GLfixed size);
typedef void (APIENTRYP PFNGLPOLYGONOFFSETXOESPROC) (GLfixed factor, GLfixed units);
typedef void (APIENTRYP PFNGLROTATEXOESPROC) (GLfixed angle, GLfixed x, GLfixed y, GLfixed z);
typedef void (APIENTRYP PFNGLSAMPLECOVERAGEOESPROC) (GLfixed value, GLboolean invert);
typedef void (APIENTRYP PFNGLSCALEXOESPROC) (GLfixed x, GLfixed y, GLfixed z);
typedef void (APIENTRYP PFNGLTEXENVXOESPROC) (GLenum target, GLenum pname, GLfixed param);
typedef void (APIENTRYP PFNGLTEXENVXVOESPROC) (GLenum target, GLenum pname, const GLfixed *params);
@@ -5185,6 +4996,7 @@ GLAPI void APIENTRY glPointParameterxvOES (GLenum pname, const GLfixed *params);
GLAPI void APIENTRY glPointSizexOES (GLfixed size);
GLAPI void APIENTRY glPolygonOffsetxOES (GLfixed factor, GLfixed units);
GLAPI void APIENTRY glRotatexOES (GLfixed angle, GLfixed x, GLfixed y, GLfixed z);
GLAPI void APIENTRY glSampleCoverageOES (GLfixed value, GLboolean invert);
GLAPI void APIENTRY glScalexOES (GLfixed x, GLfixed y, GLfixed z);
GLAPI void APIENTRY glTexEnvxOES (GLenum target, GLenum pname, GLfixed param);
GLAPI void APIENTRY glTexEnvxvOES (GLenum target, GLenum pname, const GLfixed *params);
@@ -6898,7 +6710,7 @@ typedef void (APIENTRYP PFNGLVERTEXARRAYVERTEXATTRIBLFORMATEXTPROC) (GLuint vaob
typedef void (APIENTRYP PFNGLVERTEXARRAYVERTEXATTRIBBINDINGEXTPROC) (GLuint vaobj, GLuint attribindex, GLuint bindingindex);
typedef void (APIENTRYP PFNGLVERTEXARRAYVERTEXBINDINGDIVISOREXTPROC) (GLuint vaobj, GLuint bindingindex, GLuint divisor);
typedef void (APIENTRYP PFNGLVERTEXARRAYVERTEXATTRIBLOFFSETEXTPROC) (GLuint vaobj, GLuint buffer, GLuint index, GLint size, GLenum type, GLsizei stride, GLintptr offset);
typedef void (APIENTRYP PFNGLTEXTUREPAGECOMMITMENTEXTPROC) (GLuint texture, GLint level, GLint xoffset, GLint yoffset, GLint zoffset, GLsizei width, GLsizei height, GLsizei depth, GLboolean commit);
typedef void (APIENTRYP PFNGLTEXTUREPAGECOMMITMENTEXTPROC) (GLuint texture, GLint level, GLint xoffset, GLint yoffset, GLint zoffset, GLsizei width, GLsizei height, GLsizei depth, GLboolean resident);
typedef void (APIENTRYP PFNGLVERTEXARRAYVERTEXATTRIBDIVISOREXTPROC) (GLuint vaobj, GLuint index, GLuint divisor);
#ifdef GL_GLEXT_PROTOTYPES
GLAPI void APIENTRY glMatrixLoadfEXT (GLenum mode, const GLfloat *m);
@@ -7154,7 +6966,7 @@ GLAPI void APIENTRY glVertexArrayVertexAttribLFormatEXT (GLuint vaobj, GLuint at
GLAPI void APIENTRY glVertexArrayVertexAttribBindingEXT (GLuint vaobj, GLuint attribindex, GLuint bindingindex);
GLAPI void APIENTRY glVertexArrayVertexBindingDivisorEXT (GLuint vaobj, GLuint bindingindex, GLuint divisor);
GLAPI void APIENTRY glVertexArrayVertexAttribLOffsetEXT (GLuint vaobj, GLuint buffer, GLuint index, GLint size, GLenum type, GLsizei stride, GLintptr offset);
GLAPI void APIENTRY glTexturePageCommitmentEXT (GLuint texture, GLint level, GLint xoffset, GLint yoffset, GLint zoffset, GLsizei width, GLsizei height, GLsizei depth, GLboolean commit);
GLAPI void APIENTRY glTexturePageCommitmentEXT (GLuint texture, GLint level, GLint xoffset, GLint yoffset, GLint zoffset, GLsizei width, GLsizei height, GLsizei depth, GLboolean resident);
GLAPI void APIENTRY glVertexArrayVertexAttribDivisorEXT (GLuint vaobj, GLuint index, GLuint divisor);
#endif
#endif /* GL_EXT_direct_state_access */
@@ -8818,14 +8630,6 @@ GLAPI void APIENTRY glBlendFuncSeparateINGR (GLenum sfactorRGB, GLenum dfactorRG
#define GL_INTEL_fragment_shader_ordering 1
#endif /* GL_INTEL_fragment_shader_ordering */
#ifndef GL_INTEL_framebuffer_CMAA
#define GL_INTEL_framebuffer_CMAA 1
typedef void (APIENTRYP PFNGLAPPLYFRAMEBUFFERATTACHMENTCMAAINTELPROC) (void);
#ifdef GL_GLEXT_PROTOTYPES
GLAPI void APIENTRY glApplyFramebufferAttachmentCMAAINTEL (void);
#endif
#endif /* GL_INTEL_framebuffer_CMAA */
#ifndef GL_INTEL_map_texture
#define GL_INTEL_map_texture 1
#define GL_TEXTURE_MEMORY_LAYOUT_INTEL 0x83FF
@@ -9130,65 +8934,6 @@ GLAPI void APIENTRY glBlendBarrierNV (void);
#define GL_NV_blend_square 1
#endif /* GL_NV_blend_square */
#ifndef GL_NV_command_list
#define GL_NV_command_list 1
#define GL_TERMINATE_SEQUENCE_COMMAND_NV 0x0000
#define GL_NOP_COMMAND_NV 0x0001
#define GL_DRAW_ELEMENTS_COMMAND_NV 0x0002
#define GL_DRAW_ARRAYS_COMMAND_NV 0x0003
#define GL_DRAW_ELEMENTS_STRIP_COMMAND_NV 0x0004
#define GL_DRAW_ARRAYS_STRIP_COMMAND_NV 0x0005
#define GL_DRAW_ELEMENTS_INSTANCED_COMMAND_NV 0x0006
#define GL_DRAW_ARRAYS_INSTANCED_COMMAND_NV 0x0007
#define GL_ELEMENT_ADDRESS_COMMAND_NV 0x0008
#define GL_ATTRIBUTE_ADDRESS_COMMAND_NV 0x0009
#define GL_UNIFORM_ADDRESS_COMMAND_NV 0x000A
#define GL_BLEND_COLOR_COMMAND_NV 0x000B
#define GL_STENCIL_REF_COMMAND_NV 0x000C
#define GL_LINE_WIDTH_COMMAND_NV 0x000D
#define GL_POLYGON_OFFSET_COMMAND_NV 0x000E
#define GL_ALPHA_REF_COMMAND_NV 0x000F
#define GL_VIEWPORT_COMMAND_NV 0x0010
#define GL_SCISSOR_COMMAND_NV 0x0011
#define GL_FRONT_FACE_COMMAND_NV 0x0012
typedef void (APIENTRYP PFNGLCREATESTATESNVPROC) (GLsizei n, GLuint *states);
typedef void (APIENTRYP PFNGLDELETESTATESNVPROC) (GLsizei n, const GLuint *states);
typedef GLboolean (APIENTRYP PFNGLISSTATENVPROC) (GLuint state);
typedef void (APIENTRYP PFNGLSTATECAPTURENVPROC) (GLuint state, GLenum mode);
typedef GLuint (APIENTRYP PFNGLGETCOMMANDHEADERNVPROC) (GLenum tokenID, GLuint size);
typedef GLushort (APIENTRYP PFNGLGETSTAGEINDEXNVPROC) (GLenum shadertype);
typedef void (APIENTRYP PFNGLDRAWCOMMANDSNVPROC) (GLenum primitiveMode, GLuint buffer, const GLintptr *indirects, const GLsizei *sizes, GLuint count);
typedef void (APIENTRYP PFNGLDRAWCOMMANDSADDRESSNVPROC) (GLenum primitiveMode, const GLuint64 *indirects, const GLsizei *sizes, GLuint count);
typedef void (APIENTRYP PFNGLDRAWCOMMANDSSTATESNVPROC) (GLuint buffer, const GLintptr *indirects, const GLsizei *sizes, const GLuint *states, const GLuint *fbos, GLuint count);
typedef void (APIENTRYP PFNGLDRAWCOMMANDSSTATESADDRESSNVPROC) (const GLuint64 *indirects, const GLsizei *sizes, const GLuint *states, const GLuint *fbos, GLuint count);
typedef void (APIENTRYP PFNGLCREATECOMMANDLISTSNVPROC) (GLsizei n, GLuint *lists);
typedef void (APIENTRYP PFNGLDELETECOMMANDLISTSNVPROC) (GLsizei n, const GLuint *lists);
typedef GLboolean (APIENTRYP PFNGLISCOMMANDLISTNVPROC) (GLuint list);
typedef void (APIENTRYP PFNGLLISTDRAWCOMMANDSSTATESCLIENTNVPROC) (GLuint list, GLuint segment, const void **indirects, const GLsizei *sizes, const GLuint *states, const GLuint *fbos, GLuint count);
typedef void (APIENTRYP PFNGLCOMMANDLISTSEGMENTSNVPROC) (GLuint list, GLuint segments);
typedef void (APIENTRYP PFNGLCOMPILECOMMANDLISTNVPROC) (GLuint list);
typedef void (APIENTRYP PFNGLCALLCOMMANDLISTNVPROC) (GLuint list);
#ifdef GL_GLEXT_PROTOTYPES
GLAPI void APIENTRY glCreateStatesNV (GLsizei n, GLuint *states);
GLAPI void APIENTRY glDeleteStatesNV (GLsizei n, const GLuint *states);
GLAPI GLboolean APIENTRY glIsStateNV (GLuint state);
GLAPI void APIENTRY glStateCaptureNV (GLuint state, GLenum mode);
GLAPI GLuint APIENTRY glGetCommandHeaderNV (GLenum tokenID, GLuint size);
GLAPI GLushort APIENTRY glGetStageIndexNV (GLenum shadertype);
GLAPI void APIENTRY glDrawCommandsNV (GLenum primitiveMode, GLuint buffer, const GLintptr *indirects, const GLsizei *sizes, GLuint count);
GLAPI void APIENTRY glDrawCommandsAddressNV (GLenum primitiveMode, const GLuint64 *indirects, const GLsizei *sizes, GLuint count);
GLAPI void APIENTRY glDrawCommandsStatesNV (GLuint buffer, const GLintptr *indirects, const GLsizei *sizes, const GLuint *states, const GLuint *fbos, GLuint count);
GLAPI void APIENTRY glDrawCommandsStatesAddressNV (const GLuint64 *indirects, const GLsizei *sizes, const GLuint *states, const GLuint *fbos, GLuint count);
GLAPI void APIENTRY glCreateCommandListsNV (GLsizei n, GLuint *lists);
GLAPI void APIENTRY glDeleteCommandListsNV (GLsizei n, const GLuint *lists);
GLAPI GLboolean APIENTRY glIsCommandListNV (GLuint list);
GLAPI void APIENTRY glListDrawCommandsStatesClientNV (GLuint list, GLuint segment, const void **indirects, const GLsizei *sizes, const GLuint *states, const GLuint *fbos, GLuint count);
GLAPI void APIENTRY glCommandListSegmentsNV (GLuint list, GLuint segments);
GLAPI void APIENTRY glCompileCommandListNV (GLuint list);
GLAPI void APIENTRY glCallCommandListNV (GLuint list);
#endif
#endif /* GL_NV_command_list */
#ifndef GL_NV_compute_program5
#define GL_NV_compute_program5 1
#define GL_COMPUTE_PROGRAM_NV 0x90FB
@@ -9221,17 +8966,6 @@ GLAPI void APIENTRY glSubpixelPrecisionBiasNV (GLuint xbits, GLuint ybits);
#endif
#endif /* GL_NV_conservative_raster */
#ifndef GL_NV_conservative_raster_dilate
#define GL_NV_conservative_raster_dilate 1
#define GL_CONSERVATIVE_RASTER_DILATE_NV 0x9379
#define GL_CONSERVATIVE_RASTER_DILATE_RANGE_NV 0x937A
#define GL_CONSERVATIVE_RASTER_DILATE_GRANULARITY_NV 0x937B
typedef void (APIENTRYP PFNGLCONSERVATIVERASTERPARAMETERFNVPROC) (GLenum pname, GLfloat value);
#ifdef GL_GLEXT_PROTOTYPES
GLAPI void APIENTRY glConservativeRasterParameterfNV (GLenum pname, GLfloat value);
#endif
#endif /* GL_NV_conservative_raster_dilate */
#ifndef GL_NV_copy_depth_to_color
#define GL_NV_copy_depth_to_color 1
#define GL_DEPTH_STENCIL_TO_RGBA_NV 0x886E
@@ -11111,21 +10845,6 @@ GLAPI void APIENTRY glVideoCaptureStreamParameterdvNV (GLuint video_capture_slot
#define GL_FORMAT_SUBSAMPLE_244_244_OML 0x8983
#endif /* GL_OML_subsample */
#ifndef GL_OVR_multiview
#define GL_OVR_multiview 1
#define GL_FRAMEBUFFER_ATTACHMENT_TEXTURE_NUM_VIEWS_OVR 0x9630
#define GL_FRAMEBUFFER_ATTACHMENT_TEXTURE_BASE_VIEW_INDEX_OVR 0x9632
#define GL_MAX_VIEWS_OVR 0x9631
typedef void (APIENTRYP PFNGLFRAMEBUFFERTEXTUREMULTIVIEWOVRPROC) (GLenum target, GLenum attachment, GLuint texture, GLint level, GLint baseViewIndex, GLsizei numViews);
#ifdef GL_GLEXT_PROTOTYPES
GLAPI void APIENTRY glFramebufferTextureMultiviewOVR (GLenum target, GLenum attachment, GLuint texture, GLint level, GLint baseViewIndex, GLsizei numViews);
#endif
#endif /* GL_OVR_multiview */
#ifndef GL_OVR_multiview2
#define GL_OVR_multiview2 1
#endif /* GL_OVR_multiview2 */
#ifndef GL_PGI_misc_hints
#define GL_PGI_misc_hints 1
#define GL_PREFER_DOUBLEBUFFER_HINT_PGI 0x1A1F8

View File

@@ -368,6 +368,18 @@ extern Bool glXDrawableAttribARB(Display *dpy, GLXDrawable draw, const int *attr
#endif /* GLX_ARB_render_texture */
/*
* Remove this when glxext.h is updated.
*/
#ifndef GLX_NV_float_buffer
#define GLX_NV_float_buffer 1
#define GLX_FLOAT_COMPONENTS_NV 0x20B0
#endif /* GLX_NV_float_buffer */
/*
* #?. GLX_MESA_swap_frame_usage
*/
@@ -403,6 +415,86 @@ typedef int (*PFNGLXGETSWAPINTERVALMESAPROC)(void);
#endif /* GLX_MESA_swap_control */
/*
* #?. GLX_EXT_texture_from_pixmap
* XXX not finished?
*/
#ifndef GLX_EXT_texture_from_pixmap
#define GLX_EXT_texture_from_pixmap 1
#define GLX_BIND_TO_TEXTURE_RGB_EXT 0x20D0
#define GLX_BIND_TO_TEXTURE_RGBA_EXT 0x20D1
#define GLX_BIND_TO_MIPMAP_TEXTURE_EXT 0x20D2
#define GLX_BIND_TO_TEXTURE_TARGETS_EXT 0x20D3
#define GLX_Y_INVERTED_EXT 0x20D4
#define GLX_TEXTURE_FORMAT_EXT 0x20D5
#define GLX_TEXTURE_TARGET_EXT 0x20D6
#define GLX_MIPMAP_TEXTURE_EXT 0x20D7
#define GLX_TEXTURE_FORMAT_NONE_EXT 0x20D8
#define GLX_TEXTURE_FORMAT_RGB_EXT 0x20D9
#define GLX_TEXTURE_FORMAT_RGBA_EXT 0x20DA
#define GLX_TEXTURE_1D_BIT_EXT 0x00000001
#define GLX_TEXTURE_2D_BIT_EXT 0x00000002
#define GLX_TEXTURE_RECTANGLE_BIT_EXT 0x00000004
#define GLX_TEXTURE_1D_EXT 0x20DB
#define GLX_TEXTURE_2D_EXT 0x20DC
#define GLX_TEXTURE_RECTANGLE_EXT 0x20DD
#define GLX_FRONT_LEFT_EXT 0x20DE
#define GLX_FRONT_RIGHT_EXT 0x20DF
#define GLX_BACK_LEFT_EXT 0x20E0
#define GLX_BACK_RIGHT_EXT 0x20E1
#define GLX_FRONT_EXT GLX_FRONT_LEFT_EXT
#define GLX_BACK_EXT GLX_BACK_LEFT_EXT
#define GLX_AUX0_EXT 0x20E2
#define GLX_AUX1_EXT 0x20E3
#define GLX_AUX2_EXT 0x20E4
#define GLX_AUX3_EXT 0x20E5
#define GLX_AUX4_EXT 0x20E6
#define GLX_AUX5_EXT 0x20E7
#define GLX_AUX6_EXT 0x20E8
#define GLX_AUX7_EXT 0x20E9
#define GLX_AUX8_EXT 0x20EA
#define GLX_AUX9_EXT 0x20EB
extern void glXBindTexImageEXT(Display *dpy, GLXDrawable drawable, int buffer, const int *attrib_list);
extern void glXReleaseTexImageEXT(Display *dpy, GLXDrawable drawable, int buffer);
#endif /* GLX_EXT_texture_from_pixmap */
#ifndef GLX_MESA_query_renderer
#define GLX_MESA_query_renderer 1
#define GLX_RENDERER_VENDOR_ID_MESA 0x8183
#define GLX_RENDERER_DEVICE_ID_MESA 0x8184
#define GLX_RENDERER_VERSION_MESA 0x8185
#define GLX_RENDERER_ACCELERATED_MESA 0x8186
#define GLX_RENDERER_VIDEO_MEMORY_MESA 0x8187
#define GLX_RENDERER_UNIFIED_MEMORY_ARCHITECTURE_MESA 0x8188
#define GLX_RENDERER_PREFERRED_PROFILE_MESA 0x8189
#define GLX_RENDERER_OPENGL_CORE_PROFILE_VERSION_MESA 0x818A
#define GLX_RENDERER_OPENGL_COMPATIBILITY_PROFILE_VERSION_MESA 0x818B
#define GLX_RENDERER_OPENGL_ES_PROFILE_VERSION_MESA 0x818C
#define GLX_RENDERER_OPENGL_ES2_PROFILE_VERSION_MESA 0x818D
#define GLX_RENDERER_ID_MESA 0x818E
Bool glXQueryRendererIntegerMESA(Display *dpy, int screen, int renderer, int attribute, unsigned int *value);
Bool glXQueryCurrentRendererIntegerMESA(int attribute, unsigned int *value);
const char *glXQueryRendererStringMESA(Display *dpy, int screen, int renderer, int attribute);
const char *glXQueryCurrentRendererStringMESA(int attribute);
typedef Bool (*PFNGLXQUERYRENDERERINTEGERMESAPROC) (Display *dpy, int screen, int renderer, int attribute, unsigned int *value);
typedef Bool (*PFNGLXQUERYCURRENTRENDERERINTEGERMESAPROC) (int attribute, unsigned int *value);
typedef const char *(*PFNGLXQUERYRENDERERSTRINGMESAPROC) (Display *dpy, int screen, int renderer, int attribute);
typedef const char *(*PFNGLXQUERYCURRENTRENDERERSTRINGMESAPROC) (int attribute);
#endif /* GLX_MESA_query_renderer */
/*** Should these go here, or in another header? */
/*
** GLX Events

View File

@@ -40,7 +40,14 @@
#ifndef DRI_INTERFACE_H
#define DRI_INTERFACE_H
#ifdef HAVE_LIBDRM
/* For archs with no drm.h */
#if defined(__APPLE__) || defined(__CYGWIN__) || defined(__GNU__)
#ifndef __NOT_HAVE_DRM_H
#define __NOT_HAVE_DRM_H
#endif
#endif
#ifndef __NOT_HAVE_DRM_H
#include <drm.h>
#else
typedef unsigned int drm_context_t;
@@ -495,7 +502,7 @@ struct __DRIdamageExtensionRec {
* SWRast Loader extension.
*/
#define __DRI_SWRAST_LOADER "DRI_SWRastLoader"
#define __DRI_SWRAST_LOADER_VERSION 3
#define __DRI_SWRAST_LOADER_VERSION 2
struct __DRIswrastLoaderExtensionRec {
__DRIextension base;
@@ -528,15 +535,6 @@ struct __DRIswrastLoaderExtensionRec {
void (*putImage2)(__DRIdrawable *drawable, int op,
int x, int y, int width, int height, int stride,
char *data, void *loaderPrivate);
/**
* Put image to drawable
*
* \since 3
*/
void (*getImage2)(__DRIdrawable *readable,
int x, int y, int width, int height, int stride,
char *data, void *loaderPrivate);
};
/**
@@ -1103,15 +1101,12 @@ struct __DRIdri2ExtensionRec {
/**
* Four CC formats that matches with WL_DRM_FORMAT_* from wayland_drm.h,
* GBM_FORMAT_* from gbm.h, and DRM_FORMAT_* from drm_fourcc.h. Used with
* createImageFromNames.
* Four CC formats that matches with WL_DRM_FORMAT_* from wayland_drm.h
* and GBM_FORMAT_* from gbm.h, used with createImageFromNames.
*
* \since 5
*/
#define __DRI_IMAGE_FOURCC_R8 0x20203852
#define __DRI_IMAGE_FOURCC_GR88 0x38385247
#define __DRI_IMAGE_FOURCC_RGB565 0x36314752
#define __DRI_IMAGE_FOURCC_ARGB8888 0x34325241
#define __DRI_IMAGE_FOURCC_XRGB8888 0x34325258
@@ -1146,8 +1141,6 @@ struct __DRIdri2ExtensionRec {
#define __DRI_IMAGE_COMPONENTS_Y_U_V 0x3003
#define __DRI_IMAGE_COMPONENTS_Y_UV 0x3004
#define __DRI_IMAGE_COMPONENTS_Y_XUXV 0x3005
#define __DRI_IMAGE_COMPONENTS_R 0x3006
#define __DRI_IMAGE_COMPONENTS_RG 0x3007
/**
@@ -1187,8 +1180,7 @@ enum __DRIChromaSiting {
};
/**
* \name Reasons that __DRIimageExtensionRec::createImageFromTexture or
* __DRIimageExtensionRec::createImageFromDmaBufs might fail
* \name Reasons that __DRIimageExtensionRec::createImageFromTexture might fail
*/
/*@{*/
/** Success! */
@@ -1197,14 +1189,11 @@ enum __DRIChromaSiting {
/** Memory allocation failure */
#define __DRI_IMAGE_ERROR_BAD_ALLOC 1
/** Client requested an invalid attribute */
/** Client requested an invalid attribute for a texture object */
#define __DRI_IMAGE_ERROR_BAD_MATCH 2
/** Client requested an invalid texture object */
#define __DRI_IMAGE_ERROR_BAD_PARAMETER 3
/** Client requested an invalid pitch and/or offset */
#define __DRI_IMAGE_ERROR_BAD_ACCESS 4
/*@}*/
/**
@@ -1455,11 +1444,6 @@ typedef struct __DRIDriverVtableExtensionRec {
#define __DRI2_RENDERER_OPENGL_COMPATIBILITY_PROFILE_VERSION 0x0008
#define __DRI2_RENDERER_OPENGL_ES_PROFILE_VERSION 0x0009
#define __DRI2_RENDERER_OPENGL_ES2_PROFILE_VERSION 0x000a
#define __DRI2_RENDERER_HAS_TEXTURE_3D 0x000b
/* Whether there is an sRGB format support for every supported 32-bit UNORM
* color format.
*/
#define __DRI2_RENDERER_HAS_FRAMEBUFFER_SRGB 0x000c
typedef struct __DRI2rendererQueryExtensionRec __DRI2rendererQueryExtension;
struct __DRI2rendererQueryExtensionRec {

View File

@@ -58,8 +58,8 @@ extern "C" {
#include <GL/gl.h>
#define OSMESA_MAJOR_VERSION 11
#define OSMESA_MINOR_VERSION 2
#define OSMESA_MAJOR_VERSION 10
#define OSMESA_MINOR_VERSION 0
#define OSMESA_PATCH_VERSION 0
@@ -95,18 +95,6 @@ extern "C" {
#define OSMESA_MAX_WIDTH 0x24 /* new in 4.0 */
#define OSMESA_MAX_HEIGHT 0x25 /* new in 4.0 */
/*
* Accepted in OSMesaCreateContextAttrib's attribute list.
*/
#define OSMESA_DEPTH_BITS 0x30
#define OSMESA_STENCIL_BITS 0x31
#define OSMESA_ACCUM_BITS 0x32
#define OSMESA_PROFILE 0x33
#define OSMESA_CORE_PROFILE 0x34
#define OSMESA_COMPAT_PROFILE 0x35
#define OSMESA_CONTEXT_MAJOR_VERSION 0x36
#define OSMESA_CONTEXT_MINOR_VERSION 0x37
typedef struct osmesa_context *OSMesaContext;
@@ -139,35 +127,6 @@ OSMesaCreateContextExt( GLenum format, GLint depthBits, GLint stencilBits,
GLint accumBits, OSMesaContext sharelist);
/*
* Create an Off-Screen Mesa rendering context with attribute list.
* The list is composed of (attribute, value) pairs and terminated with
* attribute==0. Supported Attributes:
*
* Attributes Values
* --------------------------------------------------------------------------
* OSMESA_FORMAT OSMESA_RGBA*, OSMESA_BGRA, OSMESA_ARGB, etc.
* OSMESA_DEPTH_BITS 0*, 16, 24, 32
* OSMESA_STENCIL_BITS 0*, 8
* OSMESA_ACCUM_BITS 0*, 16
* OSMESA_PROFILE OSMESA_COMPAT_PROFILE*, OSMESA_CORE_PROFILE
* OSMESA_CONTEXT_MAJOR_VERSION 1*, 2, 3
* OSMESA_CONTEXT_MINOR_VERSION 0+
*
* Note: * = default value
*
* We return a context version >= what's specified by OSMESA_CONTEXT_MAJOR/
* MINOR_VERSION for the given profile. For example, if you request a GL 1.4
* compat profile, you might get a GL 3.0 compat profile.
* Otherwise, null is returned if the version/profile is not supported.
*
* New in Mesa 11.2
*/
GLAPI OSMesaContext GLAPIENTRY
OSMesaCreateContextAttribs( const int *attribList, OSMesaContext sharelist );
/*
* Destroy an Off-Screen Mesa rendering context.
*

File diff suppressed because it is too large Load Diff

View File

@@ -102,8 +102,9 @@ call_once(once_flag *flag, void (*func)(void))
static inline int
cnd_broadcast(cnd_t *cond)
{
assert(cond != NULL);
return (pthread_cond_broadcast(cond) == 0) ? thrd_success : thrd_error;
if (!cond) return thrd_error;
pthread_cond_broadcast(cond);
return thrd_success;
}
// 7.25.3.2
@@ -118,16 +119,18 @@ cnd_destroy(cnd_t *cond)
static inline int
cnd_init(cnd_t *cond)
{
assert(cond != NULL);
return (pthread_cond_init(cond, NULL) == 0) ? thrd_success : thrd_error;
if (!cond) return thrd_error;
pthread_cond_init(cond, NULL);
return thrd_success;
}
// 7.25.3.4
static inline int
cnd_signal(cnd_t *cond)
{
assert(cond != NULL);
return (pthread_cond_signal(cond) == 0) ? thrd_success : thrd_error;
if (!cond) return thrd_error;
pthread_cond_signal(cond);
return thrd_success;
}
// 7.25.3.5
@@ -136,14 +139,7 @@ cnd_timedwait(cnd_t *cond, mtx_t *mtx, const xtime *xt)
{
struct timespec abs_time;
int rt;
assert(mtx != NULL);
assert(cond != NULL);
assert(xt != NULL);
abs_time.tv_sec = xt->sec;
abs_time.tv_nsec = xt->nsec;
if (!cond || !mtx || !xt) return thrd_error;
rt = pthread_cond_timedwait(cond, mtx, &abs_time);
if (rt == ETIMEDOUT)
return thrd_busy;
@@ -154,9 +150,9 @@ cnd_timedwait(cnd_t *cond, mtx_t *mtx, const xtime *xt)
static inline int
cnd_wait(cnd_t *cond, mtx_t *mtx)
{
assert(mtx != NULL);
assert(cond != NULL);
return (pthread_cond_wait(cond, mtx) == 0) ? thrd_success : thrd_error;
if (!cond || !mtx) return thrd_error;
pthread_cond_wait(cond, mtx);
return thrd_success;
}
@@ -165,7 +161,7 @@ cnd_wait(cnd_t *cond, mtx_t *mtx)
static inline void
mtx_destroy(mtx_t *mtx)
{
assert(mtx != NULL);
assert(mtx);
pthread_mutex_destroy(mtx);
}
@@ -174,7 +170,7 @@ static inline int
mtx_init(mtx_t *mtx, int type)
{
pthread_mutexattr_t attr;
assert(mtx != NULL);
if (!mtx) return thrd_error;
if (type != mtx_plain && type != mtx_timed && type != mtx_try
&& type != (mtx_plain|mtx_recursive)
&& type != (mtx_timed|mtx_recursive)
@@ -192,8 +188,9 @@ mtx_init(mtx_t *mtx, int type)
static inline int
mtx_lock(mtx_t *mtx)
{
assert(mtx != NULL);
return (pthread_mutex_lock(mtx) == 0) ? thrd_success : thrd_error;
if (!mtx) return thrd_error;
pthread_mutex_lock(mtx);
return thrd_success;
}
static inline int
@@ -206,9 +203,7 @@ thrd_yield(void);
static inline int
mtx_timedlock(mtx_t *mtx, const xtime *xt)
{
assert(mtx != NULL);
assert(xt != NULL);
if (!mtx || !xt) return thrd_error;
{
#ifdef EMULATED_THREADS_USE_NATIVE_TIMEDLOCK
struct timespec ts;
@@ -238,7 +233,7 @@ mtx_timedlock(mtx_t *mtx, const xtime *xt)
static inline int
mtx_trylock(mtx_t *mtx)
{
assert(mtx != NULL);
if (!mtx) return thrd_error;
return (pthread_mutex_trylock(mtx) == 0) ? thrd_success : thrd_busy;
}
@@ -246,8 +241,9 @@ mtx_trylock(mtx_t *mtx)
static inline int
mtx_unlock(mtx_t *mtx)
{
assert(mtx != NULL);
return (pthread_mutex_unlock(mtx) == 0) ? thrd_success : thrd_error;
if (!mtx) return thrd_error;
pthread_mutex_unlock(mtx);
return thrd_success;
}
@@ -257,7 +253,7 @@ static inline int
thrd_create(thrd_t *thr, thrd_start_t func, void *arg)
{
struct impl_thrd_param *pack;
assert(thr != NULL);
if (!thr) return thrd_error;
pack = (struct impl_thrd_param *)malloc(sizeof(struct impl_thrd_param));
if (!pack) return thrd_nomem;
pack->func = func;
@@ -333,7 +329,7 @@ thrd_yield(void)
static inline int
tss_create(tss_t *key, tss_dtor_t dtor)
{
assert(key != NULL);
if (!key) return thrd_error;
return (pthread_key_create(key, dtor) == 0) ? thrd_success : thrd_error;
}

305
include/c99/inttypes.h Normal file
View File

@@ -0,0 +1,305 @@
// ISO C9x compliant inttypes.h for Microsoft Visual Studio
// Based on ISO/IEC 9899:TC2 Committee draft (May 6, 2005) WG14/N1124
//
// Copyright (c) 2006 Alexander Chemeris
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are met:
//
// 1. Redistributions of source code must retain the above copyright notice,
// this list of conditions and the following disclaimer.
//
// 2. Redistributions in binary form must reproduce the above copyright
// notice, this list of conditions and the following disclaimer in the
// documentation and/or other materials provided with the distribution.
//
// 3. The name of the author may be used to endorse or promote products
// derived from this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED
// WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
// MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO
// EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
// PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
// OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
// WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
// OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
// ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
//
///////////////////////////////////////////////////////////////////////////////
#ifndef _MSC_VER // [
#error "Use this header only with Microsoft Visual C++ compilers!"
#endif // _MSC_VER ]
#ifndef _MSC_INTTYPES_H_ // [
#define _MSC_INTTYPES_H_
#if _MSC_VER > 1000
#pragma once
#endif
#include "stdint.h"
// 7.8 Format conversion of integer types
typedef struct {
intmax_t quot;
intmax_t rem;
} imaxdiv_t;
// 7.8.1 Macros for format specifiers
#if !defined(__cplusplus) || defined(__STDC_FORMAT_MACROS) // [ See footnote 185 at page 198
// The fprintf macros for signed integers are:
#define PRId8 "d"
#define PRIi8 "i"
#define PRIdLEAST8 "d"
#define PRIiLEAST8 "i"
#define PRIdFAST8 "d"
#define PRIiFAST8 "i"
#define PRId16 "hd"
#define PRIi16 "hi"
#define PRIdLEAST16 "hd"
#define PRIiLEAST16 "hi"
#define PRIdFAST16 "hd"
#define PRIiFAST16 "hi"
#define PRId32 "I32d"
#define PRIi32 "I32i"
#define PRIdLEAST32 "I32d"
#define PRIiLEAST32 "I32i"
#define PRIdFAST32 "I32d"
#define PRIiFAST32 "I32i"
#define PRId64 "I64d"
#define PRIi64 "I64i"
#define PRIdLEAST64 "I64d"
#define PRIiLEAST64 "I64i"
#define PRIdFAST64 "I64d"
#define PRIiFAST64 "I64i"
#define PRIdMAX "I64d"
#define PRIiMAX "I64i"
#define PRIdPTR "Id"
#define PRIiPTR "Ii"
// The fprintf macros for unsigned integers are:
#define PRIo8 "o"
#define PRIu8 "u"
#define PRIx8 "x"
#define PRIX8 "X"
#define PRIoLEAST8 "o"
#define PRIuLEAST8 "u"
#define PRIxLEAST8 "x"
#define PRIXLEAST8 "X"
#define PRIoFAST8 "o"
#define PRIuFAST8 "u"
#define PRIxFAST8 "x"
#define PRIXFAST8 "X"
#define PRIo16 "ho"
#define PRIu16 "hu"
#define PRIx16 "hx"
#define PRIX16 "hX"
#define PRIoLEAST16 "ho"
#define PRIuLEAST16 "hu"
#define PRIxLEAST16 "hx"
#define PRIXLEAST16 "hX"
#define PRIoFAST16 "ho"
#define PRIuFAST16 "hu"
#define PRIxFAST16 "hx"
#define PRIXFAST16 "hX"
#define PRIo32 "I32o"
#define PRIu32 "I32u"
#define PRIx32 "I32x"
#define PRIX32 "I32X"
#define PRIoLEAST32 "I32o"
#define PRIuLEAST32 "I32u"
#define PRIxLEAST32 "I32x"
#define PRIXLEAST32 "I32X"
#define PRIoFAST32 "I32o"
#define PRIuFAST32 "I32u"
#define PRIxFAST32 "I32x"
#define PRIXFAST32 "I32X"
#define PRIo64 "I64o"
#define PRIu64 "I64u"
#define PRIx64 "I64x"
#define PRIX64 "I64X"
#define PRIoLEAST64 "I64o"
#define PRIuLEAST64 "I64u"
#define PRIxLEAST64 "I64x"
#define PRIXLEAST64 "I64X"
#define PRIoFAST64 "I64o"
#define PRIuFAST64 "I64u"
#define PRIxFAST64 "I64x"
#define PRIXFAST64 "I64X"
#define PRIoMAX "I64o"
#define PRIuMAX "I64u"
#define PRIxMAX "I64x"
#define PRIXMAX "I64X"
#define PRIoPTR "Io"
#define PRIuPTR "Iu"
#define PRIxPTR "Ix"
#define PRIXPTR "IX"
// The fscanf macros for signed integers are:
#define SCNd8 "d"
#define SCNi8 "i"
#define SCNdLEAST8 "d"
#define SCNiLEAST8 "i"
#define SCNdFAST8 "d"
#define SCNiFAST8 "i"
#define SCNd16 "hd"
#define SCNi16 "hi"
#define SCNdLEAST16 "hd"
#define SCNiLEAST16 "hi"
#define SCNdFAST16 "hd"
#define SCNiFAST16 "hi"
#define SCNd32 "ld"
#define SCNi32 "li"
#define SCNdLEAST32 "ld"
#define SCNiLEAST32 "li"
#define SCNdFAST32 "ld"
#define SCNiFAST32 "li"
#define SCNd64 "I64d"
#define SCNi64 "I64i"
#define SCNdLEAST64 "I64d"
#define SCNiLEAST64 "I64i"
#define SCNdFAST64 "I64d"
#define SCNiFAST64 "I64i"
#define SCNdMAX "I64d"
#define SCNiMAX "I64i"
#ifdef _WIN64 // [
# define SCNdPTR "I64d"
# define SCNiPTR "I64i"
#else // _WIN64 ][
# define SCNdPTR "ld"
# define SCNiPTR "li"
#endif // _WIN64 ]
// The fscanf macros for unsigned integers are:
#define SCNo8 "o"
#define SCNu8 "u"
#define SCNx8 "x"
#define SCNX8 "X"
#define SCNoLEAST8 "o"
#define SCNuLEAST8 "u"
#define SCNxLEAST8 "x"
#define SCNXLEAST8 "X"
#define SCNoFAST8 "o"
#define SCNuFAST8 "u"
#define SCNxFAST8 "x"
#define SCNXFAST8 "X"
#define SCNo16 "ho"
#define SCNu16 "hu"
#define SCNx16 "hx"
#define SCNX16 "hX"
#define SCNoLEAST16 "ho"
#define SCNuLEAST16 "hu"
#define SCNxLEAST16 "hx"
#define SCNXLEAST16 "hX"
#define SCNoFAST16 "ho"
#define SCNuFAST16 "hu"
#define SCNxFAST16 "hx"
#define SCNXFAST16 "hX"
#define SCNo32 "lo"
#define SCNu32 "lu"
#define SCNx32 "lx"
#define SCNX32 "lX"
#define SCNoLEAST32 "lo"
#define SCNuLEAST32 "lu"
#define SCNxLEAST32 "lx"
#define SCNXLEAST32 "lX"
#define SCNoFAST32 "lo"
#define SCNuFAST32 "lu"
#define SCNxFAST32 "lx"
#define SCNXFAST32 "lX"
#define SCNo64 "I64o"
#define SCNu64 "I64u"
#define SCNx64 "I64x"
#define SCNX64 "I64X"
#define SCNoLEAST64 "I64o"
#define SCNuLEAST64 "I64u"
#define SCNxLEAST64 "I64x"
#define SCNXLEAST64 "I64X"
#define SCNoFAST64 "I64o"
#define SCNuFAST64 "I64u"
#define SCNxFAST64 "I64x"
#define SCNXFAST64 "I64X"
#define SCNoMAX "I64o"
#define SCNuMAX "I64u"
#define SCNxMAX "I64x"
#define SCNXMAX "I64X"
#ifdef _WIN64 // [
# define SCNoPTR "I64o"
# define SCNuPTR "I64u"
# define SCNxPTR "I64x"
# define SCNXPTR "I64X"
#else // _WIN64 ][
# define SCNoPTR "lo"
# define SCNuPTR "lu"
# define SCNxPTR "lx"
# define SCNXPTR "lX"
#endif // _WIN64 ]
#endif // __STDC_FORMAT_MACROS ]
// 7.8.2 Functions for greatest-width integer types
// 7.8.2.1 The imaxabs function
#define imaxabs _abs64
// 7.8.2.2 The imaxdiv function
// This is modified version of div() function from Microsoft's div.c found
// in %MSVC.NET%\crt\src\div.c
#ifdef STATIC_IMAXDIV // [
static
#else // STATIC_IMAXDIV ][
_inline
#endif // STATIC_IMAXDIV ]
imaxdiv_t __cdecl imaxdiv(intmax_t numer, intmax_t denom)
{
imaxdiv_t result;
result.quot = numer / denom;
result.rem = numer % denom;
if (numer < 0 && result.rem > 0) {
// did division wrong; must fix up
++result.quot;
result.rem -= denom;
}
return result;
}
// 7.8.2.3 The strtoimax and strtoumax functions
#define strtoimax _strtoi64
#define strtoumax _strtoui64
// 7.8.2.4 The wcstoimax and wcstoumax functions
#define wcstoimax _wcstoi64
#define wcstoumax _wcstoui64
#endif // _MSC_INTTYPES_H_ ]

View File

@@ -1,6 +1,6 @@
/**************************************************************************
*
* Copyright 2010 VMware, Inc.
* Copyright 2007-2010 VMware, Inc.
* All Rights Reserved.
*
* Permission is hereby granted, free of charge, to any person obtaining a
@@ -25,32 +25,22 @@
*
**************************************************************************/
#ifndef _STDBOOL_H_
#define _STDBOOL_H_
#include "lp_bld_format.h"
#ifndef __cplusplus
#define false 0
#define true 1
#define bool _Bool
LLVMTypeRef
lp_build_format_cache_type(struct gallivm_state *gallivm)
{
LLVMTypeRef elem_types[LP_BUILD_FORMAT_CACHE_MEMBER_COUNT];
LLVMTypeRef s;
elem_types[LP_BUILD_FORMAT_CACHE_MEMBER_DATA] =
LLVMArrayType(LLVMInt32TypeInContext(gallivm->context),
LP_BUILD_FORMAT_CACHE_SIZE * 16);
elem_types[LP_BUILD_FORMAT_CACHE_MEMBER_TAGS] =
LLVMArrayType(LLVMInt64TypeInContext(gallivm->context),
LP_BUILD_FORMAT_CACHE_SIZE);
#if LP_BUILD_FORMAT_CACHE_DEBUG
elem_types[LP_BUILD_FORMAT_CACHE_MEMBER_ACCESS_TOTAL] =
LLVMInt64TypeInContext(gallivm->context);
elem_types[LP_BUILD_FORMAT_CACHE_MEMBER_ACCESS_MISS] =
LLVMInt64TypeInContext(gallivm->context);
/* For compilers that don't have the builtin _Bool type. */
#if (defined(_MSC_VER) && _MSC_VER < 1800)
typedef unsigned char _Bool;
#endif
s = LLVMStructTypeInContext(gallivm->context, elem_types,
LP_BUILD_FORMAT_CACHE_MEMBER_COUNT, 0);
#endif /* !__cplusplus */
return s;
}
#define __bool_true_false_are_defined 1
#endif /* !_STDBOOL_H_ */

247
include/c99/stdint.h Normal file
View File

@@ -0,0 +1,247 @@
// ISO C9x compliant stdint.h for Microsoft Visual Studio
// Based on ISO/IEC 9899:TC2 Committee draft (May 6, 2005) WG14/N1124
//
// Copyright (c) 2006-2008 Alexander Chemeris
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are met:
//
// 1. Redistributions of source code must retain the above copyright notice,
// this list of conditions and the following disclaimer.
//
// 2. Redistributions in binary form must reproduce the above copyright
// notice, this list of conditions and the following disclaimer in the
// documentation and/or other materials provided with the distribution.
//
// 3. The name of the author may be used to endorse or promote products
// derived from this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED
// WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
// MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO
// EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
// PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
// OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
// WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
// OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
// ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
//
///////////////////////////////////////////////////////////////////////////////
#ifndef _MSC_VER // [
#error "Use this header only with Microsoft Visual C++ compilers!"
#endif // _MSC_VER ]
#ifndef _MSC_STDINT_H_ // [
#define _MSC_STDINT_H_
#if _MSC_VER > 1000
#pragma once
#endif
#include <limits.h>
// For Visual Studio 6 in C++ mode and for many Visual Studio versions when
// compiling for ARM we should wrap <wchar.h> include with 'extern "C++" {}'
// or compiler give many errors like this:
// error C2733: second C linkage of overloaded function 'wmemchr' not allowed
#ifdef __cplusplus
extern "C" {
#endif
# include <wchar.h>
#ifdef __cplusplus
}
#endif
// Define _W64 macros to mark types changing their size, like intptr_t.
#ifndef _W64
# if !defined(__midl) && (defined(_X86_) || defined(_M_IX86)) && _MSC_VER >= 1300
# define _W64 __w64
# else
# define _W64
# endif
#endif
// 7.18.1 Integer types
// 7.18.1.1 Exact-width integer types
// Visual Studio 6 and Embedded Visual C++ 4 doesn't
// realize that, e.g. char has the same size as __int8
// so we give up on __intX for them.
#if (_MSC_VER < 1300)
typedef signed char int8_t;
typedef signed short int16_t;
typedef signed int int32_t;
typedef unsigned char uint8_t;
typedef unsigned short uint16_t;
typedef unsigned int uint32_t;
#else
typedef signed __int8 int8_t;
typedef signed __int16 int16_t;
typedef signed __int32 int32_t;
typedef unsigned __int8 uint8_t;
typedef unsigned __int16 uint16_t;
typedef unsigned __int32 uint32_t;
#endif
typedef signed __int64 int64_t;
typedef unsigned __int64 uint64_t;
// 7.18.1.2 Minimum-width integer types
typedef int8_t int_least8_t;
typedef int16_t int_least16_t;
typedef int32_t int_least32_t;
typedef int64_t int_least64_t;
typedef uint8_t uint_least8_t;
typedef uint16_t uint_least16_t;
typedef uint32_t uint_least32_t;
typedef uint64_t uint_least64_t;
// 7.18.1.3 Fastest minimum-width integer types
typedef int8_t int_fast8_t;
typedef int16_t int_fast16_t;
typedef int32_t int_fast32_t;
typedef int64_t int_fast64_t;
typedef uint8_t uint_fast8_t;
typedef uint16_t uint_fast16_t;
typedef uint32_t uint_fast32_t;
typedef uint64_t uint_fast64_t;
// 7.18.1.4 Integer types capable of holding object pointers
#ifdef _WIN64 // [
typedef signed __int64 intptr_t;
typedef unsigned __int64 uintptr_t;
#else // _WIN64 ][
typedef _W64 signed int intptr_t;
typedef _W64 unsigned int uintptr_t;
#endif // _WIN64 ]
// 7.18.1.5 Greatest-width integer types
typedef int64_t intmax_t;
typedef uint64_t uintmax_t;
// 7.18.2 Limits of specified-width integer types
#if !defined(__cplusplus) || defined(__STDC_LIMIT_MACROS) // [ See footnote 220 at page 257 and footnote 221 at page 259
// 7.18.2.1 Limits of exact-width integer types
#define INT8_MIN ((int8_t)_I8_MIN)
#define INT8_MAX _I8_MAX
#define INT16_MIN ((int16_t)_I16_MIN)
#define INT16_MAX _I16_MAX
#define INT32_MIN ((int32_t)_I32_MIN)
#define INT32_MAX _I32_MAX
#define INT64_MIN ((int64_t)_I64_MIN)
#define INT64_MAX _I64_MAX
#define UINT8_MAX _UI8_MAX
#define UINT16_MAX _UI16_MAX
#define UINT32_MAX _UI32_MAX
#define UINT64_MAX _UI64_MAX
// 7.18.2.2 Limits of minimum-width integer types
#define INT_LEAST8_MIN INT8_MIN
#define INT_LEAST8_MAX INT8_MAX
#define INT_LEAST16_MIN INT16_MIN
#define INT_LEAST16_MAX INT16_MAX
#define INT_LEAST32_MIN INT32_MIN
#define INT_LEAST32_MAX INT32_MAX
#define INT_LEAST64_MIN INT64_MIN
#define INT_LEAST64_MAX INT64_MAX
#define UINT_LEAST8_MAX UINT8_MAX
#define UINT_LEAST16_MAX UINT16_MAX
#define UINT_LEAST32_MAX UINT32_MAX
#define UINT_LEAST64_MAX UINT64_MAX
// 7.18.2.3 Limits of fastest minimum-width integer types
#define INT_FAST8_MIN INT8_MIN
#define INT_FAST8_MAX INT8_MAX
#define INT_FAST16_MIN INT16_MIN
#define INT_FAST16_MAX INT16_MAX
#define INT_FAST32_MIN INT32_MIN
#define INT_FAST32_MAX INT32_MAX
#define INT_FAST64_MIN INT64_MIN
#define INT_FAST64_MAX INT64_MAX
#define UINT_FAST8_MAX UINT8_MAX
#define UINT_FAST16_MAX UINT16_MAX
#define UINT_FAST32_MAX UINT32_MAX
#define UINT_FAST64_MAX UINT64_MAX
// 7.18.2.4 Limits of integer types capable of holding object pointers
#ifdef _WIN64 // [
# define INTPTR_MIN INT64_MIN
# define INTPTR_MAX INT64_MAX
# define UINTPTR_MAX UINT64_MAX
#else // _WIN64 ][
# define INTPTR_MIN INT32_MIN
# define INTPTR_MAX INT32_MAX
# define UINTPTR_MAX UINT32_MAX
#endif // _WIN64 ]
// 7.18.2.5 Limits of greatest-width integer types
#define INTMAX_MIN INT64_MIN
#define INTMAX_MAX INT64_MAX
#define UINTMAX_MAX UINT64_MAX
// 7.18.3 Limits of other integer types
#ifdef _WIN64 // [
# define PTRDIFF_MIN _I64_MIN
# define PTRDIFF_MAX _I64_MAX
#else // _WIN64 ][
# define PTRDIFF_MIN _I32_MIN
# define PTRDIFF_MAX _I32_MAX
#endif // _WIN64 ]
#define SIG_ATOMIC_MIN INT_MIN
#define SIG_ATOMIC_MAX INT_MAX
#ifndef SIZE_MAX // [
# ifdef _WIN64 // [
# define SIZE_MAX _UI64_MAX
# else // _WIN64 ][
# define SIZE_MAX _UI32_MAX
# endif // _WIN64 ]
#endif // SIZE_MAX ]
// WCHAR_MIN and WCHAR_MAX are also defined in <wchar.h>
#ifndef WCHAR_MIN // [
# define WCHAR_MIN 0
#endif // WCHAR_MIN ]
#ifndef WCHAR_MAX // [
# define WCHAR_MAX _UI16_MAX
#endif // WCHAR_MAX ]
#define WINT_MIN 0
#define WINT_MAX _UI16_MAX
#endif // __STDC_LIMIT_MACROS ]
// 7.18.4 Limits of other integer types
#if !defined(__cplusplus) || defined(__STDC_CONSTANT_MACROS) // [ See footnote 224 at page 260
// 7.18.4.1 Macros for minimum-width integer constants
#define INT8_C(val) val##i8
#define INT16_C(val) val##i16
#define INT32_C(val) val##i32
#define INT64_C(val) val##i64
#define UINT8_C(val) val##ui8
#define UINT16_C(val) val##ui16
#define UINT32_C(val) val##ui32
#define UINT64_C(val) val##ui64
// 7.18.4.2 Macros for greatest-width integer constants
#define INTMAX_C INT64_C
#define UINTMAX_C UINT64_C
#endif // __STDC_CONSTANT_MACROS ]
#endif // _MSC_STDINT_H_ ]

View File

@@ -36,17 +36,17 @@
*/
#if defined(_MSC_VER)
# if _MSC_VER < 1800
# error "Microsoft Visual Studio 2013 or higher required"
# if _MSC_VER < 1500
# error "Microsoft Visual Studio 2008 or higher required"
# endif
/*
* Visual Studio will complain if we define the `inline` keyword, but
* Visual Studio 2012 will complain if we define the `inline` keyword, but
* actually it only supports the keyword on C++.
*
* To avoid this the _ALLOW_KEYWORD_MACROS must be set.
*/
# if !defined(_ALLOW_KEYWORD_MACROS)
# if (_MSC_VER >= 1700) && !defined(_ALLOW_KEYWORD_MACROS)
# define _ALLOW_KEYWORD_MACROS
# endif
@@ -81,6 +81,8 @@
/* Intel compiler supports inline keyword */
# elif defined(__WATCOMC__) && (__WATCOMC__ >= 1100)
# define inline __inline
# elif defined(__SUNPRO_C) && defined(__C99FEATURES__)
/* C99 supports inline keyword */
# elif (__STDC_VERSION__ >= 199901L)
/* C99 supports inline keyword */
# else
@@ -98,6 +100,8 @@
#ifndef restrict
# if (__STDC_VERSION__ >= 199901L)
/* C99 */
# elif defined(__SUNPRO_C) && defined(__C99FEATURES__)
/* C99 */
# elif defined(__GNUC__)
# define restrict __restrict__
# elif defined(_MSC_VER)
@@ -114,6 +118,8 @@
#ifndef __func__
# if (__STDC_VERSION__ >= 199901L)
/* C99 */
# elif defined(__SUNPRO_C) && defined(__C99FEATURES__)
/* C99 */
# elif defined(__GNUC__)
# define __func__ __FUNCTION__
# elif defined(_MSC_VER)

View File

@@ -38,16 +38,55 @@
#include "c99_compat.h"
#if defined(_MSC_VER)
/* This is to ensure that we get M_PI, etc. definitions */
#if defined(_MSC_VER) && !defined(_USE_MATH_DEFINES)
#if !defined(_USE_MATH_DEFINES)
#error _USE_MATH_DEFINES define required when building with MSVC
#endif
#if _MSC_VER < 1800
#define isfinite(x) _finite((double)(x))
#define isnan(x) _isnan((double)(x))
#endif /* _MSC_VER < 1800 */
#if !defined(_MSC_VER) && \
__STDC_VERSION__ < 199901L && \
(!defined(_XOPEN_SOURCE) || _XOPEN_SOURCE < 600) && \
!defined(__cplusplus)
#if _MSC_VER < 1800
static inline double log2( double x )
{
const double invln2 = 1.442695041;
return log( x ) * invln2;
}
static inline double
round(double x)
{
return x >= 0.0 ? floor(x + 0.5) : ceil(x - 0.5);
}
static inline float
roundf(float x)
{
return x >= 0.0f ? floorf(x + 0.5f) : ceilf(x - 0.5f);
}
#endif
#ifndef INFINITY
#include <float.h> // DBL_MAX
#define INFINITY (DBL_MAX + DBL_MAX)
#endif
#ifndef NAN
#define NAN (INFINITY - INFINITY)
#endif
#endif /* _MSC_VER */
#if (defined(_MSC_VER) && _MSC_VER < 1800) || \
(!defined(_MSC_VER) && \
__STDC_VERSION__ < 199901L && \
(!defined(_XOPEN_SOURCE) || _XOPEN_SOURCE < 600) && \
!defined(__cplusplus))
static inline long int
lrint(double d)
@@ -101,18 +140,6 @@ llrintf(float f)
return rounded;
}
static inline float
exp2f(float f)
{
return powf(2.0f, f);
}
static inline double
exp2(double d)
{
return pow(2.0, d);
}
#endif /* C99 */

View File

@@ -69,8 +69,6 @@ typedef struct ID3DPresentVtbl
HRESULT (WINAPI *SetCursor)(ID3DPresent *This, void *pBitmap, POINT *pHotspot, BOOL bShow);
HRESULT (WINAPI *SetGammaRamp)(ID3DPresent *This, const D3DGAMMARAMP *pRamp, HWND hWndOverride);
HRESULT (WINAPI *GetWindowInfo)(ID3DPresent *This, HWND hWnd, int *width, int *height, int *depth);
/* Available since version 1.1 */
BOOL (WINAPI *GetWindowOccluded)(ID3DPresent *This);
} ID3DPresentVtbl;
struct ID3DPresent
@@ -98,7 +96,6 @@ struct ID3DPresent
#define ID3DPresent_SetCursor(p,a,b,c) (p)->lpVtbl->SetCursor(p,a,b,c)
#define ID3DPresent_SetGammaRamp(p,a,b) (p)->lpVtbl->SetGammaRamp(p,a,b)
#define ID3DPresent_GetWindowInfo(p,a,b,c,d) (p)->lpVtbl->GetWindowSize(p,a,b,c,d)
#define ID3DPresent_GetWindowOccluded(p) (p)->lpVtbl->GetWindowOccluded(p)
typedef struct ID3DPresentGroupVtbl
{

View File

@@ -109,57 +109,22 @@ CHIPSET(0x162A, bdw_gt3, "Intel(R) Iris Pro P6300 (Broadwell GT3e)")
CHIPSET(0x162B, bdw_gt3, "Intel(R) Iris 6100 (Broadwell GT3)")
CHIPSET(0x162D, bdw_gt3, "Intel(R) Broadwell GT3")
CHIPSET(0x162E, bdw_gt3, "Intel(R) Broadwell GT3")
CHIPSET(0x1902, skl_gt1, "Intel(R) HD Graphics 510 (Skylake GT1)")
CHIPSET(0x1906, skl_gt1, "Intel(R) HD Graphics 510 (Skylake GT1)")
CHIPSET(0x190A, skl_gt1, "Intel(R) Skylake GT1")
CHIPSET(0x190B, skl_gt1, "Intel(R) HD Graphics 510 (Skylake GT1)")
CHIPSET(0x190E, skl_gt1, "Intel(R) Skylake GT1")
CHIPSET(0x1912, skl_gt2, "Intel(R) HD Graphics 530 (Skylake GT2)")
CHIPSET(0x1913, skl_gt2, "Intel(R) Skylake GT2f")
CHIPSET(0x1915, skl_gt2, "Intel(R) Skylake GT2f")
CHIPSET(0x1916, skl_gt2, "Intel(R) HD Graphics 520 (Skylake GT2)")
CHIPSET(0x1917, skl_gt2, "Intel(R) Skylake GT2f")
CHIPSET(0x191A, skl_gt2, "Intel(R) Skylake GT2")
CHIPSET(0x191B, skl_gt2, "Intel(R) HD Graphics 530 (Skylake GT2)")
CHIPSET(0x191D, skl_gt2, "Intel(R) HD Graphics P530 (Skylake GT2)")
CHIPSET(0x191E, skl_gt2, "Intel(R) HD Graphics 515 (Skylake GT2)")
CHIPSET(0x1921, skl_gt2, "Intel(R) HD Graphics 520 (Skylake GT2)")
CHIPSET(0x1923, skl_gt3, "Intel(R) Skylake GT3e")
CHIPSET(0x1926, skl_gt3, "Intel(R) Iris Graphics 540 (Skylake GT3e)")
CHIPSET(0x1927, skl_gt3, "Intel(R) Iris Graphics 550 (Skylake GT3e)")
CHIPSET(0x192A, skl_gt4, "Intel(R) Skylake GT4")
CHIPSET(0x192B, skl_gt3, "Intel(R) Iris Graphics 555 (Skylake GT3e)")
CHIPSET(0x192D, skl_gt3, "Intel(R) Iris Graphics P555 (Skylake GT3e)")
CHIPSET(0x1932, skl_gt4, "Intel(R) Iris Pro Graphics 580 (Skylake GT4e)")
CHIPSET(0x193A, skl_gt4, "Intel(R) Iris Pro Graphics P580 (Skylake GT4e)")
CHIPSET(0x193B, skl_gt4, "Intel(R) Iris Pro Graphics 580 (Skylake GT4e)")
CHIPSET(0x193D, skl_gt4, "Intel(R) Iris Pro Graphics P580 (Skylake GT4e)")
CHIPSET(0x5902, kbl_gt1, "Intel(R) Kabylake GT1")
CHIPSET(0x5906, kbl_gt1, "Intel(R) Kabylake GT1")
CHIPSET(0x590A, kbl_gt1, "Intel(R) Kabylake GT1")
CHIPSET(0x590B, kbl_gt1, "Intel(R) Kabylake GT1")
CHIPSET(0x590E, kbl_gt1, "Intel(R) Kabylake GT1")
CHIPSET(0x5913, kbl_gt1_5, "Intel(R) Kabylake GT1.5")
CHIPSET(0x5915, kbl_gt1_5, "Intel(R) Kabylake GT1.5")
CHIPSET(0x5917, kbl_gt1_5, "Intel(R) Kabylake GT1.5")
CHIPSET(0x5912, kbl_gt2, "Intel(R) Kabylake GT2")
CHIPSET(0x5916, kbl_gt2, "Intel(R) Kabylake GT2")
CHIPSET(0x591A, kbl_gt2, "Intel(R) Kabylake GT2")
CHIPSET(0x591B, kbl_gt2, "Intel(R) Kabylake GT2")
CHIPSET(0x591D, kbl_gt2, "Intel(R) Kabylake GT2")
CHIPSET(0x591E, kbl_gt2, "Intel(R) Kabylake GT2")
CHIPSET(0x5921, kbl_gt2, "Intel(R) Kabylake GT2F")
CHIPSET(0x5926, kbl_gt3, "Intel(R) Kabylake GT3")
CHIPSET(0x592A, kbl_gt3, "Intel(R) Kabylake GT3")
CHIPSET(0x592B, kbl_gt3, "Intel(R) Kabylake GT3")
CHIPSET(0x5932, kbl_gt4, "Intel(R) Kabylake GT4")
CHIPSET(0x593A, kbl_gt4, "Intel(R) Kabylake GT4")
CHIPSET(0x593B, kbl_gt4, "Intel(R) Kabylake GT4")
CHIPSET(0x593D, kbl_gt4, "Intel(R) Kabylake GT4")
CHIPSET(0x1902, skl_gt1, "Intel(R) Skylake DT GT1")
CHIPSET(0x1906, skl_gt1, "Intel(R) Skylake ULT GT1")
CHIPSET(0x190A, skl_gt1, "Intel(R) Skylake SRV GT1")
CHIPSET(0x190B, skl_gt1, "Intel(R) Skylake Halo GT1")
CHIPSET(0x190E, skl_gt1, "Intel(R) Skylake ULX GT1")
CHIPSET(0x1912, skl_gt2, "Intel(R) Skylake DT GT2")
CHIPSET(0x1916, skl_gt2, "Intel(R) Skylake ULT GT2")
CHIPSET(0x191A, skl_gt2, "Intel(R) Skylake SRV GT2")
CHIPSET(0x191B, skl_gt2, "Intel(R) Skylake Halo GT2")
CHIPSET(0x191D, skl_gt2, "Intel(R) Skylake WKS GT2")
CHIPSET(0x191E, skl_gt2, "Intel(R) Skylake ULX GT2")
CHIPSET(0x1921, skl_gt2, "Intel(R) Skylake ULT GT2F")
CHIPSET(0x1926, skl_gt3, "Intel(R) Skylake ULT GT3")
CHIPSET(0x192A, skl_gt3, "Intel(R) Skylake SRV GT3")
CHIPSET(0x192B, skl_gt3, "Intel(R) Skylake Halo GT3")
CHIPSET(0x22B0, chv, "Intel(R) HD Graphics (Cherryview)")
CHIPSET(0x22B1, chv, "Intel(R) HD Graphics (Cherryview)")
CHIPSET(0x22B2, chv, "Intel(R) HD Graphics (Cherryview)")
CHIPSET(0x22B3, chv, "Intel(R) HD Graphics (Cherryview)")
CHIPSET(0x0A84, bxt, "Intel(R) HD Graphics (Broxton)")
CHIPSET(0x1A84, bxt, "Intel(R) HD Graphics (Broxton)")
CHIPSET(0x5A84, bxt, "Intel(R) HD Graphics (Broxton)")

View File

@@ -63,7 +63,6 @@ CHIPSET(0x6608, OLAND_6608, OLAND)
CHIPSET(0x6610, OLAND_6610, OLAND)
CHIPSET(0x6611, OLAND_6611, OLAND)
CHIPSET(0x6613, OLAND_6613, OLAND)
CHIPSET(0x6617, OLAND_6617, OLAND)
CHIPSET(0x6620, OLAND_6620, OLAND)
CHIPSET(0x6621, OLAND_6621, OLAND)
CHIPSET(0x6623, OLAND_6623, OLAND)
@@ -157,29 +156,3 @@ CHIPSET(0x67B8, HAWAII_67B8, HAWAII)
CHIPSET(0x67B9, HAWAII_67B9, HAWAII)
CHIPSET(0x67BA, HAWAII_67BA, HAWAII)
CHIPSET(0x67BE, HAWAII_67BE, HAWAII)
CHIPSET(0x6900, ICELAND_, ICELAND)
CHIPSET(0x6901, ICELAND_, ICELAND)
CHIPSET(0x6902, ICELAND_, ICELAND)
CHIPSET(0x6903, ICELAND_, ICELAND)
CHIPSET(0x6907, ICELAND_, ICELAND)
CHIPSET(0x6920, TONGA_, TONGA)
CHIPSET(0x6921, TONGA_, TONGA)
CHIPSET(0x6928, TONGA_, TONGA)
CHIPSET(0x6929, TONGA_, TONGA)
CHIPSET(0x692B, TONGA_, TONGA)
CHIPSET(0x692F, TONGA_, TONGA)
CHIPSET(0x6930, TONGA_, TONGA)
CHIPSET(0x6938, TONGA_, TONGA)
CHIPSET(0x6939, TONGA_, TONGA)
CHIPSET(0x9870, CARRIZO_, CARRIZO)
CHIPSET(0x9874, CARRIZO_, CARRIZO)
CHIPSET(0x9875, CARRIZO_, CARRIZO)
CHIPSET(0x9876, CARRIZO_, CARRIZO)
CHIPSET(0x9877, CARRIZO_, CARRIZO)
CHIPSET(0x7300, FIJI_, FIJI)
CHIPSET(0x98E4, STONEY_, STONEY)

View File

@@ -1,2 +0,0 @@
CHIPSET(0x0010, VIRTGL, VIRTGL)
CHIPSET(0x1050, VIRTGL, VIRTGL)

View File

@@ -0,0 +1,90 @@
//
// File: vk_platform.h
//
/*
** Copyright (c) 2014-2015 The Khronos Group Inc.
**
** Permission is hereby granted, free of charge, to any person obtaining a
** copy of this software and/or associated documentation files (the
** "Materials"), to deal in the Materials without restriction, including
** without limitation the rights to use, copy, modify, merge, publish,
** distribute, sublicense, and/or sell copies of the Materials, and to
** permit persons to whom the Materials are furnished to do so, subject to
** the following conditions:
**
** The above copyright notice and this permission notice shall be included
** in all copies or substantial portions of the Materials.
**
** THE MATERIALS ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
** EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
** MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
** IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
** CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
** TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
** MATERIALS OR THE USE OR OTHER DEALINGS IN THE MATERIALS.
*/
#ifndef __VK_PLATFORM_H__
#define __VK_PLATFORM_H__
#ifdef __cplusplus
extern "C"
{
#endif // __cplusplus
/*
***************************************************************************************************
* Platform-specific directives and type declarations
***************************************************************************************************
*/
#if defined(_WIN32)
// On Windows, VKAPI should equate to the __stdcall convention
#define VKAPI __stdcall
#elif defined(__GNUC__)
// On other platforms using GCC, VKAPI stays undefined
#define VKAPI
#else
// Unsupported Platform!
#error "Unsupported OS Platform detected!"
#endif
#include <stddef.h>
#if !defined(VK_NO_STDINT_H)
#if defined(_MSC_VER) && (_MSC_VER < 1600)
typedef signed __int8 int8_t;
typedef unsigned __int8 uint8_t;
typedef signed __int16 int16_t;
typedef unsigned __int16 uint16_t;
typedef signed __int32 int32_t;
typedef unsigned __int32 uint32_t;
typedef signed __int64 int64_t;
typedef unsigned __int64 uint64_t;
#else
#include <stdint.h>
#endif
#endif // !defined(VK_NO_STDINT_H)
typedef uint64_t VkDeviceSize;
typedef uint32_t VkBool32;
typedef uint32_t VkSampleMask;
typedef uint32_t VkFlags;
#if (UINTPTR_MAX >= UINT64_MAX)
#define VK_UINTPTRLEAST64_MAX UINTPTR_MAX
typedef uintptr_t VkUintPtrLeast64;
#else
#define VK_UINTPTRLEAST64_MAX UINT64_MAX
typedef uint64_t VkUintPtrLeast64;
#endif
#ifdef __cplusplus
} // extern "C"
#endif // __cplusplus
#endif // __VK_PLATFORM_H__

View File

@@ -0,0 +1,212 @@
//
// File: vk_wsi_display.h
//
/*
** Copyright (c) 2014 The Khronos Group Inc.
**
** Permission is hereby granted, free of charge, to any person obtaining a
** copy of this software and/or associated documentation files (the
** "Materials"), to deal in the Materials without restriction, including
** without limitation the rights to use, copy, modify, merge, publish,
** distribute, sublicense, and/or sell copies of the Materials, and to
** permit persons to whom the Materials are furnished to do so, subject to
** the following conditions:
**
** The above copyright notice and this permission notice shall be included
** in all copies or substantial portions of the Materials.
**
** THE MATERIALS ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
** EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
** MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
** IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
** CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
** TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
** MATERIALS OR THE USE OR OTHER DEALINGS IN THE MATERIALS.
*/
#ifndef __VK_WSI_LUNARG_H__
#define __VK_WSI_LUNARG_H__
#include "vulkan.h"
#define VK_WSI_LUNARG_REVISION 3
#define VK_WSI_LUNARG_EXTENSION_NUMBER 1
#ifdef __cplusplus
extern "C"
{
#endif // __cplusplus
// This macro defines INT_MAX in enumerations to force compilers to use 32 bits
// to represent them. This may or may not be necessary on some compilers. The
// option to compile it out may allow compilers that warn about missing enumerants
// in switch statements to be silenced.
// Using this macro is not needed for flag bit enums because those aren't used
// as storage type anywhere.
#define VK_MAX_ENUM(Prefix) VK_##Prefix##_MAX_ENUM = 0x7FFFFFFF
// This macro defines the BEGIN_RANGE, END_RANGE, NUM, and MAX_ENUM constants for
// the enumerations.
#define VK_ENUM_RANGE(Prefix, First, Last) \
VK_##Prefix##_BEGIN_RANGE = VK_##Prefix##_##First, \
VK_##Prefix##_END_RANGE = VK_##Prefix##_##Last, \
VK_NUM_##Prefix = (VK_##Prefix##_END_RANGE - VK_##Prefix##_BEGIN_RANGE + 1), \
VK_MAX_ENUM(Prefix)
// This is a helper macro to define the value of flag bit enum values.
#define VK_BIT(bit) (1 << (bit))
// ------------------------------------------------------------------------------------------------
// Objects
VK_DEFINE_HANDLE(VkDisplayWSI)
VK_DEFINE_HANDLE(VkSwapChainWSI)
// ------------------------------------------------------------------------------------------------
// Enumeration constants
#define VK_WSI_LUNARG_ENUM(type,id) ((type)(VK_WSI_LUNARG_EXTENSION_NUMBER * -1000 + (id)))
// Extend VkPhysicalDeviceInfoType enum with extension specific constants
#define VK_PHYSICAL_DEVICE_INFO_TYPE_DISPLAY_PROPERTIES_WSI VK_WSI_LUNARG_ENUM(VkPhysicalDeviceInfoType, 0)
#define VK_PHYSICAL_DEVICE_INFO_TYPE_QUEUE_PRESENT_PROPERTIES_WSI VK_WSI_LUNARG_ENUM(VkPhysicalDeviceInfoType, 1)
// Extend VkStructureType enum with extension specific constants
#define VK_STRUCTURE_TYPE_SWAP_CHAIN_CREATE_INFO_WSI VK_WSI_LUNARG_ENUM(VkStructureType, 0)
#define VK_STRUCTURE_TYPE_PRESENT_INFO_WSI VK_WSI_LUNARG_ENUM(VkStructureType, 1)
// Extend VkImageLayout enum with extension specific constants
#define VK_IMAGE_LAYOUT_PRESENT_SOURCE_WSI VK_WSI_LUNARG_ENUM(VkImageLayout, 0)
// ------------------------------------------------------------------------------------------------
// Enumerations
typedef enum VkDisplayInfoTypeWSI_
{
// Info type for vkGetDisplayInfo()
VK_DISPLAY_INFO_TYPE_FORMAT_PROPERTIES_WSI = 0x00000003, // Return the VkFormat(s) supported for swap chains with the display
VK_ENUM_RANGE(DISPLAY_INFO_TYPE, FORMAT_PROPERTIES_WSI, FORMAT_PROPERTIES_WSI)
} VkDisplayInfoTypeWSI;
typedef enum VkSwapChainInfoTypeWSI_
{
// Info type for vkGetSwapChainInfo()
VK_SWAP_CHAIN_INFO_TYPE_PERSISTENT_IMAGES_WSI = 0x00000000, // Return information about the persistent images of the swapchain
VK_ENUM_RANGE(SWAP_CHAIN_INFO_TYPE, PERSISTENT_IMAGES_WSI, PERSISTENT_IMAGES_WSI)
} VkSwapChainInfoTypeWSI;
// ------------------------------------------------------------------------------------------------
// Flags
typedef VkFlags VkSwapModeFlagsWSI;
typedef enum VkSwapModeFlagBitsWSI_
{
VK_SWAP_MODE_FLIP_BIT_WSI = VK_BIT(0),
VK_SWAP_MODE_BLIT_BIT_WSI = VK_BIT(1),
} VkSwapModeFlagBitsWSI;
// ------------------------------------------------------------------------------------------------
// Structures
typedef struct VkDisplayPropertiesWSI_
{
VkDisplayWSI display; // Handle of the display object
VkExtent2D physicalResolution; // Max resolution for CRT?
} VkDisplayPropertiesWSI;
typedef struct VkDisplayFormatPropertiesWSI_
{
VkFormat swapChainFormat; // Format of the images of the swap chain
} VkDisplayFormatPropertiesWSI;
typedef struct VkSwapChainCreateInfoWSI_
{
VkStructureType sType; // Must be VK_STRUCTURE_TYPE_SWAP_CHAIN_CREATE_INFO_WSI
const void* pNext; // Pointer to next structure
// TBD: It is not yet clear what the use will be for the following two
// values. It seems to be needed for more-global window-system handles
// (e.g. X11 display). If not needed for the SDK, we will drop it from
// this extension, and from a future version of this header.
const void* pNativeWindowSystemHandle; // Pointer to native window system handle
const void* pNativeWindowHandle; // Pointer to native window handle
uint32_t displayCount; // Number of displays the swap chain is created for
const VkDisplayWSI* pDisplays; // displayCount number of display objects the swap chain is created for
uint32_t imageCount; // Number of images in the swap chain
VkFormat imageFormat; // Format of the images of the swap chain
VkExtent2D imageExtent; // Width and height of the images of the swap chain
uint32_t imageArraySize; // Number of layers of the images of the swap chain (needed for multi-view rendering)
VkFlags imageUsageFlags; // Usage flags for the images of the swap chain (see VkImageUsageFlags)
VkFlags swapModeFlags; // Allowed swap modes (see VkSwapModeFlagsWSI)
} VkSwapChainCreateInfoWSI;
typedef struct VkSwapChainImageInfoWSI_
{
VkImage image; // Persistent swap chain image handle
VkDeviceMemory memory; // Persistent swap chain image's memory handle
} VkSwapChainImageInfoWSI;
typedef struct VkPhysicalDeviceQueuePresentPropertiesWSI_
{
VkBool32 supportsPresent; // Tells whether the queue supports presenting
} VkPhysicalDeviceQueuePresentPropertiesWSI;
typedef struct VkPresentInfoWSI_
{
VkStructureType sType; // Must be VK_STRUCTURE_TYPE_PRESENT_INFO_WSI
const void* pNext; // Pointer to next structure
VkImage image; // Image to present
uint32_t flipInterval; // Flip interval
} VkPresentInfoWSI;
// ------------------------------------------------------------------------------------------------
// Function types
typedef VkResult (VKAPI *PFN_vkGetDisplayInfoWSI)(VkDisplayWSI display, VkDisplayInfoTypeWSI infoType, size_t* pDataSize, void* pData);
typedef VkResult (VKAPI *PFN_vkCreateSwapChainWSI)(VkDevice device, const VkSwapChainCreateInfoWSI* pCreateInfo, VkSwapChainWSI* pSwapChain);
typedef VkResult (VKAPI *PFN_vkDestroySwapChainWSI)(VkSwapChainWSI swapChain);
typedef VkResult (VKAPI *PFN_vkGetSwapChainInfoWSI)(VkSwapChainWSI swapChain, VkSwapChainInfoTypeWSI infoType, size_t* pDataSize, void* pData);
typedef VkResult (VKAPI *PFN_vkQueuePresentWSI)(VkQueue queue, const VkPresentInfoWSI* pPresentInfo);
// ------------------------------------------------------------------------------------------------
// Function prototypes
#ifdef VK_PROTOTYPES
VkResult VKAPI vkGetDisplayInfoWSI(
VkDisplayWSI display,
VkDisplayInfoTypeWSI infoType,
size_t* pDataSize,
void* pData);
VkResult VKAPI vkCreateSwapChainWSI(
VkDevice device,
const VkSwapChainCreateInfoWSI* pCreateInfo,
VkSwapChainWSI* pSwapChain);
VkResult VKAPI vkDestroySwapChainWSI(
VkSwapChainWSI swapChain);
VkResult VKAPI vkGetSwapChainInfoWSI(
VkSwapChainWSI swapChain,
VkSwapChainInfoTypeWSI infoType,
size_t* pDataSize,
void* pData);
VkResult VKAPI vkQueuePresentWSI(
VkQueue queue,
const VkPresentInfoWSI* pPresentInfo);
#endif // VK_PROTOTYPES
#ifdef __cplusplus
} // extern "C"
#endif // __cplusplus
#endif // __VK_WSI_LUNARG_H__

3047
include/vulkan/vulkan.h Normal file

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,61 @@
/*
* Copyright © 2015 Intel Corporation
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*/
#ifndef __VULKAN_INTEL_H__
#define __VULKAN_INTEL_H__
#include "vulkan.h"
#ifdef __cplusplus
extern "C"
{
#endif // __cplusplus
#define VK_STRUCTURE_TYPE_DMA_BUF_IMAGE_CREATE_INFO_INTEL 1024
typedef struct VkDmaBufImageCreateInfo_
{
VkStructureType sType; // Must be VK_STRUCTURE_TYPE_DMA_BUF_IMAGE_CREATE_INFO_INTEL
const void* pNext; // Pointer to next structure.
int fd;
VkFormat format;
VkExtent3D extent; // Depth must be 1
uint32_t strideInBytes;
} VkDmaBufImageCreateInfo;
typedef VkResult (VKAPI *PFN_vkCreateDmaBufImageINTEL)(VkDevice device, const VkDmaBufImageCreateInfo* pCreateInfo, VkDeviceMemory* pMem, VkImage* pImage);
#ifdef VK_PROTOTYPES
VkResult VKAPI vkCreateDmaBufImageINTEL(
VkDevice _device,
const VkDmaBufImageCreateInfo* pCreateInfo,
VkDeviceMemory* pMem,
VkImage* pImage);
#endif
#ifdef __cplusplus
} // extern "C"
#endif // __cplusplus
#endif // __VULKAN_INTEL_H__

View File

@@ -3,9 +3,9 @@
if BUILD_SHARED
if HAVE_COMPAT_SYMLINKS
all-local : .install-gallium-links
all-local : .libs/install-gallium-links
.install-gallium-links : $(dri_LTLIBRARIES) $(egl_LTLIBRARIES) $(lib_LTLIBRARIES)
.libs/install-gallium-links : $(dri_LTLIBRARIES) $(egl_LTLIBRARIES) $(lib_LTLIBRARIES)
$(AM_V_GEN)$(MKDIR_P) $(top_builddir)/$(LIB_DIR); \
link_dir=$(top_builddir)/$(LIB_DIR)/gallium; \
if test x$(egl_LTLIBRARIES) != x; then \
@@ -23,15 +23,4 @@ all-local : .install-gallium-links
fi; \
done && touch $@
endif
clean-local:
for f in $(notdir $(dri_LTLIBRARIES:%.la=.libs/%.$(LIB_EXT)*)) \
$(notdir $(egl_LTLIBRARIES:%.la=.libs/%.$(LIB_EXT)*)) \
$(notdir $(lib_LTLIBRARIES:%.la=.libs/%.$(LIB_EXT)*)); do \
echo $$f; \
$(RM) $(top_builddir)/$(LIB_DIR)/gallium/$$f; \
done;
rmdir $(top_builddir)/$(LIB_DIR)/gallium || true
$(RM) .install-gallium-links
endif

View File

@@ -94,8 +94,16 @@ def msvc2013_compat(env):
'-Werror=pointer-arith',
])
def msvc2008_compat(env):
msvc2013_compat(env)
if env['gcc']:
env.Append(CFLAGS = [
'-Werror=declaration-after-statement',
])
def createMSVCCompatMethods(env):
env.AddMethod(msvc2013_compat, 'MSVC2013Compat')
env.AddMethod(msvc2008_compat, 'MSVC2008Compat')
def num_jobs():
@@ -292,7 +300,6 @@ def generate(env):
# C preprocessor options
cppdefines = []
cppdefines += ['__STDC_LIMIT_MACROS', '__STDC_CONSTANT_MACROS']
if env['build'] in ('debug', 'checked'):
cppdefines += ['DEBUG']
else:
@@ -471,12 +478,20 @@ def generate(env):
# See also:
# - http://msdn.microsoft.com/en-us/library/19z1t1wy.aspx
# - cl /?
if 'MSVC_VERSION' not in env or distutils.version.LooseVersion(env['MSVC_VERSION']) < distutils.version.LooseVersion('12.0'):
# Use bundled stdbool.h and stdint.h headers for older MSVC
# versions. stdint.h was introduced in MSVC 2010, but stdbool.h
# was only introduced in MSVC 2013.
top_dir = os.path.abspath(os.path.join(os.path.dirname(__file__), '..'))
env.Append(CPPPATH = [os.path.join(top_dir, 'include/c99')])
if env['build'] == 'debug':
ccflags += [
'/Od', # disable optimizations
'/Oi', # enable intrinsic functions
]
else:
if 'MSVC_VERSION' in env and distutils.version.LooseVersion(env['MSVC_VERSION']) < distutils.version.LooseVersion('11.0'):
print 'scons: warning: Visual Studio versions prior to 2012 are known to produce incorrect code when optimizations are enabled ( https://bugs.freedesktop.org/show_bug.cgi?id=58718 )'
ccflags += [
'/O2', # optimize for speed
]

View File

@@ -106,19 +106,7 @@ def generate(env):
])
env.Prepend(LIBPATH = [os.path.join(llvm_dir, 'lib')])
# LIBS should match the output of `llvm-config --libs engine mcjit bitwriter x86asmprinter`
if llvm_version >= distutils.version.LooseVersion('3.7'):
env.Prepend(LIBS = [
'LLVMBitWriter', 'LLVMX86Disassembler', 'LLVMX86AsmParser',
'LLVMX86CodeGen', 'LLVMSelectionDAG', 'LLVMAsmPrinter',
'LLVMCodeGen', 'LLVMScalarOpts', 'LLVMProfileData',
'LLVMInstCombine', 'LLVMInstrumentation', 'LLVMTransformUtils', 'LLVMipa',
'LLVMAnalysis', 'LLVMX86Desc', 'LLVMMCDisassembler',
'LLVMX86Info', 'LLVMX86AsmPrinter', 'LLVMX86Utils',
'LLVMMCJIT', 'LLVMTarget', 'LLVMExecutionEngine',
'LLVMRuntimeDyld', 'LLVMObject', 'LLVMMCParser',
'LLVMBitReader', 'LLVMMC', 'LLVMCore', 'LLVMSupport'
])
elif llvm_version >= distutils.version.LooseVersion('3.6'):
if llvm_version >= distutils.version.LooseVersion('3.6'):
env.Prepend(LIBS = [
'LLVMBitWriter', 'LLVMX86Disassembler', 'LLVMX86AsmParser',
'LLVMX86CodeGen', 'LLVMSelectionDAG', 'LLVMAsmPrinter',

View File

@@ -19,13 +19,12 @@
# FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
# IN THE SOFTWARE.
AUTOMAKE_OPTIONS = subdir-objects
SUBDIRS = . gtest util mapi/glapi/gen mapi
# include only conditionally ?
SUBDIRS += compiler
if NEED_OPENGL_COMMON
SUBDIRS += mesa
SUBDIRS += glsl mesa
endif
SUBDIRS += loader
@@ -38,12 +37,16 @@ if HAVE_EGL_PLATFORM_WAYLAND
SUBDIRS += egl/wayland/wayland-egl egl/wayland/wayland-drm
endif
if HAVE_EGL_DRIVER_DRI2
SUBDIRS += egl/drivers/dri2
endif
if HAVE_GBM
SUBDIRS += gbm
endif
if HAVE_EGL
SUBDIRS += egl
SUBDIRS += egl/main
endif
if HAVE_GALLIUM
@@ -51,11 +54,17 @@ SUBDIRS += gallium
endif
EXTRA_DIST = \
egl/drivers/haiku \
egl/docs \
getopt hgl SConscript
AM_CFLAGS = $(VISIBILITY_CFLAGS)
AM_CXXFLAGS = $(VISIBILITY_CXXFLAGS)
if HAVE_VULKAN
SUBDIRS += vulkan
endif
AM_CPPFLAGS = \
-I$(top_srcdir)/include/ \
-I$(top_srcdir)/src/mapi/ \

View File

@@ -5,7 +5,7 @@ if env['platform'] == 'windows':
SConscript('getopt/SConscript')
SConscript('util/SConscript')
SConscript('compiler/SConscript')
SConscript('glsl/SConscript')
if env['hostonly']:
# We are just compiling the things necessary on the host for cross
@@ -31,8 +31,13 @@ SConscript('mesa/SConscript')
if not env['embedded']:
if env['platform'] not in ('cygwin', 'darwin', 'freebsd', 'haiku', 'windows'):
SConscript('glx/SConscript')
if env['platform'] not in ['darwin', 'haiku', 'sunos', 'windows']:
if env['dri']:
SConscript('egl/drivers/dri2/SConscript')
SConscript('egl/main/SConscript')
if env['platform'] == 'haiku':
SConscript('egl/SConscript')
SConscript('egl/drivers/haiku/SConscript')
SConscript('egl/main/SConscript')
if env['gles']:
SConscript('mapi/shared-glapi/SConscript')

View File

@@ -1 +0,0 @@
glsl_compiler

View File

@@ -1,67 +0,0 @@
# Mesa 3-D graphics library
#
# Copyright (C) 2015 Intel Corporation
#
# Permission is hereby granted, free of charge, to any person obtaining a
# copy of this software and associated documentation files (the "Software"),
# to deal in the Software without restriction, including without limitation
# the rights to use, copy, modify, merge, publish, distribute, sublicense,
# and/or sell copies of the Software, and to permit persons to whom the
# Software is furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included
# in all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
# THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
# FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
# DEALINGS IN THE SOFTWARE.
LOCAL_PATH := $(call my-dir)
include $(LOCAL_PATH)/Makefile.sources
# ---------------------------------------
# Build libmesa_compiler
# ---------------------------------------
include $(CLEAR_VARS)
LOCAL_SRC_FILES := $(LIBCOMPILER_FILES)
LOCAL_C_INCLUDES := \
$(MESA_TOP)/src/mapi \
$(MESA_TOP)/src/mesa \
$(MESA_TOP)/src/gallium/include \
$(MESA_TOP)/src/gallium/auxiliary
LOCAL_MODULE := libmesa_compiler
include $(MESA_COMMON_MK)
include $(BUILD_STATIC_LIBRARY)
# ---------------------------------------
# Build libmesa_nir
# ---------------------------------------
include $(CLEAR_VARS)
LOCAL_SRC_FILES := \
$(NIR_FILES)
LOCAL_C_INCLUDES := \
$(MESA_TOP)/src/mapi \
$(MESA_TOP)/src/mesa \
$(MESA_TOP)/src/gallium/include \
$(MESA_TOP)/src/gallium/auxiliary
LOCAL_STATIC_LIBRARIES := libmesa_compiler
LOCAL_MODULE := libmesa_nir
include $(LOCAL_PATH)/Android.gen.mk
include $(MESA_COMMON_MK)
include $(BUILD_STATIC_LIBRARY)

View File

@@ -1,325 +0,0 @@
#
# Copyright © 2012 Jon TURNEY
# Copyright (C) 2015 Intel Corporation
#
# Permission is hereby granted, free of charge, to any person obtaining a
# copy of this software and associated documentation files (the "Software"),
# to deal in the Software without restriction, including without limitation
# the rights to use, copy, modify, merge, publish, distribute, sublicense,
# and/or sell copies of the Software, and to permit persons to whom the
# Software is furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice (including the next
# paragraph) shall be included in all copies or substantial portions of the
# Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
# THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
# FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
# IN THE SOFTWARE.
include Makefile.sources
AM_CPPFLAGS = \
-I$(top_srcdir)/include \
-I$(top_srcdir)/src \
-I$(top_srcdir)/src/mapi \
-I$(top_srcdir)/src/mesa/ \
-I$(top_builddir)/src/compiler/glsl\
-I$(top_srcdir)/src/compiler/glsl\
-I$(top_srcdir)/src/compiler/glsl/glcpp\
-I$(top_srcdir)/src/gallium/include \
-I$(top_srcdir)/src/gallium/auxiliary \
-I$(top_srcdir)/src/gtest/include \
$(DEFINES)
AM_CFLAGS = \
$(VISIBILITY_CFLAGS) \
$(MSVC2013_COMPAT_CFLAGS)
AM_CXXFLAGS = \
$(VISIBILITY_CXXFLAGS) \
$(MSVC2013_COMPAT_CXXFLAGS)
noinst_LTLIBRARIES = libcompiler.la
libcompiler_la_SOURCES = $(LIBCOMPILER_FILES)
check_PROGRAMS =
TESTS =
BUILT_SOURCES =
CLEANFILES =
EXTRA_DIST = SConscript
EXTRA_DIST += glsl/tests glsl/glcpp/tests glsl/README \
glsl/TODO glsl/glcpp/README \
glsl/glsl_lexer.ll \
glsl/glsl_parser.yy \
glsl/glcpp/glcpp-lex.l \
glsl/glcpp/glcpp-parse.y \
glsl/Makefile.sources \
glsl/SConscript
TESTS += glsl/glcpp/tests/glcpp-test \
glsl/glcpp/tests/glcpp-test-cr-lf \
glsl/tests/blob-test \
glsl/tests/general-ir-test \
glsl/tests/optimization-test \
glsl/tests/sampler-types-test \
glsl/tests/uniform-initializer-test
TESTS_ENVIRONMENT= \
export PYTHON2=$(PYTHON2); \
export PYTHON_FLAGS=$(PYTHON_FLAGS);
check_PROGRAMS += \
glsl/glcpp/glcpp \
glsl/glsl_test \
glsl/tests/blob-test \
glsl/tests/general-ir-test \
glsl/tests/sampler-types-test \
glsl/tests/uniform-initializer-test
noinst_PROGRAMS = glsl_compiler
glsl_tests_blob_test_SOURCES = \
glsl/tests/blob_test.c
glsl_tests_blob_test_LDADD = \
glsl/libglsl.la
glsl_tests_general_ir_test_SOURCES = \
glsl/standalone_scaffolding.cpp \
glsl/tests/builtin_variable_test.cpp \
glsl/tests/invalidate_locations_test.cpp \
glsl/tests/general_ir_test.cpp \
glsl/tests/varyings_test.cpp
glsl_tests_general_ir_test_CFLAGS = \
$(PTHREAD_CFLAGS)
glsl_tests_general_ir_test_LDADD = \
$(top_builddir)/src/gtest/libgtest.la \
glsl/libglsl.la \
$(top_builddir)/src/libglsl_util.la \
$(PTHREAD_LIBS)
glsl_tests_uniform_initializer_test_SOURCES = \
glsl/tests/copy_constant_to_storage_tests.cpp \
glsl/tests/set_uniform_initializer_tests.cpp \
glsl/tests/uniform_initializer_utils.cpp \
glsl/tests/uniform_initializer_utils.h
glsl_tests_uniform_initializer_test_CFLAGS = \
$(PTHREAD_CFLAGS)
glsl_tests_uniform_initializer_test_LDADD = \
$(top_builddir)/src/gtest/libgtest.la \
glsl/libglsl.la \
$(top_builddir)/src/libglsl_util.la \
$(PTHREAD_LIBS)
glsl_tests_sampler_types_test_SOURCES = \
glsl/tests/sampler_types_test.cpp
glsl_tests_sampler_types_test_CFLAGS = \
$(PTHREAD_CFLAGS)
glsl_tests_sampler_types_test_LDADD = \
$(top_builddir)/src/gtest/libgtest.la \
glsl/libglsl.la \
$(top_builddir)/src/libglsl_util.la \
$(PTHREAD_LIBS)
noinst_LTLIBRARIES += glsl/libglsl.la glsl/libglcpp.la
glsl_libglcpp_la_LIBADD = \
$(top_builddir)/src/util/libmesautil.la
glsl_libglcpp_la_SOURCES = \
glsl/glcpp/glcpp-lex.c \
glsl/glcpp/glcpp-parse.c \
glsl/glcpp/glcpp-parse.h \
$(LIBGLCPP_FILES)
glsl_glcpp_glcpp_SOURCES = \
glsl/glcpp/glcpp.c
glsl_glcpp_glcpp_LDADD = \
glsl/libglcpp.la \
$(top_builddir)/src/libglsl_util.la \
-lm
glsl_libglsl_la_LIBADD = \
nir/libnir.la \
glsl/libglcpp.la
glsl_libglsl_la_SOURCES = \
glsl/glsl_lexer.cpp \
glsl/glsl_parser.cpp \
glsl/glsl_parser.h \
$(LIBGLSL_FILES)
glsl_compiler_SOURCES = \
$(GLSL_COMPILER_CXX_FILES)
glsl_compiler_LDADD = \
glsl/libglsl.la \
$(top_builddir)/src/libglsl_util.la \
$(top_builddir)/src/util/libmesautil.la \
$(PTHREAD_LIBS)
glsl_glsl_test_SOURCES = \
glsl/standalone_scaffolding.cpp \
glsl/test.cpp \
glsl/test_optpass.cpp \
glsl/test_optpass.h
glsl_glsl_test_LDADD = \
glsl/libglsl.la \
$(top_builddir)/src/libglsl_util.la \
$(PTHREAD_LIBS)
# We write our own rules for yacc and lex below. We'd rather use automake,
# but automake makes it especially difficult for a number of reasons:
#
# * < automake-1.12 generates .h files from .yy and .ypp files, but
# >=automake-1.12 generates .hh and .hpp files respectively. There's no
# good way of making a project that uses C++ yacc files compatible with
# both versions of automake. Strong work automake developers.
#
# * Since we're generating code from .l/.y files in a subdirectory (glcpp/)
# we'd like the resulting generated code to also go in glcpp/ for purposes
# of distribution. Automake gives no way to do this.
#
# * Since we're building multiple yacc parsers into one library (and via one
# Makefile) we have to use per-target YFLAGS. Using per-target YFLAGS causes
# automake to name the resulting generated code as <library-name>_filename.c.
# Frankly, that's ugly and we don't want a libglcpp_glcpp_parser.h file.
# In order to make build output print "LEX" and "YACC", we reproduce the
# automake variables below.
AM_V_LEX = $(am__v_LEX_$(V))
am__v_LEX_ = $(am__v_LEX_$(AM_DEFAULT_VERBOSITY))
am__v_LEX_0 = @echo " LEX " $@;
am__v_LEX_1 =
AM_V_YACC = $(am__v_YACC_$(V))
am__v_YACC_ = $(am__v_YACC_$(AM_DEFAULT_VERBOSITY))
am__v_YACC_0 = @echo " YACC " $@;
am__v_YACC_1 =
MKDIR_GEN = $(AM_V_at)$(MKDIR_P) $(@D)
YACC_GEN = $(AM_V_YACC)$(YACC) $(YFLAGS)
LEX_GEN = $(AM_V_LEX)$(LEX) $(LFLAGS)
glsl/glsl_parser.cpp glsl/glsl_parser.h: glsl/glsl_parser.yy
$(MKDIR_GEN)
$(YACC_GEN) -o $@ -p "_mesa_glsl_" --defines=$(builddir)/glsl/glsl_parser.h $(srcdir)/glsl/glsl_parser.yy
glsl/glsl_lexer.cpp: glsl/glsl_lexer.ll
$(MKDIR_GEN)
$(LEX_GEN) -o $@ $(srcdir)/glsl/glsl_lexer.ll
glsl/glcpp/glcpp-parse.c glsl/glcpp/glcpp-parse.h: glsl/glcpp/glcpp-parse.y
$(MKDIR_GEN)
$(YACC_GEN) -o $@ -p "glcpp_parser_" --defines=$(builddir)/glsl/glcpp/glcpp-parse.h $(srcdir)/glsl/glcpp/glcpp-parse.y
glsl/glcpp/glcpp-lex.c: glsl/glcpp/glcpp-lex.l
$(MKDIR_GEN)
$(LEX_GEN) -o $@ $(srcdir)/glsl/glcpp/glcpp-lex.l
# Only the parsers (specifically the header files generated at the same time)
# need to be in BUILT_SOURCES. Though if we list the parser headers YACC is
# called for the .c/.cpp file and the .h files. By listing the .c/.cpp files
# YACC is only executed once for each parser. The rest of the generated code
# will be created at the appropriate times according to standard automake
# dependency rules.
BUILT_SOURCES += \
glsl/glsl_parser.cpp \
glsl/glsl_lexer.cpp \
glsl/glcpp/glcpp-parse.c \
glsl/glcpp/glcpp-lex.c
CLEANFILES += \
glsl/glcpp/glcpp-parse.h \
glsl/glsl_parser.h \
glsl/glsl_parser.cpp \
glsl/glsl_lexer.cpp \
glsl/glcpp/glcpp-parse.c \
glsl/glcpp/glcpp-lex.c
clean-local:
$(RM) -r subtest-cr subtest-cr-lf subtest-lf subtest-lf-cr
dist-hook:
$(RM) glsl/glcpp/tests/*.out
$(RM) glsl/glcpp/tests/subtest*/*.out
noinst_LTLIBRARIES += nir/libnir.la
nir_libnir_la_CPPFLAGS = \
$(AM_CPPFLAGS) \
-I$(top_builddir)/src/compiler/nir \
-I$(top_srcdir)/src/compiler/nir
nir_libnir_la_LIBADD = \
libcompiler.la
nir_libnir_la_SOURCES = \
$(NIR_FILES) \
$(NIR_GENERATED_FILES)
PYTHON_GEN = $(AM_V_GEN)$(PYTHON2) $(PYTHON_FLAGS)
nir/nir_builder_opcodes.h: nir/nir_opcodes.py nir/nir_builder_opcodes_h.py
$(MKDIR_GEN)
$(PYTHON_GEN) $(srcdir)/nir/nir_builder_opcodes_h.py > $@ || ($(RM) $@; false)
nir/nir_constant_expressions.c: nir/nir_opcodes.py nir/nir_constant_expressions.py
$(MKDIR_GEN)
$(PYTHON_GEN) $(srcdir)/nir/nir_constant_expressions.py > $@ || ($(RM) $@; false)
nir/nir_opcodes.h: nir/nir_opcodes.py nir/nir_opcodes_h.py
$(MKDIR_GEN)
$(PYTHON_GEN) $(srcdir)/nir/nir_opcodes_h.py > $@ || ($(RM) $@; false)
nir/nir_opcodes.c: nir/nir_opcodes.py nir/nir_opcodes_c.py
$(MKDIR_GEN)
$(PYTHON_GEN) $(srcdir)/nir/nir_opcodes_c.py > $@ || ($(RM) $@; false)
nir/nir_opt_algebraic.c: nir/nir_opt_algebraic.py nir/nir_algebraic.py
$(MKDIR_GEN)
$(PYTHON_GEN) $(srcdir)/nir/nir_opt_algebraic.py > $@ || ($(RM) $@; false)
check_PROGRAMS += nir/tests/control_flow_tests
nir_tests_control_flow_tests_CPPFLAGS = \
$(AM_CPPFLAGS) \
-I$(top_builddir)/src/compiler/nir \
-I$(top_srcdir)/src/compiler/nir
nir_tests_control_flow_tests_SOURCES = \
nir/tests/control_flow_tests.cpp
nir_tests_control_flow_tests_CFLAGS = \
$(PTHREAD_CFLAGS)
nir_tests_control_flow_tests_LDADD = \
$(top_builddir)/src/gtest/libgtest.la \
nir/libnir.la \
$(top_builddir)/src/util/libmesautil.la \
$(PTHREAD_LIBS)
TESTS += nir/tests/control_flow_tests
BUILT_SOURCES += $(NIR_GENERATED_FILES)
CLEANFILES += $(NIR_GENERATED_FILES)
EXTRA_DIST += \
nir/nir_algebraic.py \
nir/nir_builder_opcodes_h.py \
nir/nir_constant_expressions.py \
nir/nir_opcodes.py \
nir/nir_opcodes_c.py \
nir/nir_opcodes_h.py \
nir/nir_opt_algebraic.py \
nir/tests \
nir/Makefile.sources

View File

@@ -1,226 +0,0 @@
LIBCOMPILER_FILES = \
builtin_type_macros.h \
glsl_types.cpp \
glsl_types.h \
nir_types.cpp \
nir_types.h \
shader_enums.c \
shader_enums.h
# libglsl
LIBGLSL_FILES = \
glsl/ast.h \
glsl/ast_array_index.cpp \
glsl/ast_expr.cpp \
glsl/ast_function.cpp \
glsl/ast_to_hir.cpp \
glsl/ast_type.cpp \
glsl/blob.c \
glsl/blob.h \
glsl/builtin_functions.cpp \
glsl/builtin_types.cpp \
glsl/builtin_variables.cpp \
glsl/glsl_parser_extras.cpp \
glsl/glsl_parser_extras.h \
glsl/glsl_symbol_table.cpp \
glsl/glsl_symbol_table.h \
glsl/hir_field_selection.cpp \
glsl/ir_basic_block.cpp \
glsl/ir_basic_block.h \
glsl/ir_builder.cpp \
glsl/ir_builder.h \
glsl/ir_clone.cpp \
glsl/ir_constant_expression.cpp \
glsl/ir.cpp \
glsl/ir.h \
glsl/ir_equals.cpp \
glsl/ir_expression_flattening.cpp \
glsl/ir_expression_flattening.h \
glsl/ir_function_can_inline.cpp \
glsl/ir_function_detect_recursion.cpp \
glsl/ir_function_inlining.h \
glsl/ir_function.cpp \
glsl/ir_hierarchical_visitor.cpp \
glsl/ir_hierarchical_visitor.h \
glsl/ir_hv_accept.cpp \
glsl/ir_import_prototypes.cpp \
glsl/ir_optimization.h \
glsl/ir_print_visitor.cpp \
glsl/ir_print_visitor.h \
glsl/ir_reader.cpp \
glsl/ir_reader.h \
glsl/ir_rvalue_visitor.cpp \
glsl/ir_rvalue_visitor.h \
glsl/ir_set_program_inouts.cpp \
glsl/ir_uniform.h \
glsl/ir_validate.cpp \
glsl/ir_variable_refcount.cpp \
glsl/ir_variable_refcount.h \
glsl/ir_visitor.h \
glsl/linker.cpp \
glsl/linker.h \
glsl/link_atomics.cpp \
glsl/link_functions.cpp \
glsl/link_interface_blocks.cpp \
glsl/link_uniforms.cpp \
glsl/link_uniform_initializers.cpp \
glsl/link_uniform_block_active_visitor.cpp \
glsl/link_uniform_block_active_visitor.h \
glsl/link_uniform_blocks.cpp \
glsl/link_varyings.cpp \
glsl/link_varyings.h \
glsl/list.h \
glsl/loop_analysis.cpp \
glsl/loop_analysis.h \
glsl/loop_controls.cpp \
glsl/loop_unroll.cpp \
glsl/lower_buffer_access.cpp \
glsl/lower_buffer_access.h \
glsl/lower_clip_distance.cpp \
glsl/lower_const_arrays_to_uniforms.cpp \
glsl/lower_discard.cpp \
glsl/lower_discard_flow.cpp \
glsl/lower_if_to_cond_assign.cpp \
glsl/lower_instructions.cpp \
glsl/lower_jumps.cpp \
glsl/lower_mat_op_to_vec.cpp \
glsl/lower_noise.cpp \
glsl/lower_offset_array.cpp \
glsl/lower_packed_varyings.cpp \
glsl/lower_named_interface_blocks.cpp \
glsl/lower_packing_builtins.cpp \
glsl/lower_subroutine.cpp \
glsl/lower_tess_level.cpp \
glsl/lower_texture_projection.cpp \
glsl/lower_variable_index_to_cond_assign.cpp \
glsl/lower_vec_index_to_cond_assign.cpp \
glsl/lower_vec_index_to_swizzle.cpp \
glsl/lower_vector.cpp \
glsl/lower_vector_derefs.cpp \
glsl/lower_vector_insert.cpp \
glsl/lower_vertex_id.cpp \
glsl/lower_output_reads.cpp \
glsl/lower_shared_reference.cpp \
glsl/lower_ubo_reference.cpp \
glsl/opt_algebraic.cpp \
glsl/opt_array_splitting.cpp \
glsl/opt_conditional_discard.cpp \
glsl/opt_constant_folding.cpp \
glsl/opt_constant_propagation.cpp \
glsl/opt_constant_variable.cpp \
glsl/opt_copy_propagation.cpp \
glsl/opt_copy_propagation_elements.cpp \
glsl/opt_dead_builtin_variables.cpp \
glsl/opt_dead_builtin_varyings.cpp \
glsl/opt_dead_code.cpp \
glsl/opt_dead_code_local.cpp \
glsl/opt_dead_functions.cpp \
glsl/opt_flatten_nested_if_blocks.cpp \
glsl/opt_flip_matrices.cpp \
glsl/opt_function_inlining.cpp \
glsl/opt_if_simplification.cpp \
glsl/opt_minmax.cpp \
glsl/opt_noop_swizzle.cpp \
glsl/opt_rebalance_tree.cpp \
glsl/opt_redundant_jumps.cpp \
glsl/opt_structure_splitting.cpp \
glsl/opt_swizzle_swizzle.cpp \
glsl/opt_tree_grafting.cpp \
glsl/opt_vectorize.cpp \
glsl/program.h \
glsl/s_expression.cpp \
glsl/s_expression.h
# glsl_compiler
GLSL_COMPILER_CXX_FILES = \
glsl/standalone_scaffolding.cpp \
glsl/standalone_scaffolding.h \
glsl/main.cpp
# libglsl generated sources
LIBGLSL_GENERATED_CXX_FILES = \
glsl/glsl_lexer.cpp \
glsl/glsl_parser.cpp
# libglcpp
LIBGLCPP_FILES = \
glsl/glcpp/glcpp.h \
glsl/glcpp/pp.c
LIBGLCPP_GENERATED_FILES = \
glsl/glcpp/glcpp-lex.c \
glsl/glcpp/glcpp-parse.c
NIR_GENERATED_FILES = \
nir/nir_builder_opcodes.h \
nir/nir_constant_expressions.c \
nir/nir_opcodes.c \
nir/nir_opcodes.h \
nir/nir_opt_algebraic.c
NIR_FILES = \
nir/glsl_to_nir.cpp \
nir/glsl_to_nir.h \
nir/nir.c \
nir/nir.h \
nir/nir_array.h \
nir/nir_builder.h \
nir/nir_clone.c \
nir/nir_constant_expressions.h \
nir/nir_control_flow.c \
nir/nir_control_flow.h \
nir/nir_control_flow_private.h \
nir/nir_dominance.c \
nir/nir_from_ssa.c \
nir/nir_gs_count_vertices.c \
nir/nir_intrinsics.c \
nir/nir_intrinsics.h \
nir/nir_instr_set.c \
nir/nir_instr_set.h \
nir/nir_liveness.c \
nir/nir_lower_alu_to_scalar.c \
nir/nir_lower_atomics.c \
nir/nir_lower_clip.c \
nir/nir_lower_global_vars_to_local.c \
nir/nir_lower_gs_intrinsics.c \
nir/nir_lower_load_const_to_scalar.c \
nir/nir_lower_locals_to_regs.c \
nir/nir_lower_idiv.c \
nir/nir_lower_io.c \
nir/nir_lower_outputs_to_temporaries.c \
nir/nir_lower_phis_to_scalar.c \
nir/nir_lower_samplers.c \
nir/nir_lower_system_values.c \
nir/nir_lower_tex.c \
nir/nir_lower_to_source_mods.c \
nir/nir_lower_two_sided_color.c \
nir/nir_lower_vars_to_ssa.c \
nir/nir_lower_var_copies.c \
nir/nir_lower_vec_to_movs.c \
nir/nir_metadata.c \
nir/nir_move_vec_src_uses_to_dest.c \
nir/nir_normalize_cubemap_coords.c \
nir/nir_opt_constant_folding.c \
nir/nir_opt_copy_propagate.c \
nir/nir_opt_cse.c \
nir/nir_opt_dce.c \
nir/nir_opt_dead_cf.c \
nir/nir_opt_gcm.c \
nir/nir_opt_global_to_local.c \
nir/nir_opt_peephole_select.c \
nir/nir_opt_remove_phis.c \
nir/nir_opt_undef.c \
nir/nir_print.c \
nir/nir_remove_dead_variables.c \
nir/nir_search.c \
nir/nir_search.h \
nir/nir_split_var_copies.c \
nir/nir_sweep.c \
nir/nir_to_ssa.c \
nir/nir_validate.c \
nir/nir_vla.h \
nir/nir_worklist.c \
nir/nir_worklist.h

View File

@@ -1,24 +0,0 @@
Import('*')
env = env.Clone()
env.MSVC2013Compat()
env.Prepend(CPPPATH = [
'#include',
'#src',
'#src/mapi',
'#src/mesa',
'#src/gallium/include',
'#src/gallium/auxiliary',
])
sources = env.ParseSourceList('Makefile.sources', 'LIBCOMPILER_FILES')
compiler = env.ConvenienceLibrary(
target = 'compiler',
source = sources
)
Export('compiler')
SConscript('glsl/SConscript')

View File

@@ -1,78 +0,0 @@
# Mesa 3-D graphics library
#
# Copyright (C) 2010-2011 Chia-I Wu <olvaffe@gmail.com>
# Copyright (C) 2010-2011 LunarG Inc.
#
# Permission is hereby granted, free of charge, to any person obtaining a
# copy of this software and associated documentation files (the "Software"),
# to deal in the Software without restriction, including without limitation
# the rights to use, copy, modify, merge, publish, distribute, sublicense,
# and/or sell copies of the Software, and to permit persons to whom the
# Software is furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included
# in all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
# THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
# FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
# DEALINGS IN THE SOFTWARE.
# included by glsl Android.mk for source generation
ifeq ($(LOCAL_MODULE_CLASS),)
LOCAL_MODULE_CLASS := STATIC_LIBRARIES
endif
intermediates := $(call local-generated-sources-dir)
LOCAL_SRC_FILES := $(LOCAL_SRC_FILES)
LOCAL_C_INCLUDES += \
$(intermediates)/glcpp \
$(LOCAL_PATH)/glcpp \
LOCAL_GENERATED_SOURCES += $(addprefix $(intermediates)/, \
$(LIBGLCPP_GENERATED_FILES) \
$(LIBGLSL_GENERATED_CXX_FILES))
define local-l-or-ll-to-c-or-cpp
@mkdir -p $(dir $@)
@echo "Mesa Lex: $(PRIVATE_MODULE) <= $<"
$(hide) $(LEX) --nounistd -o$@ $<
endef
define glsl_local-y-to-c-and-h
@mkdir -p $(dir $@)
@echo "Mesa Yacc: $(PRIVATE_MODULE) <= $<"
$(hide) $(YACC) -o $@ -p "glcpp_parser_" $<
endef
YACC_HEADER_SUFFIX := .hpp
define local-yy-to-cpp-and-h
@mkdir -p $(dir $@)
@echo "Mesa Yacc: $(PRIVATE_MODULE) <= $<"
$(hide) $(YACC) -p "_mesa_glsl_" -o $@ $<
touch $(@:$1=$(YACC_HEADER_SUFFIX))
echo '#ifndef '$(@F:$1=_h) > $(@:$1=.h)
echo '#define '$(@F:$1=_h) >> $(@:$1=.h)
cat $(@:$1=$(YACC_HEADER_SUFFIX)) >> $(@:$1=.h)
echo '#endif' >> $(@:$1=.h)
rm -f $(@:$1=$(YACC_HEADER_SUFFIX))
endef
$(intermediates)/glsl_lexer.cpp: $(LOCAL_PATH)/glsl_lexer.ll
$(call local-l-or-ll-to-c-or-cpp)
$(intermediates)/glsl_parser.cpp: $(LOCAL_PATH)/glsl_parser.yy
$(call local-yy-to-cpp-and-h,.cpp)
$(intermediates)/glcpp/glcpp-lex.c: $(LOCAL_PATH)/glcpp/glcpp-lex.l
$(call local-l-or-ll-to-c-or-cpp)
$(intermediates)/glcpp/glcpp-parse.c: $(LOCAL_PATH)/glcpp/glcpp-parse.y
$(call glsl_local-y-to-c-and-h)

View File

@@ -1,535 +0,0 @@
/*
* Copyright © 2010 Intel Corporation
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
* DEALINGS IN THE SOFTWARE.
*/
#include "ast.h"
void
ast_type_specifier::print(void) const
{
if (structure) {
structure->print();
} else {
printf("%s ", type_name);
}
if (array_specifier) {
array_specifier->print();
}
}
bool
ast_fully_specified_type::has_qualifiers(_mesa_glsl_parse_state *state) const
{
/* 'subroutine' isnt a real qualifier. */
ast_type_qualifier subroutine_only;
subroutine_only.flags.i = 0;
subroutine_only.flags.q.subroutine = 1;
subroutine_only.flags.q.subroutine_def = 1;
if (state->has_explicit_uniform_location()) {
subroutine_only.flags.q.explicit_index = 1;
}
return (this->qualifier.flags.i & ~subroutine_only.flags.i) != 0;
}
bool ast_type_qualifier::has_interpolation() const
{
return this->flags.q.smooth
|| this->flags.q.flat
|| this->flags.q.noperspective;
}
bool
ast_type_qualifier::has_layout() const
{
return this->flags.q.origin_upper_left
|| this->flags.q.pixel_center_integer
|| this->flags.q.depth_any
|| this->flags.q.depth_greater
|| this->flags.q.depth_less
|| this->flags.q.depth_unchanged
|| this->flags.q.std140
|| this->flags.q.std430
|| this->flags.q.shared
|| this->flags.q.column_major
|| this->flags.q.row_major
|| this->flags.q.packed
|| this->flags.q.explicit_location
|| this->flags.q.explicit_image_format
|| this->flags.q.explicit_index
|| this->flags.q.explicit_binding
|| this->flags.q.explicit_offset
|| this->flags.q.explicit_stream;
}
bool
ast_type_qualifier::has_storage() const
{
return this->flags.q.constant
|| this->flags.q.attribute
|| this->flags.q.varying
|| this->flags.q.in
|| this->flags.q.out
|| this->flags.q.uniform
|| this->flags.q.buffer
|| this->flags.q.shared_storage;
}
bool
ast_type_qualifier::has_auxiliary_storage() const
{
return this->flags.q.centroid
|| this->flags.q.sample
|| this->flags.q.patch;
}
/**
* This function merges both duplicate identifies within a single layout and
* multiple layout qualifiers on a single variable declaration. The
* is_single_layout_merge param is used differentiate between the two.
*/
bool
ast_type_qualifier::merge_qualifier(YYLTYPE *loc,
_mesa_glsl_parse_state *state,
const ast_type_qualifier &q,
bool is_single_layout_merge)
{
ast_type_qualifier ubo_mat_mask;
ubo_mat_mask.flags.i = 0;
ubo_mat_mask.flags.q.row_major = 1;
ubo_mat_mask.flags.q.column_major = 1;
ast_type_qualifier ubo_layout_mask;
ubo_layout_mask.flags.i = 0;
ubo_layout_mask.flags.q.std140 = 1;
ubo_layout_mask.flags.q.packed = 1;
ubo_layout_mask.flags.q.shared = 1;
ubo_layout_mask.flags.q.std430 = 1;
ast_type_qualifier ubo_binding_mask;
ubo_binding_mask.flags.i = 0;
ubo_binding_mask.flags.q.explicit_binding = 1;
ubo_binding_mask.flags.q.explicit_offset = 1;
ast_type_qualifier stream_layout_mask;
stream_layout_mask.flags.i = 0;
stream_layout_mask.flags.q.stream = 1;
/* Uniform block layout qualifiers get to overwrite each
* other (rightmost having priority), while all other
* qualifiers currently don't allow duplicates.
*/
ast_type_qualifier allowed_duplicates_mask;
allowed_duplicates_mask.flags.i =
ubo_mat_mask.flags.i |
ubo_layout_mask.flags.i |
ubo_binding_mask.flags.i;
/* Geometry shaders can have several layout qualifiers
* assigning different stream values.
*/
if (state->stage == MESA_SHADER_GEOMETRY)
allowed_duplicates_mask.flags.i |=
stream_layout_mask.flags.i;
if (is_single_layout_merge && !state->has_enhanced_layouts() &&
(this->flags.i & q.flags.i & ~allowed_duplicates_mask.flags.i) != 0) {
_mesa_glsl_error(loc, state,
"duplicate layout qualifiers used");
return false;
}
if (q.flags.q.prim_type) {
if (this->flags.q.prim_type && this->prim_type != q.prim_type) {
_mesa_glsl_error(loc, state,
"conflicting primitive type qualifiers used");
return false;
}
this->prim_type = q.prim_type;
}
if (q.flags.q.max_vertices) {
if (this->max_vertices) {
this->max_vertices->merge_qualifier(q.max_vertices);
} else {
this->max_vertices = q.max_vertices;
}
}
if (q.flags.q.subroutine_def) {
if (this->flags.q.subroutine_def) {
_mesa_glsl_error(loc, state,
"conflicting subroutine qualifiers used");
} else {
this->subroutine_list = q.subroutine_list;
}
}
if (q.flags.q.invocations) {
if (this->invocations) {
this->invocations->merge_qualifier(q.invocations);
} else {
this->invocations = q.invocations;
}
}
if (state->stage == MESA_SHADER_GEOMETRY &&
state->has_explicit_attrib_stream()) {
if (!this->flags.q.explicit_stream) {
if (q.flags.q.stream) {
this->flags.q.stream = 1;
this->stream = q.stream;
} else if (!this->flags.q.stream && this->flags.q.out) {
/* Assign default global stream value */
this->flags.q.stream = 1;
this->stream = state->out_qualifier->stream;
}
}
}
if (q.flags.q.vertices) {
if (this->vertices) {
this->vertices->merge_qualifier(q.vertices);
} else {
this->vertices = q.vertices;
}
}
if (q.flags.q.vertex_spacing) {
if (this->flags.q.vertex_spacing && this->vertex_spacing != q.vertex_spacing) {
_mesa_glsl_error(loc, state,
"conflicting vertex spacing used");
return false;
}
this->vertex_spacing = q.vertex_spacing;
}
if (q.flags.q.ordering) {
if (this->flags.q.ordering && this->ordering != q.ordering) {
_mesa_glsl_error(loc, state,
"conflicting ordering used");
return false;
}
this->ordering = q.ordering;
}
if (q.flags.q.point_mode) {
if (this->flags.q.point_mode && this->point_mode != q.point_mode) {
_mesa_glsl_error(loc, state,
"conflicting point mode used");
return false;
}
this->point_mode = q.point_mode;
}
if ((q.flags.i & ubo_mat_mask.flags.i) != 0)
this->flags.i &= ~ubo_mat_mask.flags.i;
if ((q.flags.i & ubo_layout_mask.flags.i) != 0)
this->flags.i &= ~ubo_layout_mask.flags.i;
for (int i = 0; i < 3; i++) {
if (q.flags.q.local_size & (1 << i)) {
if (this->local_size[i]) {
this->local_size[i]->merge_qualifier(q.local_size[i]);
} else {
this->local_size[i] = q.local_size[i];
}
}
}
this->flags.i |= q.flags.i;
if (q.flags.q.explicit_location)
this->location = q.location;
if (q.flags.q.explicit_index)
this->index = q.index;
if (q.flags.q.explicit_binding)
this->binding = q.binding;
if (q.flags.q.explicit_offset)
this->offset = q.offset;
if (q.precision != ast_precision_none)
this->precision = q.precision;
if (q.flags.q.explicit_image_format) {
this->image_format = q.image_format;
this->image_base_type = q.image_base_type;
}
return true;
}
bool
ast_type_qualifier::merge_out_qualifier(YYLTYPE *loc,
_mesa_glsl_parse_state *state,
const ast_type_qualifier &q,
ast_node* &node, bool create_node)
{
void *mem_ctx = state;
const bool r = this->merge_qualifier(loc, state, q, false);
if (state->stage == MESA_SHADER_GEOMETRY) {
if (q.flags.q.prim_type) {
/* Make sure this is a valid output primitive type. */
switch (q.prim_type) {
case GL_POINTS:
case GL_LINE_STRIP:
case GL_TRIANGLE_STRIP:
break;
default:
_mesa_glsl_error(loc, state, "invalid geometry shader output "
"primitive type");
break;
}
}
/* Allow future assigments of global out's stream id value */
this->flags.q.explicit_stream = 0;
} else if (state->stage == MESA_SHADER_TESS_CTRL) {
if (create_node) {
node = new(mem_ctx) ast_tcs_output_layout(*loc);
}
} else {
_mesa_glsl_error(loc, state, "out layout qualifiers only valid in "
"tessellation control or geometry shaders");
}
return r;
}
bool
ast_type_qualifier::merge_in_qualifier(YYLTYPE *loc,
_mesa_glsl_parse_state *state,
const ast_type_qualifier &q,
ast_node* &node, bool create_node)
{
void *mem_ctx = state;
bool create_gs_ast = false;
bool create_cs_ast = false;
ast_type_qualifier valid_in_mask;
valid_in_mask.flags.i = 0;
switch (state->stage) {
case MESA_SHADER_TESS_EVAL:
if (q.flags.q.prim_type) {
/* Make sure this is a valid input primitive type. */
switch (q.prim_type) {
case GL_TRIANGLES:
case GL_QUADS:
case GL_ISOLINES:
break;
default:
_mesa_glsl_error(loc, state,
"invalid tessellation evaluation "
"shader input primitive type");
break;
}
}
valid_in_mask.flags.q.prim_type = 1;
valid_in_mask.flags.q.vertex_spacing = 1;
valid_in_mask.flags.q.ordering = 1;
valid_in_mask.flags.q.point_mode = 1;
break;
case MESA_SHADER_GEOMETRY:
if (q.flags.q.prim_type) {
/* Make sure this is a valid input primitive type. */
switch (q.prim_type) {
case GL_POINTS:
case GL_LINES:
case GL_LINES_ADJACENCY:
case GL_TRIANGLES:
case GL_TRIANGLES_ADJACENCY:
break;
default:
_mesa_glsl_error(loc, state,
"invalid geometry shader input primitive type");
break;
}
}
create_gs_ast |=
q.flags.q.prim_type &&
!state->in_qualifier->flags.q.prim_type;
valid_in_mask.flags.q.prim_type = 1;
valid_in_mask.flags.q.invocations = 1;
break;
case MESA_SHADER_FRAGMENT:
valid_in_mask.flags.q.early_fragment_tests = 1;
break;
case MESA_SHADER_COMPUTE:
create_cs_ast |=
q.flags.q.local_size != 0 &&
state->in_qualifier->flags.q.local_size == 0;
valid_in_mask.flags.q.local_size = 7;
break;
default:
_mesa_glsl_error(loc, state,
"input layout qualifiers only valid in "
"geometry, fragment and compute shaders");
break;
}
/* Generate an error when invalid input layout qualifiers are used. */
if ((q.flags.i & ~valid_in_mask.flags.i) != 0) {
_mesa_glsl_error(loc, state,
"invalid input layout qualifiers used");
return false;
}
/* Input layout qualifiers can be specified multiple
* times in separate declarations, as long as they match.
*/
if (this->flags.q.prim_type) {
if (q.flags.q.prim_type &&
this->prim_type != q.prim_type) {
_mesa_glsl_error(loc, state,
"conflicting input primitive %s specified",
state->stage == MESA_SHADER_GEOMETRY ?
"type" : "mode");
}
} else if (q.flags.q.prim_type) {
state->in_qualifier->flags.q.prim_type = 1;
state->in_qualifier->prim_type = q.prim_type;
}
if (q.flags.q.invocations) {
this->flags.q.invocations = 1;
if (this->invocations) {
this->invocations->merge_qualifier(q.invocations);
} else {
this->invocations = q.invocations;
}
}
if (q.flags.q.early_fragment_tests) {
state->fs_early_fragment_tests = true;
}
if (this->flags.q.vertex_spacing) {
if (q.flags.q.vertex_spacing &&
this->vertex_spacing != q.vertex_spacing) {
_mesa_glsl_error(loc, state,
"conflicting vertex spacing specified");
}
} else if (q.flags.q.vertex_spacing) {
this->flags.q.vertex_spacing = 1;
this->vertex_spacing = q.vertex_spacing;
}
if (this->flags.q.ordering) {
if (q.flags.q.ordering &&
this->ordering != q.ordering) {
_mesa_glsl_error(loc, state,
"conflicting ordering specified");
}
} else if (q.flags.q.ordering) {
this->flags.q.ordering = 1;
this->ordering = q.ordering;
}
if (this->flags.q.point_mode) {
if (q.flags.q.point_mode &&
this->point_mode != q.point_mode) {
_mesa_glsl_error(loc, state,
"conflicting point mode specified");
}
} else if (q.flags.q.point_mode) {
this->flags.q.point_mode = 1;
this->point_mode = q.point_mode;
}
if (create_node) {
if (create_gs_ast) {
node = new(mem_ctx) ast_gs_input_layout(*loc, q.prim_type);
} else if (create_cs_ast) {
node = new(mem_ctx) ast_cs_input_layout(*loc, q.local_size);
}
}
return true;
}
bool
ast_layout_expression::process_qualifier_constant(struct _mesa_glsl_parse_state *state,
const char *qual_indentifier,
unsigned *value,
bool can_be_zero)
{
int min_value = 0;
bool first_pass = true;
*value = 0;
if (!can_be_zero)
min_value = 1;
for (exec_node *node = layout_const_expressions.head;
!node->is_tail_sentinel(); node = node->next) {
exec_list dummy_instructions;
ast_node *const_expression = exec_node_data(ast_node, node, link);
ir_rvalue *const ir = const_expression->hir(&dummy_instructions, state);
ir_constant *const const_int = ir->constant_expression_value();
if (const_int == NULL || !const_int->type->is_integer()) {
YYLTYPE loc = const_expression->get_location();
_mesa_glsl_error(&loc, state, "%s must be an integral constant "
"expression", qual_indentifier);
return false;
}
if (const_int->value.i[0] < min_value) {
YYLTYPE loc = const_expression->get_location();
_mesa_glsl_error(&loc, state, "%s layout qualifier is invalid "
"(%d < %d)", qual_indentifier,
const_int->value.i[0], min_value);
return false;
}
if (!first_pass && *value != const_int->value.u[0]) {
YYLTYPE loc = const_expression->get_location();
_mesa_glsl_error(&loc, state, "%s layout qualifier does not "
"match previous declaration (%d vs %d)",
qual_indentifier, *value, const_int->value.i[0]);
return false;
} else {
first_pass = false;
*value = const_int->value.u[0];
}
/* If the location is const (and we've verified that
* it is) then no instructions should have been emitted
* when we converted it to HIR. If they were emitted,
* then either the location isn't const after all, or
* we are emitting unnecessary instructions.
*/
assert(dummy_instructions.is_empty());
}
return true;
}

View File

@@ -1,494 +0,0 @@
/*
* Copyright (c) 2015 Intel Corporation
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
* DEALINGS IN THE SOFTWARE.
*/
/**
* \file lower_buffer_access.cpp
*
* Helper for IR lowering pass to replace dereferences of buffer object based
* shader variables with intrinsic function calls.
*
* This helper is used by lowering passes for UBOs, SSBOs and compute shader
* shared variables.
*/
#include "lower_buffer_access.h"
#include "ir_builder.h"
#include "main/macros.h"
#include "util/list.h"
#include "glsl_parser_extras.h"
using namespace ir_builder;
namespace lower_buffer_access {
static inline int
writemask_for_size(unsigned n)
{
return ((1 << n) - 1);
}
/**
* Takes a deref and recursively calls itself to break the deref down to the
* point that the reads or writes generated are contiguous scalars or vectors.
*/
void
lower_buffer_access::emit_access(void *mem_ctx,
bool is_write,
ir_dereference *deref,
ir_variable *base_offset,
unsigned int deref_offset,
bool row_major,
int matrix_columns,
unsigned int packing,
unsigned int write_mask)
{
if (deref->type->is_record()) {
unsigned int field_offset = 0;
for (unsigned i = 0; i < deref->type->length; i++) {
const struct glsl_struct_field *field =
&deref->type->fields.structure[i];
ir_dereference *field_deref =
new(mem_ctx) ir_dereference_record(deref->clone(mem_ctx, NULL),
field->name);
field_offset =
glsl_align(field_offset,
field->type->std140_base_alignment(row_major));
emit_access(mem_ctx, is_write, field_deref, base_offset,
deref_offset + field_offset,
row_major, 1, packing,
writemask_for_size(field_deref->type->vector_elements));
field_offset += field->type->std140_size(row_major);
}
return;
}
if (deref->type->is_array()) {
unsigned array_stride = packing == GLSL_INTERFACE_PACKING_STD430 ?
deref->type->fields.array->std430_array_stride(row_major) :
glsl_align(deref->type->fields.array->std140_size(row_major), 16);
for (unsigned i = 0; i < deref->type->length; i++) {
ir_constant *element = new(mem_ctx) ir_constant(i);
ir_dereference *element_deref =
new(mem_ctx) ir_dereference_array(deref->clone(mem_ctx, NULL),
element);
emit_access(mem_ctx, is_write, element_deref, base_offset,
deref_offset + i * array_stride,
row_major, 1, packing,
writemask_for_size(element_deref->type->vector_elements));
}
return;
}
if (deref->type->is_matrix()) {
for (unsigned i = 0; i < deref->type->matrix_columns; i++) {
ir_constant *col = new(mem_ctx) ir_constant(i);
ir_dereference *col_deref =
new(mem_ctx) ir_dereference_array(deref->clone(mem_ctx, NULL), col);
if (row_major) {
/* For a row-major matrix, the next column starts at the next
* element.
*/
int size_mul = deref->type->is_double() ? 8 : 4;
emit_access(mem_ctx, is_write, col_deref, base_offset,
deref_offset + i * size_mul,
row_major, deref->type->matrix_columns, packing,
writemask_for_size(col_deref->type->vector_elements));
} else {
int size_mul;
/* std430 doesn't round up vec2 size to a vec4 size */
if (packing == GLSL_INTERFACE_PACKING_STD430 &&
deref->type->vector_elements == 2 &&
!deref->type->is_double()) {
size_mul = 8;
} else {
/* std140 always rounds the stride of arrays (and matrices) to a
* vec4, so matrices are always 16 between columns/rows. With
* doubles, they will be 32 apart when there are more than 2 rows.
*
* For both std140 and std430, if the member is a
* three-'component vector with components consuming N basic
* machine units, the base alignment is 4N. For vec4, base
* alignment is 4N.
*/
size_mul = (deref->type->is_double() &&
deref->type->vector_elements > 2) ? 32 : 16;
}
emit_access(mem_ctx, is_write, col_deref, base_offset,
deref_offset + i * size_mul,
row_major, deref->type->matrix_columns, packing,
writemask_for_size(col_deref->type->vector_elements));
}
}
return;
}
assert(deref->type->is_scalar() || deref->type->is_vector());
if (!row_major) {
ir_rvalue *offset =
add(base_offset, new(mem_ctx) ir_constant(deref_offset));
unsigned mask =
is_write ? write_mask : (1 << deref->type->vector_elements) - 1;
insert_buffer_access(mem_ctx, deref, deref->type, offset, mask, -1);
} else {
unsigned N = deref->type->is_double() ? 8 : 4;
/* We're dereffing a column out of a row-major matrix, so we
* gather the vector from each stored row.
*/
assert(deref->type->base_type == GLSL_TYPE_FLOAT ||
deref->type->base_type == GLSL_TYPE_DOUBLE);
/* Matrices, row_major or not, are stored as if they were
* arrays of vectors of the appropriate size in std140.
* Arrays have their strides rounded up to a vec4, so the
* matrix stride is always 16. However a double matrix may either be 16
* or 32 depending on the number of columns.
*/
assert(matrix_columns <= 4);
unsigned matrix_stride = 0;
/* Matrix stride for std430 mat2xY matrices are not rounded up to
* vec4 size. From OpenGL 4.3 spec, section 7.6.2.2 "Standard Uniform
* Block Layout":
*
* "2. If the member is a two- or four-component vector with components
* consuming N basic machine units, the base alignment is 2N or 4N,
* respectively." [...]
* "4. If the member is an array of scalars or vectors, the base alignment
* and array stride are set to match the base alignment of a single array
* element, according to rules (1), (2), and (3), and rounded up to the
* base alignment of a vec4." [...]
* "7. If the member is a row-major matrix with C columns and R rows, the
* matrix is stored identically to an array of R row vectors with C
* components each, according to rule (4)." [...]
* "When using the std430 storage layout, shader storage blocks will be
* laid out in buffer storage identically to uniform and shader storage
* blocks using the std140 layout, except that the base alignment and
* stride of arrays of scalars and vectors in rule 4 and of structures in
* rule 9 are not rounded up a multiple of the base alignment of a vec4."
*/
if (packing == GLSL_INTERFACE_PACKING_STD430 && matrix_columns == 2)
matrix_stride = 2 * N;
else
matrix_stride = glsl_align(matrix_columns * N, 16);
const glsl_type *deref_type = deref->type->base_type == GLSL_TYPE_FLOAT ?
glsl_type::float_type : glsl_type::double_type;
for (unsigned i = 0; i < deref->type->vector_elements; i++) {
ir_rvalue *chan_offset =
add(base_offset,
new(mem_ctx) ir_constant(deref_offset + i * matrix_stride));
if (!is_write || ((1U << i) & write_mask))
insert_buffer_access(mem_ctx, deref, deref_type, chan_offset,
(1U << i), i);
}
}
}
/**
* Determine if a thing being dereferenced is row-major
*
* There is some trickery here.
*
* If the thing being dereferenced is a member of uniform block \b without an
* instance name, then the name of the \c ir_variable is the field name of an
* interface type. If this field is row-major, then the thing referenced is
* row-major.
*
* If the thing being dereferenced is a member of uniform block \b with an
* instance name, then the last dereference in the tree will be an
* \c ir_dereference_record. If that record field is row-major, then the
* thing referenced is row-major.
*/
bool
lower_buffer_access::is_dereferenced_thing_row_major(const ir_rvalue *deref)
{
bool matrix = false;
const ir_rvalue *ir = deref;
while (true) {
matrix = matrix || ir->type->without_array()->is_matrix();
switch (ir->ir_type) {
case ir_type_dereference_array: {
const ir_dereference_array *const array_deref =
(const ir_dereference_array *) ir;
ir = array_deref->array;
break;
}
case ir_type_dereference_record: {
const ir_dereference_record *const record_deref =
(const ir_dereference_record *) ir;
ir = record_deref->record;
const int idx = ir->type->field_index(record_deref->field);
assert(idx >= 0);
const enum glsl_matrix_layout matrix_layout =
glsl_matrix_layout(ir->type->fields.structure[idx].matrix_layout);
switch (matrix_layout) {
case GLSL_MATRIX_LAYOUT_INHERITED:
break;
case GLSL_MATRIX_LAYOUT_COLUMN_MAJOR:
return false;
case GLSL_MATRIX_LAYOUT_ROW_MAJOR:
return matrix || deref->type->without_array()->is_record();
}
break;
}
case ir_type_dereference_variable: {
const ir_dereference_variable *const var_deref =
(const ir_dereference_variable *) ir;
const enum glsl_matrix_layout matrix_layout =
glsl_matrix_layout(var_deref->var->data.matrix_layout);
switch (matrix_layout) {
case GLSL_MATRIX_LAYOUT_INHERITED: {
/* For interface block matrix variables we handle inherited
* layouts at HIR generation time, but we don't do that for shared
* variables, which are always column-major
*/
ir_variable *var = deref->variable_referenced();
assert((var->is_in_buffer_block() && !matrix) ||
var->data.mode == ir_var_shader_shared);
return false;
}
case GLSL_MATRIX_LAYOUT_COLUMN_MAJOR:
return false;
case GLSL_MATRIX_LAYOUT_ROW_MAJOR:
return matrix || deref->type->without_array()->is_record();
}
unreachable("invalid matrix layout");
break;
}
default:
return false;
}
}
/* The tree must have ended with a dereference that wasn't an
* ir_dereference_variable. That is invalid, and it should be impossible.
*/
unreachable("invalid dereference tree");
return false;
}
/**
* This function initializes various values that will be used later by
* emit_access when actually emitting loads or stores.
*
* Note: const_offset is an input as well as an output, clients must
* initialize it to the offset of the variable in the underlying block, and
* this function will adjust it by adding the constant offset of the member
* being accessed into that variable.
*/
void
lower_buffer_access::setup_buffer_access(void *mem_ctx,
ir_variable *var,
ir_rvalue *deref,
ir_rvalue **offset,
unsigned *const_offset,
bool *row_major,
int *matrix_columns,
const glsl_struct_field **struct_field,
unsigned packing)
{
*offset = new(mem_ctx) ir_constant(0u);
*row_major = is_dereferenced_thing_row_major(deref);
*matrix_columns = 1;
/* Calculate the offset to the start of the region of the UBO
* dereferenced by *rvalue. This may be a variable offset if an
* array dereference has a variable index.
*/
while (deref) {
switch (deref->ir_type) {
case ir_type_dereference_variable: {
deref = NULL;
break;
}
case ir_type_dereference_array: {
ir_dereference_array *deref_array = (ir_dereference_array *) deref;
unsigned array_stride;
if (deref_array->array->type->is_vector()) {
/* We get this when storing or loading a component out of a vector
* with a non-constant index. This happens for v[i] = f where v is
* a vector (or m[i][j] = f where m is a matrix). If we don't
* lower that here, it gets turned into v = vector_insert(v, i,
* f), which loads the entire vector, modifies one component and
* then write the entire thing back. That breaks if another
* thread or SIMD channel is modifying the same vector.
*/
array_stride = 4;
if (deref_array->array->type->is_double())
array_stride *= 2;
} else if (deref_array->array->type->is_matrix() && *row_major) {
/* When loading a vector out of a row major matrix, the
* step between the columns (vectors) is the size of a
* float, while the step between the rows (elements of a
* vector) is handled below in emit_ubo_loads.
*/
array_stride = 4;
if (deref_array->array->type->is_double())
array_stride *= 2;
*matrix_columns = deref_array->array->type->matrix_columns;
} else if (deref_array->type->without_array()->is_interface()) {
/* We're processing an array dereference of an interface instance
* array. The thing being dereferenced *must* be a variable
* dereference because interfaces cannot be embedded in other
* types. In terms of calculating the offsets for the lowering
* pass, we don't care about the array index. All elements of an
* interface instance array will have the same offsets relative to
* the base of the block that backs them.
*/
deref = deref_array->array->as_dereference();
break;
} else {
/* Whether or not the field is row-major (because it might be a
* bvec2 or something) does not affect the array itself. We need
* to know whether an array element in its entirety is row-major.
*/
const bool array_row_major =
is_dereferenced_thing_row_major(deref_array);
/* The array type will give the correct interface packing
* information
*/
if (packing == GLSL_INTERFACE_PACKING_STD430) {
array_stride = deref_array->type->std430_array_stride(array_row_major);
} else {
array_stride = deref_array->type->std140_size(array_row_major);
array_stride = glsl_align(array_stride, 16);
}
}
ir_rvalue *array_index = deref_array->array_index;
if (array_index->type->base_type == GLSL_TYPE_INT)
array_index = i2u(array_index);
ir_constant *const_index =
array_index->constant_expression_value(NULL);
if (const_index) {
*const_offset += array_stride * const_index->value.u[0];
} else {
*offset = add(*offset,
mul(array_index,
new(mem_ctx) ir_constant(array_stride)));
}
deref = deref_array->array->as_dereference();
break;
}
case ir_type_dereference_record: {
ir_dereference_record *deref_record = (ir_dereference_record *) deref;
const glsl_type *struct_type = deref_record->record->type;
unsigned intra_struct_offset = 0;
for (unsigned int i = 0; i < struct_type->length; i++) {
const glsl_type *type = struct_type->fields.structure[i].type;
ir_dereference_record *field_deref = new(mem_ctx)
ir_dereference_record(deref_record->record,
struct_type->fields.structure[i].name);
const bool field_row_major =
is_dereferenced_thing_row_major(field_deref);
ralloc_free(field_deref);
unsigned field_align = 0;
if (packing == GLSL_INTERFACE_PACKING_STD430)
field_align = type->std430_base_alignment(field_row_major);
else
field_align = type->std140_base_alignment(field_row_major);
intra_struct_offset = glsl_align(intra_struct_offset, field_align);
if (strcmp(struct_type->fields.structure[i].name,
deref_record->field) == 0) {
if (struct_field)
*struct_field = &struct_type->fields.structure[i];
break;
}
if (packing == GLSL_INTERFACE_PACKING_STD430)
intra_struct_offset += type->std430_size(field_row_major);
else
intra_struct_offset += type->std140_size(field_row_major);
/* If the field just examined was itself a structure, apply rule
* #9:
*
* "The structure may have padding at the end; the base offset
* of the member following the sub-structure is rounded up to
* the next multiple of the base alignment of the structure."
*/
if (type->without_array()->is_record()) {
intra_struct_offset = glsl_align(intra_struct_offset,
field_align);
}
}
*const_offset += intra_struct_offset;
deref = deref_record->record->as_dereference();
break;
}
case ir_type_swizzle: {
ir_swizzle *deref_swizzle = (ir_swizzle *) deref;
assert(deref_swizzle->mask.num_components == 1);
*const_offset += deref_swizzle->mask.x * sizeof(int);
deref = deref_swizzle->val->as_dereference();
break;
}
default:
assert(!"not reached");
deref = NULL;
break;
}
}
}
} /* namespace lower_buffer_access */

View File

@@ -1,66 +0,0 @@
/*
* Copyright (c) 2015 Intel Corporation
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
* DEALINGS IN THE SOFTWARE.
*/
/**
* \file lower_buffer_access.h
*
* Helper for IR lowering pass to replace dereferences of buffer object based
* shader variables with intrinsic function calls.
*
* This helper is used by lowering passes for UBOs, SSBOs and compute shader
* shared variables.
*/
#pragma once
#ifndef LOWER_BUFFER_ACCESS_H
#define LOWER_BUFFER_ACCESS_H
#include "ir.h"
#include "ir_rvalue_visitor.h"
namespace lower_buffer_access {
class lower_buffer_access : public ir_rvalue_enter_visitor {
public:
virtual void
insert_buffer_access(void *mem_ctx, ir_dereference *deref,
const glsl_type *type, ir_rvalue *offset,
unsigned mask, int channel) = 0;
void emit_access(void *mem_ctx, bool is_write, ir_dereference *deref,
ir_variable *base_offset, unsigned int deref_offset,
bool row_major, int matrix_columns,
unsigned int packing, unsigned int write_mask);
bool is_dereferenced_thing_row_major(const ir_rvalue *deref);
void setup_buffer_access(void *mem_ctx, ir_variable *var, ir_rvalue *deref,
ir_rvalue **offset, unsigned *const_offset,
bool *row_major, int *matrix_columns,
const glsl_struct_field **struct_field,
unsigned packing);
};
} /* namespace lower_buffer_access */
#endif /* LOWER_BUFFER_ACCESS_H */

View File

@@ -1,496 +0,0 @@
/*
* Copyright (c) 2015 Intel Corporation
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
* DEALINGS IN THE SOFTWARE.
*/
/**
* \file lower_shared_reference.cpp
*
* IR lower pass to replace dereferences of compute shader shared variables
* with intrinsic function calls.
*
* This relieves drivers of the responsibility of allocating space for the
* shared variables in the shared memory region.
*/
#include "lower_buffer_access.h"
#include "ir_builder.h"
#include "main/macros.h"
#include "util/list.h"
#include "glsl_parser_extras.h"
using namespace ir_builder;
namespace {
struct var_offset {
struct list_head node;
const ir_variable *var;
unsigned offset;
};
class lower_shared_reference_visitor :
public lower_buffer_access::lower_buffer_access {
public:
lower_shared_reference_visitor(struct gl_shader *shader)
: list_ctx(ralloc_context(NULL)), shader(shader), shared_size(0u)
{
list_inithead(&var_offsets);
}
~lower_shared_reference_visitor()
{
ralloc_free(list_ctx);
}
enum {
shared_load_access,
shared_store_access,
shared_atomic_access,
} buffer_access_type;
void insert_buffer_access(void *mem_ctx, ir_dereference *deref,
const glsl_type *type, ir_rvalue *offset,
unsigned mask, int channel);
void handle_rvalue(ir_rvalue **rvalue);
ir_visitor_status visit_enter(ir_assignment *ir);
void handle_assignment(ir_assignment *ir);
ir_call *lower_shared_atomic_intrinsic(ir_call *ir);
ir_call *check_for_shared_atomic_intrinsic(ir_call *ir);
ir_visitor_status visit_enter(ir_call *ir);
unsigned get_shared_offset(const ir_variable *);
ir_call *shared_load(void *mem_ctx, const struct glsl_type *type,
ir_rvalue *offset);
ir_call *shared_store(void *mem_ctx, ir_rvalue *deref, ir_rvalue *offset,
unsigned write_mask);
void *list_ctx;
struct gl_shader *shader;
struct list_head var_offsets;
unsigned shared_size;
bool progress;
};
unsigned
lower_shared_reference_visitor::get_shared_offset(const ir_variable *var)
{
list_for_each_entry(var_offset, var_entry, &var_offsets, node) {
if (var_entry->var == var)
return var_entry->offset;
}
struct var_offset *new_entry = rzalloc(list_ctx, struct var_offset);
list_add(&new_entry->node, &var_offsets);
new_entry->var = var;
unsigned var_align = var->type->std430_base_alignment(false);
new_entry->offset = glsl_align(shared_size, var_align);
unsigned var_size = var->type->std430_size(false);
shared_size = new_entry->offset + var_size;
return new_entry->offset;
}
void
lower_shared_reference_visitor::handle_rvalue(ir_rvalue **rvalue)
{
if (!*rvalue)
return;
ir_dereference *deref = (*rvalue)->as_dereference();
if (!deref)
return;
ir_variable *var = deref->variable_referenced();
if (!var || var->data.mode != ir_var_shader_shared)
return;
buffer_access_type = shared_load_access;
void *mem_ctx = ralloc_parent(shader->ir);
ir_rvalue *offset = NULL;
unsigned const_offset = get_shared_offset(var);
bool row_major;
int matrix_columns;
assert(var->get_interface_type() == NULL);
const unsigned packing = GLSL_INTERFACE_PACKING_STD430;
setup_buffer_access(mem_ctx, var, deref,
&offset, &const_offset,
&row_major, &matrix_columns, NULL, packing);
/* Now that we've calculated the offset to the start of the
* dereference, walk over the type and emit loads into a temporary.
*/
const glsl_type *type = (*rvalue)->type;
ir_variable *load_var = new(mem_ctx) ir_variable(type,
"shared_load_temp",
ir_var_temporary);
base_ir->insert_before(load_var);
ir_variable *load_offset = new(mem_ctx) ir_variable(glsl_type::uint_type,
"shared_load_temp_offset",
ir_var_temporary);
base_ir->insert_before(load_offset);
base_ir->insert_before(assign(load_offset, offset));
deref = new(mem_ctx) ir_dereference_variable(load_var);
emit_access(mem_ctx, false, deref, load_offset, const_offset, row_major,
matrix_columns, packing, 0);
*rvalue = deref;
progress = true;
}
void
lower_shared_reference_visitor::handle_assignment(ir_assignment *ir)
{
if (!ir || !ir->lhs)
return;
ir_rvalue *rvalue = ir->lhs->as_rvalue();
if (!rvalue)
return;
ir_dereference *deref = ir->lhs->as_dereference();
if (!deref)
return;
ir_variable *var = ir->lhs->variable_referenced();
if (!var || var->data.mode != ir_var_shader_shared)
return;
buffer_access_type = shared_store_access;
/* We have a write to a shared variable, so declare a temporary and rewrite
* the assignment so that the temporary is the LHS.
*/
void *mem_ctx = ralloc_parent(shader->ir);
const glsl_type *type = rvalue->type;
ir_variable *store_var = new(mem_ctx) ir_variable(type,
"shared_store_temp",
ir_var_temporary);
base_ir->insert_before(store_var);
ir->lhs = new(mem_ctx) ir_dereference_variable(store_var);
ir_rvalue *offset = NULL;
unsigned const_offset = get_shared_offset(var);
bool row_major;
int matrix_columns;
assert(var->get_interface_type() == NULL);
const unsigned packing = GLSL_INTERFACE_PACKING_STD430;
setup_buffer_access(mem_ctx, var, deref,
&offset, &const_offset,
&row_major, &matrix_columns, NULL, packing);
deref = new(mem_ctx) ir_dereference_variable(store_var);
ir_variable *store_offset = new(mem_ctx) ir_variable(glsl_type::uint_type,
"shared_store_temp_offset",
ir_var_temporary);
base_ir->insert_before(store_offset);
base_ir->insert_before(assign(store_offset, offset));
/* Now we have to write the value assigned to the temporary back to memory */
emit_access(mem_ctx, true, deref, store_offset, const_offset, row_major,
matrix_columns, packing, ir->write_mask);
progress = true;
}
ir_visitor_status
lower_shared_reference_visitor::visit_enter(ir_assignment *ir)
{
handle_assignment(ir);
return rvalue_visit(ir);
}
void
lower_shared_reference_visitor::insert_buffer_access(void *mem_ctx,
ir_dereference *deref,
const glsl_type *type,
ir_rvalue *offset,
unsigned mask,
int channel)
{
if (buffer_access_type == shared_store_access) {
ir_call *store = shared_store(mem_ctx, deref, offset, mask);
base_ir->insert_after(store);
} else {
ir_call *load = shared_load(mem_ctx, type, offset);
base_ir->insert_before(load);
ir_rvalue *value = load->return_deref->as_rvalue()->clone(mem_ctx, NULL);
base_ir->insert_before(assign(deref->clone(mem_ctx, NULL),
value));
}
}
static bool
compute_shader_enabled(const _mesa_glsl_parse_state *state)
{
return state->stage == MESA_SHADER_COMPUTE;
}
ir_call *
lower_shared_reference_visitor::shared_store(void *mem_ctx,
ir_rvalue *deref,
ir_rvalue *offset,
unsigned write_mask)
{
exec_list sig_params;
ir_variable *offset_ref = new(mem_ctx)
ir_variable(glsl_type::uint_type, "offset" , ir_var_function_in);
sig_params.push_tail(offset_ref);
ir_variable *val_ref = new(mem_ctx)
ir_variable(deref->type, "value" , ir_var_function_in);
sig_params.push_tail(val_ref);
ir_variable *writemask_ref = new(mem_ctx)
ir_variable(glsl_type::uint_type, "write_mask" , ir_var_function_in);
sig_params.push_tail(writemask_ref);
ir_function_signature *sig = new(mem_ctx)
ir_function_signature(glsl_type::void_type, compute_shader_enabled);
assert(sig);
sig->replace_parameters(&sig_params);
sig->is_intrinsic = true;
ir_function *f = new(mem_ctx) ir_function("__intrinsic_store_shared");
f->add_signature(sig);
exec_list call_params;
call_params.push_tail(offset->clone(mem_ctx, NULL));
call_params.push_tail(deref->clone(mem_ctx, NULL));
call_params.push_tail(new(mem_ctx) ir_constant(write_mask));
return new(mem_ctx) ir_call(sig, NULL, &call_params);
}
ir_call *
lower_shared_reference_visitor::shared_load(void *mem_ctx,
const struct glsl_type *type,
ir_rvalue *offset)
{
exec_list sig_params;
ir_variable *offset_ref = new(mem_ctx)
ir_variable(glsl_type::uint_type, "offset_ref" , ir_var_function_in);
sig_params.push_tail(offset_ref);
ir_function_signature *sig =
new(mem_ctx) ir_function_signature(type, compute_shader_enabled);
assert(sig);
sig->replace_parameters(&sig_params);
sig->is_intrinsic = true;
ir_function *f = new(mem_ctx) ir_function("__intrinsic_load_shared");
f->add_signature(sig);
ir_variable *result = new(mem_ctx)
ir_variable(type, "shared_load_result", ir_var_temporary);
base_ir->insert_before(result);
ir_dereference_variable *deref_result = new(mem_ctx)
ir_dereference_variable(result);
exec_list call_params;
call_params.push_tail(offset->clone(mem_ctx, NULL));
return new(mem_ctx) ir_call(sig, deref_result, &call_params);
}
/* Lowers the intrinsic call to a new internal intrinsic that swaps the access
* to the shared variable in the first parameter by an offset. This involves
* creating the new internal intrinsic (i.e. the new function signature).
*/
ir_call *
lower_shared_reference_visitor::lower_shared_atomic_intrinsic(ir_call *ir)
{
/* Shared atomics usually have 2 parameters, the shared variable and an
* integer argument. The exception is CompSwap, that has an additional
* integer parameter.
*/
int param_count = ir->actual_parameters.length();
assert(param_count == 2 || param_count == 3);
/* First argument must be a scalar integer shared variable */
exec_node *param = ir->actual_parameters.get_head();
ir_instruction *inst = (ir_instruction *) param;
assert(inst->ir_type == ir_type_dereference_variable ||
inst->ir_type == ir_type_dereference_array ||
inst->ir_type == ir_type_dereference_record ||
inst->ir_type == ir_type_swizzle);
ir_rvalue *deref = (ir_rvalue *) inst;
assert(deref->type->is_scalar() && deref->type->is_integer());
ir_variable *var = deref->variable_referenced();
assert(var);
/* Compute the offset to the start if the dereference
*/
void *mem_ctx = ralloc_parent(shader->ir);
ir_rvalue *offset = NULL;
unsigned const_offset = get_shared_offset(var);
bool row_major;
int matrix_columns;
assert(var->get_interface_type() == NULL);
const unsigned packing = GLSL_INTERFACE_PACKING_STD430;
buffer_access_type = shared_atomic_access;
setup_buffer_access(mem_ctx, var, deref,
&offset, &const_offset,
&row_major, &matrix_columns, NULL, packing);
assert(offset);
assert(!row_major);
assert(matrix_columns == 1);
ir_rvalue *deref_offset =
add(offset, new(mem_ctx) ir_constant(const_offset));
/* Create the new internal function signature that will take an offset
* instead of a shared variable
*/
exec_list sig_params;
ir_variable *sig_param = new(mem_ctx)
ir_variable(glsl_type::uint_type, "offset" , ir_var_function_in);
sig_params.push_tail(sig_param);
const glsl_type *type = deref->type->base_type == GLSL_TYPE_INT ?
glsl_type::int_type : glsl_type::uint_type;
sig_param = new(mem_ctx)
ir_variable(type, "data1", ir_var_function_in);
sig_params.push_tail(sig_param);
if (param_count == 3) {
sig_param = new(mem_ctx)
ir_variable(type, "data2", ir_var_function_in);
sig_params.push_tail(sig_param);
}
ir_function_signature *sig =
new(mem_ctx) ir_function_signature(deref->type,
compute_shader_enabled);
assert(sig);
sig->replace_parameters(&sig_params);
sig->is_intrinsic = true;
char func_name[64];
sprintf(func_name, "%s_shared", ir->callee_name());
ir_function *f = new(mem_ctx) ir_function(func_name);
f->add_signature(sig);
/* Now, create the call to the internal intrinsic */
exec_list call_params;
call_params.push_tail(deref_offset);
param = ir->actual_parameters.get_head()->get_next();
ir_rvalue *param_as_rvalue = ((ir_instruction *) param)->as_rvalue();
call_params.push_tail(param_as_rvalue->clone(mem_ctx, NULL));
if (param_count == 3) {
param = param->get_next();
param_as_rvalue = ((ir_instruction *) param)->as_rvalue();
call_params.push_tail(param_as_rvalue->clone(mem_ctx, NULL));
}
ir_dereference_variable *return_deref =
ir->return_deref->clone(mem_ctx, NULL);
return new(mem_ctx) ir_call(sig, return_deref, &call_params);
}
ir_call *
lower_shared_reference_visitor::check_for_shared_atomic_intrinsic(ir_call *ir)
{
exec_list& params = ir->actual_parameters;
if (params.length() < 2 || params.length() > 3)
return ir;
ir_rvalue *rvalue =
((ir_instruction *) params.get_head())->as_rvalue();
if (!rvalue)
return ir;
ir_variable *var = rvalue->variable_referenced();
if (!var || var->data.mode != ir_var_shader_shared)
return ir;
const char *callee = ir->callee_name();
if (!strcmp("__intrinsic_atomic_add", callee) ||
!strcmp("__intrinsic_atomic_min", callee) ||
!strcmp("__intrinsic_atomic_max", callee) ||
!strcmp("__intrinsic_atomic_and", callee) ||
!strcmp("__intrinsic_atomic_or", callee) ||
!strcmp("__intrinsic_atomic_xor", callee) ||
!strcmp("__intrinsic_atomic_exchange", callee) ||
!strcmp("__intrinsic_atomic_comp_swap", callee)) {
return lower_shared_atomic_intrinsic(ir);
}
return ir;
}
ir_visitor_status
lower_shared_reference_visitor::visit_enter(ir_call *ir)
{
ir_call *new_ir = check_for_shared_atomic_intrinsic(ir);
if (new_ir != ir) {
progress = true;
base_ir->replace_with(new_ir);
return visit_continue_with_parent;
}
return rvalue_visit(ir);
}
} /* unnamed namespace */
void
lower_shared_reference(struct gl_shader *shader, unsigned *shared_size)
{
if (shader->Stage != MESA_SHADER_COMPUTE)
return;
lower_shared_reference_visitor v(shader);
/* Loop over the instructions lowering references, because we take a deref
* of an shared variable array using a shared variable dereference as the
* index will produce a collection of instructions all of which have cloned
* shared variable dereferences for that array index.
*/
do {
v.progress = false;
visit_list_elements(&v, shader->ir);
} while (v.progress);
*shared_size = v.shared_size;
}

View File

@@ -1,123 +0,0 @@
/*
* Copyright © 2015 Red Hat
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
* DEALINGS IN THE SOFTWARE.
*/
/**
* \file lower_subroutine.cpp
*
* lowers subroutines to an if ladder.
*/
#include "compiler/glsl_types.h"
#include "glsl_parser_extras.h"
#include "ir.h"
#include "ir_builder.h"
using namespace ir_builder;
namespace {
class lower_subroutine_visitor : public ir_hierarchical_visitor {
public:
lower_subroutine_visitor(struct _mesa_glsl_parse_state *state)
: state(state)
{
this->progress = false;
}
ir_visitor_status visit_leave(ir_call *);
ir_call *call_clone(ir_call *call, ir_function_signature *callee);
bool progress;
struct _mesa_glsl_parse_state *state;
};
}
bool
lower_subroutine(exec_list *instructions, struct _mesa_glsl_parse_state *state)
{
lower_subroutine_visitor v(state);
visit_list_elements(&v, instructions);
return v.progress;
}
ir_call *
lower_subroutine_visitor::call_clone(ir_call *call, ir_function_signature *callee)
{
void *mem_ctx = ralloc_parent(call);
ir_dereference_variable *new_return_ref = NULL;
if (call->return_deref != NULL)
new_return_ref = call->return_deref->clone(mem_ctx, NULL);
exec_list new_parameters;
foreach_in_list(ir_instruction, ir, &call->actual_parameters) {
new_parameters.push_tail(ir->clone(mem_ctx, NULL));
}
return new(mem_ctx) ir_call(callee, new_return_ref, &new_parameters);
}
ir_visitor_status
lower_subroutine_visitor::visit_leave(ir_call *ir)
{
if (!ir->sub_var)
return visit_continue;
void *mem_ctx = ralloc_parent(ir);
ir_if *last_branch = NULL;
for (int s = this->state->num_subroutines - 1; s >= 0; s--) {
ir_rvalue *var;
ir_constant *lc = new(mem_ctx)ir_constant(s);
ir_function *fn = this->state->subroutines[s];
bool is_compat = false;
for (int i = 0; i < fn->num_subroutine_types; i++) {
if (ir->sub_var->type->without_array() == fn->subroutine_types[i]) {
is_compat = true;
break;
}
}
if (is_compat == false)
continue;
if (ir->array_idx != NULL)
var = ir->array_idx->clone(mem_ctx, NULL);
else
var = new(mem_ctx) ir_dereference_variable(ir->sub_var);
ir_function_signature *sub_sig =
fn->exact_matching_signature(this->state,
&ir->actual_parameters);
ir_call *new_call = call_clone(ir, sub_sig);
if (!last_branch)
last_branch = if_tree(equal(subr_to_int(var), lc), new_call);
else
last_branch = if_tree(equal(subr_to_int(var), lc), new_call, last_branch);
}
if (last_branch)
ir->insert_before(last_branch);
ir->remove();
return visit_continue;
}

View File

@@ -1,459 +0,0 @@
/*
* Copyright © 2014 Intel Corporation
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
* DEALINGS IN THE SOFTWARE.
*/
/**
* \file lower_tess_level.cpp
*
* This pass accounts for the difference between the way gl_TessLevelOuter
* and gl_TessLevelInner is declared in standard GLSL (as an array of
* floats), and the way it is frequently implemented in hardware (as a vec4
* and vec2).
*
* The declaration of gl_TessLevel* is replaced with a declaration
* of gl_TessLevel*MESA, and any references to gl_TessLevel* are
* translated to refer to gl_TessLevel*MESA with the appropriate
* swizzling of array indices. For instance:
*
* gl_TessLevelOuter[i]
*
* is translated into:
*
* gl_TessLevelOuterMESA[i]
*
* Since some hardware may not internally represent gl_TessLevel* as a pair
* of vec4's, this lowering pass is optional. To enable it, set the
* LowerTessLevel flag in gl_shader_compiler_options to true.
*/
#include "glsl_symbol_table.h"
#include "ir_rvalue_visitor.h"
#include "ir.h"
#include "program/prog_instruction.h" /* For WRITEMASK_* */
namespace {
class lower_tess_level_visitor : public ir_rvalue_visitor {
public:
explicit lower_tess_level_visitor(gl_shader_stage shader_stage)
: progress(false), old_tess_level_outer_var(NULL),
old_tess_level_inner_var(NULL), new_tess_level_outer_var(NULL),
new_tess_level_inner_var(NULL), shader_stage(shader_stage)
{
}
virtual ir_visitor_status visit(ir_variable *);
bool is_tess_level_array(ir_rvalue *ir);
ir_rvalue *lower_tess_level_array(ir_rvalue *ir);
virtual ir_visitor_status visit_leave(ir_assignment *);
void visit_new_assignment(ir_assignment *ir);
virtual ir_visitor_status visit_leave(ir_call *);
virtual void handle_rvalue(ir_rvalue **rvalue);
void fix_lhs(ir_assignment *);
bool progress;
/**
* Pointer to the declaration of gl_TessLevel*, if found.
*/
ir_variable *old_tess_level_outer_var;
ir_variable *old_tess_level_inner_var;
/**
* Pointer to the newly-created gl_TessLevel*MESA variables.
*/
ir_variable *new_tess_level_outer_var;
ir_variable *new_tess_level_inner_var;
/**
* Type of shader we are compiling (e.g. MESA_SHADER_TESS_CTRL)
*/
const gl_shader_stage shader_stage;
};
} /* anonymous namespace */
/**
* Replace any declaration of gl_TessLevel* as an array of floats with a
* declaration of gl_TessLevel*MESA as a vec4.
*/
ir_visitor_status
lower_tess_level_visitor::visit(ir_variable *ir)
{
if ((!ir->name) ||
((strcmp(ir->name, "gl_TessLevelInner") != 0) &&
(strcmp(ir->name, "gl_TessLevelOuter") != 0)))
return visit_continue;
assert (ir->type->is_array());
if (strcmp(ir->name, "gl_TessLevelOuter") == 0) {
if (this->old_tess_level_outer_var)
return visit_continue;
old_tess_level_outer_var = ir;
assert(ir->type->fields.array == glsl_type::float_type);
/* Clone the old var so that we inherit all of its properties */
new_tess_level_outer_var = ir->clone(ralloc_parent(ir), NULL);
/* And change the properties that we need to change */
new_tess_level_outer_var->name = ralloc_strdup(new_tess_level_outer_var,
"gl_TessLevelOuterMESA");
new_tess_level_outer_var->type = glsl_type::vec4_type;
new_tess_level_outer_var->data.max_array_access = 0;
ir->replace_with(new_tess_level_outer_var);
} else if (strcmp(ir->name, "gl_TessLevelInner") == 0) {
if (this->old_tess_level_inner_var)
return visit_continue;
old_tess_level_inner_var = ir;
assert(ir->type->fields.array == glsl_type::float_type);
/* Clone the old var so that we inherit all of its properties */
new_tess_level_inner_var = ir->clone(ralloc_parent(ir), NULL);
/* And change the properties that we need to change */
new_tess_level_inner_var->name = ralloc_strdup(new_tess_level_inner_var,
"gl_TessLevelInnerMESA");
new_tess_level_inner_var->type = glsl_type::vec2_type;
new_tess_level_inner_var->data.max_array_access = 0;
ir->replace_with(new_tess_level_inner_var);
} else {
assert(0);
}
this->progress = true;
return visit_continue;
}
/**
* Determine whether the given rvalue describes an array of floats that
* needs to be lowered to a vec4; that is, determine whether it
* matches one of the following patterns:
*
* - gl_TessLevelOuter
* - gl_TessLevelInner
*/
bool
lower_tess_level_visitor::is_tess_level_array(ir_rvalue *ir)
{
if (!ir->type->is_array())
return false;
if (ir->type->fields.array != glsl_type::float_type)
return false;
if (this->old_tess_level_outer_var) {
if (ir->variable_referenced() == this->old_tess_level_outer_var)
return true;
}
if (this->old_tess_level_inner_var) {
if (ir->variable_referenced() == this->old_tess_level_inner_var)
return true;
}
return false;
}
/**
* If the given ir satisfies is_tess_level_array(), return new ir
* representing its lowered equivalent. That is, map:
*
* - gl_TessLevelOuter => gl_TessLevelOuterMESA
* - gl_TessLevelInner => gl_TessLevelInnerMESA
*
* Otherwise return NULL.
*/
ir_rvalue *
lower_tess_level_visitor::lower_tess_level_array(ir_rvalue *ir)
{
if (!ir->type->is_array())
return NULL;
if (ir->type->fields.array != glsl_type::float_type)
return NULL;
ir_variable **new_var = NULL;
if (this->old_tess_level_outer_var) {
if (ir->variable_referenced() == this->old_tess_level_outer_var)
new_var = &this->new_tess_level_outer_var;
}
if (this->old_tess_level_inner_var) {
if (ir->variable_referenced() == this->old_tess_level_inner_var)
new_var = &this->new_tess_level_inner_var;
}
if (new_var == NULL)
return NULL;
assert(ir->as_dereference_variable());
return new(ralloc_parent(ir)) ir_dereference_variable(*new_var);
}
void
lower_tess_level_visitor::handle_rvalue(ir_rvalue **rv)
{
if (*rv == NULL)
return;
ir_dereference_array *const array_deref = (*rv)->as_dereference_array();
if (array_deref == NULL)
return;
/* Replace any expression that indexes one of the floats in gl_TessLevel*
* with an expression that indexes into one of the vec4's
* gl_TessLevel*MESA and accesses the appropriate component.
*/
ir_rvalue *lowered_vec4 =
this->lower_tess_level_array(array_deref->array);
if (lowered_vec4 != NULL) {
this->progress = true;
void *mem_ctx = ralloc_parent(array_deref);
ir_expression *const expr =
new(mem_ctx) ir_expression(ir_binop_vector_extract,
lowered_vec4,
array_deref->array_index);
*rv = expr;
}
}
void
lower_tess_level_visitor::fix_lhs(ir_assignment *ir)
{
if (ir->lhs->ir_type != ir_type_expression)
return;
void *mem_ctx = ralloc_parent(ir);
ir_expression *const expr = (ir_expression *) ir->lhs;
/* The expression must be of the form:
*
* (vector_extract gl_TessLevel*MESA, j).
*/
assert(expr->operation == ir_binop_vector_extract);
assert(expr->operands[0]->ir_type == ir_type_dereference_variable);
assert((expr->operands[0]->type == glsl_type::vec4_type) ||
(expr->operands[0]->type == glsl_type::vec2_type));
ir_dereference *const new_lhs = (ir_dereference *) expr->operands[0];
ir_constant *old_index_constant = expr->operands[1]->constant_expression_value();
if (!old_index_constant) {
ir->rhs = new(mem_ctx) ir_expression(ir_triop_vector_insert,
expr->operands[0]->type,
new_lhs->clone(mem_ctx, NULL),
ir->rhs,
expr->operands[1]);
}
ir->set_lhs(new_lhs);
if (old_index_constant) {
/* gl_TessLevel* is being accessed via a constant index. Don't bother
* creating a vector insert op. Just use a write mask.
*/
ir->write_mask = 1 << old_index_constant->get_int_component(0);
} else {
ir->write_mask = (1 << expr->operands[0]->type->vector_elements) - 1;
}
}
/**
* Replace any assignment having a gl_TessLevel* (undereferenced) as
* its LHS or RHS with a sequence of assignments, one for each component of
* the array. Each of these assignments is lowered to refer to
* gl_TessLevel*MESA as appropriate.
*/
ir_visitor_status
lower_tess_level_visitor::visit_leave(ir_assignment *ir)
{
/* First invoke the base class visitor. This causes handle_rvalue() to be
* called on ir->rhs and ir->condition.
*/
ir_rvalue_visitor::visit_leave(ir);
if (this->is_tess_level_array(ir->lhs) ||
this->is_tess_level_array(ir->rhs)) {
/* LHS or RHS of the assignment is the entire gl_TessLevel* array.
* Since we are
* reshaping gl_TessLevel* from an array of floats to a
* vec4, this isn't going to work as a bulk assignment anymore, so
* unroll it to element-by-element assignments and lower each of them.
*
* Note: to unroll into element-by-element assignments, we need to make
* clones of the LHS and RHS. This is safe because expressions and
* l-values are side-effect free.
*/
void *ctx = ralloc_parent(ir);
int array_size = ir->lhs->type->array_size();
for (int i = 0; i < array_size; ++i) {
ir_dereference_array *new_lhs = new(ctx) ir_dereference_array(
ir->lhs->clone(ctx, NULL), new(ctx) ir_constant(i));
ir_dereference_array *new_rhs = new(ctx) ir_dereference_array(
ir->rhs->clone(ctx, NULL), new(ctx) ir_constant(i));
this->handle_rvalue((ir_rvalue **) &new_rhs);
/* Handle the LHS after creating the new assignment. This must
* happen in this order because handle_rvalue may replace the old LHS
* with an ir_expression of ir_binop_vector_extract. Since this is
* not a valide l-value, this will cause an assertion in the
* ir_assignment constructor to fail.
*
* If this occurs, replace the mangled LHS with a dereference of the
* vector, and replace the RHS with an ir_triop_vector_insert.
*/
ir_assignment *const assign = new(ctx) ir_assignment(new_lhs, new_rhs);
this->handle_rvalue((ir_rvalue **) &assign->lhs);
this->fix_lhs(assign);
this->base_ir->insert_before(assign);
}
ir->remove();
return visit_continue;
}
/* Handle the LHS as if it were an r-value. Normally
* rvalue_visit(ir_assignment *) only visits the RHS, but we need to lower
* expressions in the LHS as well.
*
* This may cause the LHS to get replaced with an ir_expression of
* ir_binop_vector_extract. If this occurs, replace it with a dereference
* of the vector, and replace the RHS with an ir_triop_vector_insert.
*/
handle_rvalue((ir_rvalue **)&ir->lhs);
this->fix_lhs(ir);
return rvalue_visit(ir);
}
/**
* Set up base_ir properly and call visit_leave() on a newly created
* ir_assignment node. This is used in cases where we have to insert an
* ir_assignment in a place where we know the hierarchical visitor won't see
* it.
*/
void
lower_tess_level_visitor::visit_new_assignment(ir_assignment *ir)
{
ir_instruction *old_base_ir = this->base_ir;
this->base_ir = ir;
ir->accept(this);
this->base_ir = old_base_ir;
}
/**
* If a gl_TessLevel* variable appears as an argument in an ir_call
* expression, replace it with a temporary variable, and make sure the ir_call
* is preceded and/or followed by assignments that copy the contents of the
* temporary variable to and/or from gl_TessLevel*. Each of these
* assignments is then lowered to refer to gl_TessLevel*MESA.
*/
ir_visitor_status
lower_tess_level_visitor::visit_leave(ir_call *ir)
{
void *ctx = ralloc_parent(ir);
const exec_node *formal_param_node = ir->callee->parameters.head;
const exec_node *actual_param_node = ir->actual_parameters.head;
while (!actual_param_node->is_tail_sentinel()) {
ir_variable *formal_param = (ir_variable *) formal_param_node;
ir_rvalue *actual_param = (ir_rvalue *) actual_param_node;
/* Advance formal_param_node and actual_param_node now so that we can
* safely replace actual_param with another node, if necessary, below.
*/
formal_param_node = formal_param_node->next;
actual_param_node = actual_param_node->next;
if (!this->is_tess_level_array(actual_param))
continue;
/* User is trying to pass a whole gl_TessLevel* array to a function
* call. Since we are reshaping gl_TessLevel* from an array of floats
* to a vec4, this isn't going to work anymore, so use a temporary
* array instead.
*/
ir_variable *temp = new(ctx) ir_variable(
actual_param->type, "temp_tess_level", ir_var_temporary);
this->base_ir->insert_before(temp);
actual_param->replace_with(
new(ctx) ir_dereference_variable(temp));
if (formal_param->data.mode == ir_var_function_in
|| formal_param->data.mode == ir_var_function_inout) {
/* Copy from gl_TessLevel* to the temporary before the call.
* Since we are going to insert this copy before the current
* instruction, we need to visit it afterwards to make sure it
* gets lowered.
*/
ir_assignment *new_assignment = new(ctx) ir_assignment(
new(ctx) ir_dereference_variable(temp),
actual_param->clone(ctx, NULL));
this->base_ir->insert_before(new_assignment);
this->visit_new_assignment(new_assignment);
}
if (formal_param->data.mode == ir_var_function_out
|| formal_param->data.mode == ir_var_function_inout) {
/* Copy from the temporary to gl_TessLevel* after the call.
* Since visit_list_elements() has already decided which
* instruction it's going to visit next, we need to visit
* afterwards to make sure it gets lowered.
*/
ir_assignment *new_assignment = new(ctx) ir_assignment(
actual_param->clone(ctx, NULL),
new(ctx) ir_dereference_variable(temp));
this->base_ir->insert_after(new_assignment);
this->visit_new_assignment(new_assignment);
}
}
return rvalue_visit(ir);
}
bool
lower_tess_level(gl_shader *shader)
{
if ((shader->Stage != MESA_SHADER_TESS_CTRL) &&
(shader->Stage != MESA_SHADER_TESS_EVAL))
return false;
lower_tess_level_visitor v(shader->Stage);
visit_list_elements(&v, shader->ir);
if (v.new_tess_level_outer_var)
shader->symbols->add_variable(v.new_tess_level_outer_var);
if (v.new_tess_level_inner_var)
shader->symbols->add_variable(v.new_tess_level_inner_var);
return v.progress;
}

File diff suppressed because it is too large Load Diff

View File

@@ -1,104 +0,0 @@
/*
* Copyright © 2013 Intel Corporation
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
* DEALINGS IN THE SOFTWARE.
*/
#include "ir.h"
#include "ir_builder.h"
#include "ir_rvalue_visitor.h"
#include "ir_optimization.h"
using namespace ir_builder;
namespace {
class vector_deref_visitor : public ir_rvalue_enter_visitor {
public:
vector_deref_visitor()
: progress(false)
{
}
virtual ~vector_deref_visitor()
{
}
virtual void handle_rvalue(ir_rvalue **rv);
virtual ir_visitor_status visit_enter(ir_assignment *ir);
bool progress;
};
} /* anonymous namespace */
ir_visitor_status
vector_deref_visitor::visit_enter(ir_assignment *ir)
{
if (!ir->lhs || ir->lhs->ir_type != ir_type_dereference_array)
return ir_rvalue_enter_visitor::visit_enter(ir);
ir_dereference_array *const deref = (ir_dereference_array *) ir->lhs;
if (!deref->array->type->is_vector())
return ir_rvalue_enter_visitor::visit_enter(ir);
ir_dereference *const new_lhs = (ir_dereference *) deref->array;
ir->set_lhs(new_lhs);
ir_constant *old_index_constant = deref->array_index->constant_expression_value();
void *mem_ctx = ralloc_parent(ir);
if (!old_index_constant) {
ir->rhs = new(mem_ctx) ir_expression(ir_triop_vector_insert,
new_lhs->type,
new_lhs->clone(mem_ctx, NULL),
ir->rhs,
deref->array_index);
ir->write_mask = (1 << new_lhs->type->vector_elements) - 1;
} else {
ir->write_mask = 1 << old_index_constant->get_int_component(0);
}
return ir_rvalue_enter_visitor::visit_enter(ir);
}
void
vector_deref_visitor::handle_rvalue(ir_rvalue **rv)
{
if (*rv == NULL || (*rv)->ir_type != ir_type_dereference_array)
return;
ir_dereference_array *const deref = (ir_dereference_array *) *rv;
if (!deref->array->type->is_vector())
return;
void *mem_ctx = ralloc_parent(deref);
*rv = new(mem_ctx) ir_expression(ir_binop_vector_extract,
deref->array,
deref->array_index);
}
bool
lower_vector_derefs(gl_shader *shader)
{
vector_deref_visitor v;
visit_list_elements(&v, shader->ir);
return v.progress;
}

View File

@@ -1,71 +0,0 @@
NIR_GENERATED_FILES = \
nir_builder_opcodes.h \
nir_constant_expressions.c \
nir_opcodes.c \
nir_opcodes.h \
nir_opt_algebraic.c
NIR_FILES = \
glsl_to_nir.cpp \
glsl_to_nir.h \
nir.c \
nir.h \
nir_array.h \
nir_builder.h \
nir_clone.c \
nir_constant_expressions.h \
nir_control_flow.c \
nir_control_flow.h \
nir_control_flow_private.h \
nir_dominance.c \
nir_from_ssa.c \
nir_gs_count_vertices.c \
nir_intrinsics.c \
nir_intrinsics.h \
nir_instr_set.c \
nir_instr_set.h \
nir_liveness.c \
nir_lower_alu_to_scalar.c \
nir_lower_atomics.c \
nir_lower_clip.c \
nir_lower_global_vars_to_local.c \
nir_lower_gs_intrinsics.c \
nir_lower_load_const_to_scalar.c \
nir_lower_locals_to_regs.c \
nir_lower_idiv.c \
nir_lower_io.c \
nir_lower_outputs_to_temporaries.c \
nir_lower_phis_to_scalar.c \
nir_lower_samplers.c \
nir_lower_system_values.c \
nir_lower_tex.c \
nir_lower_to_source_mods.c \
nir_lower_two_sided_color.c \
nir_lower_vars_to_ssa.c \
nir_lower_var_copies.c \
nir_lower_vec_to_movs.c \
nir_metadata.c \
nir_move_vec_src_uses_to_dest.c \
nir_normalize_cubemap_coords.c \
nir_opt_constant_folding.c \
nir_opt_copy_propagate.c \
nir_opt_cse.c \
nir_opt_dce.c \
nir_opt_dead_cf.c \
nir_opt_gcm.c \
nir_opt_global_to_local.c \
nir_opt_peephole_select.c \
nir_opt_remove_phis.c \
nir_opt_undef.c \
nir_print.c \
nir_remove_dead_variables.c \
nir_search.c \
nir_search.h \
nir_split_var_copies.c \
nir_sweep.c \
nir_to_ssa.c \
nir_validate.c \
nir_vla.h \
nir_worklist.c \
nir_worklist.h

View File

@@ -1,663 +0,0 @@
/*
* Copyright © 2015 Red Hat
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*/
#include "nir.h"
#include "nir_control_flow_private.h"
/* Secret Decoder Ring:
* clone_foo():
* Allocate and clone a foo.
* __clone_foo():
* Clone body of foo (ie. parent class, embedded struct, etc)
*/
typedef struct {
/* maps orig ptr -> cloned ptr: */
struct hash_table *ptr_table;
/* List of phi sources. */
struct list_head phi_srcs;
/* new shader object, used as memctx for just about everything else: */
nir_shader *ns;
} clone_state;
static void
init_clone_state(clone_state *state)
{
state->ptr_table = _mesa_hash_table_create(NULL, _mesa_hash_pointer,
_mesa_key_pointer_equal);
list_inithead(&state->phi_srcs);
}
static void
free_clone_state(clone_state *state)
{
_mesa_hash_table_destroy(state->ptr_table, NULL);
}
static void *
lookup_ptr(clone_state *state, const void *ptr)
{
struct hash_entry *entry;
if (!ptr)
return NULL;
entry = _mesa_hash_table_search(state->ptr_table, ptr);
assert(entry && "Failed to find pointer!");
if (!entry)
return NULL;
return entry->data;
}
static void
store_ptr(clone_state *state, void *nptr, const void *ptr)
{
_mesa_hash_table_insert(state->ptr_table, ptr, nptr);
}
static nir_constant *
clone_constant(clone_state *state, const nir_constant *c, nir_variable *nvar)
{
nir_constant *nc = ralloc(nvar, nir_constant);
nc->value = c->value;
nc->num_elements = c->num_elements;
nc->elements = ralloc_array(nvar, nir_constant *, c->num_elements);
for (unsigned i = 0; i < c->num_elements; i++) {
nc->elements[i] = clone_constant(state, c->elements[i], nvar);
}
return nc;
}
/* NOTE: for cloning nir_variable's, bypass nir_variable_create to avoid
* having to deal with locals and globals separately:
*/
static nir_variable *
clone_variable(clone_state *state, const nir_variable *var)
{
nir_variable *nvar = rzalloc(state->ns, nir_variable);
store_ptr(state, nvar, var);
nvar->type = var->type;
nvar->name = ralloc_strdup(nvar, var->name);
nvar->data = var->data;
nvar->num_state_slots = var->num_state_slots;
nvar->state_slots = ralloc_array(nvar, nir_state_slot, var->num_state_slots);
memcpy(nvar->state_slots, var->state_slots,
var->num_state_slots * sizeof(nir_state_slot));
if (var->constant_initializer) {
nvar->constant_initializer =
clone_constant(state, var->constant_initializer, nvar);
}
nvar->interface_type = var->interface_type;
return nvar;
}
/* clone list of nir_variable: */
static void
clone_var_list(clone_state *state, struct exec_list *dst,
const struct exec_list *list)
{
exec_list_make_empty(dst);
foreach_list_typed(nir_variable, var, node, list) {
nir_variable *nvar = clone_variable(state, var);
exec_list_push_tail(dst, &nvar->node);
}
}
/* NOTE: for cloning nir_register's, bypass nir_global/local_reg_create()
* to avoid having to deal with locals and globals separately:
*/
static nir_register *
clone_register(clone_state *state, const nir_register *reg)
{
nir_register *nreg = rzalloc(state->ns, nir_register);
store_ptr(state, nreg, reg);
nreg->num_components = reg->num_components;
nreg->num_array_elems = reg->num_array_elems;
nreg->index = reg->index;
nreg->name = ralloc_strdup(nreg, reg->name);
nreg->is_global = reg->is_global;
nreg->is_packed = reg->is_packed;
/* reconstructing uses/defs/if_uses handled by nir_instr_insert() */
list_inithead(&nreg->uses);
list_inithead(&nreg->defs);
list_inithead(&nreg->if_uses);
return nreg;
}
/* clone list of nir_register: */
static void
clone_reg_list(clone_state *state, struct exec_list *dst,
const struct exec_list *list)
{
exec_list_make_empty(dst);
foreach_list_typed(nir_register, reg, node, list) {
nir_register *nreg = clone_register(state, reg);
exec_list_push_tail(dst, &nreg->node);
}
}
static void
__clone_src(clone_state *state, void *ninstr_or_if,
nir_src *nsrc, const nir_src *src)
{
nsrc->is_ssa = src->is_ssa;
if (src->is_ssa) {
nsrc->ssa = lookup_ptr(state, src->ssa);
} else {
nsrc->reg.reg = lookup_ptr(state, src->reg.reg);
if (src->reg.indirect) {
nsrc->reg.indirect = ralloc(ninstr_or_if, nir_src);
__clone_src(state, ninstr_or_if, nsrc->reg.indirect, src->reg.indirect);
}
nsrc->reg.base_offset = src->reg.base_offset;
}
}
static void
__clone_dst(clone_state *state, nir_instr *ninstr,
nir_dest *ndst, const nir_dest *dst)
{
ndst->is_ssa = dst->is_ssa;
if (dst->is_ssa) {
nir_ssa_dest_init(ninstr, ndst, dst->ssa.num_components, dst->ssa.name);
store_ptr(state, &ndst->ssa, &dst->ssa);
} else {
ndst->reg.reg = lookup_ptr(state, dst->reg.reg);
if (dst->reg.indirect) {
ndst->reg.indirect = ralloc(ninstr, nir_src);
__clone_src(state, ninstr, ndst->reg.indirect, dst->reg.indirect);
}
ndst->reg.base_offset = dst->reg.base_offset;
}
}
static nir_deref *clone_deref(clone_state *state, const nir_deref *deref,
nir_instr *ninstr, nir_deref *parent);
static nir_deref_var *
clone_deref_var(clone_state *state, const nir_deref_var *dvar,
nir_instr *ninstr)
{
nir_variable *nvar = lookup_ptr(state, dvar->var);
nir_deref_var *ndvar = nir_deref_var_create(ninstr, nvar);
if (dvar->deref.child)
ndvar->deref.child = clone_deref(state, dvar->deref.child,
ninstr, &ndvar->deref);
return ndvar;
}
static nir_deref_array *
clone_deref_array(clone_state *state, const nir_deref_array *darr,
nir_instr *ninstr, nir_deref *parent)
{
nir_deref_array *ndarr = nir_deref_array_create(parent);
ndarr->deref.type = darr->deref.type;
if (darr->deref.child)
ndarr->deref.child = clone_deref(state, darr->deref.child,
ninstr, &ndarr->deref);
ndarr->deref_array_type = darr->deref_array_type;
ndarr->base_offset = darr->base_offset;
if (ndarr->deref_array_type == nir_deref_array_type_indirect)
__clone_src(state, ninstr, &ndarr->indirect, &darr->indirect);
return ndarr;
}
static nir_deref_struct *
clone_deref_struct(clone_state *state, const nir_deref_struct *dstr,
nir_instr *ninstr, nir_deref *parent)
{
nir_deref_struct *ndstr = nir_deref_struct_create(parent, dstr->index);
ndstr->deref.type = dstr->deref.type;
if (dstr->deref.child)
ndstr->deref.child = clone_deref(state, dstr->deref.child,
ninstr, &ndstr->deref);
return ndstr;
}
static nir_deref *
clone_deref(clone_state *state, const nir_deref *dref,
nir_instr *ninstr, nir_deref *parent)
{
switch (dref->deref_type) {
case nir_deref_type_array:
return &clone_deref_array(state, nir_deref_as_array(dref),
ninstr, parent)->deref;
case nir_deref_type_struct:
return &clone_deref_struct(state, nir_deref_as_struct(dref),
ninstr, parent)->deref;
default:
unreachable("bad deref type");
return NULL;
}
}
static nir_alu_instr *
clone_alu(clone_state *state, const nir_alu_instr *alu)
{
nir_alu_instr *nalu = nir_alu_instr_create(state->ns, alu->op);
__clone_dst(state, &nalu->instr, &nalu->dest.dest, &alu->dest.dest);
nalu->dest.saturate = alu->dest.saturate;
nalu->dest.write_mask = alu->dest.write_mask;
for (unsigned i = 0; i < nir_op_infos[alu->op].num_inputs; i++) {
__clone_src(state, &nalu->instr, &nalu->src[i].src, &alu->src[i].src);
nalu->src[i].negate = alu->src[i].negate;
nalu->src[i].abs = alu->src[i].abs;
memcpy(nalu->src[i].swizzle, alu->src[i].swizzle,
sizeof(nalu->src[i].swizzle));
}
return nalu;
}
static nir_intrinsic_instr *
clone_intrinsic(clone_state *state, const nir_intrinsic_instr *itr)
{
nir_intrinsic_instr *nitr =
nir_intrinsic_instr_create(state->ns, itr->intrinsic);
unsigned num_variables = nir_intrinsic_infos[itr->intrinsic].num_variables;
unsigned num_srcs = nir_intrinsic_infos[itr->intrinsic].num_srcs;
if (nir_intrinsic_infos[itr->intrinsic].has_dest)
__clone_dst(state, &nitr->instr, &nitr->dest, &itr->dest);
nitr->num_components = itr->num_components;
memcpy(nitr->const_index, itr->const_index, sizeof(nitr->const_index));
for (unsigned i = 0; i < num_variables; i++) {
nitr->variables[i] = clone_deref_var(state, itr->variables[i],
&nitr->instr);
}
for (unsigned i = 0; i < num_srcs; i++)
__clone_src(state, &nitr->instr, &nitr->src[i], &itr->src[i]);
return nitr;
}
static nir_load_const_instr *
clone_load_const(clone_state *state, const nir_load_const_instr *lc)
{
nir_load_const_instr *nlc =
nir_load_const_instr_create(state->ns, lc->def.num_components);
memcpy(&nlc->value, &lc->value, sizeof(nlc->value));
store_ptr(state, &nlc->def, &lc->def);
return nlc;
}
static nir_ssa_undef_instr *
clone_ssa_undef(clone_state *state, const nir_ssa_undef_instr *sa)
{
nir_ssa_undef_instr *nsa =
nir_ssa_undef_instr_create(state->ns, sa->def.num_components);
store_ptr(state, &nsa->def, &sa->def);
return nsa;
}
static nir_tex_instr *
clone_tex(clone_state *state, const nir_tex_instr *tex)
{
nir_tex_instr *ntex = nir_tex_instr_create(state->ns, tex->num_srcs);
ntex->sampler_dim = tex->sampler_dim;
ntex->dest_type = tex->dest_type;
ntex->op = tex->op;
__clone_dst(state, &ntex->instr, &ntex->dest, &tex->dest);
for (unsigned i = 0; i < ntex->num_srcs; i++) {
ntex->src[i].src_type = tex->src[i].src_type;
__clone_src(state, &ntex->instr, &ntex->src[i].src, &tex->src[i].src);
}
ntex->coord_components = tex->coord_components;
ntex->is_array = tex->is_array;
ntex->is_shadow = tex->is_shadow;
ntex->is_new_style_shadow = tex->is_new_style_shadow;
ntex->component = tex->component;
ntex->texture_index = tex->texture_index;
if (tex->texture)
ntex->texture = clone_deref_var(state, tex->texture, &ntex->instr);
ntex->texture_array_size = tex->texture_array_size;
ntex->sampler_index = tex->sampler_index;
if (tex->sampler)
ntex->sampler = clone_deref_var(state, tex->sampler, &ntex->instr);
return ntex;
}
static nir_phi_instr *
clone_phi(clone_state *state, const nir_phi_instr *phi, nir_block *nblk)
{
nir_phi_instr *nphi = nir_phi_instr_create(state->ns);
__clone_dst(state, &nphi->instr, &nphi->dest, &phi->dest);
/* Cloning a phi node is a bit different from other instructions. The
* sources of phi instructions are the only time where we can use an SSA
* def before it is defined. In order to handle this, we just copy over
* the sources from the old phi instruction directly and then fix them up
* in a second pass once all the instrutions in the function have been
* properly cloned.
*
* In order to ensure that the copied sources (which are the same as the
* old phi instruction's sources for now) don't get inserted into the old
* shader's use-def lists, we have to add the phi instruction *before* we
* set up its sources.
*/
nir_instr_insert_after_block(nblk, &nphi->instr);
foreach_list_typed(nir_phi_src, src, node, &phi->srcs) {
nir_phi_src *nsrc = ralloc(nphi, nir_phi_src);
/* Just copy the old source for now. */
memcpy(nsrc, src, sizeof(*src));
/* Since we're not letting nir_insert_instr handle use/def stuff for us,
* we have to set the parent_instr manually. It doesn't really matter
* when we do it, so we might as well do it here.
*/
nsrc->src.parent_instr = &nphi->instr;
/* Stash it in the list of phi sources. We'll walk this list and fix up
* sources at the very end of clone_function_impl.
*/
list_add(&nsrc->src.use_link, &state->phi_srcs);
exec_list_push_tail(&nphi->srcs, &nsrc->node);
}
return nphi;
}
static nir_jump_instr *
clone_jump(clone_state *state, const nir_jump_instr *jmp)
{
nir_jump_instr *njmp = nir_jump_instr_create(state->ns, jmp->type);
return njmp;
}
static nir_call_instr *
clone_call(clone_state *state, const nir_call_instr *call)
{
nir_function *ncallee = lookup_ptr(state, call->callee);
nir_call_instr *ncall = nir_call_instr_create(state->ns, ncallee);
for (unsigned i = 0; i < ncall->num_params; i++)
ncall->params[i] = clone_deref_var(state, call->params[i], &ncall->instr);
ncall->return_deref = clone_deref_var(state, call->return_deref,
&ncall->instr);
return ncall;
}
static nir_instr *
clone_instr(clone_state *state, const nir_instr *instr)
{
switch (instr->type) {
case nir_instr_type_alu:
return &clone_alu(state, nir_instr_as_alu(instr))->instr;
case nir_instr_type_intrinsic:
return &clone_intrinsic(state, nir_instr_as_intrinsic(instr))->instr;
case nir_instr_type_load_const:
return &clone_load_const(state, nir_instr_as_load_const(instr))->instr;
case nir_instr_type_ssa_undef:
return &clone_ssa_undef(state, nir_instr_as_ssa_undef(instr))->instr;
case nir_instr_type_tex:
return &clone_tex(state, nir_instr_as_tex(instr))->instr;
case nir_instr_type_phi:
unreachable("Cannot clone phis with clone_instr");
case nir_instr_type_jump:
return &clone_jump(state, nir_instr_as_jump(instr))->instr;
case nir_instr_type_call:
return &clone_call(state, nir_instr_as_call(instr))->instr;
case nir_instr_type_parallel_copy:
unreachable("Cannot clone parallel copies");
default:
unreachable("bad instr type");
return NULL;
}
}
static nir_block *
clone_block(clone_state *state, struct exec_list *cf_list, const nir_block *blk)
{
/* Don't actually create a new block. Just use the one from the tail of
* the list. NIR guarantees that the tail of the list is a block and that
* no two blocks are side-by-side in the IR; It should be empty.
*/
nir_block *nblk =
exec_node_data(nir_block, exec_list_get_tail(cf_list), cf_node.node);
assert(nblk->cf_node.type == nir_cf_node_block);
assert(exec_list_is_empty(&nblk->instr_list));
/* We need this for phi sources */
store_ptr(state, nblk, blk);
nir_foreach_instr(blk, instr) {
if (instr->type == nir_instr_type_phi) {
/* Phi instructions are a bit of a special case when cloning because
* we don't want inserting the instruction to automatically handle
* use/defs for us. Instead, we need to wait until all the
* blocks/instructions are in so that we can set their sources up.
*/
clone_phi(state, nir_instr_as_phi(instr), nblk);
} else {
nir_instr *ninstr = clone_instr(state, instr);
nir_instr_insert_after_block(nblk, ninstr);
}
}
return nblk;
}
static void
clone_cf_list(clone_state *state, struct exec_list *dst,
const struct exec_list *list);
static nir_if *
clone_if(clone_state *state, struct exec_list *cf_list, const nir_if *i)
{
nir_if *ni = nir_if_create(state->ns);
__clone_src(state, ni, &ni->condition, &i->condition);
nir_cf_node_insert_end(cf_list, &ni->cf_node);
clone_cf_list(state, &ni->then_list, &i->then_list);
clone_cf_list(state, &ni->else_list, &i->else_list);
return ni;
}
static nir_loop *
clone_loop(clone_state *state, struct exec_list *cf_list, const nir_loop *loop)
{
nir_loop *nloop = nir_loop_create(state->ns);
nir_cf_node_insert_end(cf_list, &nloop->cf_node);
clone_cf_list(state, &nloop->body, &loop->body);
return nloop;
}
/* clone list of nir_cf_node: */
static void
clone_cf_list(clone_state *state, struct exec_list *dst,
const struct exec_list *list)
{
foreach_list_typed(nir_cf_node, cf, node, list) {
switch (cf->type) {
case nir_cf_node_block:
clone_block(state, dst, nir_cf_node_as_block(cf));
break;
case nir_cf_node_if:
clone_if(state, dst, nir_cf_node_as_if(cf));
break;
case nir_cf_node_loop:
clone_loop(state, dst, nir_cf_node_as_loop(cf));
break;
default:
unreachable("bad cf type");
}
}
}
static nir_function_impl *
clone_function_impl(clone_state *state, const nir_function_impl *fi,
nir_function *nfxn)
{
nir_function_impl *nfi = nir_function_impl_create(nfxn);
clone_var_list(state, &nfi->locals, &fi->locals);
clone_reg_list(state, &nfi->registers, &fi->registers);
nfi->reg_alloc = fi->reg_alloc;
nfi->num_params = fi->num_params;
nfi->params = ralloc_array(state->ns, nir_variable *, fi->num_params);
for (unsigned i = 0; i < fi->num_params; i++) {
nfi->params[i] = lookup_ptr(state, fi->params[i]);
}
nfi->return_var = lookup_ptr(state, fi->return_var);
assert(list_empty(&state->phi_srcs));
clone_cf_list(state, &nfi->body, &fi->body);
/* After we've cloned almost everything, we have to walk the list of phi
* sources and fix them up. Thanks to loops, the block and SSA value for a
* phi source may not be defined when we first encounter it. Instead, we
* add it to the phi_srcs list and we fix it up here.
*/
list_for_each_entry_safe(nir_phi_src, src, &state->phi_srcs, src.use_link) {
src->pred = lookup_ptr(state, src->pred);
assert(src->src.is_ssa);
src->src.ssa = lookup_ptr(state, src->src.ssa);
/* Remove from this list and place in the uses of the SSA def */
list_del(&src->src.use_link);
list_addtail(&src->src.use_link, &src->src.ssa->uses);
}
assert(list_empty(&state->phi_srcs));
/* All metadata is invalidated in the cloning process */
nfi->valid_metadata = 0;
return nfi;
}
static nir_function *
clone_function(clone_state *state, const nir_function *fxn, nir_shader *ns)
{
assert(ns == state->ns);
nir_function *nfxn = nir_function_create(ns, fxn->name);
/* Needed for call instructions */
store_ptr(state, nfxn, fxn);
nfxn->num_params = fxn->num_params;
nfxn->params = ralloc_array(state->ns, nir_parameter, fxn->num_params);
memcpy(nfxn->params, fxn->params, sizeof(nir_parameter) * fxn->num_params);
nfxn->return_type = fxn->return_type;
/* At first glance, it looks like we should clone the function_impl here.
* However, call instructions need to be able to reference at least the
* function and those will get processed as we clone the function_impl's.
* We stop here and do function_impls as a second pass.
*/
return nfxn;
}
nir_shader *
nir_shader_clone(void *mem_ctx, const nir_shader *s)
{
clone_state state;
init_clone_state(&state);
nir_shader *ns = nir_shader_create(mem_ctx, s->stage, s->options);
state.ns = ns;
clone_var_list(&state, &ns->uniforms, &s->uniforms);
clone_var_list(&state, &ns->inputs, &s->inputs);
clone_var_list(&state, &ns->outputs, &s->outputs);
clone_var_list(&state, &ns->globals, &s->globals);
clone_var_list(&state, &ns->system_values, &s->system_values);
/* Go through and clone functions */
foreach_list_typed(nir_function, fxn, node, &s->functions)
clone_function(&state, fxn, ns);
/* Only after all functions are cloned can we clone the actual function
* implementations. This is because nir_call_instr's need to reference the
* functions of other functions and we don't know what order the functions
* will have in the list.
*/
nir_foreach_function(s, fxn) {
nir_function *nfxn = lookup_ptr(&state, fxn);
clone_function_impl(&state, fxn->impl, nfxn);
}
clone_reg_list(&state, &ns->registers, &s->registers);
ns->reg_alloc = s->reg_alloc;
ns->info = s->info;
ns->info.name = ralloc_strdup(ns, ns->info.name);
if (ns->info.label)
ns->info.label = ralloc_strdup(ns, ns->info.label);
ns->num_inputs = s->num_inputs;
ns->num_uniforms = s->num_uniforms;
ns->num_outputs = s->num_outputs;
free_clone_state(&state);
return ns;
}

View File

@@ -1,808 +0,0 @@
/*
* Copyright © 2014 Intel Corporation
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*
* Authors:
* Connor Abbott (cwabbott0@gmail.com)
*
*/
#include "nir_control_flow_private.h"
/**
* \name Control flow modification
*
* These functions modify the control flow tree while keeping the control flow
* graph up-to-date. The invariants respected are:
* 1. Each then statement, else statement, or loop body must have at least one
* control flow node.
* 2. Each if-statement and loop must have one basic block before it and one
* after.
* 3. Two basic blocks cannot be directly next to each other.
* 4. If a basic block has a jump instruction, there must be only one and it
* must be at the end of the block.
* 5. The CFG must always be connected - this means that we must insert a fake
* CFG edge for loops with no break statement.
*
* The purpose of the second one is so that we have places to insert code during
* GCM, as well as eliminating the possibility of critical edges.
*/
/*@{*/
static bool
block_ends_in_jump(nir_block *block)
{
return !exec_list_is_empty(&block->instr_list) &&
nir_block_last_instr(block)->type == nir_instr_type_jump;
}
static inline void
block_add_pred(nir_block *block, nir_block *pred)
{
_mesa_set_add(block->predecessors, pred);
}
static inline void
block_remove_pred(nir_block *block, nir_block *pred)
{
struct set_entry *entry = _mesa_set_search(block->predecessors, pred);
assert(entry);
_mesa_set_remove(block->predecessors, entry);
}
static void
link_blocks(nir_block *pred, nir_block *succ1, nir_block *succ2)
{
pred->successors[0] = succ1;
if (succ1 != NULL)
block_add_pred(succ1, pred);
pred->successors[1] = succ2;
if (succ2 != NULL)
block_add_pred(succ2, pred);
}
static void
unlink_blocks(nir_block *pred, nir_block *succ)
{
if (pred->successors[0] == succ) {
pred->successors[0] = pred->successors[1];
pred->successors[1] = NULL;
} else {
assert(pred->successors[1] == succ);
pred->successors[1] = NULL;
}
block_remove_pred(succ, pred);
}
static void
unlink_block_successors(nir_block *block)
{
if (block->successors[1] != NULL)
unlink_blocks(block, block->successors[1]);
if (block->successors[0] != NULL)
unlink_blocks(block, block->successors[0]);
}
static void
link_non_block_to_block(nir_cf_node *node, nir_block *block)
{
if (node->type == nir_cf_node_if) {
/*
* We're trying to link an if to a block after it; this just means linking
* the last block of the then and else branches.
*/
nir_if *if_stmt = nir_cf_node_as_if(node);
nir_cf_node *last_then = nir_if_last_then_node(if_stmt);
assert(last_then->type == nir_cf_node_block);
nir_block *last_then_block = nir_cf_node_as_block(last_then);
nir_cf_node *last_else = nir_if_last_else_node(if_stmt);
assert(last_else->type == nir_cf_node_block);
nir_block *last_else_block = nir_cf_node_as_block(last_else);
if (!block_ends_in_jump(last_then_block)) {
unlink_block_successors(last_then_block);
link_blocks(last_then_block, block, NULL);
}
if (!block_ends_in_jump(last_else_block)) {
unlink_block_successors(last_else_block);
link_blocks(last_else_block, block, NULL);
}
} else {
assert(node->type == nir_cf_node_loop);
/*
* We can only get to this codepath if we're inserting a new loop, or
* at least a loop with no break statements; we can't insert break
* statements into a loop when we haven't inserted it into the CFG
* because we wouldn't know which block comes after the loop
* and therefore, which block should be the successor of the block with
* the break). Therefore, we need to insert a fake edge (see invariant
* #5).
*/
nir_loop *loop = nir_cf_node_as_loop(node);
nir_cf_node *last = nir_loop_last_cf_node(loop);
assert(last->type == nir_cf_node_block);
nir_block *last_block = nir_cf_node_as_block(last);
last_block->successors[1] = block;
block_add_pred(block, last_block);
}
}
static void
link_block_to_non_block(nir_block *block, nir_cf_node *node)
{
if (node->type == nir_cf_node_if) {
/*
* We're trying to link a block to an if after it; this just means linking
* the block to the first block of the then and else branches.
*/
nir_if *if_stmt = nir_cf_node_as_if(node);
nir_cf_node *first_then = nir_if_first_then_node(if_stmt);
assert(first_then->type == nir_cf_node_block);
nir_block *first_then_block = nir_cf_node_as_block(first_then);
nir_cf_node *first_else = nir_if_first_else_node(if_stmt);
assert(first_else->type == nir_cf_node_block);
nir_block *first_else_block = nir_cf_node_as_block(first_else);
unlink_block_successors(block);
link_blocks(block, first_then_block, first_else_block);
} else {
/*
* For similar reasons as the corresponding case in
* link_non_block_to_block(), don't worry about if the loop header has
* any predecessors that need to be unlinked.
*/
assert(node->type == nir_cf_node_loop);
nir_loop *loop = nir_cf_node_as_loop(node);
nir_cf_node *loop_header = nir_loop_first_cf_node(loop);
assert(loop_header->type == nir_cf_node_block);
nir_block *loop_header_block = nir_cf_node_as_block(loop_header);
unlink_block_successors(block);
link_blocks(block, loop_header_block, NULL);
}
}
/**
* Replace a block's successor with a different one.
*/
static void
replace_successor(nir_block *block, nir_block *old_succ, nir_block *new_succ)
{
if (block->successors[0] == old_succ) {
block->successors[0] = new_succ;
} else {
assert(block->successors[1] == old_succ);
block->successors[1] = new_succ;
}
block_remove_pred(old_succ, block);
block_add_pred(new_succ, block);
}
/**
* Takes a basic block and inserts a new empty basic block before it, making its
* predecessors point to the new block. This essentially splits the block into
* an empty header and a body so that another non-block CF node can be inserted
* between the two. Note that this does *not* link the two basic blocks, so
* some kind of cleanup *must* be performed after this call.
*/
static nir_block *
split_block_beginning(nir_block *block)
{
nir_block *new_block = nir_block_create(ralloc_parent(block));
new_block->cf_node.parent = block->cf_node.parent;
exec_node_insert_node_before(&block->cf_node.node, &new_block->cf_node.node);
struct set_entry *entry;
set_foreach(block->predecessors, entry) {
nir_block *pred = (nir_block *) entry->key;
replace_successor(pred, block, new_block);
}
/* Any phi nodes must stay part of the new block, or else their
* sourcse will be messed up. This will reverse the order of the phi's, but
* order shouldn't matter.
*/
nir_foreach_instr_safe(block, instr) {
if (instr->type != nir_instr_type_phi)
break;
exec_node_remove(&instr->node);
instr->block = new_block;
exec_list_push_head(&new_block->instr_list, &instr->node);
}
return new_block;
}
static void
rewrite_phi_preds(nir_block *block, nir_block *old_pred, nir_block *new_pred)
{
nir_foreach_instr_safe(block, instr) {
if (instr->type != nir_instr_type_phi)
break;
nir_phi_instr *phi = nir_instr_as_phi(instr);
nir_foreach_phi_src(phi, src) {
if (src->pred == old_pred) {
src->pred = new_pred;
break;
}
}
}
}
static void
insert_phi_undef(nir_block *block, nir_block *pred)
{
nir_function_impl *impl = nir_cf_node_get_function(&block->cf_node);
nir_foreach_instr(block, instr) {
if (instr->type != nir_instr_type_phi)
break;
nir_phi_instr *phi = nir_instr_as_phi(instr);
nir_ssa_undef_instr *undef =
nir_ssa_undef_instr_create(ralloc_parent(phi),
phi->dest.ssa.num_components);
nir_instr_insert_before_cf_list(&impl->body, &undef->instr);
nir_phi_src *src = ralloc(phi, nir_phi_src);
src->pred = pred;
src->src.parent_instr = &phi->instr;
src->src.is_ssa = true;
src->src.ssa = &undef->def;
list_addtail(&src->src.use_link, &undef->def.uses);
exec_list_push_tail(&phi->srcs, &src->node);
}
}
/**
* Moves the successors of source to the successors of dest, leaving both
* successors of source NULL.
*/
static void
move_successors(nir_block *source, nir_block *dest)
{
nir_block *succ1 = source->successors[0];
nir_block *succ2 = source->successors[1];
if (succ1) {
unlink_blocks(source, succ1);
rewrite_phi_preds(succ1, source, dest);
}
if (succ2) {
unlink_blocks(source, succ2);
rewrite_phi_preds(succ2, source, dest);
}
unlink_block_successors(dest);
link_blocks(dest, succ1, succ2);
}
/* Given a basic block with no successors that has been inserted into the
* control flow tree, gives it the successors it would normally have assuming
* it doesn't end in a jump instruction. Also inserts phi sources with undefs
* if necessary.
*/
static void
block_add_normal_succs(nir_block *block)
{
if (exec_node_is_tail_sentinel(block->cf_node.node.next)) {
nir_cf_node *parent = block->cf_node.parent;
if (parent->type == nir_cf_node_if) {
nir_cf_node *next = nir_cf_node_next(parent);
assert(next->type == nir_cf_node_block);
nir_block *next_block = nir_cf_node_as_block(next);
link_blocks(block, next_block, NULL);
} else {
assert(parent->type == nir_cf_node_loop);
nir_loop *loop = nir_cf_node_as_loop(parent);
nir_cf_node *head = nir_loop_first_cf_node(loop);
assert(head->type == nir_cf_node_block);
nir_block *head_block = nir_cf_node_as_block(head);
link_blocks(block, head_block, NULL);
insert_phi_undef(head_block, block);
}
} else {
nir_cf_node *next = nir_cf_node_next(&block->cf_node);
if (next->type == nir_cf_node_if) {
nir_if *next_if = nir_cf_node_as_if(next);
nir_cf_node *first_then = nir_if_first_then_node(next_if);
assert(first_then->type == nir_cf_node_block);
nir_block *first_then_block = nir_cf_node_as_block(first_then);
nir_cf_node *first_else = nir_if_first_else_node(next_if);
assert(first_else->type == nir_cf_node_block);
nir_block *first_else_block = nir_cf_node_as_block(first_else);
link_blocks(block, first_then_block, first_else_block);
} else {
assert(next->type == nir_cf_node_loop);
nir_loop *next_loop = nir_cf_node_as_loop(next);
nir_cf_node *first = nir_loop_first_cf_node(next_loop);
assert(first->type == nir_cf_node_block);
nir_block *first_block = nir_cf_node_as_block(first);
link_blocks(block, first_block, NULL);
insert_phi_undef(first_block, block);
}
}
}
static nir_block *
split_block_end(nir_block *block)
{
nir_block *new_block = nir_block_create(ralloc_parent(block));
new_block->cf_node.parent = block->cf_node.parent;
exec_node_insert_after(&block->cf_node.node, &new_block->cf_node.node);
if (block_ends_in_jump(block)) {
/* Figure out what successor block would've had if it didn't have a jump
* instruction, and make new_block have that successor.
*/
block_add_normal_succs(new_block);
} else {
move_successors(block, new_block);
}
return new_block;
}
static nir_block *
split_block_before_instr(nir_instr *instr)
{
assert(instr->type != nir_instr_type_phi);
nir_block *new_block = split_block_beginning(instr->block);
nir_foreach_instr_safe(instr->block, cur_instr) {
if (cur_instr == instr)
break;
exec_node_remove(&cur_instr->node);
cur_instr->block = new_block;
exec_list_push_tail(&new_block->instr_list, &cur_instr->node);
}
return new_block;
}
/* Splits a basic block at the point specified by the cursor. The "before" and
* "after" arguments are filled out with the blocks resulting from the split
* if non-NULL. Note that the "beginning" of the block is actually interpreted
* as before the first non-phi instruction, and it's illegal to split a block
* before a phi instruction.
*/
static void
split_block_cursor(nir_cursor cursor,
nir_block **_before, nir_block **_after)
{
nir_block *before, *after;
switch (cursor.option) {
case nir_cursor_before_block:
after = cursor.block;
before = split_block_beginning(cursor.block);
break;
case nir_cursor_after_block:
before = cursor.block;
after = split_block_end(cursor.block);
break;
case nir_cursor_before_instr:
after = cursor.instr->block;
before = split_block_before_instr(cursor.instr);
break;
case nir_cursor_after_instr:
/* We lower this to split_block_before_instr() so that we can keep the
* after-a-jump-instr case contained to split_block_end().
*/
if (nir_instr_is_last(cursor.instr)) {
before = cursor.instr->block;
after = split_block_end(cursor.instr->block);
} else {
after = cursor.instr->block;
before = split_block_before_instr(nir_instr_next(cursor.instr));
}
break;
default:
unreachable("not reached");
}
if (_before)
*_before = before;
if (_after)
*_after = after;
}
/**
* Inserts a non-basic block between two basic blocks and links them together.
*/
static void
insert_non_block(nir_block *before, nir_cf_node *node, nir_block *after)
{
node->parent = before->cf_node.parent;
exec_node_insert_after(&before->cf_node.node, &node->node);
link_block_to_non_block(before, node);
link_non_block_to_block(node, after);
}
/* walk up the control flow tree to find the innermost enclosed loop */
static nir_loop *
nearest_loop(nir_cf_node *node)
{
while (node->type != nir_cf_node_loop) {
node = node->parent;
}
return nir_cf_node_as_loop(node);
}
/*
* update the CFG after a jump instruction has been added to the end of a block
*/
void
nir_handle_add_jump(nir_block *block)
{
nir_instr *instr = nir_block_last_instr(block);
nir_jump_instr *jump_instr = nir_instr_as_jump(instr);
unlink_block_successors(block);
nir_function_impl *impl = nir_cf_node_get_function(&block->cf_node);
nir_metadata_preserve(impl, nir_metadata_none);
if (jump_instr->type == nir_jump_break ||
jump_instr->type == nir_jump_continue) {
nir_loop *loop = nearest_loop(&block->cf_node);
if (jump_instr->type == nir_jump_continue) {
nir_cf_node *first_node = nir_loop_first_cf_node(loop);
assert(first_node->type == nir_cf_node_block);
nir_block *first_block = nir_cf_node_as_block(first_node);
link_blocks(block, first_block, NULL);
} else {
nir_cf_node *after = nir_cf_node_next(&loop->cf_node);
assert(after->type == nir_cf_node_block);
nir_block *after_block = nir_cf_node_as_block(after);
link_blocks(block, after_block, NULL);
/* If we inserted a fake link, remove it */
nir_cf_node *last = nir_loop_last_cf_node(loop);
assert(last->type == nir_cf_node_block);
nir_block *last_block = nir_cf_node_as_block(last);
if (last_block->successors[1] != NULL)
unlink_blocks(last_block, after_block);
}
} else {
assert(jump_instr->type == nir_jump_return);
link_blocks(block, impl->end_block, NULL);
}
}
static void
remove_phi_src(nir_block *block, nir_block *pred)
{
nir_foreach_instr(block, instr) {
if (instr->type != nir_instr_type_phi)
break;
nir_phi_instr *phi = nir_instr_as_phi(instr);
nir_foreach_phi_src_safe(phi, src) {
if (src->pred == pred) {
list_del(&src->src.use_link);
exec_node_remove(&src->node);
}
}
}
}
/* Removes the successor of a block with a jump, and inserts a fake edge for
* infinite loops. Note that the jump to be eliminated may be free-floating.
*/
static void
unlink_jump(nir_block *block, nir_jump_type type, bool add_normal_successors)
{
nir_block *next = block->successors[0];
if (block->successors[0])
remove_phi_src(block->successors[0], block);
if (block->successors[1])
remove_phi_src(block->successors[1], block);
unlink_block_successors(block);
if (add_normal_successors)
block_add_normal_succs(block);
/* If we've just removed a break, and the block we were jumping to (after
* the loop) now has zero predecessors, we've created a new infinite loop.
*
* NIR doesn't allow blocks (other than the start block) to have zero
* predecessors. In particular, dominance assumes all blocks are reachable.
* So, we insert a "fake link" by making successors[1] point after the loop.
*
* Note that we have to do this after unlinking/recreating the block's
* successors. If we removed a "break" at the end of the loop, then
* block == last_block, so block->successors[0] would already be "next",
* and adding a fake link would create two identical successors. Doing
* this afterward works, as we'll have changed block->successors[0] to
* be the top of the loop.
*/
if (type == nir_jump_break && next->predecessors->entries == 0) {
nir_loop *loop =
nir_cf_node_as_loop(nir_cf_node_prev(&next->cf_node));
/* insert fake link */
nir_cf_node *last = nir_loop_last_cf_node(loop);
assert(last->type == nir_cf_node_block);
nir_block *last_block = nir_cf_node_as_block(last);
last_block->successors[1] = next;
block_add_pred(next, last_block);
}
}
void
nir_handle_remove_jump(nir_block *block, nir_jump_type type)
{
unlink_jump(block, type, true);
nir_function_impl *impl = nir_cf_node_get_function(&block->cf_node);
nir_metadata_preserve(impl, nir_metadata_none);
}
static void
update_if_uses(nir_cf_node *node)
{
if (node->type != nir_cf_node_if)
return;
nir_if *if_stmt = nir_cf_node_as_if(node);
if_stmt->condition.parent_if = if_stmt;
if (if_stmt->condition.is_ssa) {
list_addtail(&if_stmt->condition.use_link,
&if_stmt->condition.ssa->if_uses);
} else {
list_addtail(&if_stmt->condition.use_link,
&if_stmt->condition.reg.reg->if_uses);
}
}
/**
* Stitch two basic blocks together into one. The aggregate must have the same
* predecessors as the first and the same successors as the second.
*/
static void
stitch_blocks(nir_block *before, nir_block *after)
{
/*
* We move after into before, so we have to deal with up to 2 successors vs.
* possibly a large number of predecessors.
*
* TODO: special case when before is empty and after isn't?
*/
if (block_ends_in_jump(before)) {
assert(exec_list_is_empty(&after->instr_list));
if (after->successors[0])
remove_phi_src(after->successors[0], after);
if (after->successors[1])
remove_phi_src(after->successors[1], after);
unlink_block_successors(after);
exec_node_remove(&after->cf_node.node);
} else {
move_successors(after, before);
foreach_list_typed(nir_instr, instr, node, &after->instr_list) {
instr->block = before;
}
exec_list_append(&before->instr_list, &after->instr_list);
exec_node_remove(&after->cf_node.node);
}
}
void
nir_cf_node_insert(nir_cursor cursor, nir_cf_node *node)
{
nir_block *before, *after;
split_block_cursor(cursor, &before, &after);
if (node->type == nir_cf_node_block) {
nir_block *block = nir_cf_node_as_block(node);
exec_node_insert_after(&before->cf_node.node, &block->cf_node.node);
block->cf_node.parent = before->cf_node.parent;
/* stitch_blocks() assumes that any block that ends with a jump has
* already been setup with the correct successors, so we need to set
* up jumps here as the block is being inserted.
*/
if (block_ends_in_jump(block))
nir_handle_add_jump(block);
stitch_blocks(block, after);
stitch_blocks(before, block);
} else {
update_if_uses(node);
insert_non_block(before, node, after);
}
}
static bool
replace_ssa_def_uses(nir_ssa_def *def, void *void_impl)
{
nir_function_impl *impl = void_impl;
void *mem_ctx = ralloc_parent(impl);
nir_ssa_undef_instr *undef =
nir_ssa_undef_instr_create(mem_ctx, def->num_components);
nir_instr_insert_before_cf_list(&impl->body, &undef->instr);
nir_ssa_def_rewrite_uses(def, nir_src_for_ssa(&undef->def));
return true;
}
static void
cleanup_cf_node(nir_cf_node *node, nir_function_impl *impl)
{
switch (node->type) {
case nir_cf_node_block: {
nir_block *block = nir_cf_node_as_block(node);
/* We need to walk the instructions and clean up defs/uses */
nir_foreach_instr_safe(block, instr) {
if (instr->type == nir_instr_type_jump) {
nir_jump_type jump_type = nir_instr_as_jump(instr)->type;
unlink_jump(block, jump_type, false);
} else {
nir_foreach_ssa_def(instr, replace_ssa_def_uses, impl);
nir_instr_remove(instr);
}
}
break;
}
case nir_cf_node_if: {
nir_if *if_stmt = nir_cf_node_as_if(node);
foreach_list_typed(nir_cf_node, child, node, &if_stmt->then_list)
cleanup_cf_node(child, impl);
foreach_list_typed(nir_cf_node, child, node, &if_stmt->else_list)
cleanup_cf_node(child, impl);
list_del(&if_stmt->condition.use_link);
break;
}
case nir_cf_node_loop: {
nir_loop *loop = nir_cf_node_as_loop(node);
foreach_list_typed(nir_cf_node, child, node, &loop->body)
cleanup_cf_node(child, impl);
break;
}
case nir_cf_node_function: {
nir_function_impl *impl = nir_cf_node_as_function(node);
foreach_list_typed(nir_cf_node, child, node, &impl->body)
cleanup_cf_node(child, impl);
break;
}
default:
unreachable("Invalid CF node type");
}
}
void
nir_cf_extract(nir_cf_list *extracted, nir_cursor begin, nir_cursor end)
{
nir_block *block_begin, *block_end, *block_before, *block_after;
/* In the case where begin points to an instruction in some basic block and
* end points to the end of the same basic block, we rely on the fact that
* splitting on an instruction moves earlier instructions into a new basic
* block. If the later instructions were moved instead, then the end cursor
* would be pointing to the same place that begin used to point to, which
* is obviously not what we want.
*/
split_block_cursor(begin, &block_before, &block_begin);
split_block_cursor(end, &block_end, &block_after);
extracted->impl = nir_cf_node_get_function(&block_begin->cf_node);
exec_list_make_empty(&extracted->list);
/* Dominance and other block-related information is toast. */
nir_metadata_preserve(extracted->impl, nir_metadata_none);
nir_cf_node *cf_node = &block_begin->cf_node;
nir_cf_node *cf_node_end = &block_end->cf_node;
while (true) {
nir_cf_node *next = nir_cf_node_next(cf_node);
exec_node_remove(&cf_node->node);
cf_node->parent = NULL;
exec_list_push_tail(&extracted->list, &cf_node->node);
if (cf_node == cf_node_end)
break;
cf_node = next;
}
stitch_blocks(block_before, block_after);
}
void
nir_cf_reinsert(nir_cf_list *cf_list, nir_cursor cursor)
{
nir_block *before, *after;
split_block_cursor(cursor, &before, &after);
foreach_list_typed_safe(nir_cf_node, node, node, &cf_list->list) {
exec_node_remove(&node->node);
node->parent = before->cf_node.parent;
exec_node_insert_node_before(&after->cf_node.node, &node->node);
}
stitch_blocks(before,
nir_cf_node_as_block(nir_cf_node_next(&before->cf_node)));
stitch_blocks(nir_cf_node_as_block(nir_cf_node_prev(&after->cf_node)),
after);
}
void
nir_cf_delete(nir_cf_list *cf_list)
{
foreach_list_typed(nir_cf_node, node, node, &cf_list->list) {
cleanup_cf_node(node, cf_list->impl);
}
}

View File

@@ -1,162 +0,0 @@
/*
* Copyright © 2014 Intel Corporation
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*
* Authors:
* Connor Abbott (cwabbott0@gmail.com)
*
*/
#include "nir.h"
#pragma once
#ifdef __cplusplus
extern "C" {
#endif
/** NIR Control Flow Modification
*
* This file contains various API's that make modifying control flow in NIR,
* while maintaining the invariants checked by the validator, much easier.
* There are two parts to this:
*
* 1. Inserting control flow (if's and loops) in various places, for creating
* IR either from scratch or as part of some lowering pass.
* 2. Taking existing pieces of the IR and either moving them around or
* deleting them.
*/
/** Control flow insertion. */
/** puts a control flow node where the cursor is */
void nir_cf_node_insert(nir_cursor cursor, nir_cf_node *node);
/** puts a control flow node immediately after another control flow node */
static inline void
nir_cf_node_insert_after(nir_cf_node *node, nir_cf_node *after)
{
nir_cf_node_insert(nir_after_cf_node(node), after);
}
/** puts a control flow node immediately before another control flow node */
static inline void
nir_cf_node_insert_before(nir_cf_node *node, nir_cf_node *before)
{
nir_cf_node_insert(nir_before_cf_node(node), before);
}
/** puts a control flow node at the beginning of a list from an if, loop, or function */
static inline void
nir_cf_node_insert_begin(struct exec_list *list, nir_cf_node *node)
{
nir_cf_node_insert(nir_before_cf_list(list), node);
}
/** puts a control flow node at the end of a list from an if, loop, or function */
static inline void
nir_cf_node_insert_end(struct exec_list *list, nir_cf_node *node)
{
nir_cf_node_insert(nir_after_cf_list(list), node);
}
/** Control flow motion.
*
* These functions let you take a part of a control flow list (basically
* equivalent to a series of statement in GLSL) and "extract" it from the IR,
* so that it's a free-floating piece of IR that can be either re-inserted
* somewhere else or deleted entirely. A few notes on using it:
*
* 1. Phi nodes are considered attached to the piece of control flow that
* their sources come from. There are three places where phi nodes can
* occur, which are the three places where a block can have multiple
* predecessors:
*
* 1) After an if statement, if neither branch ends in a jump.
* 2) After a loop, if there are multiple break's.
* 3) At the beginning of a loop.
*
* For #1, the phi node is considered to be part of the if, and for #2 and
* #3 the phi node is considered to be part of the loop. This allows us to
* keep phi's intact, but it means that phi nodes cannot be separated from
* the control flow they come from. For example, extracting an if without
* extracting all the phi nodes after it is not allowed, and neither is
* extracting only some of the phi nodes at the beginning of a block. It
* also means that extracting from the beginning of a basic block actually
* means extracting from the first non-phi instruction, since there's no
* situation where extracting phi nodes without extracting what comes
* before them makes any sense.
*
* 2. Phi node sources are guaranteed to remain valid, meaning that they still
* correspond one-to-one with the predecessors of the basic block they're
* part of. In addition, the original sources will be preserved unless they
* correspond to a break or continue that was deleted. However, no attempt
* is made to ensure that SSA form is maintained. In particular, it is
* *not* guaranteed that definitions of SSA values will dominate all their
* uses after all is said and done. Either the caller must ensure that this
* is the case, or it must insert extra phi nodes to restore SSA.
*
* 3. It is invalid to move a piece of IR with a break/continue outside of the
* loop it references. Doing this will result in invalid
* successors/predecessors and phi node sources.
*
* 4. It is invalid to move a piece of IR from one function implementation to
* another.
*
* 5. Extracting a control flow list will leave lots of dangling references to
* and from other pieces of the IR. It also leaves things in a not 100%
* consistent state. This means that some things (e.g. inserting
* instructions) might not work reliably on the extracted control flow. It
* also means that extracting control flow without re-inserting it or
* deleting it is a Bad Thing (tm).
*/
typedef struct {
struct exec_list list;
nir_function_impl *impl; /* for cleaning up if the list is deleted */
} nir_cf_list;
void nir_cf_extract(nir_cf_list *extracted, nir_cursor begin, nir_cursor end);
void nir_cf_reinsert(nir_cf_list *cf_list, nir_cursor cursor);
void nir_cf_delete(nir_cf_list *cf_list);
static inline void
nir_cf_list_extract(nir_cf_list *extracted, struct exec_list *cf_list)
{
nir_cf_extract(extracted, nir_before_cf_list(cf_list),
nir_after_cf_list(cf_list));
}
/** removes a control flow node, doing any cleanup necessary */
static inline void
nir_cf_node_remove(nir_cf_node *node)
{
nir_cf_list list;
nir_cf_extract(&list, nir_before_cf_node(node), nir_after_cf_node(node));
nir_cf_delete(&list);
}
#ifdef __cplusplus
}
#endif

View File

@@ -1,93 +0,0 @@
/*
* Copyright © 2015 Intel Corporation
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*/
#include "nir.h"
#include "nir_builder.h"
static nir_intrinsic_instr *
as_intrinsic(nir_instr *instr, nir_intrinsic_op op)
{
if (instr->type != nir_instr_type_intrinsic)
return NULL;
nir_intrinsic_instr *intrin = nir_instr_as_intrinsic(instr);
if (intrin->intrinsic != op)
return NULL;
return intrin;
}
static nir_intrinsic_instr *
as_set_vertex_count(nir_instr *instr)
{
return as_intrinsic(instr, nir_intrinsic_set_vertex_count);
}
/**
* If a geometry shader emits a constant number of vertices, return the
* number of vertices. Otherwise, return -1 (unknown).
*
* This only works if you've used nir_lower_gs_intrinsics() to do vertex
* counting at the NIR level.
*/
int
nir_gs_count_vertices(const nir_shader *shader)
{
int count = -1;
nir_foreach_function(shader, function) {
if (!function->impl)
continue;
/* set_vertex_count intrinsics only appear in predecessors of the
* end block. So we don't need to walk all of them.
*/
struct set_entry *entry;
set_foreach(function->impl->end_block->predecessors, entry) {
nir_block *block = (nir_block *) entry->key;
nir_foreach_instr_reverse(block, instr) {
nir_intrinsic_instr *intrin = as_set_vertex_count(instr);
if (!intrin)
continue;
nir_const_value *val = nir_src_as_const_value(intrin->src[0]);
/* We've found a non-constant value. Bail. */
if (!val)
return -1;
if (count == -1)
count = val->i[0];
/* We've found contradictory set_vertex_count intrinsics.
* This can happen if there are early-returns in main() and
* different paths emit different numbers of vertices.
*/
if (count != val->i[0])
return -1;
}
}
}
return count;
}

View File

@@ -1,519 +0,0 @@
/*
* Copyright © 2014 Connor Abbott
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*/
#include "nir_instr_set.h"
#include "nir_vla.h"
#define HASH(hash, data) _mesa_fnv32_1a_accumulate((hash), (data))
static uint32_t
hash_src(uint32_t hash, const nir_src *src)
{
assert(src->is_ssa);
hash = HASH(hash, src->ssa);
return hash;
}
static uint32_t
hash_alu_src(uint32_t hash, const nir_alu_src *src, unsigned num_components)
{
hash = HASH(hash, src->abs);
hash = HASH(hash, src->negate);
for (unsigned i = 0; i < num_components; i++)
hash = HASH(hash, src->swizzle[i]);
hash = hash_src(hash, &src->src);
return hash;
}
static uint32_t
hash_alu(uint32_t hash, const nir_alu_instr *instr)
{
hash = HASH(hash, instr->op);
hash = HASH(hash, instr->dest.dest.ssa.num_components);
if (nir_op_infos[instr->op].algebraic_properties & NIR_OP_IS_COMMUTATIVE) {
assert(nir_op_infos[instr->op].num_inputs == 2);
uint32_t hash0 = hash_alu_src(hash, &instr->src[0],
nir_ssa_alu_instr_src_components(instr, 0));
uint32_t hash1 = hash_alu_src(hash, &instr->src[1],
nir_ssa_alu_instr_src_components(instr, 1));
/* For commutative operations, we need some commutative way of
* combining the hashes. One option would be to XOR them but that
* means that anything with two identical sources will hash to 0 and
* that's common enough we probably don't want the guaranteed
* collision. Either addition or multiplication will also work.
*/
hash = hash0 * hash1;
} else {
for (unsigned i = 0; i < nir_op_infos[instr->op].num_inputs; i++) {
hash = hash_alu_src(hash, &instr->src[i],
nir_ssa_alu_instr_src_components(instr, i));
}
}
return hash;
}
static uint32_t
hash_load_const(uint32_t hash, const nir_load_const_instr *instr)
{
hash = HASH(hash, instr->def.num_components);
hash = _mesa_fnv32_1a_accumulate_block(hash, instr->value.f,
instr->def.num_components
* sizeof(instr->value.f[0]));
return hash;
}
static int
cmp_phi_src(const void *data1, const void *data2)
{
nir_phi_src *src1 = *(nir_phi_src **)data1;
nir_phi_src *src2 = *(nir_phi_src **)data2;
return src1->pred - src2->pred;
}
static uint32_t
hash_phi(uint32_t hash, const nir_phi_instr *instr)
{
hash = HASH(hash, instr->instr.block);
/* sort sources by predecessor, since the order shouldn't matter */
unsigned num_preds = instr->instr.block->predecessors->entries;
NIR_VLA(nir_phi_src *, srcs, num_preds);
unsigned i = 0;
nir_foreach_phi_src(instr, src) {
srcs[i++] = src;
}
qsort(srcs, num_preds, sizeof(nir_phi_src *), cmp_phi_src);
for (i = 0; i < num_preds; i++) {
hash = hash_src(hash, &srcs[i]->src);
hash = HASH(hash, srcs[i]->pred);
}
return hash;
}
static uint32_t
hash_intrinsic(uint32_t hash, const nir_intrinsic_instr *instr)
{
const nir_intrinsic_info *info = &nir_intrinsic_infos[instr->intrinsic];
hash = HASH(hash, instr->intrinsic);
if (info->has_dest)
hash = HASH(hash, instr->dest.ssa.num_components);
assert(info->num_variables == 0);
hash = _mesa_fnv32_1a_accumulate_block(hash, instr->const_index,
info->num_indices
* sizeof(instr->const_index[0]));
return hash;
}
static uint32_t
hash_tex(uint32_t hash, const nir_tex_instr *instr)
{
hash = HASH(hash, instr->op);
hash = HASH(hash, instr->num_srcs);
for (unsigned i = 0; i < instr->num_srcs; i++) {
hash = HASH(hash, instr->src[i].src_type);
hash = hash_src(hash, &instr->src[i].src);
}
hash = HASH(hash, instr->coord_components);
hash = HASH(hash, instr->sampler_dim);
hash = HASH(hash, instr->is_array);
hash = HASH(hash, instr->is_shadow);
hash = HASH(hash, instr->is_new_style_shadow);
unsigned component = instr->component;
hash = HASH(hash, component);
hash = HASH(hash, instr->texture_index);
hash = HASH(hash, instr->texture_array_size);
hash = HASH(hash, instr->sampler_index);
assert(!instr->texture && !instr->sampler);
return hash;
}
/* Computes a hash of an instruction for use in a hash table. Note that this
* will only work for instructions where instr_can_rewrite() returns true, and
* it should return identical hashes for two instructions that are the same
* according nir_instrs_equal().
*/
static uint32_t
hash_instr(const void *data)
{
const nir_instr *instr = data;
uint32_t hash = _mesa_fnv32_1a_offset_bias;
switch (instr->type) {
case nir_instr_type_alu:
hash = hash_alu(hash, nir_instr_as_alu(instr));
break;
case nir_instr_type_load_const:
hash = hash_load_const(hash, nir_instr_as_load_const(instr));
break;
case nir_instr_type_phi:
hash = hash_phi(hash, nir_instr_as_phi(instr));
break;
case nir_instr_type_intrinsic:
hash = hash_intrinsic(hash, nir_instr_as_intrinsic(instr));
break;
case nir_instr_type_tex:
hash = hash_tex(hash, nir_instr_as_tex(instr));
break;
default:
unreachable("Invalid instruction type");
}
return hash;
}
bool
nir_srcs_equal(nir_src src1, nir_src src2)
{
if (src1.is_ssa) {
if (src2.is_ssa) {
return src1.ssa == src2.ssa;
} else {
return false;
}
} else {
if (src2.is_ssa) {
return false;
} else {
if ((src1.reg.indirect == NULL) != (src2.reg.indirect == NULL))
return false;
if (src1.reg.indirect) {
if (!nir_srcs_equal(*src1.reg.indirect, *src2.reg.indirect))
return false;
}
return src1.reg.reg == src2.reg.reg &&
src1.reg.base_offset == src2.reg.base_offset;
}
}
}
static bool
nir_alu_srcs_equal(const nir_alu_instr *alu1, const nir_alu_instr *alu2,
unsigned src1, unsigned src2)
{
if (alu1->src[src1].abs != alu2->src[src2].abs ||
alu1->src[src1].negate != alu2->src[src2].negate)
return false;
for (unsigned i = 0; i < nir_ssa_alu_instr_src_components(alu1, src1); i++) {
if (alu1->src[src1].swizzle[i] != alu2->src[src2].swizzle[i])
return false;
}
return nir_srcs_equal(alu1->src[src1].src, alu2->src[src2].src);
}
/* Returns "true" if two instructions are equal. Note that this will only
* work for the subset of instructions defined by instr_can_rewrite(). Also,
* it should only return "true" for instructions that hash_instr() will return
* the same hash for (ignoring collisions, of course).
*/
static bool
nir_instrs_equal(const nir_instr *instr1, const nir_instr *instr2)
{
if (instr1->type != instr2->type)
return false;
switch (instr1->type) {
case nir_instr_type_alu: {
nir_alu_instr *alu1 = nir_instr_as_alu(instr1);
nir_alu_instr *alu2 = nir_instr_as_alu(instr2);
if (alu1->op != alu2->op)
return false;
/* TODO: We can probably acutally do something more inteligent such
* as allowing different numbers and taking a maximum or something
* here */
if (alu1->dest.dest.ssa.num_components != alu2->dest.dest.ssa.num_components)
return false;
if (nir_op_infos[alu1->op].algebraic_properties & NIR_OP_IS_COMMUTATIVE) {
assert(nir_op_infos[alu1->op].num_inputs == 2);
return (nir_alu_srcs_equal(alu1, alu2, 0, 0) &&
nir_alu_srcs_equal(alu1, alu2, 1, 1)) ||
(nir_alu_srcs_equal(alu1, alu2, 0, 1) &&
nir_alu_srcs_equal(alu1, alu2, 1, 0));
} else {
for (unsigned i = 0; i < nir_op_infos[alu1->op].num_inputs; i++) {
if (!nir_alu_srcs_equal(alu1, alu2, i, i))
return false;
}
}
return true;
}
case nir_instr_type_tex: {
nir_tex_instr *tex1 = nir_instr_as_tex(instr1);
nir_tex_instr *tex2 = nir_instr_as_tex(instr2);
if (tex1->op != tex2->op)
return false;
if (tex1->num_srcs != tex2->num_srcs)
return false;
for (unsigned i = 0; i < tex1->num_srcs; i++) {
if (tex1->src[i].src_type != tex2->src[i].src_type ||
!nir_srcs_equal(tex1->src[i].src, tex2->src[i].src)) {
return false;
}
}
if (tex1->coord_components != tex2->coord_components ||
tex1->sampler_dim != tex2->sampler_dim ||
tex1->is_array != tex2->is_array ||
tex1->is_shadow != tex2->is_shadow ||
tex1->is_new_style_shadow != tex2->is_new_style_shadow ||
tex1->component != tex2->component ||
tex1->texture_index != tex2->texture_index ||
tex1->texture_array_size != tex2->texture_array_size ||
tex1->sampler_index != tex2->sampler_index) {
return false;
}
/* Don't support un-lowered sampler derefs currently. */
assert(!tex1->texture && !tex1->sampler &&
!tex2->texture && !tex2->sampler);
return true;
}
case nir_instr_type_load_const: {
nir_load_const_instr *load1 = nir_instr_as_load_const(instr1);
nir_load_const_instr *load2 = nir_instr_as_load_const(instr2);
if (load1->def.num_components != load2->def.num_components)
return false;
return memcmp(load1->value.f, load2->value.f,
load1->def.num_components * sizeof(*load2->value.f)) == 0;
}
case nir_instr_type_phi: {
nir_phi_instr *phi1 = nir_instr_as_phi(instr1);
nir_phi_instr *phi2 = nir_instr_as_phi(instr2);
if (phi1->instr.block != phi2->instr.block)
return false;
nir_foreach_phi_src(phi1, src1) {
nir_foreach_phi_src(phi2, src2) {
if (src1->pred == src2->pred) {
if (!nir_srcs_equal(src1->src, src2->src))
return false;
break;
}
}
}
return true;
}
case nir_instr_type_intrinsic: {
nir_intrinsic_instr *intrinsic1 = nir_instr_as_intrinsic(instr1);
nir_intrinsic_instr *intrinsic2 = nir_instr_as_intrinsic(instr2);
const nir_intrinsic_info *info =
&nir_intrinsic_infos[intrinsic1->intrinsic];
if (intrinsic1->intrinsic != intrinsic2->intrinsic ||
intrinsic1->num_components != intrinsic2->num_components)
return false;
if (info->has_dest && intrinsic1->dest.ssa.num_components !=
intrinsic2->dest.ssa.num_components)
return false;
for (unsigned i = 0; i < info->num_srcs; i++) {
if (!nir_srcs_equal(intrinsic1->src[i], intrinsic2->src[i]))
return false;
}
assert(info->num_variables == 0);
for (unsigned i = 0; i < info->num_indices; i++) {
if (intrinsic1->const_index[i] != intrinsic2->const_index[i])
return false;
}
return true;
}
case nir_instr_type_call:
case nir_instr_type_jump:
case nir_instr_type_ssa_undef:
case nir_instr_type_parallel_copy:
default:
unreachable("Invalid instruction type");
}
return false;
}
static bool
src_is_ssa(nir_src *src, void *data)
{
(void) data;
return src->is_ssa;
}
static bool
dest_is_ssa(nir_dest *dest, void *data)
{
(void) data;
return dest->is_ssa;
}
/* This function determines if uses of an instruction can safely be rewritten
* to use another identical instruction instead. Note that this function must
* be kept in sync with hash_instr() and nir_instrs_equal() -- only
* instructions that pass this test will be handed on to those functions, and
* conversely they must handle everything that this function returns true for.
*/
static bool
instr_can_rewrite(nir_instr *instr)
{
/* We only handle SSA. */
if (!nir_foreach_dest(instr, dest_is_ssa, NULL) ||
!nir_foreach_src(instr, src_is_ssa, NULL))
return false;
switch (instr->type) {
case nir_instr_type_alu:
case nir_instr_type_load_const:
case nir_instr_type_phi:
return true;
case nir_instr_type_tex: {
nir_tex_instr *tex = nir_instr_as_tex(instr);
/* Don't support un-lowered sampler derefs currently. */
if (tex->texture || tex->sampler)
return false;
return true;
}
case nir_instr_type_intrinsic: {
const nir_intrinsic_info *info =
&nir_intrinsic_infos[nir_instr_as_intrinsic(instr)->intrinsic];
return (info->flags & NIR_INTRINSIC_CAN_ELIMINATE) &&
(info->flags & NIR_INTRINSIC_CAN_REORDER) &&
info->num_variables == 0; /* not implemented yet */
}
case nir_instr_type_call:
case nir_instr_type_jump:
case nir_instr_type_ssa_undef:
return false;
case nir_instr_type_parallel_copy:
default:
unreachable("Invalid instruction type");
}
return false;
}
static nir_ssa_def *
nir_instr_get_dest_ssa_def(nir_instr *instr)
{
switch (instr->type) {
case nir_instr_type_alu:
assert(nir_instr_as_alu(instr)->dest.dest.is_ssa);
return &nir_instr_as_alu(instr)->dest.dest.ssa;
case nir_instr_type_load_const:
return &nir_instr_as_load_const(instr)->def;
case nir_instr_type_phi:
assert(nir_instr_as_phi(instr)->dest.is_ssa);
return &nir_instr_as_phi(instr)->dest.ssa;
case nir_instr_type_intrinsic:
assert(nir_instr_as_intrinsic(instr)->dest.is_ssa);
return &nir_instr_as_intrinsic(instr)->dest.ssa;
case nir_instr_type_tex:
assert(nir_instr_as_tex(instr)->dest.is_ssa);
return &nir_instr_as_tex(instr)->dest.ssa;
default:
unreachable("We never ask for any of these");
}
}
static bool
cmp_func(const void *data1, const void *data2)
{
return nir_instrs_equal(data1, data2);
}
struct set *
nir_instr_set_create(void *mem_ctx)
{
return _mesa_set_create(mem_ctx, hash_instr, cmp_func);
}
void
nir_instr_set_destroy(struct set *instr_set)
{
_mesa_set_destroy(instr_set, NULL);
}
bool
nir_instr_set_add_or_rewrite(struct set *instr_set, nir_instr *instr)
{
if (!instr_can_rewrite(instr))
return false;
struct set_entry *entry = _mesa_set_search(instr_set, instr);
if (entry) {
nir_ssa_def *def = nir_instr_get_dest_ssa_def(instr);
nir_ssa_def *new_def =
nir_instr_get_dest_ssa_def((nir_instr *) entry->key);
nir_ssa_def_rewrite_uses(def, nir_src_for_ssa(new_def));
return true;
}
_mesa_set_add(instr_set, instr);
return false;
}
void
nir_instr_set_remove(struct set *instr_set, nir_instr *instr)
{
if (!instr_can_rewrite(instr))
return;
struct set_entry *entry = _mesa_set_search(instr_set, instr);
if (entry)
_mesa_set_remove(instr_set, entry);
}

View File

@@ -1,62 +0,0 @@
/*
* Copyright © 2014 Connor Abbott
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*/
#pragma once
#include "nir.h"
/**
* This file defines functions for creating, destroying, and manipulating an
* "instruction set," which is an abstraction for finding duplicate
* instructions using a hash set. Note that the question of whether an
* instruction is actually a duplicate (e.g. whether it has any side effects)
* is handled transparently. The user can pass any instruction to
* nir_instr_set_add_or_rewrite() and nir_instr_set_remove(), and if the
* instruction isn't safe to rewrite or isn't supported, it's silently
* removed.
*/
/*@{*/
/** Creates an instruction set, using a given ralloc mem_ctx */
struct set *nir_instr_set_create(void *mem_ctx);
/** Destroys an instruction set. */
void nir_instr_set_destroy(struct set *instr_set);
/**
* Adds an instruction to an instruction set if it doesn't exist, or if it
* does already exist, rewrites all uses of it to point to the other
* already-inserted instruction. Returns 'true' if the uses of the instruction
* were rewritten.
*/
bool nir_instr_set_add_or_rewrite(struct set *instr_set, nir_instr *instr);
/**
* Removes an instruction from an instruction set, so that other instructions
* won't be merged with it.
*/
void nir_instr_set_remove(struct set *instr_set, nir_instr *instr);
/*@}*/

View File

@@ -1,317 +0,0 @@
/*
* Copyright © 2014 Intel Corporation
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*
* Authors:
* Connor Abbott (cwabbott0@gmail.com)
*
*/
/**
* This header file defines all the available intrinsics in one place. It
* expands to a list of macros of the form:
*
* INTRINSIC(name, num_srcs, src_components, has_dest, dest_components,
* num_variables, num_indices, idx0, idx1, idx2, flags)
*
* Which should correspond one-to-one with the nir_intrinsic_info structure. It
* is included in both ir.h to create the nir_intrinsic enum (with members of
* the form nir_intrinsic_(name)) and and in opcodes.c to create
* nir_intrinsic_infos, which is a const array of nir_intrinsic_info structures
* for each intrinsic.
*/
#define ARR(...) { __VA_ARGS__ }
INTRINSIC(load_var, 0, ARR(), true, 0, 1, 0, xx, xx, xx, NIR_INTRINSIC_CAN_ELIMINATE)
INTRINSIC(store_var, 1, ARR(0), false, 0, 1, 1, WRMASK, xx, xx, 0)
INTRINSIC(copy_var, 0, ARR(), false, 0, 2, 0, xx, xx, xx, 0)
/*
* Interpolation of input. The interp_var_at* intrinsics are similar to the
* load_var intrinsic acting an a shader input except that they interpolate
* the input differently. The at_sample and at_offset intrinsics take an
* aditional source that is a integer sample id or a vec2 position offset
* respectively.
*/
INTRINSIC(interp_var_at_centroid, 0, ARR(0), true, 0, 1, 0, xx, xx, xx,
NIR_INTRINSIC_CAN_ELIMINATE | NIR_INTRINSIC_CAN_REORDER)
INTRINSIC(interp_var_at_sample, 1, ARR(1), true, 0, 1, 0, xx, xx, xx,
NIR_INTRINSIC_CAN_ELIMINATE | NIR_INTRINSIC_CAN_REORDER)
INTRINSIC(interp_var_at_offset, 1, ARR(2), true, 0, 1, 0, xx, xx, xx,
NIR_INTRINSIC_CAN_ELIMINATE | NIR_INTRINSIC_CAN_REORDER)
/*
* Ask the driver for the size of a given buffer. It takes the buffer index
* as source.
*/
INTRINSIC(get_buffer_size, 1, ARR(1), true, 1, 0, 0, xx, xx, xx,
NIR_INTRINSIC_CAN_ELIMINATE | NIR_INTRINSIC_CAN_REORDER)
/*
* a barrier is an intrinsic with no inputs/outputs but which can't be moved
* around/optimized in general
*/
#define BARRIER(name) INTRINSIC(name, 0, ARR(), false, 0, 0, 0, xx, xx, xx, 0)
BARRIER(barrier)
BARRIER(discard)
/*
* Memory barrier with semantics analogous to the memoryBarrier() GLSL
* intrinsic.
*/
BARRIER(memory_barrier)
/*
* Shader clock intrinsic with semantics analogous to the clock2x32ARB()
* GLSL intrinsic.
* The latter can be used as code motion barrier, which is currently not
* feasible with NIR.
*/
INTRINSIC(shader_clock, 0, ARR(), true, 1, 0, 0, xx, xx, xx, NIR_INTRINSIC_CAN_ELIMINATE)
/*
* Memory barrier with semantics analogous to the compute shader
* groupMemoryBarrier(), memoryBarrierAtomicCounter(), memoryBarrierBuffer(),
* memoryBarrierImage() and memoryBarrierShared() GLSL intrinsics.
*/
BARRIER(group_memory_barrier)
BARRIER(memory_barrier_atomic_counter)
BARRIER(memory_barrier_buffer)
BARRIER(memory_barrier_image)
BARRIER(memory_barrier_shared)
/** A conditional discard, with a single boolean source. */
INTRINSIC(discard_if, 1, ARR(1), false, 0, 0, 0, xx, xx, xx, 0)
/**
* Basic Geometry Shader intrinsics.
*
* emit_vertex implements GLSL's EmitStreamVertex() built-in. It takes a single
* index, which is the stream ID to write to.
*
* end_primitive implements GLSL's EndPrimitive() built-in.
*/
INTRINSIC(emit_vertex, 0, ARR(), false, 0, 0, 1, STREAM_ID, xx, xx, 0)
INTRINSIC(end_primitive, 0, ARR(), false, 0, 0, 1, STREAM_ID, xx, xx, 0)
/**
* Geometry Shader intrinsics with a vertex count.
*
* Alternatively, drivers may implement these intrinsics, and use
* nir_lower_gs_intrinsics() to convert from the basic intrinsics.
*
* These maintain a count of the number of vertices emitted, as an additional
* unsigned integer source.
*/
INTRINSIC(emit_vertex_with_counter, 1, ARR(1), false, 0, 0, 1, STREAM_ID, xx, xx, 0)
INTRINSIC(end_primitive_with_counter, 1, ARR(1), false, 0, 0, 1, STREAM_ID, xx, xx, 0)
INTRINSIC(set_vertex_count, 1, ARR(1), false, 0, 0, 0, xx, xx, xx, 0)
/*
* Atomic counters
*
* The *_var variants take an atomic_uint nir_variable, while the other,
* lowered, variants take a constant buffer index and register offset.
*/
#define ATOMIC(name, flags) \
INTRINSIC(atomic_counter_##name##_var, 0, ARR(), true, 1, 1, 0, xx, xx, xx, flags) \
INTRINSIC(atomic_counter_##name, 1, ARR(1), true, 1, 0, 1, BASE, xx, xx, flags)
ATOMIC(inc, 0)
ATOMIC(dec, 0)
ATOMIC(read, NIR_INTRINSIC_CAN_ELIMINATE)
/*
* Image load, store and atomic intrinsics.
*
* All image intrinsics take an image target passed as a nir_variable. Image
* variables contain a number of memory and layout qualifiers that influence
* the semantics of the intrinsic.
*
* All image intrinsics take a four-coordinate vector and a sample index as
* first two sources, determining the location within the image that will be
* accessed by the intrinsic. Components not applicable to the image target
* in use are undefined. Image store takes an additional four-component
* argument with the value to be written, and image atomic operations take
* either one or two additional scalar arguments with the same meaning as in
* the ARB_shader_image_load_store specification.
*/
INTRINSIC(image_load, 2, ARR(4, 1), true, 4, 1, 0, xx, xx, xx,
NIR_INTRINSIC_CAN_ELIMINATE)
INTRINSIC(image_store, 3, ARR(4, 1, 4), false, 0, 1, 0, xx, xx, xx, 0)
INTRINSIC(image_atomic_add, 3, ARR(4, 1, 1), true, 1, 1, 0, xx, xx, xx, 0)
INTRINSIC(image_atomic_min, 3, ARR(4, 1, 1), true, 1, 1, 0, xx, xx, xx, 0)
INTRINSIC(image_atomic_max, 3, ARR(4, 1, 1), true, 1, 1, 0, xx, xx, xx, 0)
INTRINSIC(image_atomic_and, 3, ARR(4, 1, 1), true, 1, 1, 0, xx, xx, xx, 0)
INTRINSIC(image_atomic_or, 3, ARR(4, 1, 1), true, 1, 1, 0, xx, xx, xx, 0)
INTRINSIC(image_atomic_xor, 3, ARR(4, 1, 1), true, 1, 1, 0, xx, xx, xx, 0)
INTRINSIC(image_atomic_exchange, 3, ARR(4, 1, 1), true, 1, 1, 0, xx, xx, xx, 0)
INTRINSIC(image_atomic_comp_swap, 4, ARR(4, 1, 1, 1), true, 1, 1, 0, xx, xx, xx, 0)
INTRINSIC(image_size, 0, ARR(), true, 4, 1, 0, xx, xx, xx,
NIR_INTRINSIC_CAN_ELIMINATE | NIR_INTRINSIC_CAN_REORDER)
INTRINSIC(image_samples, 0, ARR(), true, 1, 1, 0, xx, xx, xx,
NIR_INTRINSIC_CAN_ELIMINATE | NIR_INTRINSIC_CAN_REORDER)
/*
* SSBO atomic intrinsics
*
* All of the SSBO atomic memory operations read a value from memory,
* compute a new value using one of the operations below, write the new
* value to memory, and return the original value read.
*
* All operations take 3 sources except CompSwap that takes 4. These
* sources represent:
*
* 0: The SSBO buffer index.
* 1: The offset into the SSBO buffer of the variable that the atomic
* operation will operate on.
* 2: The data parameter to the atomic function (i.e. the value to add
* in ssbo_atomic_add, etc).
* 3: For CompSwap only: the second data parameter.
*/
INTRINSIC(ssbo_atomic_add, 3, ARR(1, 1, 1), true, 1, 0, 0, xx, xx, xx, 0)
INTRINSIC(ssbo_atomic_imin, 3, ARR(1, 1, 1), true, 1, 0, 0, xx, xx, xx, 0)
INTRINSIC(ssbo_atomic_umin, 3, ARR(1, 1, 1), true, 1, 0, 0, xx, xx, xx, 0)
INTRINSIC(ssbo_atomic_imax, 3, ARR(1, 1, 1), true, 1, 0, 0, xx, xx, xx, 0)
INTRINSIC(ssbo_atomic_umax, 3, ARR(1, 1, 1), true, 1, 0, 0, xx, xx, xx, 0)
INTRINSIC(ssbo_atomic_and, 3, ARR(1, 1, 1), true, 1, 0, 0, xx, xx, xx, 0)
INTRINSIC(ssbo_atomic_or, 3, ARR(1, 1, 1), true, 1, 0, 0, xx, xx, xx, 0)
INTRINSIC(ssbo_atomic_xor, 3, ARR(1, 1, 1), true, 1, 0, 0, xx, xx, xx, 0)
INTRINSIC(ssbo_atomic_exchange, 3, ARR(1, 1, 1), true, 1, 0, 0, xx, xx, xx, 0)
INTRINSIC(ssbo_atomic_comp_swap, 4, ARR(1, 1, 1, 1), true, 1, 0, 0, xx, xx, xx, 0)
/*
* CS shared variable atomic intrinsics
*
* All of the shared variable atomic memory operations read a value from
* memory, compute a new value using one of the operations below, write the
* new value to memory, and return the original value read.
*
* All operations take 2 sources except CompSwap that takes 3. These
* sources represent:
*
* 0: The offset into the shared variable storage region that the atomic
* operation will operate on.
* 1: The data parameter to the atomic function (i.e. the value to add
* in shared_atomic_add, etc).
* 2: For CompSwap only: the second data parameter.
*/
INTRINSIC(shared_atomic_add, 2, ARR(1, 1), true, 1, 0, 0, xx, xx, xx, 0)
INTRINSIC(shared_atomic_imin, 2, ARR(1, 1), true, 1, 0, 0, xx, xx, xx, 0)
INTRINSIC(shared_atomic_umin, 2, ARR(1, 1), true, 1, 0, 0, xx, xx, xx, 0)
INTRINSIC(shared_atomic_imax, 2, ARR(1, 1), true, 1, 0, 0, xx, xx, xx, 0)
INTRINSIC(shared_atomic_umax, 2, ARR(1, 1), true, 1, 0, 0, xx, xx, xx, 0)
INTRINSIC(shared_atomic_and, 2, ARR(1, 1), true, 1, 0, 0, xx, xx, xx, 0)
INTRINSIC(shared_atomic_or, 2, ARR(1, 1), true, 1, 0, 0, xx, xx, xx, 0)
INTRINSIC(shared_atomic_xor, 2, ARR(1, 1), true, 1, 0, 0, xx, xx, xx, 0)
INTRINSIC(shared_atomic_exchange, 2, ARR(1, 1), true, 1, 0, 0, xx, xx, xx, 0)
INTRINSIC(shared_atomic_comp_swap, 3, ARR(1, 1, 1), true, 1, 0, 0, xx, xx, xx, 0)
#define SYSTEM_VALUE(name, components, num_indices, idx0, idx1, idx2) \
INTRINSIC(load_##name, 0, ARR(), true, components, 0, num_indices, \
idx0, idx1, idx2, \
NIR_INTRINSIC_CAN_ELIMINATE | NIR_INTRINSIC_CAN_REORDER)
SYSTEM_VALUE(front_face, 1, 0, xx, xx, xx)
SYSTEM_VALUE(vertex_id, 1, 0, xx, xx, xx)
SYSTEM_VALUE(vertex_id_zero_base, 1, 0, xx, xx, xx)
SYSTEM_VALUE(base_vertex, 1, 0, xx, xx, xx)
SYSTEM_VALUE(instance_id, 1, 0, xx, xx, xx)
SYSTEM_VALUE(base_instance, 1, 0, xx, xx, xx)
SYSTEM_VALUE(draw_id, 1, 0, xx, xx, xx)
SYSTEM_VALUE(sample_id, 1, 0, xx, xx, xx)
SYSTEM_VALUE(sample_pos, 2, 0, xx, xx, xx)
SYSTEM_VALUE(sample_mask_in, 1, 0, xx, xx, xx)
SYSTEM_VALUE(primitive_id, 1, 0, xx, xx, xx)
SYSTEM_VALUE(invocation_id, 1, 0, xx, xx, xx)
SYSTEM_VALUE(tess_coord, 3, 0, xx, xx, xx)
SYSTEM_VALUE(tess_level_outer, 4, 0, xx, xx, xx)
SYSTEM_VALUE(tess_level_inner, 2, 0, xx, xx, xx)
SYSTEM_VALUE(patch_vertices_in, 1, 0, xx, xx, xx)
SYSTEM_VALUE(local_invocation_id, 3, 0, xx, xx, xx)
SYSTEM_VALUE(work_group_id, 3, 0, xx, xx, xx)
SYSTEM_VALUE(user_clip_plane, 4, 1, UCP_ID, xx, xx)
SYSTEM_VALUE(num_work_groups, 3, 0, xx, xx, xx)
SYSTEM_VALUE(helper_invocation, 1, 0, xx, xx, xx)
/*
* Load operations pull data from some piece of GPU memory. All load
* operations operate in terms of offsets into some piece of theoretical
* memory. Loads from externally visible memory (UBO and SSBO) simply take a
* byte offset as a source. Loads from opaque memory (uniforms, inputs, etc.)
* take a base+offset pair where the base (const_index[0]) gives the location
* of the start of the variable being loaded and and the offset source is a
* offset into that variable.
*
* Some load operations such as UBO/SSBO load and per_vertex loads take an
* additional source to specify which UBO/SSBO/vertex to load from.
*
* The exact address type depends on the lowering pass that generates the
* load/store intrinsics. Typically, this is vec4 units for things such as
* varying slots and float units for fragment shader inputs. UBO and SSBO
* offsets are always in bytes.
*/
#define LOAD(name, srcs, num_indices, idx0, idx1, idx2, flags) \
INTRINSIC(load_##name, srcs, ARR(1, 1, 1, 1), true, 0, 0, num_indices, idx0, idx1, idx2, flags)
/* src[] = { offset }. const_index[] = { base } */
LOAD(uniform, 1, 1, BASE, xx, xx, NIR_INTRINSIC_CAN_ELIMINATE | NIR_INTRINSIC_CAN_REORDER)
/* src[] = { buffer_index, offset }. No const_index */
LOAD(ubo, 2, 0, xx, xx, xx, NIR_INTRINSIC_CAN_ELIMINATE | NIR_INTRINSIC_CAN_REORDER)
/* src[] = { offset }. const_index[] = { base } */
LOAD(input, 1, 1, BASE, xx, xx, NIR_INTRINSIC_CAN_ELIMINATE | NIR_INTRINSIC_CAN_REORDER)
/* src[] = { vertex, offset }. const_index[] = { base } */
LOAD(per_vertex_input, 2, 1, BASE, xx, xx, NIR_INTRINSIC_CAN_ELIMINATE | NIR_INTRINSIC_CAN_REORDER)
/* src[] = { buffer_index, offset }. No const_index */
LOAD(ssbo, 2, 0, xx, xx, xx, NIR_INTRINSIC_CAN_ELIMINATE)
/* src[] = { offset }. const_index[] = { base } */
LOAD(output, 1, 1, BASE, xx, xx, NIR_INTRINSIC_CAN_ELIMINATE)
/* src[] = { vertex, offset }. const_index[] = { base } */
LOAD(per_vertex_output, 2, 1, BASE, xx, xx, NIR_INTRINSIC_CAN_ELIMINATE)
/* src[] = { offset }. const_index[] = { base } */
LOAD(shared, 1, 1, BASE, xx, xx, NIR_INTRINSIC_CAN_ELIMINATE)
/*
* Stores work the same way as loads, except now the first source is the value
* to store and the second (and possibly third) source specify where to store
* the value. SSBO and shared memory stores also have a write mask as
* const_index[0].
*/
#define STORE(name, srcs, num_indices, idx0, idx1, idx2, flags) \
INTRINSIC(store_##name, srcs, ARR(0, 1, 1, 1), false, 0, 0, num_indices, idx0, idx1, idx2, flags)
/* src[] = { value, offset }. const_index[] = { base, write_mask } */
STORE(output, 2, 2, BASE, WRMASK, xx, 0)
/* src[] = { value, vertex, offset }. const_index[] = { base, write_mask } */
STORE(per_vertex_output, 3, 2, BASE, WRMASK, xx, 0)
/* src[] = { value, block_index, offset }. const_index[] = { write_mask } */
STORE(ssbo, 3, 1, WRMASK, xx, xx, 0)
/* src[] = { value, offset }. const_index[] = { base, write_mask } */
STORE(shared, 2, 2, BASE, WRMASK, xx, 0)
LAST_INTRINSIC(store_shared)

Some files were not shown because too many files have changed in this diff Show More