Compare commits

..

310 Commits

Author SHA1 Message Date
Emil Velikov
4a0bd3dcff docs: Add sha256sums for the 10.5.9 release
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-07-04 12:48:39 +01:00
Emil Velikov
7f40d08374 Add release notes for the 10.5.9 release
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-07-04 12:09:10 +01:00
Emil Velikov
8c8006099f Update version to 10.5.9
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-07-04 12:04:06 +01:00
Mario Kleiner
81a0920c3a winsys/radeon: Use dup fd as key in drm-winsys hash table to fix ZaphodHeads.
Same problem and fix as for nouveau's ZaphodHeads trouble.

See patch ...

"nouveau: Use dup fd as key in drm-winsys hash table to fix ZaphodHeads."

... for reference.

Cc: "10.3 10.4 10.5 10.6" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
(cherry picked from commit 28dda47ae4)
2015-07-04 12:03:15 +01:00
Ben Widawsky
80e283b053 i965/skl: Use 1 register for uniform pull constant payload
When under dispatch_width=16 the previous code would allocate 2 registers for
the payload when only one is needed. This manifested itself through bugs on SKL
which needs to mess with this instruction.

Ken thought this might impact shader-db, but apparently it doesn't

Backported to 10.5.8 from (upstream uses alloc.allocate()):
commit 6e62a52865
Author: Ben Widawsky <benjamin.widawsky@intel.com>
Date:   Thu Feb 19 15:49:34 2015 -0800

    i965/skl: Use 1 register for uniform pull constant payload

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89118
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=88999
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Tested-by: Timo Aaltonen <timo.aaltonen@canonical.com>
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
2015-07-01 15:22:41 +01:00
Ilia Mirkin
999b2b2716 nv50/ir: copy joinAt when splitting both before and after
The current implementation only moves the joinAt when splitting after
the given instruction, not before it. So if you have a BB with

  foo
  instr
  bar
  joinat

and thus with joinAt set, we end up first splitting before instr, at
which point the instr's bb is updated to the new bb. Since that bb
doesn't have a joinAt set (despite containing one), when splitting after
the instr, there is nothing to copy over. Since the joinat will be in
the "split" bb irrespective of whether we're splitting before or after
the instruction, move it over in either case.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91124
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 5dcb28c3d2)
2015-07-01 15:22:41 +01:00
Tapani Pälli
81ac47605f glsl: validate sampler array indexing for 'constant-index-expression'
Desktop GLSL < 130 and GLSL ES < 300 allow sampler array indexing where
index can contain a loop induction variable. This extra check will warn
during linking if some of the indexes could not be turned in to constant
expressions.

v2: warning instead of error for backends that did not enable
    EmitNoIndirectSampler option (have dynamic indexing)

Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Cc: "10.5" and "10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 9350ea6979)
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>

Conflicts:
	src/glsl/linker.cpp
2015-07-01 15:22:41 +01:00
Tapani Pälli
f8ea1430ae i965: use EmitNoIndirectSampler for gen < 7
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Cc: "10.5" and "10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 8852e26e93)
[Emil Velikov: move the hunk to brw_initialize_context_constants()]
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>

Conflicts:
	src/mesa/drivers/dri/i965/brw_shader.cpp
2015-07-01 15:22:41 +01:00
Tapani Pälli
c11a8c1c33 mesa/st: use EmitNoIndirectSampler if !ARB_gpu_shader5
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Cc: "10.5" and "10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit f17c8c287f)
2015-07-01 15:22:40 +01:00
Tapani Pälli
4c80d44311 i915: use EmitNoIndirectSampler
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Cc: "10.5" and "10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 2dc2b12ed1)
2015-07-01 15:22:40 +01:00
Tapani Pälli
8e79cf6ba0 mesa/glsl: new compiler option EmitNoIndirectSampler
Patch provides new compiler option for backend to force unroll loops
that have non-constant expression indexing on sampler arrays.

This makes sure that we can never end up with a shader that uses loop
induction variable as sampler array index but does not unroll because
of having too much instructions. This would not work without dynamic
indexing support.

v2: change option name as EmitNoIndirectSampler

Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Cc: "10.5" and "10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit e4512e1581)
2015-07-01 15:22:40 +01:00
Tapani Pälli
5bec733641 glsl: Allow dynamic sampler array indexing with GLSL ES < 3.00
Dynamic indexing of sampler arrays is prohibited by GLSL ES 3.00.
Earlier versions allow 'constant-index-expression' indexing, where
index can contain a loop induction variable.

Patch allows dynamic indexing for sampler arrays when GLSL ES < 3.00.
This change makes 'sampler-array-index.frag' parser test in Piglit
pass + fishgl.com works when running Chrome on OpenGL ES 2.0 backend

v2: small change and some more commit message (Tapani)
v3: refactor checks to make it more readable (Ian Romanick)
v4: change warning comment in GLSL ES case (Curro)

Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Signed-off-by: Kalyan Kondapally <kalyan.kondapally@intel.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Cc: "10.5" and "10.6" <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=84225
(cherry picked from commit edb8383c98)
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>

Conflicts:
	src/glsl/ast_array_index.cpp
2015-07-01 15:22:40 +01:00
Ilia Mirkin
5985de6f1a nv50/ir: fix emission of address reg in 3rd source
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91056
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit d5f1253b0c)
2015-07-01 15:22:40 +01:00
Mario Kleiner
cdc7aa1038 nouveau: Use dup fd as key in drm-winsys hash table to fix ZaphodHeads.
The dup'ed fd owned by the nouveau_screen for a device node
must also be used as key for the winsys hash table, instead
of using the original fd passed in for a screen, to make
multi-x-screen ZaphodHeads configurations work on nouveau.

The original fd's lifetime differs from that of the nouveau_screen stored
in the hash. The hash key is the fd, and in order to compare hash entries
we fstat them, so the fd must be around for as long as the screen is.

This is an extension of the fix in commit a59f2bb1 (nouveau: dup fd
before passing it to device).

Cc: "10.3 10.4 10.5 10.6" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
(cherry picked from commit a98600b0eb)
2015-07-01 15:22:40 +01:00
Ilia Mirkin
16b6fa972d nv50/ir: propagate modifier to right arg when const-folding mad
An immediate has to be the second arg of an ADD operation. However we
were mistakenly propagating the modifier of the non-folded value to the
folded immediate argument.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91117
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit ad62ec8316)
2015-07-01 15:22:40 +01:00
Michel Dänzer
3c4570e3fc winsys/radeon: Unmap GPU VM address range when destroying BO
But only when doing so is safe according to the
RADEON_INFO_VA_UNMAP_WORKING kernel query.

This avoids kernel GPU VM address range conflicts when the BO has other
references than the GEM handle being closed, e.g. when the BO is shared.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90537
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90873

Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Christian König <christian.koenig@amd.com>
(cherry picked from commit 7796e8889a)
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>

Conflicts:
	src/gallium/winsys/radeon/drm/radeon_drm_winsys.c
2015-07-01 15:22:40 +01:00
Boyan Ding
2843e2a5c8 egl/x11: Remove duplicate call to dri2_x11_add_configs_for_visuals
The call to dri2_x11_add_configs_for_visuals (previously
dri2_add_configs_for_visuals) was moved downwards in commit f8c5b8a1,
but appeared again in its original position after its rename in
d019cd81. Remove it.

Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Boyan Ding <boyan.j.ding@gmail.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Chad Versace <chad.versace@intel.com>
(cherry picked from commit 3fa9bb81ec)
2015-07-01 15:22:40 +01:00
Ilia Mirkin
36250def23 nv50,nvc0: make sure to pushbuf_refn before putting bo into pushbuf_data
Without first running the bo through pushbuf_refn, the nouveau drm
library will have uninitialized structures regarding this bo, and will
insert incorrect data.

This fixes supertuxkart 0.9 crash on start (where it ends up doing a lot
of indirect draws).

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 78d58e6425)
2015-07-01 15:22:39 +01:00
Ilia Mirkin
8bcc8f4b48 nvc0: always put all tfb bufs into bufctx
Since we clear the TFB bufctx binding point above, we need to put all of
the active tfb's back in, even if they haven't changed since last time.
Otherwise the tfb may get moved into sysmem and the underlying mapping
will generate write errors.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 9fcbf515b4)
2015-07-01 15:22:39 +01:00
Ilia Mirkin
54ac592009 glsl: binding point is a texture unit, which is a combined space
This fixes compilation failures in Dota 2 Reborn where a texture unit
binding point was used that was numerically higher than the max
per stage.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Chris Forbes <chrisf@ijw.co.nz>
Reviewed-by: Timothy Arceri <t_arceri@yahoo.com.au>
Tested-by: Nick Sarnie <commendsarnex@gmail.com>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit fccf012adc)
2015-07-01 15:22:39 +01:00
Emil Velikov
7b528f0d78 gbm: do not (over)link against libglapi.so
The whole of GBM does not rely on even a single symbol from the GL
dispatch library, unsuprisingly. The only need for it comes from the
unresolved symbols in the DRI modules, which are now correctly handled
with Frank's commit.

Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
(cherry picked from commit a0dc6b7824)
2015-07-01 15:22:39 +01:00
Frank Henigman
8c748fa125 gbm: dlopen libglapi so gbm_create_device works
Dri driver libs are not linked to pull in libglapi so gbm_create_device()
fails when it tries to dlopen them (unless the application is linked
with something that does pull in libglapi, like libGL).
Until dri drivers can be fixed properly, dlopen libglapi before trying
to dlopen them.

Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Frank Henigman <fjhenigman@google.com>
[Emil Velikov: Drop misleading bugzilla link, mention that libname differs]
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>

(cherry picked from commit 828f13330c)
2015-07-01 15:22:39 +01:00
Emil Velikov
6bb51be018 configure: error out when building libEGL without shared-glapi
The latter is a hard requirement and without it we'll error out later
on in the build.

Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
(cherry picked from commit 994be5143a)
2015-07-01 15:22:39 +01:00
Emil Velikov
a939a4b9a6 configure: error out when building backend-less libEGL
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
(cherry picked from commit ddc886b5bf)
2015-07-01 15:22:39 +01:00
Emil Velikov
e724322f39 configure: warn about shared_glapi & xlib-glx only when both are set
Printing out the message when shared_glapi is disabled only leads to
confusion.

Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
(cherry picked from commit 6d744aaf4e)
2015-07-01 15:22:39 +01:00
Ben Widawsky
9df2ae3305 i965/gen9: Implement Push Constant Buffer workaround
This implements a workaround (exact excerpt as a comment in the code). The docs
specify [clearly, after you struggle for a while] that the offset isn't relative
to state base. This actually makes sense. This fixes hangs on SKL.

Buffer #0 is meant to be used for normal uniforms.
Buffer #1 is typically used for gather constants when using RS.
Buffer #1-#3 could be used to push a bunch of UBO data which would just be
  somewhere in memory, and not relative to the dynamic state.

NOTE: I've moved away from the ternary operator for the new gen9 conditions.
Admittedly it's probably not great to do this, but I really want to fix this all
up in the subsequent patch and doing it here makes that diff a lot nicer. I want
to split out the gen8/9 code to make the function a bit more readable, but to
keep this easily cherry-pickable I am doing this fix first. If we decide not to
merge the cleanup patch then I can revisit this.

Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Tested-by: Valtteri Rantala <Valtteri.rantala@intel.com>
(cherry picked from commit 90754d2df0)
2015-07-01 15:22:39 +01:00
Ilia Mirkin
e82c46de11 mesa: add GL_PROGRAM_PIPELINE support in KHR_debug calls
This was apparently missed when ARB_sso support was added.
Add label support to pipeline objects just like all the other
debug-related objects.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Timothy Arceri <t_arceri@yahoo.com.au>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 770f141866)
2015-07-01 15:22:39 +01:00
Ilia Mirkin
fc6caeb9ad glsl: add version checks to conditionals for builtin variable enablement
A number of builtin variables have checks based on the extension being
enabled, but were missing enablement via a higher GLSL version.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Timothy Arceri <t_arceri@yahoo.com.au>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit b6e238023c)
2015-07-01 15:22:38 +01:00
Chris Wilson
e3bcc87755 i965: Export format comparison for blitting between miptrees
Since the introduction of

commit 536003c11e
Author: Boyan Ding <boyan.j.ding@gmail.com>
Date:   Wed Mar 25 19:36:54 2015 +0800

    i965: Add XRGB8888 format to intel_screen_make_configs

winsys buffers no longer have an alpha channel. This causes
_mesa_format_matches_format_and_type() to reject previously working BGRA
uploads from using the BLT fast path. Instead of using the generic
routine for matching formats exactly, export the slightly more relaxed
check from intel_miptree_blit() which importantly allows the blitter
routine to apply a small number of format conversions.

References: https://bugs.freedesktop.org/show_bug.cgi?id=90839
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Jason Ekstrand <jason@jlekstrand.net>
Cc: Alexander Monakov <amonakov@gmail.com>
Cc: Kristian Høgsberg <krh@bitplanet.net>
Cc: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit 922c0c9fd5)
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>

Conflicts:
	src/mesa/drivers/dri/i965/intel_pixel_draw.c
2015-07-01 15:22:38 +01:00
Chris Wilson
22d7f3ceca i915: Blit RGBX<->RGBA drawpixels
The blitter already has code to accommodate filling in the alpha channel
for BGRX destination formats, so expand this to also allow filling the
alpha channgel in RGBX formats.

More importantly for the next patch is moving the test into its own
function for the purpose of exporting the check to the callers.

v2: Fix alpha expansion as spotted by Alexander with the fix suggested by
Kenneth

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Jason Ekstrand <jason@jlekstrand.net>
Cc: Alexander Monakov <amonakov@gmail.com>
Cc: Kristian Høgsberg <krh@bitplanet.net>
Cc: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by Kenneth Graunke <kenneth@whitecape.org>
Cc: mesa-stable@lists.freedesktop.org

(cherry picked from commit c2d0606827)
2015-07-01 15:22:38 +01:00
Chris Wilson
074122fbe3 i965: Fix HW blitter pitch limits
The BLT pitch is specified in bytes for linear surfaces and in dwords
for tiled surfaces. In both cases the programmable limit is 32,767, so
adjust the check to compensate for the effect of tiling.

v2: Tweak whitespace for functions (Kenneth)

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Kristian Høgsberg <krh@bitplanet.net>
Cc: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by Kenneth Graunke <kenneth@whitecape.org>
Cc: mesa-stable@lists.freedesktop.org

(cherry picked from commit 8da79b8378)
2015-07-01 15:22:38 +01:00
Emil Velikov
a81b1d5512 docs: Add sha256sums for the 10.5.8 release
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-06-20 16:37:16 +01:00
Emil Velikov
24b043aab7 Add release notes for the 10.5.8 release
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-06-20 15:14:45 +01:00
Emil Velikov
2de5d0762a Update version to 10.5.8
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-06-20 15:09:23 +01:00
Boyan Ding
e6f1a49809 egl/x11: Set version of swrastLoader to 2
which it actually implements instead of the newest version defined in
dri_interface.h

Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Boyan Ding <boyan.j.ding@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
(cherry picked from commit 997fc807b2)
2015-06-18 13:44:09 +01:00
Ilia Mirkin
4ab83eee98 nvc0/ir: can't have a join on a load with an indirect source
Triggers an INVALID_OPCODE warning on GK208. Seems rare enough to not
warrant verification on other chips. Fixes the new piglits:

  ubo_array_indexing/fs-nonuniform-control-flow.shader_test
  ubo_array_indexing/vs-nonuniform-control-flow.shader_test

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 36e3eb6a95)
2015-06-18 13:44:03 +01:00
Ilia Mirkin
512927bc02 nv50,nvc0: clamp uniform size to 64k
The state tracker will pass through requests from buggy applications
which will have the buffer size larger than the max allowed (64k). Clamp
the size to 64k so that we don't get errors when uploading the constbuf
data.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 8b24388647)
2015-06-18 13:43:56 +01:00
Ilia Mirkin
b26bac8b1e nvc0/ir: fix collection of first uses for texture barrier insertion
One of the places we have to insert texbars is in situations where the
result of the tex gets overwritten by a different instruction (e.g. in a
conditional statement). However in some situations it can actually
appear as though the original tex itself is an overwriting instruction.
This can naturally never really happen, so just ignore the tex
instruction when it comes up.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90347
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit a2af42c1d2)
2015-06-18 13:43:48 +01:00
Erik Faye-Lund
20fd2abdd6 mesa: build xmlconfig to a separate static library
As we use the file from both the dri modules and loader, we end up with
multiple definition of the symbols provided in our gallium dri  modules.
Additionally we compile the file twice.

Resolve both issues, effectively enabling the build on toolchains which
don't support -Wl,--allow-multiple-definition.

v2: [Emil Velikov]
 - Fix the Scons/Android build.
 - Resolve libgbm build issues (bring back the missing -lm)

Cc: Julien Isorce <j.isorce@samsung.com>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90310
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90905
Acked-by: Matt Turner <mattst88@gmail.com>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
(cherry picked from commit 634f200256)
2015-06-18 13:43:40 +01:00
Jason Ekstrand
bb00457f49 i965/fs: Don't let the EOT send message interfere with the MRF hack
Previously, we just put the message for the EOT send as high in the file as
it would go.  This is because the register pre-filling hardware will stop
all over the early registers in the file in preparation for the next thread
while you're still sending the last message.  However, if something happens
to spill, then the MRF hack interferes with the EOT send message and, if
things aren't scheduled nicely, will stomp on it.

Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90520
Reviewed-by: Neil Roberts <neil@linux.intel.com>
(cherry picked from commit 86e5afbfee)
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>

Conflicts:
	src/mesa/drivers/dri/i965/brw_fs_reg_allocate.cpp
2015-06-18 13:42:59 +01:00
Roland Scheidegger
f6e743ea38 draw: (trivial) fix NULL pointer dereference
This probably got broken when the samplers were converted to be indexed
by shader type.
Seen when looking at bug 89819 though I'm not sure if that really was what
the bug was about...

Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>

Reviewed-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit 6e5970ffee)
2015-06-18 13:29:04 +01:00
Ben Widawsky
1812014fe8 i965: Disable compaction for EOT send messages
AFAICT, there is no real way to make sure a send message with EOT is properly
ignored from compact, nor can I see a way to actually encode EOT while
compacting. Before the single send optimization we'd always bail because we hit
the is_immediate && !is_compactable_immediate case. However, with single send,
is_immediate is not true, and so we end up trying to compact the un-compactible.

Without this, any compacting single send instruction will hang because the EOT
isn't there. I am not sure how I didn't hit this when I originally enabled the
optimization.  I didn't check if some surrounding code changed.

I know Neil and Matt were both looking into this. I did a quick search and
didn't see any patches out there to handle this. Please ignore if this has
already been sent by someone. (Direct me to it and I will review it).

Reported-by: Neil Roberts <neil@linux.intel.com>
Reported-by: Mark Janes <mark.a.janes@intel.com>
Tested-by: Mark Janes <mark.a.janes@intel.com>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Reviewed-by: Matt Turner <mattst88@gmail.com>
(cherry picked from commit b307921c3f)
2015-06-18 13:28:52 +01:00
Francisco Jerez
631414a507 i965: Don't compact instructions with unmapped bits.
Some instruction bits don't have a mapping defined to any compacted
instruction field.  If they're ever set and we end up compacting the
instruction they will be forced to zero.  Avoid using compaction in such
cases.

v2: Align multiple lines of an expression to the same column.  Change
    conditional compaction of 3-source instructions to an
    assertion. (Matt)
v3: The 3-source instruction bit 105 is part of SourceIndex on CHV.
    Add assertion that reserved bit 7 is not set. (Matt)
    Document overlap with UIP and 64-bit immediate fields.
v4: Make some more unmapped bit checks assertions. (Matt)

Reviewed-by: Matt Turner <mattst88@gmail.com>
(cherry picked from commit 35a77a148f)
2015-06-18 13:28:40 +01:00
Marek Olšák
4a2d7fbde4 egl: fix setting context flags
Cc: 10.6 10.5 10.4 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Chad Versace <chad.versace@intel.com>
(cherry picked from commit f9f894447e)
2015-06-18 12:51:08 +01:00
Emil Velikov
eb3a704bb0 docs: Add sha256sums for the 10.5.7 release
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-06-07 11:45:25 +01:00
Emil Velikov
495bcbc48c Add release notes for the 10.5.7 release
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-06-07 11:13:19 +01:00
Emil Velikov
95b81ee9e8 Update version to 10.5.7
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-06-07 11:11:38 +01:00
Ilia Mirkin
07aa6d78bb nv30: avoid doing extra work on clear and hitting unexpected states
Clearing can happen at a time when various state objects are incoherent
and not ready for a draw. Some of the validation functions don't handle
this well, so only flush the framebuffer state. This has the advantage
of also not doing extra work.

This works around some crashes that can happen when clearing.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Tobias Klausmann <tobias.johannes.klausmann@mni.thm.de>
(cherry picked from commit aba3392541)
2015-06-03 12:28:58 +01:00
Marek Olšák
70816a66d5 st/dri: fix postprocessing crash when there's no depth buffer
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89131

Cc: 10.6 10.5 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit 25e9ae2b79)
2015-06-03 12:06:08 +01:00
Ben Widawsky
f8bb4a814d i965: Emit 3DSTATE_MULTISAMPLE before WM_HZ_OP (gen8+)
Starting with GEN8, there is documentation that the multisample state command
must be emitted before the 3DSTATE_WM_HZ_OP command any time the multisample
count changes. The 3DSTATE_WM_HZ_OP packet gets emitted as a result of a
intel_hix_exec(), which is called upon a fast clear and/or a resolve. This can
happen before the state atoms are checked, and so the multisample state must be
put directly in the function.

v1:
- In v0, I was always emitting the command, but Ken came up with the condition to
determine whether or not the sample count actually changed.
- Ken's recommendation was to set brw->num_multisamples after emitting
3DSTATE_MULTISAMPLE. This doesn't work. I put my best guess as to why in the XXX
(it was causing 7 regressions on BDW).

v2:
Flag NEW_MULTISAMPLE state. As Ken found, in state upload we check for the
multisample change to determine whether or not to emit certain packets. Since
the hiz code doesn't actually care about the number of multisamples, set the
flag and let the later code take care of it.

Jenkins results:
http://otc-mesa-ci.jf.intel.com/view/dev/job/bwidawsk/136/

Fixes around 200 piglit tests on SKL. I'm somewhat surprised that it seems to
have no impact on BDW as the restriction is needed there as well.

Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Reviewed-by: Neil Roberts <neil@linux.intel.com> (v0)
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> (v2)
(cherry picked from commit e2d84d99f5)
2015-06-03 12:06:01 +01:00
Emil Velikov
8c56fc9275 cherry-ignore: add clover build fix not applicable for 10.5
The 10.5 codebase uses a compat/wrapper implementation for the string
class. As such it already explicitly includes the string header.

Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-06-03 12:05:43 +01:00
Ilia Mirkin
6338bd6f70 nv30: falling back to draw path for edgeflag does no good
The problem is that the EDGEFLAG has to be toggled at vertex submission
time. This can be done from either the draw or the regular paths. Avoid
falling back to draw just because there's an edgeflag.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 3ec1815285)
2015-06-03 11:53:33 +01:00
Ilia Mirkin
7087a507bf nv30/draw: switch varying hookup logic to know about texcoords
Commit 8acaf862df switched things over to use TEXCOORD instead of
GENERIC, but did not update the nv30 swtnl draw paths. This teaches the
draw logic about TEXCOORD.

Among other things, this fixes a crash in demos/arbocclude when using
swtnl. Curiously enough, the point-sprite piglit works without this.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 25be70462d)
2015-06-03 11:53:26 +01:00
Ilia Mirkin
93650a80e9 nv30/draw: allocate vertex buffers in gart
These are only used once per draw, so it makes sense to keep them in
GART. Also take this opportunity to modernize the buffer mapping API
usage.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Ben Skeggs <bskeggs@redhat.com>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit c3d36a2e1a)
2015-06-03 11:53:18 +01:00
Ilia Mirkin
23c0c94ac3 nv30/draw: only use the DMA1 object (GART) if the bo is not in VRAM
Instead of always having it in the data, let the bo placement decide it.
This fixes glxgears with swtnl forced on.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Ben Skeggs <bskeggs@redhat.com>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit fdad7dfbda)
2015-06-03 11:53:04 +01:00
Ilia Mirkin
efaf906415 nv30/draw: fix indexed draws with swtnl path and a resource index buffer
The map = assignment was missing.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 3600439897)
2015-06-03 11:52:57 +01:00
Ilia Mirkin
adee8f1ca5 glsl: avoid leaking linked gl_shader when there's a late linker error
This makes piglit mixing-clip-distance-and-clip-vertex-disallowed have 0
definitely lost blocks with valgrind. (Same non-0 number of possibly
lost blocks though.)

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Tobias Klausmann <tobias.johannes.klausmann@mni.thm.de>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 5646f0f18a)
2015-06-03 11:52:50 +01:00
Ilia Mirkin
56d13627eb st/mesa: don't leak glsl_to_tgsi object on link failure
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit bb973723a5)
2015-06-03 11:52:39 +01:00
Ilia Mirkin
a10c9c6725 nv30/draw: draw expects constbuf size in bytes, not vec4 units
This fixes glxgears with NV30_SWTNL=1 forced on. Probably fixes a bunch
of other situations where we fall back to the swtnl path.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 147816375d)
2015-06-03 11:52:32 +01:00
Ilia Mirkin
2d610edf5c nv30/draw: avoid leaving stale pointers in draw state
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 89585edf3c)
2015-06-03 11:52:25 +01:00
Ilia Mirkin
0f7c9815a0 nv30: fix clip plane uploads and enable changes
nv30_validate_clip depends on the rasterizer state. Also we should
upload all the new clip planes on change since next time the plane data
won't have changed, but the enables might.

This fixes fixed-clip-enables and vs-clip-vertex-enables shader tests.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Tobias Klausmann <tobias.johannes.klausmann@mni.thm.de>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 7518fc3c66)
2015-06-03 11:52:19 +01:00
Ilia Mirkin
ece96b4954 nv30: avoid leaking render state and draw shaders
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 9870ed05dd)
2015-06-03 11:52:10 +01:00
Ilia Mirkin
e692b6428c nv30: don't leak fragprog consts
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 605ce36d7f)
2015-06-03 11:51:37 +01:00
Ilia Mirkin
7d76e89053 nv50/ir: avoid messing up arg1 of PFETCH
There can be scenarios where the "indirect" arg of a PFETCH becomes
known, and so the code will attempt to propagate it. Use this
opportunity to just fold it into the first argument, and prevent the
load propagation pass from touching PFETCH further.

This fixes gs-input-array-vec4-index-rd.shader_test and
vs-output-array-vec4-index-wr-before-gs.shader_test on nvc0 at least.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Tobias Klausmann <tobias.johannes.klausmann@mni.thm.de>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit fa7f9f123b)
2015-06-03 11:51:19 +01:00
Ilia Mirkin
60e81b47e5 nvc0: a geometry shader can have up to 1024 vertices output
The 1024 is already reported everywhere, not sure where this 0x1ff came
from.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 921917c8d8)
2015-06-03 11:51:09 +01:00
Jason Ekstrand
c04aaa5d99 i965/fs: Fix implied_mrf_writes for scratch writes
We build the entire message in the generator so all the MRF writes are
implied.

Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
(cherry picked from commit 6ca67f62e8)
2015-06-03 11:50:58 +01:00
Ilia Mirkin
a642ec0551 nvc0/ir: LOAD's can't be used for shader inputs
We forgot to convert to VFETCH in case of indirect access. Fix that.

This avoids crashes on the new gs-input-array-vec4-index-rd and
vs-output-array-vec4-index-wr-before-gs but they still fail.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 217301843a)
2015-06-03 11:50:43 +01:00
Ilia Mirkin
c13b16029d nv50/ir: guess that the constant offset is the starting slot of array
When we get something like IN[ADDR[0].x+5], we will now guess that we
should look at IN[5] for the "base" information.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 0bab3962f5)
2015-06-03 11:50:32 +01:00
Ilia Mirkin
58b9e11084 nvc0/ir: set ftz when sources are floats, not just destinations
In the case of a compare, the destination might be a predicate, but we
still want to flush denorms.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit d1eea18a59)
2015-06-03 11:50:20 +01:00
Emil Velikov
d88de40d7a get-pick-list.sh: Require explicit "10.5" for nominating stable patches
A nomination unadorned with a specific version is now interpreted as
being aimed at the 10.6 branch, which was recently opened.

Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-06-03 11:49:19 +01:00
Emil Velikov
8cb28bc49d docs: Add sha256sums for the 10.5.6 release
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-05-24 10:43:31 +01:00
Emil Velikov
b1cf9cfb16 Add release notes for the 10.5.6 release
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-05-23 09:02:41 +01:00
Emil Velikov
eba7b7cf68 Update version to 10.5.6
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-05-23 08:58:02 +01:00
Emil Velikov
fe521a083f targets/osmesa: drop the -module tag from LDFLAGS
Gallium equivalent of commit 06ff751f97f(darwin: Fix install name of
libOSMesa)

Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
(cherry picked from commit 36438f0db6)
2015-05-20 22:16:56 +01:00
Jeremy Huddleston Sequoia
86b7e91ae9 darwin: Fix install name of libOSMesa
Passing -module to glibtool causes the resulting library to be called
libSomething.so rather than libSomething.dylib on darwin.

Regardless if libOSMesa is a library or a module, it has been used as
the former for quite some time. Update the build to reflect that and
resolve the naming issue.

Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
[Emil Velikov: Tweak the commit message.]
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>

(cherry picked from commit 06ff751f97)
2015-05-20 22:16:48 +01:00
Jeremy Huddleston Sequoia
0a26544985 swrast: Build fix for darwin
Fixes regression from commit 64b1dc4449

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90147
Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
CC: Emil Velikov <emil.l.velikov@gmail.com>
CC: jon.turney@dronecode.org.uk
CC: ionic@macports.org
(cherry picked from commit 5b2d3480f5)
Nominated-by: Emil Velikov <emil.l.velikov@gmail.com>

Squashed with commit

swrast: Build fix for Solaris

Fixes regression from commit 5b2d3480f5

Cc: "10.5 10.6" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
(cherry picked from commit 31cd2d75dc)
2015-05-20 22:16:18 +01:00
Emil Velikov
eb1b41302d egl/main: fix EGL_KHR_get_all_proc_addresses
The extension requires that the address of the core functions should be
available via eglGetProcAddress. Currently the list is guarded by
_EGL_GET_CORE_ADDRESSES, which was only set for the scons (windows)
build.

Unconditionally enable it for all the builds (automake, android and
haiku) considering that the extension is not platform specific and is
always enabled.

v2: Drop the _EGL_GET_CORE_ADDRESSES macro altogether.

Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
(cherry picked from commit 448e01b291)

Conflicts:
	src/egl/main/eglapi.c
2015-05-20 22:14:52 +01:00
Rob Clark
b1c045c62f freedreno: fix bug in tile/slot calculation
This was causing corruption with hw binning on a306.  Unlikely that it
is a306 specific, but rather the smaller gmem size resulted in different
tile configuration which was triggering the bug at certain resolutions.

Signed-off-by: Rob Clark <robclark@freedesktop.org>
Cc: "10.4" and "10.5" and "10.6" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 4925c35660)
2015-05-20 22:10:34 +01:00
Rob Clark
21dd729dcc freedreno: enable a306
Whitelist adreno 306 (as found in msm8916/apq8016).  Works pretty much
out of the box, although the smaller GMEM size requires more tiles to
fit 1920x1080, so bump up the max # of tiles as well.

Since it is just whitelist + trivial change, it makes sense to land on
all the active release branches.

Note that a305c ends up with gpu-id "306", hence a306 ends up with
gpu-id of "307".  Apparently that is what happens when you let the
marketing dept name things.

Cc: "10.4" and "10.5" and "10.6" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Rob Clark <robclark@freedesktop.org>
(cherry picked from commit fcc7d6323b)
2015-05-20 22:10:07 +01:00
Laura Ekstrand
104dc7447d main: Complete error conditions for glInvalidate*Framebuffer.
Signed-off-by: Fredrik Höglund <fredrik@kde.org>
Reviewed-by: Fredrik Höglund <fredrik@kde.org>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit b4368ac09d)
2015-05-20 22:09:51 +01:00
Laura Ekstrand
0475deac26 main: Fix an error generated by FramebufferTexture
gl*FramebufferTexture should generate GL_INVALID_VALUE when the
texture doesn't exist.

[Fredrik: Split this change out from the next commit]

Signed-off-by: Fredrik Höglund <fredrik@kde.org>
Reviewed-by: Fredrik Höglund <fredrik@kde.org>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 69bdc9dcb8)
2015-05-20 22:09:20 +01:00
Fredrik Höglund
7f7e74b536 mesa: Generate GL_INVALID_VALUE in framebuffer_texture when layer < 0
Signed-off-by: Fredrik Höglund <fredrik@kde.org>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 8ba7ad8abc)
2015-05-20 22:09:07 +01:00
Fredrik Höglund
a2624f2d2e main: Require that the texture exists in framebuffer_texture
Generate GL_INVALID_OPERATION if the texture hasn't been created.

Signed-off-by: Fredrik Höglund <fredrik@kde.org>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit f9f5c82284)
2015-05-20 22:08:47 +01:00
Francisco Jerez
a6e1495aa7 i965: Fix PBO cache coherency issue after _mesa_meta_pbo_GetTexSubImage().
This problem can easily be reproduced with a number of
ARB_shader_image_load_store piglit tests, which use a buffer object as
PBO for a pixel transfer operation and later on bind the same buffer
to the pipeline as shader image -- The problem is not exclusive to
images though, and is likely to affect other kinds of buffer objects
that can be bound to the 3D pipeline, including vertex, index,
uniform, atomic counter buffers, etc.

CC: 10.5 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
(cherry picked from commit d247615e0d)
2015-05-20 22:08:30 +01:00
Ilia Mirkin
005def20f4 nvc0: switch mechanism for shader eviction to be a while loop
This aligns it to work similarly to nv50. However there's no library
code there, so the whole thing can be freed. Here we end up with an
allocated node that's not attached to a specific program.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=86792
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit d06ce2f1df)
2015-05-20 22:07:00 +01:00
Alex Deucher
4ad41f45b5 radeonsi: add new bonaire pci id
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit 71ba30f778)
2015-05-20 22:06:42 +01:00
Francisco Jerez
907ee1494f clover: Implement locking of the wait_count, _chain and _status members of event.
Tested-by: Tom Stellard <thomas.stellard@amd.com>
CC: 10.5 <mesa-stable@lists.freedesktop.org>
(cherry picked from commit a533d4edf1)
2015-05-20 22:06:33 +01:00
Francisco Jerez
b4045dd5e9 clover: Wrap event::_status in a method to prevent unlocked access.
Tested-by: Tom Stellard <thomas.stellard@amd.com>
CC: 10.5 <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 4022a468b2)
2015-05-20 22:06:26 +01:00
Francisco Jerez
f25af5317f clover: Refactor event::trigger and ::abort to prevent deadlock and reentrancy issues.
Refactor ::trigger and ::abort to split out the operations that access
concurrently modified data members and require locking from the
recursive and possibly re-entrant part of these methods.  This will
avoid some deadlock situations when locking is implemented.

Tested-by: Tom Stellard <thomas.stellard@amd.com>
CC: 10.5 <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 2232b929fd)
2015-05-20 22:06:06 +01:00
Ilia Mirkin
1353ba5397 st/mesa: make sure to create a "clean" bool when doing i2b
i2b has to work for all integers, not just 1. INEG would not necessarily
result with all bits set, which is something that other operations can
rely on by e.g. using AND (or INEG for b2i).

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit 2b5355c8ab)
2015-05-20 22:05:58 +01:00
Tom Stellard
097e70202b clover: Fix a bug with multi-threaded events v2
It was possible for some events never to get triggered if one thread
was creating events and another threads was waiting for them.

This patch consolidates soft_event::wait() and hard_event::wait()
into event::wait() so that hard_event objects will now wait for
all their dependencies to be submitted before flushing the command
queue.

v2:
  - Rename variables
  - Use mutable varibales so we can keep event::wait() const
  - Open code signalled() call so mutex can be atted to signalled
    without deadlocking.

CC: 10.5 <mesa-stable@lists.freedesktop.org>

Reviewed-by: Francisco Jerez <currojerez@riseup.net>
(cherry picked from commit 9c4dc98b29)
2015-05-20 22:05:52 +01:00
Tom Stellard
5bb7ee4fd6 clover: Add a mutex to guard queue::queued_events
This fixes a potential crash where on a sequence like this:

Thread 0: Check if queue is not empty.
Thread 1: Remove item from queue, making it empty.
Thread 0: Do something assuming queue is not empty.

CC: 10.5 <mesa-stable@lists.freedesktop.org>

Reviewed-by: Francisco Jerez <currojerez@riseup.net>
(cherry picked from commit f546902d95)
2015-05-20 22:05:46 +01:00
Axel Davy
0024b85cc6 glx/dri3: Add additional check for gpu offloading case
Checks blitImage is implemented.
Initially having the __DRIimageExtension extension
at version 9 at least meant blitImage was supported.
However some implementation do advertise version >= 9
without implementing it.

CC: 10.5 <mesa-stable@lists.freedesktop.org>

Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>

Signed-off-by: Axel Davy <axel.davy@ens.fr>
(cherry picked from commit c4ff6d00cd)
2015-05-20 22:05:38 +01:00
Axel Davy
ace7d9afc9 egl/wayland: properly destroy wayland objects
the wl_registry and the wl_queue allocated weren't destroyed.

CC: 10.5 <mesa-stable@lists.freedesktop.org>

Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>

Signed-off-by: Axel Davy <axel.davy@ens.fr>
(cherry picked from commit 6aaf09b93b)
2015-05-20 22:05:31 +01:00
Marta Lofstedt
3dc08d4d89 main: glGetIntegeri_v fails for GL_VERTEX_BINDING_STRIDE
The return type for GL_VERTEX_BINDING_STRIDE is missing,
this cause glGetIntegeri_v to fail.

Signed-off-by: Marta Lofstedt <marta.lofstedt@linux.intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 4a8cd2799c)
2015-05-20 22:05:17 +01:00
Ilia Mirkin
8a3ed2dee7 nv50/ir: only enable mul saturate on G200+
Commit 44673512a8 enabled support for saturating fmul. However
experimentally this does not seem to work on the older chips. Restrict
the feature to G200 (NVA0) and later.

Reported-by: Pierre Moreau <pierre.morrow@free.fr>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90350
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Tested-by: Pierre Moreau <pierre.morrow@free.fr>
Reviewed-by: Tobias Klausmann <tobias.johannes.klausmann@mni.thm.de>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit da136dc07d)
2015-05-20 22:05:06 +01:00
Ilia Mirkin
d47d799b5f nvc0: reset the instanced elements state when doing blit using 3d engine
Since we update num_vtxelts here, we could otherwise end up with stale
instancing information in the upper bits which wouldn't otherwise get
reset. (Also we run the risk of the previous draw having set the first
element as instanced.)

This appears as one of the causes for the test pointed out in fdo#90363
to fail on nvc0.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90363
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit 7892210400)
2015-05-20 22:04:59 +01:00
Ilia Mirkin
f3804a0051 nvc0: keep track of PGRAPH state in nvc0_screen
See identical commit for nv50. Destroying the current context and then
creating a new one or switching to another existing context would cause
the "current" state to not be properly initialized, so we save it off in
the screen.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit e9b1ea29bf)
2015-05-20 22:04:52 +01:00
Ilia Mirkin
113e2bd732 nv50: keep track of PGRAPH state in nv50_screen
Normally this is kept in nv50_context, and on switching the active
context, the state is copied from the previous context. However when the
last context is destroyed, this is lost, and a new context might later
be created. When the currently-active context is destroyed, save its
state in the screen, and restore it when setting the current context.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90363
Reported-by: Matteo Bruni <matteo.mystral@gmail.com>
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Tested-by: Matteo Bruni <matteo.mystral@gmail.com>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit f617029db3)
2015-05-20 22:04:45 +01:00
Ilia Mirkin
e63d65b235 nv50/ir: only propagate saturate up if some actual folding took place
The former logic would copy the saturate up to any mul with an immediate
if there was a subsequent mul with a saturate. However we only want to
do that if we collapsed 2 muls by multiplying their immediates (or were
able to put the immediate in as a post-multiplier).

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit c4ac09e30e)
2015-05-20 22:04:33 +01:00
Roland Scheidegger
4504642ccc draw: (trivial) fix out-of-bounds vector initialization
Was off-by-one. llvm says inserting an element with an index higher than the
number of elements yields undefined results. Previously such inserts were
ignored but as of llvm revision 235854 the vector gets replaced with undef,
causing failures.
This fixes piglit gl-3.2-layered-rendering-gl-layer, as mentioned in
https://llvm.org/bugs/show_bug.cgi?id=23424.

Reviewed-by: Brian Paul <brianp@vmware.com>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit b8a1495106)
2015-05-20 22:04:24 +01:00
Tim Rowley
be7b998a1f mesa: fix shininess check for ffvertex_prog v2
Switch to using VERT_BIT_GENERIC macro, as varying_vp_inputs is a
bitmask.

Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit ce01c0af70)
2015-05-20 22:04:06 +01:00
Emil Velikov
8ee1a1c08b docs: Add sha256 sums for the 10.5.5 release
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-05-11 22:02:03 +01:00
Emil Velikov
d88fb40505 Add release notes for the 10.5.5 release
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-05-11 20:19:33 +01:00
Emil Velikov
0d425c413f Update version to 10.5.5
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-05-11 20:14:23 +01:00
Boyan Ding
536003c11e i965: Add XRGB8888 format to intel_screen_make_configs
Some application, such as drm backend of weston, uses XRGB8888 config as
default. i965 doesn't provide this format, but before commit 65c8965d,
the drm platform of EGL takes ARGB8888 as XRGB8888. Now that commit
65c8965d makes EGL recognize format correctly so weston won't start
because it can't find XRGB8888. Add XRGB8888 format to i965 just as
other drivers do.

Cc: mesa-stable@lists.freedesktop.org
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89689
Signed-off-by: Boyan Ding <boyan.j.ding@gmail.com>
Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
(cherry picked from commit 28090b30dd)
2015-05-05 15:05:50 +01:00
Ilia Mirkin
071402cc3b nv50/ir: fix asFlow() const helper for OP_JOIN
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit db269ae495)
2015-05-05 15:05:42 +01:00
Ilia Mirkin
b4c973da03 nvc0/ir: fix predicated PFETCH emission
src1 would contain the predicate, which would get emitted as a register
source by an undiscerning srcId helper. Work around this in the same way
as in emitTEX.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit a9d08a250a)

Squashed with commit

nvc0/ir: fix predicated PFETCH for real

Commit a9d08a250 accidentally didn't make use of the new src1 variable.
Use it.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit 33f0d1138d)
2015-05-05 15:05:03 +01:00
Ilia Mirkin
a27d03dd9f gk110/ir: fix set with a register dest to not auto-set the abs flag
This was causing src0 to always have the absolute value flag set.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit 515ac907e6)
2015-05-05 14:58:58 +01:00
Emil Velikov
0b06729591 r300: do not link against libdrm_intel
Accidentally added since the introduction of the file.

Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
(cherry picked from commit b124dc2b70)
2015-05-05 14:58:46 +01:00
Ilia Mirkin
dfe88d4f52 nvc0/ir: flush denorms to zero in non-compute shaders
This will set the FTZ flag (flush denorms to zero) on all opcodes that
can take it.

This resolves issues in Unigine Heaven 4.0 where there were solid-filled
boxes popping up.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89455
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
(cherry picked from commit 6fe0d4f035)
2015-05-05 14:58:39 +01:00
Kenneth Graunke
1176e5862a i965: Disallow linear blits that are not cacheline aligned.
The BLT engine on Gen8+ requires linear surfaces to be cacheline
aligned.  This restriction was added as part of converting the BLT to
use 48-bit addressing.

The main user, intel_emit_linear_blit, now handles this properly.
But we might also have linear miptrees; just refuse to blit those.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=88521
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit 5957da1edb)
2015-05-05 14:58:16 +01:00
Kenneth Graunke
716b5f7cbe i965: Make intel_emit_linear_blit handle Gen8+ alignment restrictions.
The BLT engine on Gen8+ requires linear surfaces to be cacheline
aligned.  This restriction was added as part of converting the BLT to
use 48-bit addressing.

intel_emit_linear_blit needs to handle blits that are not cacheline
aligned, as we use it for arbitrary glBufferSubData calls and subrange
mappings.

Since intel_emit_linear_blit uses 1 byte per pixel, we can use the src/dst
pixel X offset field to represent the unaligned portion, and subtract
that from the address so it's cacheline aligned.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=88521
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit 8c17d53823)
2015-05-05 14:56:15 +01:00
Roland Scheidegger
cf60503378 draw: fix prim ids when there's no gs
We were resetting the prim id count for each run of the prim assembler,
hence this only worked when the draw calls were very small (the exact limit
depending on the vertex size), since larger draw calls get split up.
So, do the same as we do already if there's a gs, reset it to zero explicitly
for every new instance (this possibly could use the same variable but that
isn't doable without some heavy refactoring and I'm not sure it makes sense).

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90130.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>

CC: <mesa-stable@lists.freedesktop.org>
(cherry picked from commit f2a7fd9943)
2015-05-05 14:54:47 +01:00
Emil Velikov
adb47b5b27 docs: Add sha256 sums for the 10.5.4 release
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-04-24 22:51:25 +01:00
Emil Velikov
ea0d1f575c Add release notes for the 10.5.4 release
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-04-24 22:27:09 +01:00
Emil Velikov
776e47a305 Update version to 10.5.4
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-04-24 22:22:47 +01:00
Brian Paul
e2dd554651 glsl: rewrite glsl_type::record_key_hash() to avoid buffer overflow
This should be more efficient than the previous snprintf() solution.
But more importantly, it avoids a buffer overflow bug that could result
in crashes or unpredictable results when processing very large interface
blocks.

For the app in question, key->length = 103 for some interfaces.  The check
if size >= sizeof(hash_key) was insufficient to prevent overflows of the
hash_key[128] array because it didn't account for the terminating zero.
In this case, this caused the call to hash_table_string_hash() to return
different results for identical inputs, and then shader linking failed.

This new solution also takes all structure fields into account instead
of just the first 15 when sizeof(pointer)==8.

Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
(cherry picked from commit 31667e6237)
2015-04-22 16:17:10 +01:00
Emil Velikov
0f5ec7250d android: mesa: fix the path of the SSE4_1 optimisations
Commit dd6f641303c(mesa: Build with subdir-objects.) removed the SRCDIR
variable, but forgot to update all references of it.

v2: Fix path - must be relative to LOCAL_PATH. (Chih-Wei)

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Chih-Wei Huang <cwhuang@linux.org.tw>
(cherry picked from commit 669cfc267a)
2015-04-22 16:17:03 +01:00
Mauro Rossi
a2785a23e6 android: add inital NIR build
Required by the i965 driver.

v2:
 - Split out the nir_builder_opcodes.h rules.
 - Do not unconditionally hide the python command - use $(hide)
 - Use LOCAL_EXPORT_C_INCLUDE_DIRS to manage includes for the generated
sources.

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
[Emil Velikov: Split from a larger commit, v2]
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Chih-Wei Huang <cwhuang@linux.org.tw>

(cherry picked from commit 06619749a1)
2015-04-22 16:16:57 +01:00
Emil Velikov
2ad5a068fa android: dri: link against libmesa_util
The dri modules depend on symbols provided by it.

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Chih-Wei Huang <cwhuang@linux.org.tw>
(cherry picked from commit 618885f71f)
2015-04-22 16:16:45 +01:00
Emil Velikov
1dfbc95b76 android: add $(mesa_top)/src/mesa/main to the includes list
Required by the format_{un,}pack rework. Otherwise the build will fail
to locate the respective headers - format_{un,}pack.h

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Chih-Wei Huang <cwhuang@linux.org.tw>
(cherry picked from commit 0afbd2df04)
2015-04-22 16:16:38 +01:00
Emil Velikov
215f28c695 android: add HAVE__BUILTIN_* and HAVE_FUNC_ATTRIBUTE_* defines
All of those are available on gcc 4.5 and later with the current android
build using gcc 4.7.

Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Chih-Wei Huang <cwhuang@linux.org.tw>
(cherry picked from commit 39a175e0c7)
2015-04-22 16:16:30 +01:00
Emil Velikov
95633d09e8 android: dri/common: conditionally include drm_cflags/set __NOT_HAVE_DRM_H
Otherwise we'll fail to find the drm.h header.

Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
(cherry picked from commit 8d90bfb724)
2015-04-22 16:16:23 +01:00
Emil Velikov
33c6a4c018 android: egl: add libsync_cflags to the build
... via local_shared_libraries. Otherwise the sync/sync.h header won't
be found.

Note: 10.5 and earlier will need similar change in st/egl.

v2: Append the library to the local_shared_libraries list. (Chih-Wei)

Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Chih-Wei Huang <cwhuang@linux.org.tw>
(cherry picked from commit 2d06791f6f)
2015-04-22 16:16:15 +01:00
Mauro Rossi
548fc14d36 android: mesa: generate the format_{un,}pack.[ch] sources
Missed out with commit e1fdcddafe9(mesa: Autogenerate format_unpack.c)

v2: Conditionaly print the python commands - s/@/$(hide) / (Chih-Wei)

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
[Emil Velikov: Split our from a larger commit.]
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>

(cherry picked from commit 5f7081eb90)
2015-04-22 16:16:07 +01:00
Emil Velikov
c614fef092 android: add $(mesa_top)/src include to the whole of mesa
Many parts of mesa already have the include with others depending on it
but it's missing. Add it once at the top makefile and be done with it.

Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Chih-Wei Huang <cwhuang@linux.org.tw>
(cherry picked from commit 6fb8017866)
2015-04-22 16:16:00 +01:00
Emil Velikov
8ba18d75d6 android: use LOCAL_SHARED_LIBRARIES over TARGET_OUT_HEADERS
... to manage the LIBDRM*_CFLAGS. The former is the recommended approach
by the Android build system developers while the latter has been
depreciated for quite some time.

Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
(cherry picked from commit 86919352e3)
2015-04-22 16:15:50 +01:00
Kenneth Graunke
597d85c6b7 drirc: Add "Second Life" quirk (allow_glsl_extension_directive_midshader).
Appears to fix shader compilation.  Tested by starting the client,
dragging the "quality and speed" slider back and forth, and watching the
console output - instead of piles of "shader failed to compile", the CPU
seems to be busy compiling shaders.  I haven't actually tried to play.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=69226
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=71591
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit 00bf7d2e9c)
2015-04-22 16:14:51 +01:00
Marek Olšák
ef0e335c14 glsl_to_tgsi: don't use a potentially-undefined immediate for ir_query_levels
Cc: 10.4 10.5 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit dcc74d47c4)
2015-04-22 16:14:33 +01:00
Marek Olšák
f8966a07d9 glsl_to_tgsi: fix out-of-bounds constant access and crash for uniforms
This fixes piglit shaders@glsl-fs-uniform-array-loop-unroll with immediate
shader compilation - it's a compiler test, so it has never been translated
to TGSI before.

Cc: 10.4 10.5 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit 14c5bc3b9a)
2015-04-22 16:14:05 +01:00
Kristian Høgsberg
ee63b31594 i965: Rewrite ir_tex to ir_txl with lod 0 for vertex shaders
The ir_tex opcode turns into a sample or sample_c message, which will try to
compute derivatives to determine the lod. This produces garbage for
non-fragment shaders where the sample coordinates don't correspond to
subspans.

We fix this by rewriting the opcode from ir_tex to ir_txl and setting the
lod to 0.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89457
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Kristian Høgsberg <kristian.h.kristensen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
(cherry picked from commit 993a6288f7)
2015-04-22 16:13:50 +01:00
Ian Romanick
1ebb42a6b2 nir: Fix typo in "ushr by 0" algebraic replacement
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Cc: "10.5" <mesa-stable@lists.freedestkop.org>
(cherry picked from commit bc672e261c)
2015-04-22 16:13:38 +01:00
Kenneth Graunke
a968edae6c i965: Fix software primitive restart with indirect draws.
new_prim was declared as a stack variable within a nested scope; we
tried to retain a pointer to that data beyond the scope, which is bogus.

GCC with -O1 eliminated most of the code that set new_prim's fields.

Move the declaration to fix the bug.

v2: Also fix new_ib (thanks to Matt Turner and Ben Widawsky).

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=81025
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit 406df68736)
2015-04-22 16:13:04 +01:00
Dave Airlie
f44e518364 st/mesa: align cube map arrays layers
We create textures internally for texsubimage, and we use
the values from sub image to create a new texture, however
we don't align these to valid sizes, and cube map arrays
must have an array size aligned to 6.

This fixes texsubimage cube_map_array on CAYMAN at least,
(it was causing  GPU hang and bad values), it probably
also fixes it on radeonsi and evergreen.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89957
Tested-by: Tom Stellard <thomas.stellard@amd.com>
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
(cherry picked from commit cc5860e407)
2015-04-22 16:12:51 +01:00
Dave Airlie
b2a6b01082 st/mesa: convert sub image for cube map arrays to 2d arrays for upload
Since we can subimage upload a number of cube map array layers,
that aren't a complete cube map array, we should specify things
as a 2D array and blit from that.

Suggested by Ilia Mirkin as an alternate fix for texsubimage
cube map array issues.

seems to work just as well.

Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
(cherry picked from commit 5ed79312ed)
2015-04-22 16:12:22 +01:00
Mathias Froehlich
abdae1c4d1 i965: Flush batchbuffer containing the query on glQueryCounter.
This change fixes a regression with timer queries introduced with
commit 3eb6258. There the pending batchbuffer is flushed
only if glEndQuery is executed. This present change adds such
a flush to glQueryCounter which also schedules a value query
just like glEndQuery does. The patch fixes GPU timer queries
going mad from within osgviewer.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Mathias Froehlich <Mathias.Froehlich@web.de>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit 1e1d5456ba)
2015-04-22 16:12:08 +01:00
Emil Velikov
f73954e0a7 radeonsi: remove unused si_dump_key()
Accidentally added with commit 64d0f0e3b24(radeonsi: Cache
LLVMTargetMachineRef in context instead of in screen)

Reported-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-04-22 16:10:51 +01:00
Emil Velikov
65776421fe docs: Add 256 sums for the 10.5.3 release
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-04-12 23:10:42 +01:00
Emil Velikov
c4b8bff6e2 Add release notes for the 10.5.3 release
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-04-12 22:29:06 +01:00
Emil Velikov
b17312cac2 Update version to 10.5.3
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-04-12 22:21:07 +01:00
Emil Velikov
37d925a635 dist: add the VG depedencies into the tarball
Otherwise the scons build will fail.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89905
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-04-12 22:18:30 +01:00
Emil Velikov
b3bb6b1963 configure.ac: error out if python/mako is not found when required
In case of using a distribution tarball (or a dirty git tree) one can
have the generated sources locally. Make configure.ac error out
otherwise, to alert that about the unmet requirement(s) of python/mako.

v2: Check only for a single file for each dependency.

Suggested-by: Matt Turner <mattst88@gmail.com>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
(cherry picked from commit 4008975e6f)
2015-04-09 12:28:58 +01:00
Ville Syrjälä
42854fdf2c i965: Fix URB size for CHV
Increase the device info .urb.size for CHV to match the default URB
size (192kB).

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
(cherry picked from commit 970dc23603)
2015-04-08 19:56:45 +01:00
Marcin Ślusarz
64bb117f6d nouveau: synchronize "scratch runout" destruction with the command stream
When nvc0_push_vbo calls nouveau_scratch_done it does not mean
scratch buffers can be freed immediately. It means "when hardware
advances to this place in the command stream the scratch buffers
can be freed".

To fix it, just postpone scratch runout destruction after current
fence is signalled.

The bug existed for a very long time. Nobody noticed, because
"scratch runout" code path is rarely executed.

Fixes hang at the very beginning of first mission in "Serious Sam 3"
on nve7/gk107. It manifested as:

nouveau E[   PFIFO][0000:01:00.0] read fault at 0x000a9e0000 [PTE] from GR/GPC0/PE_2 on channel 0x007f853000 [Sam3[17056]]

Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
(cherry picked from commit f9e2295560)
2015-04-08 19:09:02 +01:00
Ilia Mirkin
b7c44cd5ab nv50,nvc0: limit the y-tiling of 3d textures to the first level's tiling
We limit y-tiling to 0x20 when depth is involved. However the function is
run for each miplevel, and the hardware expects miplevel 0 to have the
highest tiling settings. Perform the y-tiling limit on all levels of a
3d texture, not just the ones that have depth.

Fixes:
  texelFetch fs sampler3D 98x129x1-98x129x9

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Tested-by: Nick Tenney <nick.tenney@gmail.com> # GT216
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit ae720c66cb)
2015-04-08 18:45:33 +01:00
Iago Toral Quiroga
a02e05f0fa i965: Do not render primitives in non-zero streams then TF is disabled
Haswell hardware seems to ignore Render Stream Select bits from
3DSTATE_STREAMOUT packet when the SOL stage is disabled even if
the PRM says otherwise. Because of this, all primitives are sent
down the pipeline for rasterization, which is wrong. If SOL is
enabled, Render Stream Select is honored and primitives bound to
non-zero streams are discarded after stream output.

Since the only purpose of primives sent to non-zero streams is to
be recorded by transform feedback, we can simply discard all geometry
bound to non-zero streams then transform feedback is disabled
to prevent it from ever reaching the rasterization stage.

Notice that this patch introduces a small change in the behavior we
get when a geometry shader emits more vertices than the maximum declared:
before, a vertex that was emitted to a non-zero stream when TF was
disabled would still count for the purposes of checking that we don't
exceed the maximum number of output vertices declared by the shader. With
this change, these vertices are completely ignored and won't increase
the output vertex count, making more room for other (hopefully more
useful) vertices.

Fixes piglit test arb_gpu_shader5-emitstreamvertex_nodraw on Haswell
and Broadwell.

v2 (Ken): Drop is_haswell check in favor of doing this unconditionally.
Broadwell needs the workaround as well, and it doesn't hurt to do it in
general.  Also tweak comments - the Haswell PRM does actually mention
this ("Command Reference: Instructions" page 797).

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=83962
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit 2042a2f961)
2015-04-08 18:45:33 +01:00
Kenneth Graunke
11e7ae0741 i965: Add forgotten multi-stream code to Gen8 SOL state.
Fixes Piglit's arb_gpu_shader5-xfb-streams-without-invocations.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Chris Forbes <chrisf@ijw.co.nz>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit f368d0fa1f)
2015-04-08 18:45:33 +01:00
Kenneth Graunke
b6ce705dac i965: Fix instanced geometry shaders on Gen8+.
Jordan added this in commit 741782b594 for
Gen7 platforms.  I missed this when adding the Broadwell code.

Fixes Piglit's spec/arb_gpu_shader5/invocation-id-{basic,in-separate-gs}
with MESA_EXTENSION_OVERRIDE=GL_ARB_gpu_shader5 set.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Chris Forbes <chrisf@ijw.co.nz>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit f9e5dc0a85)
2015-04-08 18:45:33 +01:00
Ilia Mirkin
71367a1bb4 nv50: allocate more offset space for occlusion queries
Commit 1a170980a0 started writing to q->data[4]/[5] but kept the
per-query space at 16, which meant that in some cases we would write
past the end of the buffer. Rotate by 32, like nvc0 does. This ensures
that we always have 32 bytes in front of us, and the data writes will go
within the allocated space.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89679
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Tested-by: Nick Tenney <nick.tenney@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Tobias Klausmann <tobias.johannes.klausmann@mni.thm.de>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit ba353935a3)
2015-04-08 18:45:33 +01:00
Emil Velikov
f3b2698f6a xmlpool: don't forget to ship the MOS
This will allow us to finally remove python from the build time
dependencies list. Considering that you're building from a release
tarball of course :-)

Cc: Bernd Kuhls <bernd.kuhls@t-online.de>
Reported-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
(cherry picked from commit a665b9b3c8)
2015-04-08 18:45:32 +01:00
Tom Stellard
88e05a251f clover: Return CL_BUILD_ERROR for CL_PROGRAM_BUILD_STATUS when compilation fails v2
v2:
  - Don't use _errs map

Cc: 10.5 10.4 <mesa-stable@lists.freedesktop.org>

Reviewed-by: Francisco Jerez <currojerez@riseup.net>
(cherry picked from commit fda7558057)
2015-04-08 18:45:32 +01:00
Michel Dänzer
64d0f0e3b2 radeonsi: Cache LLVMTargetMachineRef in context instead of in screen
Fixes a crash in genymotion with several threads compiling shaders
concurrently.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89746

Cc: 10.5 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
(cherry picked from commit d64adc3a79)

Conflicts:
	src/gallium/drivers/radeonsi/si_shader.c
2015-04-08 18:45:32 +01:00
Ilia Mirkin
9a3a2479d4 freedreno/a3xx: point size should not be divided by 2
The division is probably a holdover from the days when the fixed point
inline functions generated by headergen were broken.

Also reduce the maximum point size to 4092 (vs 4096), which is what the
blob does.

Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
(cherry picked from commit 7fc5da8b93)
2015-04-08 18:45:32 +01:00
Ilia Mirkin
61fc1295af freedreno/a3xx: fix 3d texture layout
The SZ2 field contains the layer size of a lower miplevel. It only
contains 4 bits, which limits the maximum layer size it can describe. In
situations where the next miplevel would be too big, the hardware
appears to keep minifying the size until it hits one of that size.
Unfortunately the hardware's ideas about sizes can differ from
freedreno's which can still lead to issues. Minimize those by stopping
to minify as soon as possible.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 738c8319ac)
2015-04-08 18:45:32 +01:00
Ilia Mirkin
ac46cf31fc nv50/ir/gk110: fix offset flag position for TXD opcode
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
(cherry picked from commit 58030a8f99)
2015-04-08 18:45:32 +01:00
Ilia Mirkin
c8d962c205 nv50/ir: take postFactor into account when doing peephole optimizations
Multiply operations can have a post-factor on them, which other ops
don't support. Only perform the peephole optimizations when there is no
post-factor involved.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89758
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
(cherry picked from commit 49b86007aa)
2015-04-08 18:45:32 +01:00
Ilia Mirkin
407365e375 st/mesa: update arrays when the current attrib has been updated
Fixes the recently-sent gl-2.0-vertex-const-attr piglit test. Makes sure
to revalidate arrays when only the current attribute has been updated
via glVertexAttrib*.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89754
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 9d1b5febb6)
2015-04-08 18:45:32 +01:00
Dave Airlie
712466fb53 st_glsl_to_tgsi: only do mov copy propagation on temps (v2)
Don't propagate ARRAYs

This should fix:
https://bugs.freedesktop.org/show_bug.cgi?id=89759

v2: just specify arrays so we get input propagation
Signed-off-by: Dave Airlie <airlied@redhat.com>
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
(cherry picked from commit 91e3533481)
2015-04-08 18:45:32 +01:00
Emil Velikov
ff87ae1e00 docs: Add sha256 sums for the 10.5.2 release
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-28 18:50:31 +00:00
Emil Velikov
5e59f895c4 Add release notes for the 10.5.2 release
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-28 18:42:51 +00:00
Emil Velikov
ebbfa79755 Update version to 10.5.2
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-28 18:36:03 +00:00
Emil Velikov
fda3bc1e03 cherry-ignore: add commit non applicable for 10.5
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-28 18:33:58 +00:00
Samuel Iglesias Gonsalvez
e98909b056 configure: Introduce new output variable to ax_check_python_mako_module.m4
This output variables gives more flexibility for future changes
in autoconf to detect if it is needed to auto-generate files and
check for the auto-generation dependencies.

It is still returning error when Python is not installed.

Signed-off-by: Samuel Iglesias Gonsalvez <siglesias@igalia.com>
Reviewed-by: Kai Wasserbäch <kai@dev.carbon-project.org>
(cherry picked from commit ced9425327)

Squashed with commit

configure.ac: move AC_MSG_RESULT reporting back into the m4 macro

The one who does AC_MSG_CHECKING should provide the AC_MSG_RESULT.

Fixes: ced9425327 (configure: Introduce new output variable to
ax_check_python_mako_module.m4"

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89328
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Samuel Iglesias Gonsalvez <siglesias@igalia.com>
(cherry picked from commit 248eb54eb6)
2015-03-26 01:47:12 +00:00
Anuj Phogat
d83d2ea9a6 glsl: Generate link error for non-matching gl_FragCoord redeclarations
in different fragment shaders. This also applies to a case when gl_FragCoord
is redeclared with no layout qualifiers in one fragment shader and not
declared but used in other fragment shader.

Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Khronos Bug#12957
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Chris Forbes <chrisf@ijw.co.nz>

(cherry picked from commit d8208312a3)
2015-03-26 01:47:12 +00:00
Mario Kleiner
d6413ed98f mapi: Make private copies of name strings provided by client.
glXGetProcAddress("glFoo") ends up in stub_add_dynamic() to
create dynamic stubs for dynamic functions. stub_add_dynamic()
doesn't store the caller provided name string "Foo" in a mesa
private copy, but just stores a pointer to the "glFoo" string
passed to glXGetProcAddress - a pointer into arbitrary memory
outside mesa's control.

If the caller passes some dynamically allocated/changing
memory buffer to glXGetProcAddress(), or the caller gets unmapped
from memory, e.g., some dynamically loaded application
plugin which uses OpenGL, this ends badly - with a dangling
pointer.

strdup() the name string provided by the client to avoid
this problem.

Cc: "10.3 10.4 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit 1110113a7f)
2015-03-26 01:47:11 +00:00
Tom Stellard
3147f0bd25 clover: Return 0 as storage size for local kernel args that are not set v2
The storage size for local kernel args can be queried before the
arguments are set by using the CL_KERNEL_LOCAL_MEM_SIZE param
of clGetKernelWorkGroupInfo().

The spec says that if local kernel arguments have not been specified,
then we should assume their size is 0.

v2:
  - Implement using c++11 member initialization.

Reviewed-by: Jan Vesely <jan.vesely@rutgers.edu>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>

Cc: 10.5 10.4 <mesa-stable@lists.freedesktop.org>
(cherry picked from commit dfb1ae9d91)
2015-03-26 01:47:11 +00:00
Tapani Pälli
c2760f0a16 glsl: fix names in lower_constant_arrays_to_uniforms
Patch changes lowering pass to use unique name for each uniform
so that arrays from different stages cannot end up having same
name.

v2: instead of global counter, use pointer to achieve
    unique name (Kenneth Graunke)

Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89590
Reviewed-by: Chris Forbes <chrisf@ijw.co.nz>
Cc: 10.5 10.4 <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 3cf99701ba)
2015-03-26 01:47:11 +00:00
Francisco Jerez
859b4afc97 i965: Set nr_params to the number of uniform components in the VS/GS path.
Both do_vs_prog and do_gs_prog initialize brw_stage_prog_data::nr_params to
the number of uniform *vectors* required by the shader rather than the number
of uniform components, contradicting the comment.  This is inconsistent with
what the state upload code and scalar path expect but it happens to work until
Gen8 because vec4_visitor interprets it as a number of vectors on construction
and later on overwrites its original value with the number of uniform
components referenced by the shader.

Also there's no need to add the number of samplers, they're not actually
passed in as uniforms.

Fixes a memory corruption issue on BDW with SIMD8 VS.

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
(cherry picked from commit fd149628e1)
[Emil Velikov: s/DIV_ROUND_UP/CEILING/]
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-26 01:46:35 +00:00
Marek Olšák
d33bf8150f radeonsi: increase coords array size for radeon_llvm_emit_prepare_cube_coords
radeon_llvm_emit_prepare_cube_coords uses coords[4] in some cases (TXB2 etc.)

Discovered by Coverity. Reported by Ilia Mirkin.

Cc: 10.5 10.4 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
(cherry picked from commit a984abdad3)
2015-03-25 21:48:25 +00:00
Mario Kleiner
8ebda1f144 glx: Handle out-of-sequence swap completion events correctly. (v2)
The code for emitting INTEL_swap_events swap completion
events needs to translate from 32-Bit sbc on the wire to
64-Bit sbc for the events and handle wraparound accordingly.

It assumed that events would be sent by the server in the
order their corresponding swap requests were emitted from
the client, iow. sbc count should be always increasing. This
was correct for DRI2.

This is not always the case under the DRI3/Present backend,
where the Present extension can execute presents and send out
completion events in a different order than the submission
order of the present requests, due to client code specifying
targetMSC target vblank counts which are not strictly
monotonically increasing. This confused the wraparound
handling. This patch fixes the problem by handling 32-Bit
wraparound in both directions. As long as successive swap
completion events real 64-Bit sbc's don't differ by more
than 2^30, this should be able to do the right thing.

How this is supposed to work:

awire->sbc contains the low 32-Bits of the true 64-Bit sbc
of the current swap event, transmitted over the wire.

glxDraw->lastEventSbc contains the low 32-Bits of the 64-Bit
sbc of the most recently processed swap event.

glxDraw->eventSbcWrap is a 64-Bit offset which tracks the upper
32-Bits of the current sbc. The final 64-Bit output sbc
aevent->sbc is computed from the sum of awire->sbc and
glxDraw->eventSbcWrap.

Under DRI3/Present, swap completion events can be received
slightly out of order due to non-monotic targetMsc specified
by client code, e.g., present request submission:

Submission sbc:   1   2   3
targetMsc:        10  11  9

Reception of completion events:
Completion sbc:   3   1   2

The completion sequence 3, 1, 2 would confuse the old wraparound
handling made for DRI2 as 1 < 3 --> Assumes a 32-Bit wraparound
has happened when it hasn't.

The client can queue multiple present requests, in the case of
Mesa up to n requests for n-buffered rendering, e.g., n =  2-4 in
the current Mesa GLX DRI3/Present implementation. In the case of
direct Pixmap presents via xcb_present_pixmap() the number n is
limited by the amount of memory available.

We reasonably assume that the number of outstanding requests n is
much less than 2 billion due to memory contraints and common sense.
Therefore while the order of received sbc's can be a bit scrambled,
successive 64-Bit sbc's won't deviate by much, a given sbc may be
a few counts lower or higher than the previous received sbc.

Therefore any large difference between the incoming awire->sbc and
the last recorded glxDraw->lastEventSbc will be due to 32-Bit
wraparound and we need to adapt glxDraw->eventSbcWrap accordingly
to adjust the upper 32-Bits of the sbc.

Two cases, correponding to the two if-statements in the patch:

a) Previous sbc event was below the last 2^32 boundary, in the previous
glxDraw->eventSbcWrap epoch, the new sbc event is in the next 2^32
epoch, therefore the low 32-Bit awire->sbc wrapped around to zero,
or close to zero --> awire->sbc is apparently much lower than the
glxDraw->lastEventSbc recorded for the previous epoch

--> We need to increment glxDraw->eventSbcWrap by 2^32 to adjust
the current epoch to be one higher than the previous one.

--> Case a) also handles the old DRI2 behaviour.

b) Previous sbc event was above closest 2^32 boundary, but now a
late event from the previous 2^32 epoch arrives, with a true sbc
that belongs to the previous 2^32 segment, so the awire->sbc of
this late event has a high count close to 2^32, whereas
glxDraw->lastEventSbc is closer to zero --> awire->sbc is much
greater than glXDraw->lastEventSbc.

--> We need to decrement glxDraw->eventSbcWrap by 2^32 to adjust
the current epoch back to the previous lower epoch of this late
completion event.

We assume such a wraparound to a higher (a) epoch or lower (b)
epoch has happened if awire->sbc and glxDraw->lastEventSbc differ
by more than 2^30 counts, as such a difference can only happen
on wraparound, or if somehow 2^30 present requests would be pending
for a given drawable inside the server, which is rather unlikely.

v2: Explain the reason for this patch and the new wraparound handling
    much more extensive in commit message, no code change wrt. initial
    version.

Cc: "10.3 10.4 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
(cherry picked from commit cc5ddd584d)
2015-03-25 21:46:50 +00:00
Emil Velikov
0410d9b186 auxiliary/os: fix the android build - s/drm_munmap/os_munmap/
Squash this silly typo introduced with commit c63eb5dd5ec(auxiliary/os: get
the mmap/munmap wrappers working with android)

Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit 55f0c0a29f)
2015-03-25 21:35:32 +00:00
Emil Velikov
af3e6e2858 loader: include <sys/stat.h> for non-sysfs builds
Required by fstat(), otherwise we'll error out due to implicit function
declaration.

Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89530
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reported-by: Vadim Rutkovsky <vrutkovs@redhat.com>
Tested-by: Vadim Rutkovsky <vrutkovs@redhat.com>
(cherry picked from commit 771cd266b9)
2015-03-25 21:33:55 +00:00
Felix Janda
29810e43da c11/threads: Use PTHREAD_MUTEX_RECURSIVE by default
Previously PTHREAD_MUTEX_RECURSIVE_NP had been used on linux for
compatibility with old glibc. Since mesa defines __GNU_SOURCE__
on linux PTHREAD_MUTEX_RECURSIVE is also available since at least
1998. So we can unconditionally use the portable version
PTHREAD_MUTEX_RECURSIVE.

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=88534
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
(cherry picked from commit aead7fe2e2)
2015-03-25 21:33:26 +00:00
Rob Clark
2e0f2ad59c freedreno: update generated headers
Fix a3xx texture layer-size.

Signed-off-by: Rob Clark <robclark@freedesktop.org>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit e92bc6b38e)
2015-03-25 21:32:57 +00:00
Ilia Mirkin
411f975a81 freedreno: fix slice pitch calculations
For example if width were 65, the first slice would get 96 while the
second would get 32. However the hardware appears to expect the second
pitch to be 64, based on halving the 96 (and aligning up to 32).

This fixes texelFetch piglit tests on a3xx below a certain size. Going
higher they break again, but most likely due to unrelated reasons.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Rob Clark <robclark@freedesktop.org>
(cherry picked from commit 620e29b748)
2015-03-25 21:32:01 +00:00
Ilia Mirkin
3fa76f3f79 freedreno/a3xx: use the same layer size for all slices
We only program in one layer size per texture, so that means that all
levels must share one size. This makes the piglit test

bin/texelFetch fs sampler2DArray

have the same breakage as its non-array version instead of being
completely off, and makes

bin/ext_texture_array-gen-mipmap

start passing.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Rob Clark <robclark@freedesktop.org>
(cherry picked from commit 89b26d5a36)
2015-03-25 21:30:29 +00:00
Samuel Iglesias Gonsalvez
5e572b1cce glsl: optimize (0 cmp x + y) into (-x cmp y).
The optimization done by commit 34ec1a24d did not take it into account.

Fixes:

dEQP-GLES3.functional.shaders.random.all_features.fragment.20

Signed-off-by: Samuel Iglesias Gonsalvez <siglesias@igalia.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit b43bbfa90a)
2015-03-25 21:30:06 +00:00
Emil Velikov
2beab3c01c st/egl: don't ship the dri2.c link at the tarball
During 'make dist' the path of the symbolic link (x11/dri2.c) becomes
too long, and tar converts it to hard one. To make it more complicated
on Haiku tar errors out (due to lack of hardlink support) rather than
falling back to the next best thing.
So remove the symlink from git, and disable the scons x11_drm egl code.
The offending code is not build with either automake nor android.

Brian, Jose would you have any objections against this ? I was
playing around to get the symlink resolved, although I could not get the
dependency tracking resolved, so env.Command() was never executed :-\

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89680
Cc: mesa-stable@lists.freedesktop.org
Cc: Alexander von Gluck IV <kallisti5@unixzen.com>
Cc: Brian Paul <brianp@vmware.com>
Acked-by: Jose Fonseca <jfonseca@vmware.com>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-25 21:28:08 +00:00
Emil Velikov
d80bc65016 automake: add missing egl files to the tarball
Namely the Haiku EGL driver backend and the SConscript for the dri2 EGL
driver backend.

Cc: Alexander von Gluck IV <kallisti5@unixzen.com>
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
(cherry picked from commit 5dc573e5de)
2015-03-25 21:04:19 +00:00
Emil Velikov
2abba086ca docs: Add sha256 sums for the 10.5.1 release
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-13 23:32:12 +00:00
Emil Velikov
11c0ff60ef Add release notes for the 10.5.1 release
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-13 22:35:01 +00:00
Emil Velikov
0f32ac3920 Update version to 10.5.1
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-13 22:32:35 +00:00
Rob Clark
ce13666feb freedreno/ir3: fix failed assert in grouping
Turns out there are scenarios where we need to insert mov's in "front"
of an input.  Triggered by shaders like:

  VERT
  DCL IN[0]
  DCL IN[1]
  DCL OUT[0], POSITION
  DCL OUT[1], GENERIC[9]
  DCL SAMP[0]
  DCL TEMP[0], LOCAL
    0: MOV TEMP[0].xy, IN[1].xyyy
    1: MOV TEMP[0].w, IN[1].wwww
    2: TXF TEMP[0], TEMP[0], SAMP[0], 1D_ARRAY
    3: MOV OUT[1], TEMP[0]
    4: MOV OUT[0], IN[0]
    5: END

Signed-off-by: Rob Clark <robclark@freedesktop.org>
(cherry picked from commit 27648efa20)
2015-03-12 12:45:48 +00:00
Rob Clark
065a24bd96 freedreno/ir3: handle flat bypass for a4xx
We may not need this for later a4xx patchlevels, but we do at least need
this for patchlevel 0.  Bypass bary.f for fetching varyings when flat
shading is needed (rather than configure via cmdstream).  This requires
a special dummy bary.f w/ (ei) flag to signal to scheduler when all
varyings are consumed.  And requires shader variants based on rasterizer
flatshade state to handle TGSI_INTERPOLATE_COLOR.

Signed-off-by: Rob Clark <robclark@freedesktop.org>
(cherry picked from commit e9f2abe349)
2015-03-12 12:45:48 +00:00
Rob Clark
1dec8bbb8e freedreno/ir3: add support for memory (cat6) instructions
Scheduled basically the same as texture (cat5) instructions, using (sy)
flag for synchronization.

Signed-off-by: Rob Clark <robclark@freedesktop.org>
(cherry picked from commit 9d732d3125)
2015-03-12 12:45:45 +00:00
Rob Clark
af4d1096f7 freedreno/ir3: fix up cat6 instruction encodings
I think there is at least one more sub-encoding, but these two should be
enough to cover the common load/store instructions.

Signed-off-by: Rob Clark <robclark@freedesktop.org>
(cherry picked from commit 20b50a0712)
2015-03-12 12:38:54 +00:00
Rob Clark
645d7f464b freedreno/a4xx: aniso filtering
Signed-off-by: Rob Clark <robclark@freedesktop.org>
(cherry picked from commit dd70e78674)
2015-03-12 12:38:34 +00:00
Rob Clark
80c4ba0c83 freedreno: update generated headers
Signed-off-by: Rob Clark <robclark@freedesktop.org>
(cherry picked from commit c70097ae86)
2015-03-12 12:38:21 +00:00
Rob Clark
aca5fdae06 freedreno/a4xx: set PC_PRIM_VTX_CNTL.VAROUT properly
Fixes xonotic, some webgl stuff, and really pretty much anything with
more than 4 varyings.

Signed-off-by: Rob Clark <robclark@freedesktop.org>
(cherry picked from commit 51e335742e)
2015-03-12 12:37:54 +00:00
Rob Clark
7abc57b669 freedreno: update generated headers
Signed-off-by: Rob Clark <robclark@freedesktop.org>
(cherry picked from commit fb1301e40a)

Conflicts:
	src/gallium/drivers/freedreno/a3xx/a3xx.xml.h
2015-03-12 12:37:08 +00:00
Rob Clark
20ea65beb3 freedreno/a4xx: bit of cleanup
Signed-off-by: Rob Clark <robclark@freedesktop.org>
(cherry picked from commit bdf023482a)
2015-03-12 12:34:50 +00:00
Rob Clark
38777e1345 freedreno/a2xx: fix increment in assert
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=88883
Signed-off-by: Rob Clark <robclark@freedesktop.org>
(cherry picked from commit 6855226653)
2015-03-12 12:33:42 +00:00
Iago Toral Quiroga
4de2f25083 i965: Fix out-of-bounds accesses into pull_constant_loc array
The piglit test glsl-fs-uniform-array-loop-unroll.shader_test was designed
to do an out of bounds access into an uniform array to make sure that we
handle that situation gracefully inside the driver, however, as Ken describes
in bug 79202, Valgrind reports that this is leading to an out-of-bounds access
in fs_visitor::demote_pull_constants().

Before accessing the pull_constant_loc array we should make sure that
the uniform we are trying to access is valid.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=79202
Reviewed-by: Matt Turner <mattst88@gmail.com>
(cherry picked from commit 6ac1bc90c4)
Nominated-by: Matt Turner <mattst88@gmail.com>
2015-03-11 18:37:18 +00:00
Kenneth Graunke
fbd06fe65c i965/fs: Don't issue FB writes for bound but unwritten color targets.
We used to loop over all color attachments, and emit FB writes for each
one, even if the shader didn't write to a corresponding output variable.
Those color attachments would be filled with garbage (undefined values).

Football Manager binds a framebuffer with 4 color attachments, but draws
to it using a shader that only writes to gl_FragData[0..2].  This meant
that color attachment 3 would be filled with garbage, resulting in
rendering artifacts.  Now we skip writing to it, fixing rendering.

Writes to gl_FragColor initialize outputs[0..nr_color_regions-1] to
GRFs, while writes to gl_FragData[i] initialize outputs[i].

Thanks to Jason Ekstrand for tracking this down.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=86747
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit e95969cd95)

Conflicts:
	src/mesa/drivers/dri/i965/brw_fs_visitor.cpp
2015-03-11 18:35:39 +00:00
Kenneth Graunke
c232d765af i965/fs: Make emit_shader_time_end() insert before EOT.
Previously, we emitted the shader-time epilogue from emit_fb_writes(),
during the middle of looping through color regions (or emit_urb_writes
for the VS).  This is duplicated several times and rather awkward.

I need to fix a bug in our FB write handling, and it will be a lot
easier if we move emit_shader_time_end() out of there.

Now, we simply emit FB writes/URB writes, and subsequently have
emit_shader_time_end() insert instructions before the final SEND with
EOT.  Not only is this simpler, it's actually a slight improvement:
we now include the MOVs to set up the final FB write payload in our
shader-time measurements.

Note that INTEL_DEBUG=shader_time only exists on Gen7+, and uses
send-from-GRF.  (In the past, we might have hit trouble where both
attempt to use MRFs for messages; that's not a problem now.)

v2: Rebase on v3 of the previous patch and other shader_time fixes.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com> [v1]
Acked-by: Matt Turner <mattst88@gmail.com>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit 4ebeb71573)

Conflicts:
	src/mesa/drivers/dri/i965/brw_fs.cpp
2015-03-11 18:29:49 +00:00
Kenneth Graunke
0d625e1ae7 i965/fs: Make get_timestamp() pass back the MOV rather than emitting it.
This makes another part of the INTEL_DEBUG=shader_time code emittable
at arbitrary locations, rather than just at the end of the instruction
stream.

v2: Don't lose smear!  Caught by Topi Pohjolainen.
v3: Don't set smear on the destination of the MOV.  Thanks Topi!

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit e43af8d09f)
2015-03-11 18:23:21 +00:00
Kenneth Graunke
e9e182658d i965/fs: Make emit_shader_time_write return rather than emit.
Instead of emit_shader_time_write, we now do emit(SHADER_TIME_ADD(...)).
The advantage is that we can also insert a shader time write at an
arbitrary location in the instruction stream, rather than being
restricted to emitting at the end.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit bea854c7f3)
2015-03-11 18:23:15 +00:00
Kenneth Graunke
82ef4994dd i965/fs: Set smear on shader_time diff register.
The ADD(diff, diff, fs_reg(-2u)) instruction reads diff, which is a
width 1 register.  We need to read it as <0,1,0> with a subreg of 0,
which is what smear accomplishes.

Fixes assertion:
brw_eu_emit.c:285: validate_reg: Assertion `hstride == 0' failed.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=86974
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit f1adc45dbe)

Conflicts:
	src/mesa/drivers/dri/i965/brw_fs.cpp
2015-03-11 18:22:08 +00:00
Kenneth Graunke
c3fc8b2870 i965/fs: Set force_writemask_all on shader_time instructions.
These computations don't have anything to do with the currently
executing channels, so they should use force_writemask_all.

This fixes assert failures.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=86974
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit ef9cc7d0c1)

Conflicts:
	src/mesa/drivers/dri/i965/brw_fs.cpp
2015-03-11 18:21:10 +00:00
Marek Olšák
aea510a95f r300g: fix sRGB->sRGB blits
Cc: 10.5 10.4 <mesa-stable@lists.freedesktop.org>
(cherry picked from commit c939231e72)
2015-03-11 18:14:43 +00:00
Marek Olšák
c898d5c931 r300g: fix a crash when resolving into an sRGB texture
Cc: 10.5 10.4 <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 9953586af2)
2015-03-11 18:14:38 +00:00
Marek Olšák
32a7f119dc r300g: fix RGTC1 and LATC1 SNORM formats
Cc: 10.5 10.4 <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 74a757f92f)
2015-03-11 18:14:31 +00:00
Stefan Dösinger
578ac079c7 r300g: Fix the ATI1N swizzle (RGTC1 and LATC1)
This fixes the GL_COMPRESSED_RED_RGTC1 part of piglit's rgtc-teximage-01
test as well as the precision part of Wine's 3dc format test (fd.o bug
89156).

The Z component seems to contain a lower precision version of the
result, probably a temporary value from the decompression computation.
The Y and W component contain different data that depends on the input
values as well, but I could not make sense of them (Not that I tried
very hard).

GL_COMPRESSED_SIGNED_RED_RGTC1 still seems to have precision problems in
piglit, and both formats are affected by a compiler bug if they're
sampled by the shader with a swizzle other than .xyzw. Wine uses .xxxx,
which returns random garbage.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89156
Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Cc: 10.5 10.4 <mesa-stable@lists.freedesktop.org>
(cherry picked from commit f710b99071)
2015-03-11 18:14:25 +00:00
Rob Clark
0ea3c150cf freedreno/ir3: fix silly typo for binning pass shaders
Was resulting in gl_PointSize write being optimized out, causing
particle system type shaders to hang if hw binning enabled.

Fixes neverball, OGLES2ParticleSystem, etc.

Signed-off-by: Rob Clark <robclark@freedesktop.org>
(cherry picked from commit 60096ed906)
2015-03-11 18:12:31 +00:00
Ilia Mirkin
b542424ab8 freedreno/ir3: get the # of miplevels from getinfo
This fixes ARB_texture_query_levels to actually return the desired
value.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Rob Clark <robclark@freedesktop.org>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit cb3eb43ad6)
2015-03-11 18:12:19 +00:00
Ilia Mirkin
d8ed6aa474 freedreno/ir3: fix array count returned by TXQ
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Rob Clark <robclark@freedesktop.org>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 8ac957a51c)
2015-03-11 18:12:14 +00:00
Ilia Mirkin
5b1bd4fc9e freedreno: move fb state copy after checking for size change
Fixes: 1f3ca56b ("freedreno: use util_copy_framebuffer_state()")
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Rob Clark <robclark@freedesktop.org>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit f3dfe6513c)
2015-03-11 18:12:08 +00:00
Kenneth Graunke
cddbb3a7ba glsl: Mark array access when copying to a temporary for the ?: operator.
Piglit's spec/glsl-1.20/compiler/structure-and-array-operations/
array-selection.vert test contains the following code:

   gl_Position = (pick_from_a_or_b ? a : b)[i];

where "a" and "b" are uniform vec4[2] variables.

ast_to_hir creates a temporary vec4[2] variable, conditional_tmp, and
generates an if-block to copy one or the other:

   (declare (temporary) (array vec4 2) conditional_tmp)
   (if (var_ref pick_from_a_or_b)
     ((assign () (var_ref conditional_tmp) (var_ref a)))
     ((assign () (var_ref conditional_tmp) (var_ref b))))

However, we failed to update max_array_access for "a" and "b", so it
remained 0 - here, the whole array is being accessed.  At link time,
update_array_sizes() used this bogus information to change the types
of "a" and "b" to vec4[1].  We then had assignments from a vec4[1] to
a vec4[2], which is highly illegal.

This tripped assertions in nir_split_var_copies with scalar VS.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit 9f1e250e77)
2015-03-11 18:11:56 +00:00
Neil Roberts
e4d3bd6855 meta: Fix the y offset for 1D_ARRAY in _mesa_meta_pbo_TexSubImage
The yoffset needs to be interpreted as a slice offset for 1D array
textures. This patch implements that by moving the yoffset into
zoffset similar to how it moves the height into depth.

Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 7286a68991)
2015-03-11 18:11:47 +00:00
Neil Roberts
614e7ebdad meta: Allow GL_UN/PACK_IMAGE_HEIGHT in _mesa_meta_pbo_Get/TexSubImage
Now that a layered source PBO is interpreted as a single tall 2D image
it's quite easy to accept the image height packing option by just
creating an image that is tall enough to include the image padding.

I'm not sure whether the image height property should affect 1D_ARRAY
textures. My intuition and interpretation of the GL spec (which is a
bit vague) would be that it shouldn't. However the software fallback
path in Mesa uses the property for packing but not for unpacking. The
binary NVidia driver uses it for both. This patch doesn't use it for
either case so it is different from the software fallback. There is
some discussion about this here:

http://lists.freedesktop.org/archives/mesa-dev/2015-February/077925.html

This is tested by the texsubimage Piglit test with the array and pbo
arguments. Previously this test was skipping this code path because it
always sets the image height.

I've also tested it by modifying the getteximage-targets test. It
wasn't using this code path before because it was using the default
texture object so this code couldn't successfully create a frame
buffer. I also modified it to add some image padding with the image
height in the PBO.

Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit a08bff1e98)
2015-03-11 18:11:40 +00:00
Neil Roberts
7f32fa0dcb Revert "common: Fix PBOs for 1D_ARRAY."
This reverts commit 546aba143d.

I think the changes to the calls to glBlitFramebuffer from this patch
are no different to what it was doing previously because it used to
set height to 1 before doing the blits. However it was introducing
some problems with the blit for layer 0 because this was no longer
special cased. It didn't fix problems with the yoffset which needs to
be interpreted as a slice offset. I think a better solution would be
to modify the original if statement to cope with the yoffset.

Conflicts:
	src/mesa/drivers/common/meta_tex_subimage.c

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
(cherry picked from commit 7d10d2feee)
2015-03-11 18:11:10 +00:00
Neil Roberts
a15de1ae1a meta: In pbo_{Get,}TexSubImage don't repeatedly rebind the source tex
A layered PBO image is now interpreted as a single tall 2D image so
the z argument in _mesa_meta_bind_fbo_image is ignored. Therefore this
was just redundantly rebinding the same image repeatedly.

Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
(cherry picked from commit a44606eb81)
2015-03-11 18:10:44 +00:00
Matt Turner
31fcb21ef5 i965: Avoid applying negate to wrong MAD source.
For some given GLSL IR like (+ (neg x) (* 1.2 x)), the try_emit_mad
function would see that one of the +'s sources was a negate expression
and set mul_negate = true without confirming that it was actually a
multiply.

Cc: 10.5 <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89315
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89095
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
(cherry picked from commit d528907fd2)
[Emil Velikov: drop the changes in brw_vec4_visitor.cpp]
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>

Conflicts:
	src/mesa/drivers/dri/i965/brw_fs_visitor.cpp
	src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
2015-03-07 18:23:15 +00:00
Laura Ekstrand
0cd8e357e3 main: Fix target checking for CopyTexSubImage*D.
This fixes a dEQP test failure.  In the test,
glCopyTexSubImage2D was called with target = 0 and failed to throw
INVALID ENUM. This failure was caused by _mesa_get_current_tex_object(ctx,
target) being called before the target checking.  To remedy this, target
checking was separated from the main error-checking function and
called prior to _mesa_get_current_tex_object.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89312

Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
(cherry picked from commit ca65764d60)
2015-03-07 18:13:19 +00:00
Laura Ekstrand
8b4db9c687 main: Fix target checking for CompressedTexSubImage*D.
This fixes a dEQP test failure.  In the test,
glCompressedTexSubImage2D was called with target = 0 and failed to throw
INVALID ENUM. This failure was caused by _mesa_get_current_tex_object(ctx,
target) being called before the target checking.  To remedy this, target
checking was made into its own function and called prior to
_mesa_get_current_tex_object.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89311

Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
(cherry picked from commit 549078cb5a)
2015-03-07 18:12:42 +00:00
Frank Henigman
b0400a58db intel: fix EGLImage renderbuffer _BaseFormat
Correctly set _BaseFormat field when creating a gl_renderbuffer
with EGLImage storage.

Change-Id: I8c9f7302d18b617f54fa68304d8ffee087ed8a77
Signed-off-by: Frank Henigman <fjhenigman@google.com>
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>
Reviewed-by: Chad Versace <chad.versace@intel.com>
(cherry picked from commit e43729943e)
Nominated-by: Chad Versace <chad.versace@intel.com>
2015-03-07 18:08:10 +00:00
Matt Turner
ef1c87ba09 Revert SHA1 additions.
The shader-cache isn't finished, so the configure checks are a bit
premature and will only stand to confuse users of Mesa 10.5.0.

This is a squash of the follow four reverts:

   Revert "Rename sha1.c and sha1.h to mesa-sha1.c and mesa-sha1.h"
   Revert "configure: Add machinery for --enable-shader-cache (and --disable-shader-cache)"
   Revert "sha1: Fix gcry_md_hd_t typo."
   Revert "mesa: Add mesa SHA-1 functions"

Reviewed-by: Carl Worth <cworth@cworth.org>
2015-03-07 18:04:42 +00:00
Andrey Sudnik
a71223eb59 i965/vec4: Don't lose the saturate modifier in copy propagation.
Cc: 10.4, 10.5 <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89224
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
(cherry picked from commit 0dfec59a27)
2015-03-07 17:54:27 +00:00
Kenneth Graunke
47a3ae1f20 i965: Split Gen4-5 BlitFramebuffer code; prefer BLT over Meta.
A while back I switched intel_blit_framebuffer to prefer Meta over the
BLT.  This meant that Gen8 platforms would start using the 3D engine
for blits, just like we do on Gen6-7.5.

However, I hadn't considered Gen4-5 when making that change.  The BLT
engine appears to be substantially faster on 965GM than using Meta to
drive the 3D engine.  This isn't too surprising: original Gen4 doesn't
support tile offsets (that came on G45), and the level/layer fields
don't work for cubemap rendering, so for inconvenient miplevel
alignments, we end up blitting or copying data to/from temporaries
in order to render to it.  We may as well just use the blitter.

I chose to use the BLT on Gen4-5 because they use the same ring for
both 3D and BLT; Gen6+ splits it out.

Fixes regressions on 965GM due to botched tile offset code (we should
fix those properly as well, but they're longstanding bugs - for now,
put things back to the status quo).

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89430
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit aa0705c06c)
2015-03-07 17:54:17 +00:00
Matt Turner
dbf974636a i965: Tell intel_get_memcpy() which direction the memcpy() is going.
The SSSE3 swizzling code was written for fast uploads to the GPU and
assumed the destination was always 16-byte aligned. When we began using
this code for fast downloads as well we didn't do anything to account
for the fact that the destination pointer given by glReadPixels() or
glGetTexImage() is not guaranteed to be suitably aligned.

With SSSE3 enabled (at compile-time), some applications would crash when
an SSE aligned-store instruction tried to store to an unaligned
destination (or an assertion that the destination is aligned would
trigger).

To remedy this, tell intel_get_memcpy() whether we're uploading or
downloading so that it can select whether to assume the destination or
source is aligned, respectively.

Cc: 10.5 <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89416
Tested-by: Uriy Zhuravlev <stalkerg@gmail.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
(cherry picked from commit 2e4c95dfe2)
2015-03-07 17:53:59 +00:00
Emil Velikov
35909f0e12 mapi: fix commit 90411b56f6
Handle static glapi as well.

Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-07 17:53:15 +00:00
Emil Velikov
b83333ac5b cherry-ignore: ignore a few more commits picked without -x
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-07 17:50:29 +00:00
Emil Velikov
30079d6846 egl/main: no longer export internal function
With the split of the gallium egl module we had previously it required
access to some of the internal functions. As the only build (automake)
that did this no longer builds it we can now appropriately hide those
functions.

Cc: 10.5 <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit dd438ae34b)
2015-03-07 17:46:30 +00:00
Matt Turner
e5eab59b57 i965/fs: Don't propagate cmod to inst with different type.
Cc: 10.5 <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89317
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
(cherry picked from commit 1e128e9b69)
2015-03-07 17:22:30 +00:00
Matt Turner
57f54b067b r300g: Check return value of snprintf().
Would have at least prevented the crash the previous patch fixed.

Cc: 10.4, 10.5 <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.gentoo.org/show_bug.cgi?id=540970
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
(cherry picked from commit ade0b580e7)
2015-03-07 17:22:22 +00:00
Matt Turner
e0670715bc r300g: Use PATH_MAX instead of limiting ourselves to 100 chars.
When built with Gentoo's package manager, the Mesa source directory
exists seven directories deep. The path to the .test file is too long
and is silently truncated, leading to a crash. Just use PATH_MAX.

Cc: 10.4, 10.5 <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.gentoo.org/show_bug.cgi?id=540970
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
(cherry picked from commit f5e2aa1324)
2015-03-07 17:22:13 +00:00
Daniel Stone
8357abf4c3 egl: Take alpha bits into account when selecting GBM formats
This fixes piglit when using PIGLIT_PLATFORM=gbm

Tom Stellard:
  - Fix ARGB2101010 format

Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Chad Versace <chad.versace@intel.com>
(cherry picked from commit 65c8965d03)
2015-03-07 17:22:05 +00:00
Marc-Andre Lureau
b0f60c891a gallium/auxiliary/indices: fix start param
Since commit 28f3f8d, indices generator take a start parameter. However, some
index values have been left to start at 0.

This fixes the glean/fbo test with the virgl driver, and copytexsubimage
with freedreno.

Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 073a5d2e84)
2015-03-07 17:21:58 +00:00
Matt Turner
73efd2891f i965/fs: Don't use backend_visitor::instructions after creating the CFG.
This is a fix for a regression introduced in commit a9f8296d ("i965/fs:
Preserve the CFG in a few more places.").

The errata this code works around is described in a comment before the function:

   "[DevBW, DevCL] Errata: A destination register from a send can not be
    used as a destination register until after it has been sourced by an
    instruction with a different destination register.

The framebuffer write's sources must be in message registers, which SEND
instructions cannot have as a destination. There's no way for this
errata to affect anything at the end of the program. Just remove the
code.

Cc: 10.4, 10.5 <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=84613
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
(cherry picked from commit e214000f25)
2015-03-07 17:21:41 +00:00
Jason Ekstrand
73f8e342a7 main/base_tex_format: Properly handle STENCIL_INDEX1/4/16
This takes "fbo-stencil blit GL_STENCIL_INDEX1/4/16" from crash to pass on
BDW.

Cc: 10.5 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
(cherry picked from commit c4925d7f3b)
2015-03-07 17:21:34 +00:00
Jason Ekstrand
53d66c0c7c meta/TexSubImage: Stash everything other than PIXEL_TRANSFER/store in meta_begin
Previously, there were bugs where if the app set a scissor it could affect
the area of the texture that was downloaded.  There was also potential that
the framebuffer SRGB state could affect downloads.  This ensures that those
will get saved/restored and can't affect the texture download.

Cc: 10.5 <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89292
Reviewed-by: Neil Roberts <neil@linux.intel.com>
(cherry picked from commit b1ab02d9c0)
2015-03-07 17:21:24 +00:00
Matt Turner
560fb4ee45 i965: Consider scratch writes to have side effects.
We could do better by tracking scratch reads and writes.

Cc: 10.5 <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=88793
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
(cherry picked from commit da20bf068e)
2015-03-07 17:21:12 +00:00
Matt Turner
446aa309e1 mesa: Correct backwards NULL check.
Cc: "10.4, 10.5" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
(cherry picked from commit 491d42135a)
2015-03-07 17:20:53 +00:00
Matt Turner
4c45e23938 mesa: Free memory allocated for luminance in readpixels.
Cc: "10.4, 10.5" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
(cherry picked from commit 87109acbed)
2015-03-07 17:20:46 +00:00
Matt Turner
9267820be6 mesa: Indent break statements and add a missing one.
Always indenting break statements makes spotting missing ones easier.

Cc: "10.4, 10.5" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
(cherry picked from commit 2b2fa18652)
2015-03-07 17:20:40 +00:00
Chris Forbes
b730f6e18e i965/gs: Check newly-generated GS-out VUE map against correct stage
Previously, we compared our new GS-out VUE map to the existing *VS*-out
VUE map, which is bogus.

This would mostly manifest as redundant dirty flagging where the GS is
in use but the VS and GS output layouts differ; but there is a scary
case where we would fail to flag a GS-out layout change if it happened
to match the VS-out layout.

Signed-off-by: Chris Forbes <chrisf@ijw.co.nz>
Cc: "10.5, 10.4" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=88885
(cherry picked from commit b51ff50a76)
2015-03-07 17:20:28 +00:00
Matt Turner
f5fbed4831 i965/vec4: Fix implementation of i2b.
I broke this in commit 2881b123d. I must have misread i2b as b2i.

Cc: 10.5 <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=88246
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
(cherry picked from commit 43ef2657a0)
2015-03-07 16:59:07 +00:00
Ian Romanick
f7cc5fb2cb i965/fs/nir: Use emit_math for nir_op_fpow
It appears that all the other instructions that need it already use it.
This one just got missed.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit b8a1637119)
2015-03-07 16:58:59 +00:00
Emil Velikov
0d3e4ed134 docs: Add sha256 sums for the 10.5.0 release
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-06 23:58:47 +00:00
Emil Velikov
97357d475f docs: Update 10.5.0 release notes
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-06 22:55:59 +00:00
Emil Velikov
c899144da6 Bump version to 10.5.0 (final)
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-06 22:46:49 +00:00
Emil Velikov
b973acc093 xmlpool: make sure we ship options.h
The header is included in ../xmlpool.h. With the latter of which used
directly in a number of places in mesa.
Note that we can also add it (alongside t_option.h) to noinst_HEADERS,
but neither solution fixes the issue that brough us here - namely:
Do not regenerate the headers, if it already exists.

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-02 15:50:32 +00:00
Emil Velikov
90411b56f6 mapi: fix shared-glapi dependency tracking
I.e. add shared-glapi/glapi_mapi_tmp.h to the SOURCES list. Otherwise
there will be no knowledge that the file is required by others for the
build. Thus autotools won't pick it up for the distribution tarball.

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-02 15:50:32 +00:00
Emil Velikov
19422e433c mesa: drop Makefile from get_hash.h dependency list
Not required. Additionally this had the side effect of generating the
file, despite it's existence.

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-02 15:50:32 +00:00
Emil Velikov
1b00847bb2 mesa: fix dependency tracking of generated sources
Some of the files generated were not in the SOURCES variable, thus
although generated prior to compilation the dependency tracking was
incomplete. The latter of which resulted in the files missing from the
distribution tarball.

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-02 15:50:32 +00:00
Emil Velikov
a0264d4076 mesa: rename format_info.c to format_info.h
The file is auto-generated, and #included by formats.c. Let's rename it
to reflect the latter. This will also help up fix the dependency
tracking by adding it to the _SOURCES variable, without the side effect
of it being compiled (twice).

Cc: "10.4, 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-02 15:50:25 +00:00
Emil Velikov
c7d4987897 mesa/main: update .gitignore
Drop the no longer present get_es{1,2}.c from the list.

Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-03-02 13:00:55 +00:00
Emil Velikov
c32d835281 Increment version to 10.5.0-rc3
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-02-28 00:04:30 +00:00
Jonathan Gray
54cc3cdff4 auxilary/os: correct sysctl use in os_get_total_physical_memory()
The length argument passed to sysctl was the size of the pointer
not the type.  The result of this is sysctl calls would fail on
32 bit BSD/Mac OS X.

Additionally the wrong pointer was passed as an argument to store
the result of the sysctl call.

Cc: "10.4, 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Jonathan Gray <jsg@jsg.id.au>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
(cherry picked from commit 7983a3d2e0)
2015-02-27 23:29:29 +00:00
Dave Airlie
c0f425bb81 r600g/sb: treat undefined values like constants
When we schedule an instructions with undefined value, we
eventually will use 0, which is a constant, however sb wasn't
taking this into account and creating ops with illegal scalar
swizzles.

this replaces my fix for op3 in t slots.

Reviewed-by: Glenn Kennard <glenn.kennard@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
(cherry picked from commit e8e4437ed0)
2015-02-27 23:14:00 +00:00
Neil Roberts
7c448551eb i965/skl: Implement WaDisable1DDepthStencil
Skylake+ doesn't support setting a depth buffer to a 1D surface but it
does allow pretending it's a 2D texture with a height of 1 instead.

This fixes the GL_DEPTH_COMPONENT_* tests of the copyteximage piglit
test (and also seems to avoid a subsequent GPU hang).

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89037
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
(cherry picked from commit 5b29b2922a)
Nominated-by: Ian Romanick <idr@freedesktop.org>
2015-02-27 23:12:15 +00:00
Tom Stellard
3e54be2e8a pipe-loader: Fix build with dri drivers enabled, and vl state trackers disabled
Configure arguments:

./configure --disable-dri3 --disable-xvmc --enable-opencl
            --with-gallium-drivers=r300,r600,radeonsi
            --with-egl-platforms=drm

Build error:

make[3]: *** No rule to make target
`../../../../src/gallium/auxiliary/libgalliumvlwinsys.la', needed by
`pipe_r300.la'.  Stop.

Cc: "10.5" <mesa-stable@lists.freedestkop.org>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
(cherry picked from commit ed07255149)
2015-02-27 23:05:03 +00:00
Matt Turner
6f52686d77 glsl: Rewrite and fix min/max to saturate optimization.
There were some bugs, and the code was really difficult to follow. We
would optimize

   min(max(x, b), 1.0) into max(sat(x), b)

but not pay attention to the order of min/max and also do

   max(min(x, b), 1.0) into max(sat(x), b)

Corrects four shaders from Champions of Regnum that do

   min(max(x, 1), 10)

and corrects rendering of Mass Effect under VMware Workstation.

Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89180
Reviewed-by: Abdiel Janulgue <abdiel.janulgue@linux.intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
(cherry picked from commit cb25087c7b)
2015-02-27 23:04:51 +00:00
Andreas Boll
1527f569e5 glx: Fix returned values of GLX_RENDERER_PREFERRED_PROFILE_MESA
If the renderer supports the core profile the query returned incorrectly
0x8 as value, because it was using (1U << __DRI_API_OPENGL_CORE) for the
returned value.

The same happened with the compatibility profile. It returned 0x1
(1U << __DRI_API_OPENGL) instead of 0x2.

Internal DRI defines:
   dri_interface.h: #define __DRI_API_OPENGL       0
   dri_interface.h: #define __DRI_API_OPENGL_CORE  3

Those two bits are supposed for internal usage only and should be
translated to GLX_CONTEXT_CORE_PROFILE_BIT_ARB (0x1) for a preferred
core context profile and GLX_CONTEXT_COMPATIBILITY_PROFILE_BIT_ARB (0x2)
for a preferred compatibility context profile.

This patch implements the above translation in the glx module.

v2: Fix the incorrect behavior in the glx module

Cc: "10.3 10.4 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Andreas Boll <andreas.boll.dev@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
(cherry picked from commit 6d164f65c5)
2015-02-27 23:04:34 +00:00
Laura Ekstrand
dfbaa8a843 common: Fix PBOs for 1D_ARRAY.
Corrects the way that _mesa_meta_pbo_TexSubImage and
_mesa_meta_pbo_GetTexSubImage handle 1D_ARRAY textures.  Fixes a failure in
the Piglit arb_direct_state_access/gettextureimage-targets test.

Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Tested-by: Laura Ekstrand <laura@jlekstrand.net>

Cc: "10.4, 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 546aba143d)
2015-02-27 23:03:30 +00:00
Laura Ekstrand
9e8446beb1 common: Correct PBO 2D_ARRAY handling.
Changes PBO uploads and downloads to use a tall (height * depth) 2D texture
for blitting. This fixes the bug where 2D_ARRAY, 3D, and CUBE_MAP_ARRAY
textures are not properly uploaded and downloaded.

Removes the option to use a 2D ARRAY texture for the PBO during upload and
download.  This option didn't work because the miptree couldn't be set up
reliably.

v2: Review from Jason Ekstrand and Neil Roberts:
   -Delete the depth parameter from create_texture_for_pbo
   -Abandon the option to create a 2D ARRAY texture in create_texture_for_pbo

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>

Cc: "10.4, 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit ccc5ce6f72)
2015-02-27 23:03:12 +00:00
Laura Ekstrand
9e5af14163 common: Correct texture init for meta pbo uploads and downloads.
This moves the line setting immutability for the texture to after
_mesa_initialize_texture_object so that the initializer function will not
cancel it out. Moreover, because of the ARB_texture_view extension, immutable
textures must have NumLayers > 0, or depth will equal (0-1)=0xFFFFFFFF during
SURFACE_STATE setup, which triggers assertions.

v2: Review from Kenneth Graunke:
   - Include more explanation in the commit message.
   - Make texture setup bug fixes into a separate patch.

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>

Cc: "10.4, 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 06084652fe)
2015-02-27 23:02:01 +00:00
Leo Liu
ed7272ade9 st/omx/dec/h264: fix picture out-of-order with poc type 0 v2
poc counter should be reset with IDR frame,
otherwise there would be a re-order issue with
frames before and after IDR

v2: add commit message

Signed-off-by: Leo Liu <leo.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 9c7b343bc0)
2015-02-27 23:01:40 +00:00
Emil Velikov
3592cbb21f Increment version to 10.5.0-rc2
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-02-24 15:35:17 +00:00
Emil Velikov
5e026a2f63 install-lib-links: remove the .install-lib-links file
With earlier commit (install-lib-links: don't depend on .libs directory)
we moved the location of the file from .libs/ to the current dir.
Although we did not attribute that in the former case autotools was
doing us a favour and removing the file. Explicitly remove the file at
clean-local time, otherwise we'll end up with dangling files.

Cc: "10.3 10.4 10.5" <mesa-stable@lists.freedesktop.org>
Cc: Matt Turner <mattst88@gmail.com>
Cc: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
(cherry picked from commit fece147be5)
2015-02-24 15:34:25 +00:00
Eduardo Lima Mitev
04dc301669 mesa: Fix error validating args for TexSubImage3D
The zoffset and depth values were not being considered when calling
error_check_subtexture_dimensions().

Fixes 2 dEQP tests:
* dEQP-GLES3.functional.negative_api.texture.texsubimage3d_neg_offset
* dEQP-GLES3.functional.negative_api.texture.texsubimage3d_invalid_offset

Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Cc: "10.4 10.5" <mesa-stable@lists.freedestkop.org>
(cherry picked from commit 2aa71e9485)
2015-02-24 12:38:40 +00:00
Vivek Kasireddy
4952a37697 egl, wayland: RGB565 format support on Back-buffer
In current code, color format is always hardcoded to
__DRI_IMAGE_FORMAT_ARGB8888 when buffer or DRI image is
allocated in function calls, get_back_bo and dri2_get_buffers,
regardless of current target's color format. This problem
may leads to incorrect render pitch calculation, which
eventually ends up with wrong offset of pixels in
the frame buffer when the image is in different color format
from dri surf's, especially with different bpp. (e.g. RGB565-16bpp)

Attached code patch simply adds RGB565 and XRGB8888 cases to two
functions noted above to resolve the issue.

v2: added a case of XRGB8888, format and bpp selection is done
    via switch-case (not "if-else" anymore)

Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
Signed-off-by: Dongwon Kim <dongwon.kim@intel.com>
Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 1e96eece30)
2015-02-24 12:38:21 +00:00
Matt Turner
aac12f4542 i965: Link test programs with gtest before pthreads.
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.gentoo.org/show_bug.cgi?id=540962
(cherry picked from commit 0b6d43e329)
2015-02-24 12:38:04 +00:00
Marek Olšák
760407ff9e radeonsi: fix point sprites
Broken by a27b74819a.

This fix is critical and should be ported to stable ASAP.

Cc: 10.5 10.4 <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 7820a11e3d)

Squashed with commit

radeonsi: fix a warning caused by previous commit

Cc: 10.5 10.4 <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 050bf75c8b)
2015-02-24 12:37:26 +00:00
Marek Olšák
3cb2120088 vbo: fix an unitialized-variable warning
It looks like a bug to me.

Cc: 10.5 10.4 10.3 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit 0feb0b7373)
2015-02-24 12:36:28 +00:00
Marek Olšák
b97e7ad355 radeonsi: don't use SQC_CACHES to flush ICACHE and KCACHE on SI
This reverts 73c2b0d18c.

It doesn't seem to be reliable. It's probably missing a wait packet or
something, because it's just a register write and doesn't wait for anything.
SURFACE_SYNC at least seems to wait until the flush is done. Just guessing.

Let's not complicate things and revert this.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=88561

Cc: 10.5 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 7692704b14)
2015-02-24 12:36:10 +00:00
Matt Turner
bf563d59b2 i965/vec4: Add and use byte-MOV instruction for unpack 4x8.
Previously we were using a B/UB source in an Align16 instruction, which
is illegal. It for some reason works on all platforms, except Broadwell.

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=86811
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
(cherry picked from commit e0137fd6f7)
2015-02-24 12:35:46 +00:00
Matt Turner
7f7c35e854 i965/fs: Consider MOV.SAT to interfere if it has a source modifier.
The saturate propagation pass recognizes that the second instruction
below does not interfere with an attempt to propagate the saturate
modifier from instruction 3 to 1.

 1:  add(8)     dst0   src0  src1
 2:  mov.sat(8) dst1   dst0
 3:  mov.sat(8) dst2   dst0

Unfortunately, we did not consider the case of instruction 2 having a
source modifier on dst0. Take for instance:

 1:  add(8)     dst0   src0  src1
 2:  mov.sat(8) dst1  -dst0
 3:  mov.sat(8) dst2   dst0

Consider such an instruction to interfere. Increase instruction counts
in Anomaly 2, which could be a bug fix depending on the values the first
instruction produces.

instructions in affected programs:     53228 -> 53934 (1.33%)
HURT:                                  360

Cc: <mesa-stable@lists.freedesktop.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
(cherry picked from commit 7f8dd91d16)
2015-02-24 12:35:17 +00:00
Matt Turner
b554a10e82 i965/fs: Use fs_inst::overwrites_reg() in saturate propagation.
This is safer and matches the conditional_mod propagation pass.

Cc: <mesa-stable@lists.freedesktop.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
(cherry picked from commit 871ad3f08b)
2015-02-24 12:34:51 +00:00
Matt Turner
11cab7b36a i965/fs: Add unit tests for saturate propagation pass.
Cc: <mesa-stable@lists.freedesktop.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
(cherry picked from commit bf3389ec49)
2015-02-24 12:34:43 +00:00
Brian Paul
80ec8ee3a1 st/mesa: fix sampler view reference counting bug in glDraw/CopyPixels
Use pipe_sampler_view_reference() instead of ordinary assignment.
Also add a new sanity check assertion.

Fixes piglit gl-1.0-drawpixels-color-index test crash.  But note
that the test still fails.

Cc: "10.4, 10.5" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
(cherry picked from commit 62a8883f32)
2015-02-24 12:34:11 +00:00
Brian Paul
d7fe0d9ce7 swrast: fix multiple color buffer writing
If a fragment program wrote to more than one color buffer, the
first fragment color got replicated to all dest buffers.  This
fixes 5 piglit FBO tests, including fbo-drawbuffers-arbfp.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=45348
Cc: "10.4, 10.5" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
(cherry picked from commit 89c96afe3c)
2015-02-24 12:33:16 +00:00
Lucas Stach
afe00ddc20 install-lib-links: don't depend on .libs directory
This snippet can be included in Makefiles that may, depending on the
project configuration, not actually build any installable libraries.

In that case we don't have anything to depend on and this part of
the makefile may be executed before the .libs directory is created,
so do not depend on it being there.

Cc: "10.3 10.4 10.5" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
(cherry picked from commit 5c1aac17ad)
2015-02-24 12:32:55 +00:00
Emil Velikov
b9b8c10273 auxiliary/vl: honour the DRI2PROTO_CFLAGS
Otherwise for non-default installations the build will fail to find the
headers and error out.

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
(cherry picked from commit 8a71fd8d49)
2015-02-24 12:32:46 +00:00
Emil Velikov
7f72b69c62 auxiliary/vl: Build vl_winsys_dri.c only when needed.
With commit c39dbfdd0f7(auxiliary/vl: bring back the VL code for the dri
targets) we did not fully consider users of dri-swrast alone. Thus we
ended up trying to compile the dri2 specific code on platform which lack
it - Cygwin for example.

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Reported-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
(cherry picked from commit dd7b6670a2)
2015-02-24 12:32:11 +00:00
Emil Velikov
9bd832c24c automake: Use AM_DISTCHECK_CONFIGURE_FLAGS
Currently we use DISTCHECK_CONFIGURE_FLAGS, which is reserved for
the user. As with other variables, one should use the AM_ variable
within the makefile.

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
(cherry picked from commit 3018c4a56a)
2015-02-24 12:30:04 +00:00
Michel Dänzer
ae275a653a Revert "radeon/llvm: enable unsafe math for graphics shaders"
This reverts commit 0e9cdedd2e.

It caused the grass to disappear in The Talos Principle.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89069
Cc: "10.5 10.4" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
(cherry picked from commit 4db985a5fa)
2015-02-24 12:29:33 +00:00
Kenneth Graunke
86aaa10ce7 i965: Prefer Meta over the BLT for BlitFramebuffer.
There's some debate about whether we should use Meta or BLORP,
but either should run circles around the BLT engine.

In particular, this means that Gen8+ will use the 3D engine for blits,
like we do on Gen6-7.

Improves performance in "copypixrate -blit -back" (from Mesa demos)
by 232.037% +/- 3.15795% (n=10) on Broadwell GT3e.

v2: Rebase on Laura's changes.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit d523fefa75)
2015-02-24 12:28:49 +00:00
Alan Coopersmith
3e513ea65c Make _mesa_swizzle_and_convert argument types in .c match those in .h
Caused Solaris Studio compilers to fail to build with errors about
incompatible function redefinitions.

Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit b7ce7c00e3)
2015-02-24 12:28:39 +00:00
Alan Coopersmith
2f8e96f71d Use __typeof instead of typeof with Solaris Studio compilers
While the C compiler accepts typeof, C++ requires __typeof.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=86944
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit 4671dca0ee)
2015-02-24 12:28:06 +00:00
Alan Coopersmith
c65ca6c815 Avoid fighting with Solaris headers over isnormal()
When compiling in C99 or C++11 modes, Solaris defines isnormal() as
a macro via <math.h>, which causes the function definition to become
too mangled to compile.

Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit d602fbd861)
2015-02-24 12:23:44 +00:00
Alan Coopersmith
4fa129cd9b Remove extraneous ; after DECL_TYPE usage
The macro is defined to provide a trailing ; so this caused the expansion
to end in ";;" which made the Solaris Studio compilers issue warnings for
every line of:
  "builtin_type_macros.h", line 113: Warning: extra ";" ignored.
for every file that included the header, filling build logs with thousands
of useless warnings.

Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit 815b3bd096)
2015-02-24 12:23:34 +00:00
Alan Coopersmith
4c2e35e0aa Bracket arguments to tr so they work with Solaris tr
https://www.gnu.org/savannah-checkouts/gnu/autoconf/manual/autoconf-2.69/html_node/Limitations-of-Usual-Tools.html#index-g_t_0040command_007btr_007d-1842

Without this fix, egl fails to build on Solaris, with the error:

<command-line>:0:22: error: '_EGL_PLATFORM_x11' undeclared (first use in this function)
egldisplay.c:207:31: note: in expansion of macro '_EGL_NATIVE_PLATFORM'
             native_platform = _EGL_NATIVE_PLATFORM;
                               ^

Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 60ad5103b9)
2015-02-24 12:23:15 +00:00
Kenneth Graunke
474f33e690 glsl: Reduce memory consumption of copy propagation passes.
opt_copy_propagation and opt_copy_propagation_elements create new ACP
and Kill sets each time they enter a new control flow block.  For if
blocks, they also copy the entire existing ACP set contents into the
new set.

When we exit the control flow block, we discard the new sets.  However,
we weren't freeing them - so they lived on until the pass finished.
This can waste a lot of memory (57MB on one pessimal shader).

This patch makes the pass allocate ACP entries using this->acp as the
memory context, and Kill entries out of this->kill.  It also steals
kill entries when moving them from the inner kill list to the parent.

It then frees the lists, including their contents.

v2: Move ralloc_free(this->acp) just before this->acp = orig_acp
    (suggested by Eric Anholt).

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Cc: "10.5 10.4" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 76960a55e6)
2015-02-24 12:22:58 +00:00
Laura Ekstrand
20471e9423 main: Fixed _mesa_GetCompressedTexImage_sw to copy slices correctly.
Previously array textures were not working with GetCompressedTextureImage,
leading to failures in the test
arb_direct_state_access/getcompressedtextureimage.c.

Tested-by: Laura Ekstrand <laura@jlekstrand.net>
Reviewed-by: Brian Paul <brianp@vmware.com>

Cc: "10.4, 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 92163482bd)
2015-02-24 12:22:43 +00:00
Marek Olšák
f0a736160b radeonsi: fix a crash if a stencil ref state is set before a DSA state
+ minor indentation fixes

Discovered by Axel Davy.

This can't be reproduced with any app, because all state trackers set a DSA
state first.

Cc: 10.5 10.4 10.3 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Axel Davy <axel.davy@ens.fr>
(cherry picked from commit 2ead74888a)
2015-02-24 12:22:22 +00:00
Marek Olšák
3a2043fddd mesa: fix AtomicBuffer typo in _mesa_DeleteBuffers
Cc: 10.5 10.4 10.3 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit e8625a29fe)
2015-02-24 12:22:07 +00:00
Marek Olšák
39d823fa58 radeonsi: small fix in SPI state
Cc: 10.5 10.4 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
(cherry picked from commit a27b74819a)
2015-02-24 12:21:55 +00:00
Neil Roberts
7bcdde5174 meta: Fix saving the results of the current occlusion query
When restoring the current state in _mesa_meta_end it was previously trying to
copy the on-going sample count of the current occlusion query into the new
query after restarting it so that the driver will continue adding to the
previous value. This wouldn't work for two reasons. Firstly, the query might
not be ready yet so the Result member will usually be zero. Secondly the saved
query is stored as a pointer to the query object, not a copy of the struct, so
it is actually restarting the exact same object. Copying the result value is
just copying between identical addresses with no effect. The call to
_mesa_BeginQuery will have always reset it back to zero.

This patch fixes it by making it actually wait for the query object to be
ready before grabbing the previous result. The downside of doing this is that
it could introduce a stall but I think this situation is unlikely so it might
not matter too much. A better solution might be to introduce a real
suspend/resume mechanism to the driver interface. This could be implemented in
the i965 driver by saving the depth count multiple times like it does in the
i945 driver.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=88248
Reviewed-by: Carl Worth <cworth@cworth.org>
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit bb77745681)
2015-02-24 12:21:33 +00:00
Iago Toral Quiroga
f2e513b2c0 mesa: Fix element count for byte-swaps in texstore, readpix and texgetimage
Some old format conversion code in pack.c implemented byte-swapping like this:

GLint comps = _mesa_components_in_format(dstFormat);
GLint swapSize = _mesa_sizeof_packed_type(dstType);
if (swapSize == 2)
   _mesa_swap2((GLushort *) dstAddr, n * comps);
else if (swapSize == 4)
   _mesa_swap4((GLuint *) dstAddr, n * comps);

where n is the pixel count. But this is incorrect for packed formats,
where _mesa_sizeof_packed_type is already returning the size of a pixel
instead of the size of a single component, so multiplying this by the
number of components in the format results in a larger element count
for _mesa_swap than we want.

Unfortunately, we followed the same implementation for byte-swapping
in the rewrite of the format conversion code for texstore, readpixels
and texgetimage.

This patch computes the correct element counts for _mesa_swap calls
by computing the bytes per pixel in the image and dividing that by the
swap size to obtain the number of swaps required per pixel. Then multiplies
that by the number of pixels in the image to obtain the swap count that
we need to use.

Also, when handling byte-swapping in texstore_rgba, we were ignoring
the image's depth. This patch fixes this too.

Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit ba426522dd)
2015-02-24 12:20:05 +00:00
Iago Toral Quiroga
c6b0a1638c mesa: Handle transferOps in texstore_rgba
In the recent rewrite of the format conversion code we did not handle this.
This patch adds the missing support.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89068
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 4b249d2eed)
2015-02-24 12:18:01 +00:00
Matt Turner
694d656551 i965/vec4/vp: Use vec4_visitor::CMP.
... instead of emit(BRW_OPCODE_CMP, ...). In commit 6b3a301f I changed
vec4_visitor::CMP to set the destination's type to that of src0. In the
following commit (2335153f) I removed an apparently now unnecessary work
around for Gen8 that did the same thing.

But there was a single place that emitted a CMP instruction without
using the vec4_visitor::CMP function. Use it there.

And change dst_null_d to dst_null_f for good measure, since ARB vp
doesn't have integers.

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89032
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
(cherry picked from commit 72b9f8db2a)
2015-02-24 12:17:20 +00:00
Ilia Mirkin
c38068f5c1 nvc0: allow holes in xfb target lists
Tested with a modified xfb-streams test which outputs to streams 0, 2,
and 3.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 854eb06bee)
2015-02-24 12:16:45 +00:00
Ilia Mirkin
88d47387b8 st/mesa: treat resource-less xfb buffers as if they weren't there
If a transform feedback buffer's size is 0, st_bufferobj_data doesn't
end up creating a buffer for it. There's no point in trying to write to
such a buffer, so just pretend as if it's not really there.

This fixes arb_gpu_shader5-xfb-streams-without-invocations on nvc0.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 80d373ed5b)
2015-02-24 12:16:28 +00:00
Ilia Mirkin
c08e9605c2 nvc0: bail out of 2d blits with non-A8_UNORM alpha formats
This fixes the teximage-colors uploads with GL_ALPHA format and
non-GL_UNSIGNED_BYTE type.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 68e4f3f572)
2015-02-24 12:16:12 +00:00
Tapani Pälli
fe98850c5f mesa: fix OES_texture_float texture render target behavior
Current implementation allowed usage of unsized type texture GL_FLOAT
and GL_HALF_FLOAT as a render target as this was 'expected behavior' by
WEBGL_oes_texture_float and is also allowed by the oes-texture-float
WebGL test. However this broke some ES3 conformance tests that do not
accept such behavior. Patch sets such an fbo incomplete as expected by
the ES3 conformance tests. Textures with sized types like RGBA32F will
still continue to work as render targets.

v2: code style cleanups (Ian Romanick, Matt Turner)

Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=88905
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit e333035c47)
2015-02-24 12:16:01 +00:00
Carl Worth
6f14955b71 Revert use of Mesa IR optimizer for ARB_fragment_programs
Commit f82f2fb3dc added use of the Mesa
IR optimizer for both ARB_fragment_program and ARB_vertex_program, but
only justified the vertex-program portions with measured performance
improvements.

Meanwhile, the optimizer was seen to generate hundreds of unused
immediates without discarding them, causing failures.

Discard the use of the optimizer for now to fix the regression. (In
the future, we anticpate things moving from Mesa IR to NIR for better
optimization anyway.)

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=82477

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>

CC: "10.3 10.4 10.5" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit 55a57834bf)
2015-02-24 12:15:15 +00:00
Emil Velikov
8ab5a28bd6 cherry-ignore: ignore commits picked without -x
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-02-24 12:14:20 +00:00
Kenneth Graunke
4d6bc2c0ff i965: Fix integer border color on Haswell.
+82 Piglits - 100% of border color tests now pass on Haswell.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Chris Forbes <chrisf@ijw.co.nz>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit 08a06b6b89)
2015-02-24 12:10:01 +00:00
Kenneth Graunke
631e76500d i965: Use a gl_color_union for sampler border color.
This should have no effect, but will make it easier to implement other
bug fixes.

v2: Eliminate "unsigned one" local; just use the value where necessary.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit e1e73443c5)
2015-02-24 12:09:47 +00:00
Kenneth Graunke
28650815ac i965: Override swizzles for integer luminance formats.
The hardware's integer luminance formats are completely unusable;
currently we fall back to RGBA.  This means we need to override
the texture swizzle to obtain the XXX1 values expected for luminance
formats.

Fixes spec/EXT_texture_integer/texwrap formats bordercolor [swizzled]
on Broadwell - 100% of border color tests now pass on Broadwell.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Chris Forbes <chrisf@ijw.co.nz>
Cc: mesa-stable@lists.freedesktop.org
(cherry picked from commit 8cb18760cc)
2015-02-24 12:09:18 +00:00
Emil Velikov
b9518a41d4 Increment version to 10.5.0-rc1
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-02-12 14:00:41 +00:00
Emil Velikov
550d7c26e7 auxiliary/vl: bring back the VL code for the dri targets
With commit c642e87d9f4(auxiliary/vl: rework the build of the VL code)
we split out the VL code into a separate static library that was meant
to be used by the VL targets alone - va, vdpau, xvmc.

The commit failed to consider the way we handle vdpau-gl interop and
broke it. Bring back the functionality by keeping the vl <> vl_stub
separation as requrested by Christian.

v2: Update the omx target as well. Update mesa-stable email address.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=86837
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Tested-by: Andy Furniss <adf.lists@gmail.com>
2015-02-12 13:29:37 +00:00
Emil Velikov
716886c338 configure: rework wayland_scanner handling(fix make distcheck)
Currently having the wayland-scanner is optional, which causes problems
when autotools parses through the makefiles, and tries to generate all
the BUILT_SOURCES.

As the config option --with-egl-platform=wayland is not the default, we
won't end up setting the WAYLAND_SCANNER variable, which in turn will
cause some files to not get generated.

There has been a wayland-scanner package as of wayland 1.2 which
provides a variable for the scanner binary, so let's use that one and
fall back to manually searching via AC_PATH_PROG when needed.

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
Reviewed-by: Matt Turner <mattst88@gmail.com>
2015-02-12 13:29:36 +00:00
Emil Velikov
8898b68a3f nir: add missing header to the sources list
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
2015-02-12 13:29:35 +00:00
Emil Velikov
93675c7aac nir: resolve nir.h dependency list (fix make distcheck)
Use nir/nir_opcodes.h as is (w/o the absolute path), as it is the target
name used to generate the actual file. Otherwise the target is missing,
the file won't get generated and the build will fail.

Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
2015-02-12 13:29:24 +00:00
Axel Davy
6555c00b19 egl: Soften several HAVE_DRM_PLATFORM to HAVE_LIBDRM
To fix build when libdrm is not found,
commit a594cec7e3 did put several
parts of egl code under #ifdef HAVE_DRM_PLATFORM.

HAVE_DRM_PLATFORM means the egl drm platform is being built.
What should have been used instead is HAVE_LIBDRM.

At a few locations, the HAVE_DRM_PLATFORM introduced
have already been replaced by HAVE_LIBDRM, this patch
replaces the remaining occurences.

This patch makes for example EGL_EXT_image_dma_buf_import
be advertised by egl under x11 when the drm egl platform
is not built, whereas previously it required the drm egl
platform to be built.

Signed-off-by: Axel Davy <axel.davy@ens.fr>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
2015-02-12 13:29:10 +00:00
Jeremy Huddleston Sequoia
d03de1dd7d darwin: build fix
xfont.c:237:14: error: implicit declaration of function 'GetGLXDRIDrawable' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
   glxdraw = GetGLXDRIDrawable(CC->currentDpy, CC->currentDrawable);
             ^
Fixes regression from 291be28476

Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
(cherry picked from commit e68b67b53f)
2015-02-11 00:23:48 -08:00
Jeremy Huddleston Sequoia
b1b7b5b068 darwin: build fix
../../../src/mesa/main/compiler.h:47:10: fatal error: 'util/macros.h' file not found

Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
(cherry picked from commit 1c67a5687a)
2015-02-10 20:35:24 -08:00
12334 changed files with 954273 additions and 4850078 deletions

View File

View File

View File

View File

View File

View File

View File

View File

View File

View File

View File

@@ -1,2 +0,0 @@
# Vendored code
src/amd/vulkan/radix_sort/*

View File

@@ -1,9 +0,0 @@
# The following files are opted into `ninja clang-format` and
# enforcement in the CI.
src/gallium/drivers/i915
src/gallium/targets/teflon/**/*
src/amd/vulkan/**/*
src/amd/compiler/**/*
src/egl/**/*
src/etnaviv/isa/**/*

View File

@@ -1,18 +1,12 @@
((nil . ((show-trailing-whitespace . t)))
(prog-mode
((prog-mode
(indent-tabs-mode . nil)
(tab-width . 8)
(c-basic-offset . 3)
(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)))
(whitespace-style face indentation)
(whitespace-line-column . 79)
(eval ignore-errors
(require 'whitespace)
(whitespace-mode 1)))
)
(makefile-mode (indent-tabs-mode . t))
)

View File

@@ -1,44 +0,0 @@
# To use this config on you editor, follow the instructions at:
# http://editorconfig.org
root = true
[*]
charset = utf-8
insert_final_newline = true
tab_width = 8
[*.{c,h,cpp,hpp,cc,hh,y,yy}]
indent_style = space
indent_size = 3
max_line_length = 78
[{Makefile*,*.mk}]
indent_style = tab
[*.py]
indent_style = space
indent_size = 4
[*.yml]
indent_style = space
indent_size = 2
[*.rst]
indent_style = space
indent_size = 3
[*.patch]
trim_trailing_whitespace = false
[{meson.build,meson_options.txt}]
indent_style = space
indent_size = 2
[*.ps1]
indent_style = space
indent_size = 2
[*.rs]
indent_style = space
indent_size = 4

View File

@@ -1,67 +0,0 @@
# List of commits to ignore when using `git blame`.
# Enable with:
# git config blame.ignoreRevsFile .git-blame-ignore-revs
#
# Per git-blame(1):
# Ignore revisions listed in the file, one unabbreviated object name
# per line, in git-blame. Whitespace and comments beginning with # are
# ignored.
#
# Please keep these in chronological order :)
#
# You can add a new commit with the following command:
# git log -1 --pretty=format:'%n# %s%n%H%n' >> .git-blame-ignore-revs $COMMIT
# pvr: Fix clang-format error.
0ad5b0a74ef73f5fcbe1406ad9d57fe5dc00a5b1
# panfrost: Fix up some formatting for clang-format
a4705afe63412498d13ded73cba969c66be67907
# asahi: clang-format the world again
26c51bb8d8a33098b1990425a391f56ffba5728c
# perfetto: Add a .clang-format for the directory.
da78d5d729b1800136dd713b68492cb339993f4a
# panfrost/winsys: Clang-format
c90f036516a5376002be6550a917e8bad6a8a3b8
# panfrost: Re-run clang-format
4ccf174009af6732cbffa5d8ebb4687da7517505
# panvk: Clang-format
c7bf3b69ebc8f2252dbf724a4de638e6bb2ac402
# pan/mdg: Fix icky formatting
133af0d6c945d3aaca8989edd15283a2b7dcc6c7
# mapi: clang-format _glapi_add_dispatch()
30332529663268a6406e910848e906e725e6fda7
# radv: reformat according to its .clang-format
8b319c6db8bd93603b18bd783eb75225fcfd51b7
# aco: reformat according to its .clang-format
6b21653ab4d3a67e711fe10e3d403128b6d26eb2
# egl: re-format using clang-format
2f670d89db038d5a29f6b72732fd7ad63dfaf4c6
# panfrost: clang-format the tree
0afd691f29683f6e9dde60f79eca094373521806
# aco: Format.
1e2639026fec7069806449f9ba2a124ce4eb5569
# radv: Format.
59c501ca353f8ec9d2717c98af2bfa1a1dbf4d75
# pvr: clang-format fixes
953c04ebd39c52d457301bdd8ac803949001da2d
# freedreno: Re-indent
2d439343ea1aee146d4ce32800992cd389bd505d
# ir3: Reformat source with clang-format
177138d8cb0b4f6a42ef0a1f8593e14d79f17c54

11
.gitattributes vendored
View File

@@ -1,7 +1,4 @@
*.csv eol=crlf
* text=auto
*.jpg binary
*.png binary
*.gif binary
*.ico binary
*.cl gitlab-language=c
*.dsp -crlf
*.dsw -crlf
*.sln -crlf
*.vcproj -crlf

View File

@@ -1,60 +0,0 @@
name: macOS-CI
on: push
permissions:
contents: read
jobs:
macOS-CI:
strategy:
matrix:
glx_option: ['dri', 'xlib']
runs-on: macos-11
env:
GALLIUM_DUMP_CPU: true
MESON_EXEC: /Users/runner/Library/Python/3.11/bin/meson
steps:
- name: Checkout
uses: actions/checkout@v3
- name: Install Dependencies
run: |
cat > Brewfile <<EOL
brew "bison"
brew "expat"
brew "gettext"
brew "libx11"
brew "libxcb"
brew "libxdamage"
brew "libxext"
brew "molten-vk"
brew "ninja"
brew "pkg-config"
brew "python@3.10"
EOL
brew update
brew bundle --verbose
- name: Install Mako and meson
run: pip3 install --user mako meson
- name: Configure
run: |
cat > native_config <<EOL
[binaries]
llvm-config = '/usr/local/opt/llvm/bin/llvm-config'
EOL
$MESON_EXEC . build --native-file=native_config -Dmoltenvk-dir=$(brew --prefix molten-vk) -Dbuild-tests=true -Dosmesa=true -Dgallium-drivers=swrast,zink -Dglx=${{ matrix.glx_option }}
- name: Build
run: $MESON_EXEC compile -C build
- name: Test
run: $MESON_EXEC test -C build --print-errorlogs
- name: Install
run: $MESON_EXEC install -C build --destdir $PWD/install
- name: 'Upload Artifact'
if: always()
uses: actions/upload-artifact@v3
with:
name: macos-${{ matrix.glx_option }}-result
path: |
build/meson-logs/
install/
retention-days: 5

49
.gitignore vendored
View File

@@ -1,6 +1,47 @@
.vscode*
*.a
*.dll
*.exe
*.ilk
*.la
*.lo
*.log
*.o
*.obj
*.os
*.pc
*.pdb
*.pyc
*.pyo
*.out
/build
.venv/
*.so
*.so.*
*.sw[a-z]
*.tar
*.tar.bz2
*.tar.gz
*.tar.xz
*.trs
*.zip
*~
depend
depend.bak
bin/ltmain.sh
lib
lib64
configure
configure.lineno
autom4te.cache
aclocal.m4
config.log
config.status
cscope*
.scon*
config.py
build
libtool
manifest.txt
.dir-locals.el
.deps/
.dirstamp
.libs/
Makefile
Makefile.in

View File

@@ -1,351 +0,0 @@
# Types of CI pipelines:
# | pipeline name | context | description |
# |----------------------|-----------|-------------------------------------------------------------|
# | merge pipeline | mesa/mesa | pipeline running for an MR; if it passes the MR gets merged |
# | pre-merge pipeline | mesa/mesa | same as above, except its status doesn't affect the MR |
# | post-merge pipeline | mesa/mesa | pipeline immediately after merging |
# | fork pipeline | fork | pipeline running in a user fork |
# | scheduled pipeline | mesa/mesa | nightly pipelines, running every morning at 4am UTC |
# | direct-push pipeline | mesa/mesa | when commits are pushed directly to mesa/mesa, bypassing Marge and its gating pipeline |
#
# Note that the release branches maintained by the release manager fall under
# the "direct push" category.
#
# "context" indicates the permissions that the jobs get; notably, any
# container created in mesa/mesa gets pushed immediately for everyone to use
# as soon as the image tag change is merged.
#
# Merge pipelines contain all jobs that must pass before the MR can be merged.
# Pre-merge pipelines contain the exact same jobs as merge pipelines.
# Post-merge pipelines contain *only* the `pages` job that deploys the new
# version of the website.
# Fork pipelines contain everything.
# Scheduled pipelines only contain the container+build jobs, and some extra
# test jobs (typically "full" variants of pre-merge jobs that only run 1/X
# test cases), but not a repeat of the merge pipeline jobs.
# Direct-push pipelines contain the same jobs as merge pipelines.
workflow:
rules:
# do not duplicate pipelines on merge pipelines
- if: $CI_COMMIT_BRANCH && $CI_OPEN_MERGE_REQUESTS && $CI_PIPELINE_SOURCE == "push"
when: never
# merge pipeline
- if: &is-merge-attempt $GITLAB_USER_LOGIN == "marge-bot" && $CI_PIPELINE_SOURCE == "merge_request_event"
variables:
KERNEL_IMAGE_BASE: https://${S3_HOST}/${S3_KERNEL_BUCKET}/${KERNEL_REPO}/${KERNEL_TAG}
MESA_CI_PERFORMANCE_ENABLED: 1
VALVE_INFRA_VANGOGH_JOB_PRIORITY: "" # Empty tags are ignored by gitlab
# post-merge pipeline
- if: &is-post-merge $GITLAB_USER_LOGIN == "marge-bot" && $CI_PIPELINE_SOURCE == "push"
# nightly pipeline
- if: &is-scheduled-pipeline $CI_PIPELINE_SOURCE == "schedule"
variables:
KERNEL_IMAGE_BASE: https://${S3_HOST}/${S3_KERNEL_BUCKET}/${KERNEL_REPO}/${KERNEL_TAG}
JOB_PRIORITY: 50
VALVE_INFRA_VANGOGH_JOB_PRIORITY: priority:low
# pipeline for direct pushes that bypassed the CI
- if: &is-direct-push $CI_PROJECT_NAMESPACE == "mesa" && $CI_PIPELINE_SOURCE == "push" && $GITLAB_USER_LOGIN != "marge-bot"
variables:
KERNEL_IMAGE_BASE: https://${S3_HOST}/${S3_KERNEL_BUCKET}/${KERNEL_REPO}/${KERNEL_TAG}
JOB_PRIORITY: 40
VALVE_INFRA_VANGOGH_JOB_PRIORITY: priority:low
# pre-merge or fork pipeline
- if: $FORCE_KERNEL_TAG != null
variables:
KERNEL_IMAGE_BASE: https://${S3_HOST}/${S3_KERNEL_BUCKET}/${KERNEL_REPO}/${FORCE_KERNEL_TAG}
JOB_PRIORITY: 50
VALVE_INFRA_VANGOGH_JOB_PRIORITY: priority:low
- if: $FORCE_KERNEL_TAG == null
variables:
KERNEL_IMAGE_BASE: https://${S3_HOST}/${S3_KERNEL_BUCKET}/${KERNEL_REPO}/${KERNEL_TAG}
JOB_PRIORITY: 50
VALVE_INFRA_VANGOGH_JOB_PRIORITY: priority:low
variables:
FDO_UPSTREAM_REPO: mesa/mesa
MESA_TEMPLATES_COMMIT: &ci-templates-commit d5aa3941aa03c2f716595116354fb81eb8012acb
CI_PRE_CLONE_SCRIPT: |-
set -o xtrace
wget -q -O download-git-cache.sh ${CI_PROJECT_URL}/-/raw/${CI_COMMIT_SHA}/.gitlab-ci/download-git-cache.sh
bash download-git-cache.sh
rm download-git-cache.sh
set +o xtrace
S3_JWT_FILE: /s3_jwt
S3_HOST: s3.freedesktop.org
# This bucket is used to fetch the kernel image
S3_KERNEL_BUCKET: mesa-rootfs
# Bucket for git cache
S3_GITCACHE_BUCKET: git-cache
# Bucket for the pipeline artifacts pushed to S3
S3_ARTIFACTS_BUCKET: artifacts
# Buckets for traces
S3_TRACIE_RESULTS_BUCKET: mesa-tracie-results
S3_TRACIE_PUBLIC_BUCKET: mesa-tracie-public
S3_TRACIE_PRIVATE_BUCKET: mesa-tracie-private
# per-pipeline artifact storage on MinIO
PIPELINE_ARTIFACTS_BASE: ${S3_HOST}/${S3_ARTIFACTS_BUCKET}/${CI_PROJECT_PATH}/${CI_PIPELINE_ID}
# per-job artifact storage on MinIO
JOB_ARTIFACTS_BASE: ${PIPELINE_ARTIFACTS_BASE}/${CI_JOB_ID}
# reference images stored for traces
PIGLIT_REPLAY_REFERENCE_IMAGES_BASE: "${S3_HOST}/${S3_TRACIE_RESULTS_BUCKET}/$FDO_UPSTREAM_REPO"
# For individual CI farm status see .ci-farms folder
# Disable farm with `git mv .ci-farms{,-disabled}/$farm_name`
# Re-enable farm with `git mv .ci-farms{-disabled,}/$farm_name`
# NEVER MIX FARM MAINTENANCE WITH ANY OTHER CHANGE IN THE SAME MERGE REQUEST!
ARTIFACTS_BASE_URL: https://${CI_PROJECT_ROOT_NAMESPACE}.${CI_PAGES_DOMAIN}/-/${CI_PROJECT_NAME}/-/jobs/${CI_JOB_ID}/artifacts
# Python scripts for structured logger
PYTHONPATH: "$PYTHONPATH:$CI_PROJECT_DIR/install"
# Drop once deqp-runner is upgraded to > 0.18.0
MESA_VK_ABORT_ON_DEVICE_LOSS: 1
# Avoid the wall of "Unsupported SPIR-V capability" warnings in CI job log, hiding away useful output
MESA_SPIRV_LOG_LEVEL: error
default:
id_tokens:
S3_JWT:
aud: https://s3.freedesktop.org
before_script:
- >
export SCRIPTS_DIR=$(mktemp -d) &&
curl -L -s --retry 4 -f --retry-all-errors --retry-delay 60 -O --output-dir "${SCRIPTS_DIR}" "${CI_PROJECT_URL}/-/raw/${CI_COMMIT_SHA}/.gitlab-ci/setup-test-env.sh" &&
. ${SCRIPTS_DIR}/setup-test-env.sh &&
echo -n "${S3_JWT}" > "${S3_JWT_FILE}" &&
unset CI_JOB_JWT S3_JWT # Unsetting vulnerable env variables
after_script:
# Work around https://gitlab.com/gitlab-org/gitlab/-/issues/20338
- find -name '*.log' -exec mv {} {}.txt \;
- >
set +x
test -e "${S3_JWT_FILE}" &&
export S3_JWT="$(<${S3_JWT_FILE})" &&
rm "${S3_JWT_FILE}"
# Retry when job fails. Failed jobs can be found in the Mesa CI Daily Reports:
# https://gitlab.freedesktop.org/mesa/mesa/-/issues/?sort=created_date&state=opened&label_name%5B%5D=CI%20daily
retry:
max: 1
# Ignore runner_unsupported, stale_schedule, archived_failure, or
# unmet_prerequisites
when:
- api_failure
- runner_system_failure
- script_failure
- job_execution_timeout
- scheduler_failure
- data_integrity_failure
- unknown_failure
stages:
- sanity
- container
- git-archive
- build-x86_64
- build-misc
- code-validation
- amd
- intel
- nouveau
- arm
- broadcom
- freedreno
- etnaviv
- software-renderer
- layered-backends
- deploy
include:
- project: 'freedesktop/ci-templates'
ref: 16bc29078de5e0a067ff84a1a199a3760d3b3811
file:
- '/templates/ci-fairy.yml'
- project: 'freedesktop/ci-templates'
ref: *ci-templates-commit
file:
- '/templates/alpine.yml'
- '/templates/debian.yml'
- '/templates/fedora.yml'
- local: '.gitlab-ci/image-tags.yml'
- local: '.gitlab-ci/lava/lava-gitlab-ci.yml'
- local: '.gitlab-ci/container/gitlab-ci.yml'
- local: '.gitlab-ci/build/gitlab-ci.yml'
- local: '.gitlab-ci/test/gitlab-ci.yml'
- local: '.gitlab-ci/farm-rules.yml'
- local: '.gitlab-ci/test-source-dep.yml'
- local: 'docs/gitlab-ci.yml'
- local: 'src/**/ci/gitlab-ci.yml'
# YAML anchors for rule conditions
# --------------------------------
.rules-anchors:
# Pre-merge pipeline
- &is-pre-merge '$CI_PIPELINE_SOURCE == "merge_request_event"'
.never-post-merge-rules:
rules:
- if: *is-post-merge
when: never
.container+build-rules:
rules:
# Run when re-enabling a disabled farm, but not when disabling it
- !reference [.disable-farm-mr-rules, rules]
# Never run immediately after merging, as we just ran everything
- !reference [.never-post-merge-rules, rules]
# Build everything in merge pipelines, if any files affecting the pipeline
# were changed
- if: *is-merge-attempt
changes: &all_paths
- VERSION
- bin/git_sha1_gen.py
- bin/install_megadrivers.py
- bin/symbols-check.py
# GitLab CI
- .gitlab-ci.yml
- .gitlab-ci/**/*
- .ci-farms/*
# Meson
- meson*
- build-support/**/*
- subprojects/**/*
# clang format
- .clang-format
- .clang-format-include
- .clang-format-ignore
# Source code
- include/**/*
- src/**/*
when: on_success
# Same as above, but for pre-merge pipelines
- if: *is-pre-merge
changes:
*all_paths
when: manual
# Skip everything for pre-merge and merge pipelines which don't change
# anything in the build
- if: *is-merge-attempt
when: never
- if: *is-pre-merge
when: never
# Build everything after someone bypassed the CI
- if: *is-direct-push
when: on_success
# Build everything in scheduled pipelines
- if: *is-scheduled-pipeline
when: on_success
# Allow building everything in fork pipelines, but build nothing unless
# manually triggered
- when: manual
.ci-deqp-artifacts:
artifacts:
name: "mesa_${CI_JOB_NAME}"
when: always
untracked: false
paths:
# Watch out! Artifacts are relative to the build dir.
# https://gitlab.com/gitlab-org/gitlab-ce/commit/8788fb925706cad594adf6917a6c5f6587dd1521
- artifacts
- _build/meson-logs/*.txt
- _build/meson-logs/strace
# Git archive
make git archive:
extends:
- .fdo.ci-fairy
stage: git-archive
rules:
- !reference [.scheduled_pipeline-rules, rules]
# ensure we are running on packet
tags:
- packet.net
script:
# Compactify the .git directory
- git gc --aggressive
# Download & cache the perfetto subproject as well.
- rm -rf subprojects/perfetto ; mkdir -p subprojects/perfetto && curl https://android.googlesource.com/platform/external/perfetto/+archive/$(grep 'revision =' subprojects/perfetto.wrap | cut -d ' ' -f3).tar.gz | tar zxf - -C subprojects/perfetto
# compress the current folder
- tar -cvzf ../$CI_PROJECT_NAME.tar.gz .
- ci-fairy s3cp --token-file "${S3_JWT_FILE}" ../$CI_PROJECT_NAME.tar.gz https://$S3_HOST/git-cache/$CI_PROJECT_NAMESPACE/$CI_PROJECT_NAME/$CI_PROJECT_NAME.tar.gz
# Sanity checks of MR settings and commit logs
sanity:
extends:
- .fdo.ci-fairy
stage: sanity
rules:
- if: *is-pre-merge
when: on_success
- when: never
variables:
GIT_STRATEGY: none
script:
# ci-fairy check-commits --junit-xml=check-commits.xml
- ci-fairy check-merge-request --require-allow-collaboration --junit-xml=check-merge-request.xml
- |
set -eu
image_tags=(
DEBIAN_BASE_TAG
DEBIAN_BUILD_TAG
DEBIAN_X86_64_TEST_ANDROID_TAG
DEBIAN_X86_64_TEST_GL_TAG
DEBIAN_X86_64_TEST_VK_TAG
ALPINE_X86_64_BUILD_TAG
ALPINE_X86_64_LAVA_SSH_TAG
FEDORA_X86_64_BUILD_TAG
KERNEL_ROOTFS_TAG
KERNEL_TAG
PKG_REPO_REV
WINDOWS_X64_MSVC_TAG
WINDOWS_X64_BUILD_TAG
WINDOWS_X64_TEST_TAG
)
for var in "${image_tags[@]}"
do
if [ "$(echo -n "${!var}" | wc -c)" -gt 20 ]
then
echo "$var is too long; please make sure it is at most 20 chars."
exit 1
fi
done
artifacts:
when: on_failure
reports:
junit: check-*.xml
tags:
- placeholder-job
mr-label-maker-test:
extends:
- .fdo.ci-fairy
stage: sanity
rules:
- !reference [.mr-label-maker-rules, rules]
variables:
GIT_STRATEGY: fetch
timeout: 10m
script:
- set -eu
- python3 -m venv .venv
- source .venv/bin/activate
- pip install git+https://gitlab.freedesktop.org/freedesktop/mr-label-maker
- mr-label-maker --dry-run --mr $CI_MERGE_REQUEST_IID
# Jobs that need to pass before spending hardware resources on further testing
.required-for-hardware-jobs:
needs:
- job: clang-format
optional: true
- job: rustfmt
optional: true

View File

@@ -1,82 +0,0 @@
# Note: skips lists for CI are just a list of lines that, when
# non-zero-length and not starting with '#', will regex match to
# delete lines from the test list. Be careful.
# This test checks the driver's reported conformance version against the
# version of the CTS we're running. This check fails every few months
# and everyone has to go and bump the number in every driver.
# Running this check only makes sense while preparing a conformance
# submission, so skip it in the regular CI.
dEQP-VK.api.driver_properties.conformance_version
# Exclude this test which might fail when a new extension is implemented.
dEQP-VK.info.device_extensions
# These are tremendously slow (pushing toward a minute), and aren't
# reliable to be run in parallel with other tests due to CPU-side timing.
dEQP-GLES[0-9]*.functional.flush_finish.*
# piglit: WGL is Windows-only
wgl@.*
# These are sensitive to CPU timing, and would need to be run in isolation
# on the system rather than in parallel with other tests.
glx@glx_arb_sync_control@timing.*
# This test is not built with waffle, while we do build tests with waffle
spec@!opengl 1.1@windowoverlap
# These tests all read from the front buffer after a swap. Given that we
# run piglit tests in parallel in Mesa CI, and don't have a compositor
# running, the frontbuffer reads may end up with undefined results from
# windows overlapping us.
#
# Piglit does mark these tests as not to be run in parallel, but deqp-runner
# doesn't respect that. We need to extend deqp-runner to allow some tests to be
# marked as single-threaded and run after the rayon loop if we want to support
# them.
#
# Note that "glx-" tests don't appear in x11-skips.txt because they can be
# run even if PIGLIT_PLATFORM=gbm (for example)
glx@glx-copy-sub-buffer.*
# A majority of the tests introduced in CTS 1.3.7.0 are experiencing failures and flakes.
# Disable these tests until someone with a more deeper understanding of EGL examines them.
#
# Note: on sc8280xp/a690 I get identical results (same passes and fails)
# between freedreno, zink, and llvmpipe, so I believe this is either a
# deqp bug or egl/wayland bug, rather than driver issue.
#
# With llvmpipe, the failing tests have the error message:
#
# "Illegal sampler view creation without bind flag"
#
# which might be a hint. (But some passing tests also have the same
# error message.)
#
# more context from David Heidelberg on IRC: the deqp commit where these
# started failing is: https://github.com/KhronosGroup/VK-GL-CTS/commit/79b25659bcbced0cfc2c3fe318951c585f682abe
# prior to that they were skipping.
wayland-dEQP-EGL.functional.color_clears.single_context.gles1.other
wayland-dEQP-EGL.functional.color_clears.single_context.gles2.other
wayland-dEQP-EGL.functional.color_clears.single_context.gles3.other
wayland-dEQP-EGL.functional.color_clears.multi_context.gles1.other
wayland-dEQP-EGL.functional.color_clears.multi_context.gles2.other
wayland-dEQP-EGL.functional.color_clears.multi_context.gles3.other
wayland-dEQP-EGL.functional.color_clears.multi_context.gles1_gles2.other
wayland-dEQP-EGL.functional.color_clears.multi_context.gles1_gles2_gles3.other
wayland-dEQP-EGL.functional.color_clears.multi_thread.gles1.other
wayland-dEQP-EGL.functional.color_clears.multi_thread.gles2.other
wayland-dEQP-EGL.functional.color_clears.multi_thread.gles3.other
wayland-dEQP-EGL.functional.color_clears.multi_thread.gles1_gles2.other
wayland-dEQP-EGL.functional.color_clears.multi_thread.gles1_gles2_gles3.other
# Seems to be the same is as wayland-dEQP-EGL.functional.color_clears.*
wayland-dEQP-EGL.functional.render.single_context.gles2.other
wayland-dEQP-EGL.functional.render.single_context.gles3.other
wayland-dEQP-EGL.functional.render.multi_context.gles2.other
wayland-dEQP-EGL.functional.render.multi_context.gles3.other
wayland-dEQP-EGL.functional.render.multi_context.gles2_gles3.other
wayland-dEQP-EGL.functional.render.multi_thread.gles2.other
wayland-dEQP-EGL.functional.render.multi_thread.gles3.other
wayland-dEQP-EGL.functional.render.multi_thread.gles2_gles3.other

View File

@@ -1,67 +0,0 @@
version: 1
# Rules to match for a machine to qualify
target:
id: '{{ ci_runner_id }}'
timeouts:
first_console_activity: # This limits the time it can take to receive the first console log
minutes: {{ timeout_first_minutes }}
retries: {{ timeout_first_retries }}
console_activity: # Reset every time we receive a message from the logs
minutes: {{ timeout_minutes }}
retries: {{ timeout_retries }}
boot_cycle:
minutes: {{ timeout_boot_minutes }}
retries: {{ timeout_boot_retries }}
overall: # Maximum time the job can take, not overrideable by the "continue" deployment
minutes: {{ timeout_overall_minutes }}
retries: 0
# no retries possible here
console_patterns:
session_end:
regex: >-
{{ session_end_regex }}
{% if session_reboot_regex %}
session_reboot:
regex: >-
{{ session_reboot_regex }}
{% endif %}
job_success:
regex: >-
{{ job_success_regex }}
job_warn:
regex: >-
{{ job_warn_regex }}
# Environment to deploy
deployment:
# Initial boot
start:
kernel:
url: '{{ kernel_url }}'
cmdline: >
SALAD.machine_id={{ '{{' }} machine_id }}
console={{ '{{' }} local_tty_device }},115200 earlyprintk=vga,keep
loglevel={{ log_level }} no_hash_pointers
b2c.service="--privileged --tls-verify=false --pid=host docker://{{ '{{' }} fdo_proxy_registry }}/gfx-ci/ci-tron/telegraf:latest" b2c.hostname=dut-{{ '{{' }} machine.full_name }}
b2c.container="-ti --tls-verify=false docker://{{ '{{' }} fdo_proxy_registry }}/gfx-ci/ci-tron/machine-registration:latest check"
b2c.ntp_peer=10.42.0.1 b2c.pipefail b2c.cache_device=auto b2c.poweroff_delay={{ poweroff_delay }}
b2c.minio="gateway,{{ '{{' }} minio_url }},{{ '{{' }} job_bucket_access_key }},{{ '{{' }} job_bucket_secret_key }}"
b2c.volume="{{ '{{' }} job_bucket }}-results,mirror=gateway/{{ '{{' }} job_bucket }},pull_on=pipeline_start,push_on=changes,overwrite{% for excl in job_volume_exclusions %},exclude={{ excl }}{% endfor %},remove,expiration=pipeline_end,preserve"
{% for volume in volumes %}
b2c.volume={{ volume }}
{% endfor %}
b2c.container="-v {{ '{{' }} job_bucket }}-results:{{ working_dir }} -w {{ working_dir }} {% for mount_volume in mount_volumes %} -v {{ mount_volume }}{% endfor %} --tls-verify=false docker://{{ local_container }} {{ container_cmd }}"
{% if kernel_cmdline_extras is defined %}
{{ kernel_cmdline_extras }}
{% endif %}
initramfs:
url: '{{ initramfs_url }}'
{% if dtb_url is defined %}
dtb:
url: '{{ dtb_url }}'
{% endif %}

View File

@@ -1,55 +0,0 @@
#!/usr/bin/env python3
# Copyright © 2022 Valve 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.
from jinja2 import Environment, FileSystemLoader
from os import environ, path
# Pass all the environment variables prefixed by B2C_
values = {
key.removeprefix("B2C_").lower(): environ[key]
for key in environ if key.startswith("B2C_")
}
env = Environment(loader=FileSystemLoader(path.dirname(values['job_template'])),
trim_blocks=True, lstrip_blocks=True)
template = env.get_template(path.basename(values['job_template']))
values['ci_job_id'] = environ['CI_JOB_ID']
values['ci_runner_id'] = environ['CI_RUNNER_ID']
values['job_volume_exclusions'] = [excl for excl in values['job_volume_exclusions'].split(",") if excl]
values['working_dir'] = environ['CI_PROJECT_DIR']
# Use the gateway's pull-through registry caches to reduce load on fd.o.
values['local_container'] = environ['IMAGE_UNDER_TEST']
values['local_container'] = values['local_container'].replace(
'registry.freedesktop.org',
'{{ fdo_proxy_registry }}'
)
if 'kernel_cmdline_extras' not in values:
values['kernel_cmdline_extras'] = ''
with open(path.splitext(path.basename(values['job_template']))[0], "w") as f:
f.write(template.render(values))

View File

@@ -1,2 +0,0 @@
[*.sh]
indent_size = 2

View File

@@ -1,13 +0,0 @@
#!/bin/sh
# Init entrypoint for bare-metal devices; calls common init code.
# First stage: very basic setup to bring up network and /dev etc
/init-stage1.sh
# Second stage: run jobs
test $? -eq 0 && /init-stage2.sh
# Wait until the job would have timed out anyway, so we don't spew a "init
# exited" panic.
sleep 6000

View File

@@ -1,17 +0,0 @@
#!/bin/bash
# shellcheck disable=SC2086 # we want word splitting
if [ -z "$BM_POE_INTERFACE" ]; then
echo "Must supply the PoE Interface to power down"
exit 1
fi
if [ -z "$BM_POE_ADDRESS" ]; then
echo "Must supply the PoE Switch host"
exit 1
fi
SNMP_KEY="1.3.6.1.4.1.9.9.402.1.2.1.1.1.$BM_POE_INTERFACE"
SNMP_OFF="i 4"
snmpset -v2c -r 3 -t 30 -cmesaci "$BM_POE_ADDRESS" "$SNMP_KEY" $SNMP_OFF

View File

@@ -1,22 +0,0 @@
#!/bin/bash
# shellcheck disable=SC2086 # we want word splitting
if [ -z "$BM_POE_INTERFACE" ]; then
echo "Must supply the PoE Interface to power up"
exit 1
fi
if [ -z "$BM_POE_ADDRESS" ]; then
echo "Must supply the PoE Switch host"
exit 1
fi
set -ex
SNMP_KEY="1.3.6.1.4.1.9.9.402.1.2.1.1.1.$BM_POE_INTERFACE"
SNMP_ON="i 1"
SNMP_OFF="i 4"
snmpset -v2c -r 3 -t 10 -cmesaci "$BM_POE_ADDRESS" "$SNMP_KEY" $SNMP_OFF
sleep 3s
snmpset -v2c -r 3 -t 10 -cmesaci "$BM_POE_ADDRESS" "$SNMP_KEY" $SNMP_ON

View File

@@ -1,124 +0,0 @@
#!/bin/bash
# shellcheck disable=SC1091 # The relative paths in this file only become valid at runtime.
# shellcheck disable=SC2034
# shellcheck disable=SC2086 # we want word splitting
# Boot script for Chrome OS devices attached to a servo debug connector, using
# NFS and TFTP to boot.
# We're run from the root of the repo, make a helper var for our paths
BM=$CI_PROJECT_DIR/install/bare-metal
CI_COMMON=$CI_PROJECT_DIR/install/common
CI_INSTALL=$CI_PROJECT_DIR/install
# Runner config checks
if [ -z "$BM_SERIAL" ]; then
echo "Must set BM_SERIAL in your gitlab-runner config.toml [[runners]] environment"
echo "This is the CPU serial device."
exit 1
fi
if [ -z "$BM_SERIAL_EC" ]; then
echo "Must set BM_SERIAL in your gitlab-runner config.toml [[runners]] environment"
echo "This is the EC serial device for controlling board power"
exit 1
fi
if [ ! -d /nfs ]; then
echo "NFS rootfs directory needs to be mounted at /nfs by the gitlab runner"
exit 1
fi
if [ ! -d /tftp ]; then
echo "TFTP directory for this board needs to be mounted at /tftp by the gitlab runner"
exit 1
fi
# job config checks
if [ -z "$BM_KERNEL" ]; then
echo "Must set BM_KERNEL to your board's kernel FIT image"
exit 1
fi
if [ -z "$BM_ROOTFS" ]; then
echo "Must set BM_ROOTFS to your board's rootfs directory in the job's variables"
exit 1
fi
if [ -z "$BM_CMDLINE" ]; then
echo "Must set BM_CMDLINE to your board's kernel command line arguments"
exit 1
fi
set -ex
# Clear out any previous run's artifacts.
rm -rf results/
mkdir -p results
# Create the rootfs in the NFS directory. rm to make sure it's in a pristine
# state, since it's volume-mounted on the host.
rsync -a --delete $BM_ROOTFS/ /nfs/
mkdir -p /nfs/results
. $BM/rootfs-setup.sh /nfs
# Put the kernel/dtb image and the boot command line in the tftp directory for
# the board to find. For normal Mesa development, we build the kernel and
# store it in the docker container that this script is running in.
#
# However, container builds are expensive, so when you're hacking on the
# kernel, it's nice to be able to skip the half hour container build and plus
# moving that container to the runner. So, if BM_KERNEL is a URL, fetch it
# instead of looking in the container. Note that the kernel build should be
# the output of:
#
# make Image.lzma
#
# mkimage \
# -A arm64 \
# -f auto \
# -C lzma \
# -d arch/arm64/boot/Image.lzma \
# -b arch/arm64/boot/dts/qcom/sdm845-cheza-r3.dtb \
# cheza-image.img
rm -rf /tftp/*
if echo "$BM_KERNEL" | grep -q http; then
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
$BM_KERNEL -o /tftp/vmlinuz
elif [ -n "${FORCE_KERNEL_TAG}" ]; then
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
"${FDO_HTTP_CACHE_URI:-}${KERNEL_IMAGE_BASE}/${DEBIAN_ARCH}/${BM_KERNEL}" -o /tftp/vmlinuz
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
"${FDO_HTTP_CACHE_URI:-}${KERNEL_IMAGE_BASE}/${DEBIAN_ARCH}/modules.tar.zst" -o modules.tar.zst
tar --keep-directory-symlink --zstd -xf modules.tar.zst -C "/nfs/"
rm modules.tar.zst &
else
cp /baremetal-files/"$BM_KERNEL" /tftp/vmlinuz
fi
echo "$BM_CMDLINE" > /tftp/cmdline
set +e
STRUCTURED_LOG_FILE=job_detail.json
python3 $CI_INSTALL/custom_logger.py ${STRUCTURED_LOG_FILE} --update dut_job_type "${DEVICE_TYPE}"
python3 $CI_INSTALL/custom_logger.py ${STRUCTURED_LOG_FILE} --update farm "${FARM}"
python3 $CI_INSTALL/custom_logger.py ${STRUCTURED_LOG_FILE} --create-dut-job dut_name "${CI_RUNNER_DESCRIPTION}"
python3 $CI_INSTALL/custom_logger.py ${STRUCTURED_LOG_FILE} --update-dut-time submit "${CI_JOB_STARTED_AT}"
python3 $BM/cros_servo_run.py \
--cpu $BM_SERIAL \
--ec $BM_SERIAL_EC \
--test-timeout ${TEST_PHASE_TIMEOUT:-20}
ret=$?
python3 $CI_INSTALL/custom_logger.py ${STRUCTURED_LOG_FILE} --close-dut-job
python3 $CI_INSTALL/custom_logger.py ${STRUCTURED_LOG_FILE} --close
set -e
# Bring artifacts back from the NFS dir to the build dir where gitlab-runner
# will look for them.
cp -Rp /nfs/results/. results/
if [ -f "${STRUCTURED_LOG_FILE}" ]; then
cp -p ${STRUCTURED_LOG_FILE} results/
echo "Structured log file is available at https://${CI_PROJECT_ROOT_NAMESPACE}.pages.freedesktop.org/-/${CI_PROJECT_NAME}/-/jobs/${CI_JOB_ID}/artifacts/results/${STRUCTURED_LOG_FILE}"
fi
exit $ret

View File

@@ -1,168 +0,0 @@
#!/usr/bin/env python3
#
# Copyright © 2020 Google LLC
# SPDX-License-Identifier: MIT
import argparse
import re
import sys
from custom_logger import CustomLogger
from serial_buffer import SerialBuffer
class CrosServoRun:
def __init__(self, cpu, ec, test_timeout, logger):
self.cpu_ser = SerialBuffer(
cpu, "results/serial.txt", "R SERIAL-CPU> ")
# Merge the EC serial into the cpu_ser's line stream so that we can
# effectively poll on both at the same time and not have to worry about
self.ec_ser = SerialBuffer(
ec, "results/serial-ec.txt", "R SERIAL-EC> ", line_queue=self.cpu_ser.line_queue)
self.test_timeout = test_timeout
self.logger = logger
def close(self):
self.ec_ser.close()
self.cpu_ser.close()
def ec_write(self, s):
print("W SERIAL-EC> %s" % s)
self.ec_ser.serial.write(s.encode())
def cpu_write(self, s):
print("W SERIAL-CPU> %s" % s)
self.cpu_ser.serial.write(s.encode())
def print_error(self, message):
RED = '\033[0;31m'
NO_COLOR = '\033[0m'
print(RED + message + NO_COLOR)
self.logger.update_status_fail(message)
def run(self):
# Flush any partial commands in the EC's prompt, then ask for a reboot.
self.ec_write("\n")
self.ec_write("reboot\n")
bootloader_done = False
self.logger.create_job_phase("boot")
tftp_failures = 0
# This is emitted right when the bootloader pauses to check for input.
# Emit a ^N character to request network boot, because we don't have a
# direct-to-netboot firmware on cheza.
for line in self.cpu_ser.lines(timeout=120, phase="bootloader"):
if re.search("load_archive: loading locale_en.bin", line):
self.cpu_write("\016")
bootloader_done = True
break
# The Cheza firmware seems to occasionally get stuck looping in
# this error state during TFTP booting, possibly based on amount of
# network traffic around it, but it'll usually recover after a
# reboot. Currently mostly visible on google-freedreno-cheza-14.
if re.search("R8152: Bulk read error 0xffffffbf", line):
tftp_failures += 1
if tftp_failures >= 10:
self.print_error(
"Detected intermittent tftp failure, restarting run.")
return 1
# If the board has a netboot firmware and we made it to booting the
# kernel, proceed to processing of the test run.
if re.search("Booting Linux", line):
bootloader_done = True
break
# The Cheza boards have issues with failing to bring up power to
# the system sometimes, possibly dependent on ambient temperature
# in the farm.
if re.search("POWER_GOOD not seen in time", line):
self.print_error(
"Detected intermittent poweron failure, abandoning run.")
return 1
if not bootloader_done:
self.print_error("Failed to make it through bootloader, abandoning run.")
return 1
self.logger.create_job_phase("test")
for line in self.cpu_ser.lines(timeout=self.test_timeout, phase="test"):
if re.search("---. end Kernel panic", line):
return 1
# There are very infrequent bus errors during power management transitions
# on cheza, which we don't expect to be the case on future boards.
if re.search("Kernel panic - not syncing: Asynchronous SError Interrupt", line):
self.print_error(
"Detected cheza power management bus error, abandoning run.")
return 1
# If the network device dies, it's probably not graphics's fault, just try again.
if re.search("NETDEV WATCHDOG", line):
self.print_error(
"Detected network device failure, abandoning run.")
return 1
# These HFI response errors started appearing with the introduction
# of piglit runs. CosmicPenguin says:
#
# "message ID 106 isn't a thing, so likely what happened is that we
# got confused when parsing the HFI queue. If it happened on only
# one run, then memory corruption could be a possible clue"
#
# Given that it seems to trigger randomly near a GPU fault and then
# break many tests after that, just restart the whole run.
if re.search("a6xx_hfi_send_msg.*Unexpected message id .* on the response queue", line):
self.print_error(
"Detected cheza power management bus error, abandoning run.")
return 1
if re.search("coreboot.*bootblock starting", line):
self.print_error(
"Detected spontaneous reboot, abandoning run.")
return 1
if re.search("arm-smmu 5040000.iommu: TLB sync timed out -- SMMU may be deadlocked", line):
self.print_error("Detected cheza MMU fail, abandoning run.")
return 1
result = re.search("hwci: mesa: (\S*)", line)
if result:
if result.group(1) == "pass":
self.logger.update_dut_job("status", "pass")
return 0
else:
self.logger.update_status_fail("test fail")
return 1
self.print_error(
"Reached the end of the CPU serial log without finding a result")
return 1
def main():
parser = argparse.ArgumentParser()
parser.add_argument('--cpu', type=str,
help='CPU Serial device', required=True)
parser.add_argument(
'--ec', type=str, help='EC Serial device', required=True)
parser.add_argument(
'--test-timeout', type=int, help='Test phase timeout (minutes)', required=True)
args = parser.parse_args()
logger = CustomLogger("job_detail.json")
logger.update_dut_time("start", None)
servo = CrosServoRun(args.cpu, args.ec, args.test_timeout * 60, logger)
retval = servo.run()
# power down the CPU on the device
servo.ec_write("power off\n")
logger.update_dut_time("end", None)
servo.close()
sys.exit(retval)
if __name__ == '__main__':
main()

View File

@@ -1,10 +0,0 @@
#!/bin/bash
relay=$1
if [ -z "$relay" ]; then
echo "Must supply a relay arg"
exit 1
fi
"$CI_PROJECT_DIR"/install/bare-metal/eth008-power-relay.py "$ETH_HOST" "$ETH_PORT" off "$relay"

View File

@@ -1,28 +0,0 @@
#!/usr/bin/python3
import sys
import socket
host = sys.argv[1]
port = sys.argv[2]
mode = sys.argv[3]
relay = sys.argv[4]
msg = None
if mode == "on":
msg = b'\x20'
else:
msg = b'\x21'
msg += int(relay).to_bytes(1, 'big')
msg += b'\x00'
c = socket.create_connection((host, int(port)))
c.sendall(msg)
data = c.recv(1)
c.close()
if data[0] == b'\x01':
print('Command failed')
sys.exit(1)

View File

@@ -1,12 +0,0 @@
#!/bin/bash
relay=$1
if [ -z "$relay" ]; then
echo "Must supply a relay arg"
exit 1
fi
"$CI_PROJECT_DIR"/install/bare-metal/eth008-power-relay.py "$ETH_HOST" "$ETH_PORT" off "$relay"
sleep 5
"$CI_PROJECT_DIR"/install/bare-metal/eth008-power-relay.py "$ETH_HOST" "$ETH_PORT" on "$relay"

View File

@@ -1,31 +0,0 @@
#!/bin/bash
set -e
STRINGS=$(mktemp)
ERRORS=$(mktemp)
trap 'rm $STRINGS; rm $ERRORS;' EXIT
FILE=$1
shift 1
while getopts "f:e:" opt; do
case $opt in
f) echo "$OPTARG" >> "$STRINGS";;
e) echo "$OPTARG" >> "$STRINGS" ; echo "$OPTARG" >> "$ERRORS";;
*) exit
esac
done
shift $((OPTIND -1))
echo "Waiting for $FILE to say one of following strings"
cat "$STRINGS"
while ! grep -E -wf "$STRINGS" "$FILE"; do
sleep 2
done
if grep -E -wf "$ERRORS" "$FILE"; then
exit 1
fi

View File

@@ -1,167 +0,0 @@
#!/bin/bash
# shellcheck disable=SC1091 # The relative paths in this file only become valid at runtime.
# shellcheck disable=SC2034
# shellcheck disable=SC2086 # we want word splitting
. "$SCRIPTS_DIR"/setup-test-env.sh
BM=$CI_PROJECT_DIR/install/bare-metal
CI_COMMON=$CI_PROJECT_DIR/install/common
if [ -z "$BM_SERIAL" ] && [ -z "$BM_SERIAL_SCRIPT" ]; then
echo "Must set BM_SERIAL OR BM_SERIAL_SCRIPT in your gitlab-runner config.toml [[runners]] environment"
echo "BM_SERIAL:"
echo " This is the serial device to talk to for waiting for fastboot to be ready and logging from the kernel."
echo "BM_SERIAL_SCRIPT:"
echo " This is a shell script to talk to for waiting for fastboot to be ready and logging from the kernel."
exit 1
fi
if [ -z "$BM_POWERUP" ]; then
echo "Must set BM_POWERUP in your gitlab-runner config.toml [[runners]] environment"
echo "This is a shell script that should reset the device and begin its boot sequence"
echo "such that it pauses at fastboot."
exit 1
fi
if [ -z "$BM_POWERDOWN" ]; then
echo "Must set BM_POWERDOWN in your gitlab-runner config.toml [[runners]] environment"
echo "This is a shell script that should power off the device."
exit 1
fi
if [ -z "$BM_FASTBOOT_SERIAL" ]; then
echo "Must set BM_FASTBOOT_SERIAL in your gitlab-runner config.toml [[runners]] environment"
echo "This must be the a stable-across-resets fastboot serial number."
exit 1
fi
if [ -z "$BM_KERNEL" ]; then
echo "Must set BM_KERNEL to your board's kernel vmlinuz or Image.gz in the job's variables:"
exit 1
fi
if [ -z "$BM_DTB" ]; then
echo "Must set BM_DTB to your board's DTB file in the job's variables:"
exit 1
fi
if [ -z "$BM_ROOTFS" ]; then
echo "Must set BM_ROOTFS to your board's rootfs directory in the job's variables:"
exit 1
fi
if echo $BM_CMDLINE | grep -q "root=/dev/nfs"; then
BM_FASTBOOT_NFSROOT=1
fi
set -ex
# Clear out any previous run's artifacts.
rm -rf results/
mkdir -p results/
if [ -n "$BM_FASTBOOT_NFSROOT" ]; then
# Create the rootfs in the NFS directory. rm to make sure it's in a pristine
# state, since it's volume-mounted on the host.
rsync -a --delete $BM_ROOTFS/ /nfs/
mkdir -p /nfs/results
. $BM/rootfs-setup.sh /nfs
# Root on NFS, no need for an inintramfs.
rm -f rootfs.cpio.gz
touch rootfs.cpio
gzip rootfs.cpio
else
# Create the rootfs in a temp dir
rsync -a --delete $BM_ROOTFS/ rootfs/
. $BM/rootfs-setup.sh rootfs
# Finally, pack it up into a cpio rootfs. Skip the vulkan CTS since none of
# these devices use it and it would take up space in the initrd.
if [ -n "$PIGLIT_PROFILES" ]; then
EXCLUDE_FILTER="deqp|arb_gpu_shader5|arb_gpu_shader_fp64|arb_gpu_shader_int64|glsl-4.[0123456]0|arb_tessellation_shader"
else
EXCLUDE_FILTER="piglit|python"
fi
pushd rootfs
find -H . | \
grep -E -v "external/(openglcts|vulkancts|amber|glslang|spirv-tools)" |
grep -E -v "traces-db|apitrace|renderdoc" | \
grep -E -v $EXCLUDE_FILTER | \
cpio -H newc -o | \
xz --check=crc32 -T4 - > $CI_PROJECT_DIR/rootfs.cpio.gz
popd
fi
if echo "$BM_KERNEL $BM_DTB" | grep -q http; then
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
"$BM_KERNEL" -o kernel
# FIXME: modules should be supplied too
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
"$BM_DTB" -o dtb
cat kernel dtb > Image.gz-dtb
elif [ -n "${FORCE_KERNEL_TAG}" ]; then
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
"${FDO_HTTP_CACHE_URI:-}${KERNEL_IMAGE_BASE}/${DEBIAN_ARCH}/${BM_KERNEL}" -o kernel
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
"${FDO_HTTP_CACHE_URI:-}${KERNEL_IMAGE_BASE}/${DEBIAN_ARCH}/modules.tar.zst" -o modules.tar.zst
if [ -n "$BM_DTB" ]; then
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
"${FDO_HTTP_CACHE_URI:-}${KERNEL_IMAGE_BASE}/${DEBIAN_ARCH}/${BM_DTB}.dtb" -o dtb
fi
cat kernel dtb > Image.gz-dtb || echo "No DTB available, using pure kernel."
rm kernel
tar --keep-directory-symlink --zstd -xf modules.tar.zst -C "$BM_ROOTFS/"
rm modules.tar.zst &
else
cat /baremetal-files/"$BM_KERNEL" /baremetal-files/"$BM_DTB".dtb > Image.gz-dtb
cp /baremetal-files/"$BM_DTB".dtb dtb
fi
export PATH=$BM:$PATH
mkdir -p artifacts
mkbootimg.py \
--kernel Image.gz-dtb \
--ramdisk rootfs.cpio.gz \
--dtb dtb \
--cmdline "$BM_CMDLINE" \
$BM_MKBOOT_PARAMS \
--header_version 2 \
-o artifacts/fastboot.img
rm Image.gz-dtb dtb
# Start background command for talking to serial if we have one.
if [ -n "$BM_SERIAL_SCRIPT" ]; then
$BM_SERIAL_SCRIPT > results/serial-output.txt &
while [ ! -e results/serial-output.txt ]; do
sleep 1
done
fi
set +e
$BM/fastboot_run.py \
--dev="$BM_SERIAL" \
--test-timeout ${TEST_PHASE_TIMEOUT:-20} \
--fbserial="$BM_FASTBOOT_SERIAL" \
--powerup="$BM_POWERUP" \
--powerdown="$BM_POWERDOWN"
ret=$?
set -e
if [ -n "$BM_FASTBOOT_NFSROOT" ]; then
# Bring artifacts back from the NFS dir to the build dir where gitlab-runner
# will look for them.
cp -Rp /nfs/results/. results/
fi
exit $ret

View File

@@ -1,159 +0,0 @@
#!/usr/bin/env python3
#
# Copyright © 2020 Google LLC
#
# 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.
import argparse
import subprocess
import re
from serial_buffer import SerialBuffer
import sys
import threading
class FastbootRun:
def __init__(self, args, test_timeout):
self.powerup = args.powerup
self.ser = SerialBuffer(
args.dev, "results/serial-output.txt", "R SERIAL> ")
self.fastboot = "fastboot boot -s {ser} artifacts/fastboot.img".format(
ser=args.fbserial)
self.test_timeout = test_timeout
def close(self):
self.ser.close()
def print_error(self, message):
RED = '\033[0;31m'
NO_COLOR = '\033[0m'
print(RED + message + NO_COLOR)
def logged_system(self, cmd, timeout=60):
print("Running '{}'".format(cmd))
try:
return subprocess.call(cmd, shell=True, timeout=timeout)
except subprocess.TimeoutExpired:
self.print_error("timeout, abandoning run.")
return 1
def run(self):
if ret := self.logged_system(self.powerup):
return ret
fastboot_ready = False
for line in self.ser.lines(timeout=2 * 60, phase="bootloader"):
if re.search("[Ff]astboot: [Pp]rocessing commands", line) or \
re.search("Listening for fastboot command on", line):
fastboot_ready = True
break
if re.search("data abort", line):
self.print_error(
"Detected crash during boot, abandoning run.")
return 1
if not fastboot_ready:
self.print_error(
"Failed to get to fastboot prompt, abandoning run.")
return 1
if ret := self.logged_system(self.fastboot):
return ret
print_more_lines = -1
for line in self.ser.lines(timeout=self.test_timeout, phase="test"):
if print_more_lines == 0:
return 1
if print_more_lines > 0:
print_more_lines -= 1
if re.search("---. end Kernel panic", line):
return 1
# The db820c boards intermittently reboot. Just restart the run
# when if we see a reboot after we got past fastboot.
if re.search("PON REASON", line):
self.print_error(
"Detected spontaneous reboot, abandoning run.")
return 1
# db820c sometimes wedges around iommu fault recovery
if re.search("watchdog: BUG: soft lockup - CPU.* stuck", line):
self.print_error(
"Detected kernel soft lockup, abandoning run.")
return 1
# If the network device dies, it's probably not graphics's fault, just try again.
if re.search("NETDEV WATCHDOG", line):
self.print_error(
"Detected network device failure, abandoning run.")
return 1
# A3xx recovery doesn't quite work. Sometimes the GPU will get
# wedged and recovery will fail (because power can't be reset?)
# This assumes that the jobs are sufficiently well-tested that GPU
# hangs aren't always triggered, so just try again. But print some
# more lines first so that we get better information on the cause
# of the hang. Once a hang happens, it's pretty chatty.
if "[drm:adreno_recover] *ERROR* gpu hw init failed: -22" in line:
self.print_error(
"Detected GPU hang, abandoning run.")
if print_more_lines == -1:
print_more_lines = 30
result = re.search("hwci: mesa: (\S*)", line)
if result:
if result.group(1) == "pass":
return 0
else:
return 1
self.print_error(
"Reached the end of the CPU serial log without finding a result, abandoning run.")
return 1
def main():
parser = argparse.ArgumentParser()
parser.add_argument(
'--dev', type=str, help='Serial device (otherwise reading from serial-output.txt)')
parser.add_argument('--powerup', type=str,
help='shell command for rebooting', required=True)
parser.add_argument('--powerdown', type=str,
help='shell command for powering off', required=True)
parser.add_argument('--fbserial', type=str,
help='fastboot serial number of the board', required=True)
parser.add_argument('--test-timeout', type=int,
help='Test phase timeout (minutes)', required=True)
args = parser.parse_args()
fastboot = FastbootRun(args, args.test_timeout * 60)
retval = fastboot.run()
fastboot.close()
fastboot.logged_system(args.powerdown)
sys.exit(retval)
if __name__ == '__main__':
main()

View File

@@ -1,10 +0,0 @@
#!/bin/bash
relay=$1
if [ -z "$relay" ]; then
echo "Must supply a relay arg"
exit 1
fi
"$CI_PROJECT_DIR"/install/bare-metal/google-power-relay.py off "$relay"

View File

@@ -1,19 +0,0 @@
#!/usr/bin/python3
import sys
import serial
mode = sys.argv[1]
relay = sys.argv[2]
# our relays are "off" means "board is powered".
mode_swap = {
"on": "off",
"off": "on",
}
mode = mode_swap[mode]
ser = serial.Serial('/dev/ttyACM0', 115200, timeout=2)
command = "relay {} {}\n\r".format(mode, relay)
ser.write(command.encode())
ser.close()

View File

@@ -1,12 +0,0 @@
#!/bin/bash
relay=$1
if [ -z "$relay" ]; then
echo "Must supply a relay arg"
exit 1
fi
"$CI_PROJECT_DIR"/install/bare-metal/google-power-relay.py off "$relay"
sleep 5
"$CI_PROJECT_DIR"/install/bare-metal/google-power-relay.py on "$relay"

View File

@@ -1,569 +0,0 @@
#!/usr/bin/env python3
#
# Copyright 2015, The Android Open Source Project
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""Creates the boot image."""
from argparse import (ArgumentParser, ArgumentTypeError,
FileType, RawDescriptionHelpFormatter)
from hashlib import sha1
from os import fstat
from struct import pack
import array
import collections
import os
import re
import subprocess
import tempfile
# Constant and structure definition is in
# system/tools/mkbootimg/include/bootimg/bootimg.h
BOOT_MAGIC = 'ANDROID!'
BOOT_MAGIC_SIZE = 8
BOOT_NAME_SIZE = 16
BOOT_ARGS_SIZE = 512
BOOT_EXTRA_ARGS_SIZE = 1024
BOOT_IMAGE_HEADER_V1_SIZE = 1648
BOOT_IMAGE_HEADER_V2_SIZE = 1660
BOOT_IMAGE_HEADER_V3_SIZE = 1580
BOOT_IMAGE_HEADER_V3_PAGESIZE = 4096
BOOT_IMAGE_HEADER_V4_SIZE = 1584
BOOT_IMAGE_V4_SIGNATURE_SIZE = 4096
VENDOR_BOOT_MAGIC = 'VNDRBOOT'
VENDOR_BOOT_MAGIC_SIZE = 8
VENDOR_BOOT_NAME_SIZE = BOOT_NAME_SIZE
VENDOR_BOOT_ARGS_SIZE = 2048
VENDOR_BOOT_IMAGE_HEADER_V3_SIZE = 2112
VENDOR_BOOT_IMAGE_HEADER_V4_SIZE = 2128
VENDOR_RAMDISK_TYPE_NONE = 0
VENDOR_RAMDISK_TYPE_PLATFORM = 1
VENDOR_RAMDISK_TYPE_RECOVERY = 2
VENDOR_RAMDISK_TYPE_DLKM = 3
VENDOR_RAMDISK_NAME_SIZE = 32
VENDOR_RAMDISK_TABLE_ENTRY_BOARD_ID_SIZE = 16
VENDOR_RAMDISK_TABLE_ENTRY_V4_SIZE = 108
# Names with special meaning, mustn't be specified in --ramdisk_name.
VENDOR_RAMDISK_NAME_BLOCKLIST = {b'default'}
PARSER_ARGUMENT_VENDOR_RAMDISK_FRAGMENT = '--vendor_ramdisk_fragment'
def filesize(f):
if f is None:
return 0
try:
return fstat(f.fileno()).st_size
except OSError:
return 0
def update_sha(sha, f):
if f:
sha.update(f.read())
f.seek(0)
sha.update(pack('I', filesize(f)))
else:
sha.update(pack('I', 0))
def pad_file(f, padding):
pad = (padding - (f.tell() & (padding - 1))) & (padding - 1)
f.write(pack(str(pad) + 'x'))
def get_number_of_pages(image_size, page_size):
"""calculates the number of pages required for the image"""
return (image_size + page_size - 1) // page_size
def get_recovery_dtbo_offset(args):
"""calculates the offset of recovery_dtbo image in the boot image"""
num_header_pages = 1 # header occupies a page
num_kernel_pages = get_number_of_pages(filesize(args.kernel), args.pagesize)
num_ramdisk_pages = get_number_of_pages(filesize(args.ramdisk),
args.pagesize)
num_second_pages = get_number_of_pages(filesize(args.second), args.pagesize)
dtbo_offset = args.pagesize * (num_header_pages + num_kernel_pages +
num_ramdisk_pages + num_second_pages)
return dtbo_offset
def write_header_v3_and_above(args):
if args.header_version > 3:
boot_header_size = BOOT_IMAGE_HEADER_V4_SIZE
else:
boot_header_size = BOOT_IMAGE_HEADER_V3_SIZE
args.output.write(pack(f'{BOOT_MAGIC_SIZE}s', BOOT_MAGIC.encode()))
# kernel size in bytes
args.output.write(pack('I', filesize(args.kernel)))
# ramdisk size in bytes
args.output.write(pack('I', filesize(args.ramdisk)))
# os version and patch level
args.output.write(pack('I', (args.os_version << 11) | args.os_patch_level))
args.output.write(pack('I', boot_header_size))
# reserved
args.output.write(pack('4I', 0, 0, 0, 0))
# version of boot image header
args.output.write(pack('I', args.header_version))
args.output.write(pack(f'{BOOT_ARGS_SIZE + BOOT_EXTRA_ARGS_SIZE}s',
args.cmdline))
if args.header_version >= 4:
# The signature used to verify boot image v4.
args.output.write(pack('I', BOOT_IMAGE_V4_SIGNATURE_SIZE))
pad_file(args.output, BOOT_IMAGE_HEADER_V3_PAGESIZE)
def write_vendor_boot_header(args):
if filesize(args.dtb) == 0:
raise ValueError('DTB image must not be empty.')
if args.header_version > 3:
vendor_ramdisk_size = args.vendor_ramdisk_total_size
vendor_boot_header_size = VENDOR_BOOT_IMAGE_HEADER_V4_SIZE
else:
vendor_ramdisk_size = filesize(args.vendor_ramdisk)
vendor_boot_header_size = VENDOR_BOOT_IMAGE_HEADER_V3_SIZE
args.vendor_boot.write(pack(f'{VENDOR_BOOT_MAGIC_SIZE}s',
VENDOR_BOOT_MAGIC.encode()))
# version of boot image header
args.vendor_boot.write(pack('I', args.header_version))
# flash page size
args.vendor_boot.write(pack('I', args.pagesize))
# kernel physical load address
args.vendor_boot.write(pack('I', args.base + args.kernel_offset))
# ramdisk physical load address
args.vendor_boot.write(pack('I', args.base + args.ramdisk_offset))
# ramdisk size in bytes
args.vendor_boot.write(pack('I', vendor_ramdisk_size))
args.vendor_boot.write(pack(f'{VENDOR_BOOT_ARGS_SIZE}s',
args.vendor_cmdline))
# kernel tags physical load address
args.vendor_boot.write(pack('I', args.base + args.tags_offset))
# asciiz product name
args.vendor_boot.write(pack(f'{VENDOR_BOOT_NAME_SIZE}s', args.board))
# header size in bytes
args.vendor_boot.write(pack('I', vendor_boot_header_size))
# dtb size in bytes
args.vendor_boot.write(pack('I', filesize(args.dtb)))
# dtb physical load address
args.vendor_boot.write(pack('Q', args.base + args.dtb_offset))
if args.header_version > 3:
vendor_ramdisk_table_size = (args.vendor_ramdisk_table_entry_num *
VENDOR_RAMDISK_TABLE_ENTRY_V4_SIZE)
# vendor ramdisk table size in bytes
args.vendor_boot.write(pack('I', vendor_ramdisk_table_size))
# number of vendor ramdisk table entries
args.vendor_boot.write(pack('I', args.vendor_ramdisk_table_entry_num))
# vendor ramdisk table entry size in bytes
args.vendor_boot.write(pack('I', VENDOR_RAMDISK_TABLE_ENTRY_V4_SIZE))
# bootconfig section size in bytes
args.vendor_boot.write(pack('I', filesize(args.vendor_bootconfig)))
pad_file(args.vendor_boot, args.pagesize)
def write_header(args):
if args.header_version > 4:
raise ValueError(
f'Boot header version {args.header_version} not supported')
if args.header_version in {3, 4}:
return write_header_v3_and_above(args)
ramdisk_load_address = ((args.base + args.ramdisk_offset)
if filesize(args.ramdisk) > 0 else 0)
second_load_address = ((args.base + args.second_offset)
if filesize(args.second) > 0 else 0)
args.output.write(pack(f'{BOOT_MAGIC_SIZE}s', BOOT_MAGIC.encode()))
# kernel size in bytes
args.output.write(pack('I', filesize(args.kernel)))
# kernel physical load address
args.output.write(pack('I', args.base + args.kernel_offset))
# ramdisk size in bytes
args.output.write(pack('I', filesize(args.ramdisk)))
# ramdisk physical load address
args.output.write(pack('I', ramdisk_load_address))
# second bootloader size in bytes
args.output.write(pack('I', filesize(args.second)))
# second bootloader physical load address
args.output.write(pack('I', second_load_address))
# kernel tags physical load address
args.output.write(pack('I', args.base + args.tags_offset))
# flash page size
args.output.write(pack('I', args.pagesize))
# version of boot image header
args.output.write(pack('I', args.header_version))
# os version and patch level
args.output.write(pack('I', (args.os_version << 11) | args.os_patch_level))
# asciiz product name
args.output.write(pack(f'{BOOT_NAME_SIZE}s', args.board))
args.output.write(pack(f'{BOOT_ARGS_SIZE}s', args.cmdline))
sha = sha1()
update_sha(sha, args.kernel)
update_sha(sha, args.ramdisk)
update_sha(sha, args.second)
if args.header_version > 0:
update_sha(sha, args.recovery_dtbo)
if args.header_version > 1:
update_sha(sha, args.dtb)
img_id = pack('32s', sha.digest())
args.output.write(img_id)
args.output.write(pack(f'{BOOT_EXTRA_ARGS_SIZE}s', args.extra_cmdline))
if args.header_version > 0:
if args.recovery_dtbo:
# recovery dtbo size in bytes
args.output.write(pack('I', filesize(args.recovery_dtbo)))
# recovert dtbo offset in the boot image
args.output.write(pack('Q', get_recovery_dtbo_offset(args)))
else:
# Set to zero if no recovery dtbo
args.output.write(pack('I', 0))
args.output.write(pack('Q', 0))
# Populate boot image header size for header versions 1 and 2.
if args.header_version == 1:
args.output.write(pack('I', BOOT_IMAGE_HEADER_V1_SIZE))
elif args.header_version == 2:
args.output.write(pack('I', BOOT_IMAGE_HEADER_V2_SIZE))
if args.header_version > 1:
if filesize(args.dtb) == 0:
raise ValueError('DTB image must not be empty.')
# dtb size in bytes
args.output.write(pack('I', filesize(args.dtb)))
# dtb physical load address
args.output.write(pack('Q', args.base + args.dtb_offset))
pad_file(args.output, args.pagesize)
return img_id
class AsciizBytes:
"""Parses a string and encodes it as an asciiz bytes object.
>>> AsciizBytes(bufsize=4)('foo')
b'foo\\x00'
>>> AsciizBytes(bufsize=4)('foob')
Traceback (most recent call last):
...
argparse.ArgumentTypeError: Encoded asciiz length exceeded: max 4, got 5
"""
def __init__(self, bufsize):
self.bufsize = bufsize
def __call__(self, arg):
arg_bytes = arg.encode() + b'\x00'
if len(arg_bytes) > self.bufsize:
raise ArgumentTypeError(
'Encoded asciiz length exceeded: '
f'max {self.bufsize}, got {len(arg_bytes)}')
return arg_bytes
class VendorRamdiskTableBuilder:
"""Vendor ramdisk table builder.
Attributes:
entries: A list of VendorRamdiskTableEntry namedtuple.
ramdisk_total_size: Total size in bytes of all ramdisks in the table.
"""
VendorRamdiskTableEntry = collections.namedtuple( # pylint: disable=invalid-name
'VendorRamdiskTableEntry',
['ramdisk_path', 'ramdisk_size', 'ramdisk_offset', 'ramdisk_type',
'ramdisk_name', 'board_id'])
def __init__(self):
self.entries = []
self.ramdisk_total_size = 0
self.ramdisk_names = set()
def add_entry(self, ramdisk_path, ramdisk_type, ramdisk_name, board_id):
# Strip any trailing null for simple comparison.
stripped_ramdisk_name = ramdisk_name.rstrip(b'\x00')
if stripped_ramdisk_name in VENDOR_RAMDISK_NAME_BLOCKLIST:
raise ValueError(
f'Banned vendor ramdisk name: {stripped_ramdisk_name}')
if stripped_ramdisk_name in self.ramdisk_names:
raise ValueError(
f'Duplicated vendor ramdisk name: {stripped_ramdisk_name}')
self.ramdisk_names.add(stripped_ramdisk_name)
if board_id is None:
board_id = array.array(
'I', [0] * VENDOR_RAMDISK_TABLE_ENTRY_BOARD_ID_SIZE)
else:
board_id = array.array('I', board_id)
if len(board_id) != VENDOR_RAMDISK_TABLE_ENTRY_BOARD_ID_SIZE:
raise ValueError('board_id size must be '
f'{VENDOR_RAMDISK_TABLE_ENTRY_BOARD_ID_SIZE}')
with open(ramdisk_path, 'rb') as f:
ramdisk_size = filesize(f)
self.entries.append(self.VendorRamdiskTableEntry(
ramdisk_path, ramdisk_size, self.ramdisk_total_size, ramdisk_type,
ramdisk_name, board_id))
self.ramdisk_total_size += ramdisk_size
def write_ramdisks_padded(self, fout, alignment):
for entry in self.entries:
with open(entry.ramdisk_path, 'rb') as f:
fout.write(f.read())
pad_file(fout, alignment)
def write_entries_padded(self, fout, alignment):
for entry in self.entries:
fout.write(pack('I', entry.ramdisk_size))
fout.write(pack('I', entry.ramdisk_offset))
fout.write(pack('I', entry.ramdisk_type))
fout.write(pack(f'{VENDOR_RAMDISK_NAME_SIZE}s',
entry.ramdisk_name))
fout.write(entry.board_id)
pad_file(fout, alignment)
def write_padded_file(f_out, f_in, padding):
if f_in is None:
return
f_out.write(f_in.read())
pad_file(f_out, padding)
def parse_int(x):
return int(x, 0)
def parse_os_version(x):
match = re.search(r'^(\d{1,3})(?:\.(\d{1,3})(?:\.(\d{1,3}))?)?', x)
if match:
a = int(match.group(1))
b = c = 0
if match.lastindex >= 2:
b = int(match.group(2))
if match.lastindex == 3:
c = int(match.group(3))
# 7 bits allocated for each field
assert a < 128
assert b < 128
assert c < 128
return (a << 14) | (b << 7) | c
return 0
def parse_os_patch_level(x):
match = re.search(r'^(\d{4})-(\d{2})(?:-(\d{2}))?', x)
if match:
y = int(match.group(1)) - 2000
m = int(match.group(2))
# 7 bits allocated for the year, 4 bits for the month
assert 0 <= y < 128
assert 0 < m <= 12
return (y << 4) | m
return 0
def parse_vendor_ramdisk_type(x):
type_dict = {
'none': VENDOR_RAMDISK_TYPE_NONE,
'platform': VENDOR_RAMDISK_TYPE_PLATFORM,
'recovery': VENDOR_RAMDISK_TYPE_RECOVERY,
'dlkm': VENDOR_RAMDISK_TYPE_DLKM,
}
if x.lower() in type_dict:
return type_dict[x.lower()]
return parse_int(x)
def get_vendor_boot_v4_usage():
return """vendor boot version 4 arguments:
--ramdisk_type {none,platform,recovery,dlkm}
specify the type of the ramdisk
--ramdisk_name NAME
specify the name of the ramdisk
--board_id{0..15} NUMBER
specify the value of the board_id vector, defaults to 0
--vendor_ramdisk_fragment VENDOR_RAMDISK_FILE
path to the vendor ramdisk file
These options can be specified multiple times, where each vendor ramdisk
option group ends with a --vendor_ramdisk_fragment option.
Each option group appends an additional ramdisk to the vendor boot image.
"""
def parse_vendor_ramdisk_args(args, args_list):
"""Parses vendor ramdisk specific arguments.
Args:
args: An argparse.Namespace object. Parsed results are stored into this
object.
args_list: A list of argument strings to be parsed.
Returns:
A list argument strings that are not parsed by this method.
"""
parser = ArgumentParser(add_help=False)
parser.add_argument('--ramdisk_type', type=parse_vendor_ramdisk_type,
default=VENDOR_RAMDISK_TYPE_NONE)
parser.add_argument('--ramdisk_name',
type=AsciizBytes(bufsize=VENDOR_RAMDISK_NAME_SIZE),
required=True)
for i in range(VENDOR_RAMDISK_TABLE_ENTRY_BOARD_ID_SIZE):
parser.add_argument(f'--board_id{i}', type=parse_int, default=0)
parser.add_argument(PARSER_ARGUMENT_VENDOR_RAMDISK_FRAGMENT, required=True)
unknown_args = []
vendor_ramdisk_table_builder = VendorRamdiskTableBuilder()
if args.vendor_ramdisk is not None:
vendor_ramdisk_table_builder.add_entry(
args.vendor_ramdisk.name, VENDOR_RAMDISK_TYPE_PLATFORM, b'', None)
while PARSER_ARGUMENT_VENDOR_RAMDISK_FRAGMENT in args_list:
idx = args_list.index(PARSER_ARGUMENT_VENDOR_RAMDISK_FRAGMENT) + 2
vendor_ramdisk_args = args_list[:idx]
args_list = args_list[idx:]
ramdisk_args, extra_args = parser.parse_known_args(vendor_ramdisk_args)
ramdisk_args_dict = vars(ramdisk_args)
unknown_args.extend(extra_args)
ramdisk_path = ramdisk_args.vendor_ramdisk_fragment
ramdisk_type = ramdisk_args.ramdisk_type
ramdisk_name = ramdisk_args.ramdisk_name
board_id = [ramdisk_args_dict[f'board_id{i}']
for i in range(VENDOR_RAMDISK_TABLE_ENTRY_BOARD_ID_SIZE)]
vendor_ramdisk_table_builder.add_entry(ramdisk_path, ramdisk_type,
ramdisk_name, board_id)
if len(args_list) > 0:
unknown_args.extend(args_list)
args.vendor_ramdisk_total_size = (vendor_ramdisk_table_builder
.ramdisk_total_size)
args.vendor_ramdisk_table_entry_num = len(vendor_ramdisk_table_builder
.entries)
args.vendor_ramdisk_table_builder = vendor_ramdisk_table_builder
return unknown_args
def parse_cmdline():
version_parser = ArgumentParser(add_help=False)
version_parser.add_argument('--header_version', type=parse_int, default=0)
if version_parser.parse_known_args()[0].header_version < 3:
# For boot header v0 to v2, the kernel commandline field is split into
# two fields, cmdline and extra_cmdline. Both fields are asciiz strings,
# so we minus one here to ensure the encoded string plus the
# null-terminator can fit in the buffer size.
cmdline_size = BOOT_ARGS_SIZE + BOOT_EXTRA_ARGS_SIZE - 1
else:
cmdline_size = BOOT_ARGS_SIZE + BOOT_EXTRA_ARGS_SIZE
parser = ArgumentParser(formatter_class=RawDescriptionHelpFormatter,
epilog=get_vendor_boot_v4_usage())
parser.add_argument('--kernel', type=FileType('rb'),
help='path to the kernel')
parser.add_argument('--ramdisk', type=FileType('rb'),
help='path to the ramdisk')
parser.add_argument('--second', type=FileType('rb'),
help='path to the second bootloader')
parser.add_argument('--dtb', type=FileType('rb'), help='path to the dtb')
dtbo_group = parser.add_mutually_exclusive_group()
dtbo_group.add_argument('--recovery_dtbo', type=FileType('rb'),
help='path to the recovery DTBO')
dtbo_group.add_argument('--recovery_acpio', type=FileType('rb'),
metavar='RECOVERY_ACPIO', dest='recovery_dtbo',
help='path to the recovery ACPIO')
parser.add_argument('--cmdline', type=AsciizBytes(bufsize=cmdline_size),
default='', help='kernel command line arguments')
parser.add_argument('--vendor_cmdline',
type=AsciizBytes(bufsize=VENDOR_BOOT_ARGS_SIZE),
default='',
help='vendor boot kernel command line arguments')
parser.add_argument('--base', type=parse_int, default=0x10000000,
help='base address')
parser.add_argument('--kernel_offset', type=parse_int, default=0x00008000,
help='kernel offset')
parser.add_argument('--ramdisk_offset', type=parse_int, default=0x01000000,
help='ramdisk offset')
parser.add_argument('--second_offset', type=parse_int, default=0x00f00000,
help='second bootloader offset')
parser.add_argument('--dtb_offset', type=parse_int, default=0x01f00000,
help='dtb offset')
parser.add_argument('--os_version', type=parse_os_version, default=0,
help='operating system version')
parser.add_argument('--os_patch_level', type=parse_os_patch_level,
default=0, help='operating system patch level')
parser.add_argument('--tags_offset', type=parse_int, default=0x00000100,
help='tags offset')
parser.add_argument('--board', type=AsciizBytes(bufsize=BOOT_NAME_SIZE),
default='', help='board name')
parser.add_argument('--pagesize', type=parse_int,
choices=[2**i for i in range(11, 15)], default=2048,
help='page size')
parser.add_argument('--id', action='store_true',
help='print the image ID on standard output')
parser.add_argument('--header_version', type=parse_int, default=0,
help='boot image header version')
parser.add_argument('-o', '--output', type=FileType('wb'),
help='output file name')
parser.add_argument('--gki_signing_algorithm',
help='GKI signing algorithm to use')
parser.add_argument('--gki_signing_key',
help='path to RSA private key file')
parser.add_argument('--gki_signing_signature_args',
help='other hash arguments passed to avbtool')
parser.add_argument('--gki_signing_avbtool_path',
help='path to avbtool for boot signature generation')
parser.add_argument('--vendor_boot', type=FileType('wb'),
help='vendor boot output file name')
parser.add_argument('--vendor_ramdisk', type=FileType('rb'),
help='path to the vendor ramdisk')
parser.add_argument('--vendor_bootconfig', type=FileType('rb'),
help='path to the vendor bootconfig file')
args, extra_args = parser.parse_known_args()
if args.vendor_boot is not None and args.header_version > 3:
extra_args = parse_vendor_ramdisk_args(args, extra_args)
if len(extra_args) > 0:
raise ValueError(f'Unrecognized arguments: {extra_args}')
if args.header_version < 3:
args.extra_cmdline = args.cmdline[BOOT_ARGS_SIZE-1:]
args.cmdline = args.cmdline[:BOOT_ARGS_SIZE-1] + b'\x00'
assert len(args.cmdline) <= BOOT_ARGS_SIZE
assert len(args.extra_cmdline) <= BOOT_EXTRA_ARGS_SIZE
return args
def add_boot_image_signature(args, pagesize):
"""Adds the boot image signature.
Note that the signature will only be verified in VTS to ensure a
generic boot.img is used. It will not be used by the device
bootloader at boot time. The bootloader should only verify
the boot vbmeta at the end of the boot partition (or in the top-level
vbmeta partition) via the Android Verified Boot process, when the
device boots.
"""
args.output.flush() # Flush the buffer for signature calculation.
# Appends zeros if the signing key is not specified.
if not args.gki_signing_key or not args.gki_signing_algorithm:
zeros = b'\x00' * BOOT_IMAGE_V4_SIGNATURE_SIZE
args.output.write(zeros)
pad_file(args.output, pagesize)
return
avbtool = 'avbtool' # Used from otatools.zip or Android build env.
# We need to specify the path of avbtool in build/core/Makefile.
# Because avbtool is not guaranteed to be in $PATH there.
if args.gki_signing_avbtool_path:
avbtool = args.gki_signing_avbtool_path
# Need to specify a value of --partition_size for avbtool to work.
# We use 64 MB below, but avbtool will not resize the boot image to
# this size because --do_not_append_vbmeta_image is also specified.
avbtool_cmd = [
avbtool, 'add_hash_footer',
'--partition_name', 'boot',
'--partition_size', str(64 * 1024 * 1024),
'--image', args.output.name,
'--algorithm', args.gki_signing_algorithm,
'--key', args.gki_signing_key,
'--salt', 'd00df00d'] # TODO: use a hash of kernel/ramdisk as the salt.
# Additional arguments passed to avbtool.
if args.gki_signing_signature_args:
avbtool_cmd += args.gki_signing_signature_args.split()
# Outputs the signed vbmeta to a separate file, then append to boot.img
# as the boot signature.
with tempfile.TemporaryDirectory() as temp_out_dir:
boot_signature_output = os.path.join(temp_out_dir, 'boot_signature')
avbtool_cmd += ['--do_not_append_vbmeta_image',
'--output_vbmeta_image', boot_signature_output]
subprocess.check_call(avbtool_cmd)
with open(boot_signature_output, 'rb') as boot_signature:
if filesize(boot_signature) > BOOT_IMAGE_V4_SIGNATURE_SIZE:
raise ValueError(
f'boot sigature size is > {BOOT_IMAGE_V4_SIGNATURE_SIZE}')
write_padded_file(args.output, boot_signature, pagesize)
def write_data(args, pagesize):
write_padded_file(args.output, args.kernel, pagesize)
write_padded_file(args.output, args.ramdisk, pagesize)
write_padded_file(args.output, args.second, pagesize)
if args.header_version > 0 and args.header_version < 3:
write_padded_file(args.output, args.recovery_dtbo, pagesize)
if args.header_version == 2:
write_padded_file(args.output, args.dtb, pagesize)
if args.header_version >= 4:
add_boot_image_signature(args, pagesize)
def write_vendor_boot_data(args):
if args.header_version > 3:
builder = args.vendor_ramdisk_table_builder
builder.write_ramdisks_padded(args.vendor_boot, args.pagesize)
write_padded_file(args.vendor_boot, args.dtb, args.pagesize)
builder.write_entries_padded(args.vendor_boot, args.pagesize)
write_padded_file(args.vendor_boot, args.vendor_bootconfig,
args.pagesize)
else:
write_padded_file(args.vendor_boot, args.vendor_ramdisk, args.pagesize)
write_padded_file(args.vendor_boot, args.dtb, args.pagesize)
def main():
args = parse_cmdline()
if args.vendor_boot is not None:
if args.header_version not in {3, 4}:
raise ValueError(
'--vendor_boot not compatible with given header version')
if args.header_version == 3 and args.vendor_ramdisk is None:
raise ValueError('--vendor_ramdisk missing or invalid')
write_vendor_boot_header(args)
write_vendor_boot_data(args)
if args.output is not None:
if args.second is not None and args.header_version > 2:
raise ValueError(
'--second not compatible with given header version')
img_id = write_header(args)
if args.header_version > 2:
write_data(args, BOOT_IMAGE_HEADER_V3_PAGESIZE)
else:
write_data(args, args.pagesize)
if args.id and img_id is not None:
print('0x' + ''.join(f'{octet:02x}' for octet in img_id))
if __name__ == '__main__':
main()

View File

@@ -1,16 +0,0 @@
#!/bin/bash
if [ -z "$BM_POE_INTERFACE" ]; then
echo "Must supply the PoE Interface to power up"
exit 1
fi
if [ -z "$BM_POE_ADDRESS" ]; then
echo "Must supply the PoE Switch host"
exit 1
fi
SNMP_KEY="SNMPv2-SMI::mib-2.105.1.1.1.3.1.$((${BM_POE_BASE:-0} + BM_POE_INTERFACE))"
SNMP_OFF="i 2"
flock /var/run/poe.lock -c "snmpset -v2c -r 3 -t 30 -cmesaci $BM_POE_ADDRESS $SNMP_KEY $SNMP_OFF"

View File

@@ -1,19 +0,0 @@
#!/bin/bash
if [ -z "$BM_POE_INTERFACE" ]; then
echo "Must supply the PoE Interface to power up"
exit 1
fi
if [ -z "$BM_POE_ADDRESS" ]; then
echo "Must supply the PoE Switch host"
exit 1
fi
SNMP_KEY="SNMPv2-SMI::mib-2.105.1.1.1.3.1.$((${BM_POE_BASE:-0} + BM_POE_INTERFACE))"
SNMP_ON="i 1"
SNMP_OFF="i 2"
flock /var/run/poe.lock -c "snmpset -v2c -r 3 -t 30 -cmesaci $BM_POE_ADDRESS $SNMP_KEY $SNMP_OFF"
sleep 3s
flock /var/run/poe.lock -c "snmpset -v2c -r 3 -t 30 -cmesaci $BM_POE_ADDRESS $SNMP_KEY $SNMP_ON"

View File

@@ -1,229 +0,0 @@
#!/bin/bash
# shellcheck disable=SC1091
# shellcheck disable=SC2034
# shellcheck disable=SC2059
# shellcheck disable=SC2086 # we want word splitting
. "$SCRIPTS_DIR"/setup-test-env.sh
# Boot script for devices attached to a PoE switch, using NFS for the root
# filesystem.
# We're run from the root of the repo, make a helper var for our paths
BM=$CI_PROJECT_DIR/install/bare-metal
CI_COMMON=$CI_PROJECT_DIR/install/common
CI_INSTALL=$CI_PROJECT_DIR/install
# Runner config checks
if [ -z "$BM_SERIAL" ]; then
echo "Must set BM_SERIAL in your gitlab-runner config.toml [[runners]] environment"
echo "This is the serial port to listen the device."
exit 1
fi
if [ -z "$BM_POE_ADDRESS" ]; then
echo "Must set BM_POE_ADDRESS in your gitlab-runner config.toml [[runners]] environment"
echo "This is the PoE switch address to connect for powering up/down devices."
exit 1
fi
if [ -z "$BM_POE_INTERFACE" ]; then
echo "Must set BM_POE_INTERFACE in your gitlab-runner config.toml [[runners]] environment"
echo "This is the PoE switch interface where the device is connected."
exit 1
fi
if [ -z "$BM_POWERUP" ]; then
echo "Must set BM_POWERUP in your gitlab-runner config.toml [[runners]] environment"
echo "This is a shell script that should power up the device and begin its boot sequence."
exit 1
fi
if [ -z "$BM_POWERDOWN" ]; then
echo "Must set BM_POWERDOWN in your gitlab-runner config.toml [[runners]] environment"
echo "This is a shell script that should power off the device."
exit 1
fi
if [ ! -d /nfs ]; then
echo "NFS rootfs directory needs to be mounted at /nfs by the gitlab runner"
exit 1
fi
if [ ! -d /tftp ]; then
echo "TFTP directory for this board needs to be mounted at /tftp by the gitlab runner"
exit 1
fi
# job config checks
if [ -z "$BM_ROOTFS" ]; then
echo "Must set BM_ROOTFS to your board's rootfs directory in the job's variables"
exit 1
fi
if [ -z "$BM_BOOTFS" ] && { [ -z "$BM_KERNEL" ] || [ -z "$BM_DTB" ]; } ; then
echo "Must set /boot files for the TFTP boot in the job's variables or set kernel and dtb"
exit 1
fi
if [ -z "$BM_CMDLINE" ]; then
echo "Must set BM_CMDLINE to your board's kernel command line arguments"
exit 1
fi
set -ex
date +'%F %T'
# Clear out any previous run's artifacts.
rm -rf results/
mkdir -p results
# Create the rootfs in the NFS directory. rm to make sure it's in a pristine
# state, since it's volume-mounted on the host.
rsync -a --delete $BM_ROOTFS/ /nfs/
date +'%F %T'
# If BM_BOOTFS is an URL, download it
if echo $BM_BOOTFS | grep -q http; then
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
"${FDO_HTTP_CACHE_URI:-}$BM_BOOTFS" -o /tmp/bootfs.tar
BM_BOOTFS=/tmp/bootfs.tar
fi
date +'%F %T'
# If BM_BOOTFS is a file, assume it is a tarball and uncompress it
if [ -f "${BM_BOOTFS}" ]; then
mkdir -p /tmp/bootfs
tar xf $BM_BOOTFS -C /tmp/bootfs
BM_BOOTFS=/tmp/bootfs
fi
# If BM_KERNEL and BM_DTS is present
if [ -n "${FORCE_KERNEL_TAG}" ]; then
if [ -z "${BM_KERNEL}" ] || [ -z "${BM_DTB}" ]; then
echo "This machine cannot be tested with external kernel since BM_KERNEL or BM_DTB missing!"
exit 1
fi
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
"${FDO_HTTP_CACHE_URI:-}${KERNEL_IMAGE_BASE}/${DEBIAN_ARCH}/${BM_KERNEL}" -o "${BM_KERNEL}"
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
"${FDO_HTTP_CACHE_URI:-}${KERNEL_IMAGE_BASE}/${DEBIAN_ARCH}/${BM_DTB}.dtb" -o "${BM_DTB}.dtb"
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
"${FDO_HTTP_CACHE_URI:-}${KERNEL_IMAGE_BASE}/${DEBIAN_ARCH}/modules.tar.zst" -o modules.tar.zst
fi
date +'%F %T'
# Install kernel modules (it could be either in /lib/modules or
# /usr/lib/modules, but we want to install in the latter)
if [ -n "${FORCE_KERNEL_TAG}" ]; then
tar --keep-directory-symlink --zstd -xf modules.tar.zst -C /nfs/
rm modules.tar.zst &
elif [ -n "${BM_BOOTFS}" ]; then
[ -d $BM_BOOTFS/usr/lib/modules ] && rsync -a $BM_BOOTFS/usr/lib/modules/ /nfs/usr/lib/modules/
[ -d $BM_BOOTFS/lib/modules ] && rsync -a $BM_BOOTFS/lib/modules/ /nfs/lib/modules/
else
echo "No modules!"
fi
date +'%F %T'
# Install kernel image + bootloader files
if [ -n "${FORCE_KERNEL_TAG}" ] || [ -z "$BM_BOOTFS" ]; then
mv "${BM_KERNEL}" "${BM_DTB}.dtb" /tftp/
else # BM_BOOTFS
rsync -aL --delete $BM_BOOTFS/boot/ /tftp/
fi
date +'%F %T'
# Set up the pxelinux config for Jetson Nano
mkdir -p /tftp/pxelinux.cfg
cat <<EOF >/tftp/pxelinux.cfg/default-arm-tegra210-p3450-0000
PROMPT 0
TIMEOUT 30
DEFAULT primary
MENU TITLE jetson nano boot options
LABEL primary
MENU LABEL CI kernel on TFTP
LINUX Image
FDT tegra210-p3450-0000.dtb
APPEND \${cbootargs} $BM_CMDLINE
EOF
# Set up the pxelinux config for Jetson TK1
cat <<EOF >/tftp/pxelinux.cfg/default-arm-tegra124-jetson-tk1
PROMPT 0
TIMEOUT 30
DEFAULT primary
MENU TITLE jetson TK1 boot options
LABEL primary
MENU LABEL CI kernel on TFTP
LINUX zImage
FDT tegra124-jetson-tk1.dtb
APPEND \${cbootargs} $BM_CMDLINE
EOF
# Create the rootfs in the NFS directory
. $BM/rootfs-setup.sh /nfs
date +'%F %T'
echo "$BM_CMDLINE" > /tftp/cmdline.txt
# Add some options in config.txt, if defined
if [ -n "$BM_BOOTCONFIG" ]; then
printf "$BM_BOOTCONFIG" >> /tftp/config.txt
fi
set +e
STRUCTURED_LOG_FILE=job_detail.json
python3 $CI_INSTALL/custom_logger.py ${STRUCTURED_LOG_FILE} --update dut_job_type "${DEVICE_TYPE}"
python3 $CI_INSTALL/custom_logger.py ${STRUCTURED_LOG_FILE} --update farm "${FARM}"
ATTEMPTS=3
first_attempt=True
while [ $((ATTEMPTS--)) -gt 0 ]; do
python3 $CI_INSTALL/custom_logger.py ${STRUCTURED_LOG_FILE} --create-dut-job dut_name "${CI_RUNNER_DESCRIPTION}"
# Update subtime time to CI_JOB_STARTED_AT only for the first run
if [ "$first_attempt" = "True" ]; then
python3 $CI_INSTALL/custom_logger.py ${STRUCTURED_LOG_FILE} --update-dut-time submit "${CI_JOB_STARTED_AT}"
else
python3 $CI_INSTALL/custom_logger.py ${STRUCTURED_LOG_FILE} --update-dut-time submit
fi
python3 $BM/poe_run.py \
--dev="$BM_SERIAL" \
--powerup="$BM_POWERUP" \
--powerdown="$BM_POWERDOWN" \
--test-timeout ${TEST_PHASE_TIMEOUT:-20}
ret=$?
if [ $ret -eq 2 ]; then
echo "Did not detect boot sequence, retrying..."
python3 $CI_INSTALL/custom_logger.py ${STRUCTURED_LOG_FILE} --close-dut-job
first_attempt=False
else
ATTEMPTS=0
fi
done
python3 $CI_INSTALL/custom_logger.py ${STRUCTURED_LOG_FILE} --close-dut-job
python3 $CI_INSTALL/custom_logger.py ${STRUCTURED_LOG_FILE} --close
set -e
date +'%F %T'
# Bring artifacts back from the NFS dir to the build dir where gitlab-runner
# will look for them.
cp -Rp /nfs/results/. results/
if [ -f "${STRUCTURED_LOG_FILE}" ]; then
cp -p ${STRUCTURED_LOG_FILE} results/
echo "Structured log file is available at ${ARTIFACTS_BASE_URL}/results/${STRUCTURED_LOG_FILE}"
fi
date +'%F %T'
exit $ret

View File

@@ -1,127 +0,0 @@
#!/usr/bin/env python3
#
# Copyright © 2020 Igalia, S.L.
#
# 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.
import argparse
import os
import re
import sys
import threading
from custom_logger import CustomLogger
from serial_buffer import SerialBuffer
class PoERun:
def __init__(self, args, test_timeout, logger):
self.powerup = args.powerup
self.powerdown = args.powerdown
self.ser = SerialBuffer(
args.dev, "results/serial-output.txt", "")
self.test_timeout = test_timeout
self.logger = logger
def print_error(self, message):
RED = '\033[0;31m'
NO_COLOR = '\033[0m'
print(RED + message + NO_COLOR)
self.logger.update_status_fail(message)
def logged_system(self, cmd):
print("Running '{}'".format(cmd))
return os.system(cmd)
def run(self):
if self.logged_system(self.powerup) != 0:
self.logger.update_status_fail("powerup failed")
return 1
boot_detected = False
self.logger.create_job_phase("boot")
for line in self.ser.lines(timeout=5 * 60, phase="bootloader"):
if re.search("Booting Linux", line):
boot_detected = True
break
if not boot_detected:
self.print_error(
"Something wrong; couldn't detect the boot start up sequence")
return 1
self.logger.create_job_phase("test")
for line in self.ser.lines(timeout=self.test_timeout, phase="test"):
if re.search("---. end Kernel panic", line):
self.logger.update_status_fail("kernel panic")
return 1
# Binning memory problems
if re.search("binner overflow mem", line):
self.print_error("Memory overflow in the binner; GPU hang")
return 1
if re.search("nouveau 57000000.gpu: bus: MMIO read of 00000000 FAULT at 137000", line):
self.print_error("nouveau jetson boot bug, abandoning run.")
return 1
# network fail on tk1
if re.search("NETDEV WATCHDOG:.* transmit queue 0 timed out", line):
self.print_error("nouveau jetson tk1 network fail, abandoning run.")
return 1
result = re.search("hwci: mesa: (\S*)", line)
if result:
if result.group(1) == "pass":
self.logger.update_dut_job("status", "pass")
return 0
else:
self.logger.update_status_fail("test fail")
return 1
self.print_error(
"Reached the end of the CPU serial log without finding a result")
return 1
def main():
parser = argparse.ArgumentParser()
parser.add_argument('--dev', type=str,
help='Serial device to monitor', required=True)
parser.add_argument('--powerup', type=str,
help='shell command for rebooting', required=True)
parser.add_argument('--powerdown', type=str,
help='shell command for powering off', required=True)
parser.add_argument(
'--test-timeout', type=int, help='Test phase timeout (minutes)', required=True)
args = parser.parse_args()
logger = CustomLogger("job_detail.json")
logger.update_dut_time("start", None)
poe = PoERun(args, args.test_timeout * 60, logger)
retval = poe.run()
poe.logged_system(args.powerdown)
logger.update_dut_time("end", None)
sys.exit(retval)
if __name__ == '__main__':
main()

View File

@@ -1,37 +0,0 @@
#!/usr/bin/env bash
# shellcheck disable=SC2086 # we want word splitting
rootfs_dst=$1
mkdir -p $rootfs_dst/results
# Set up the init script that brings up the system.
cp $BM/bm-init.sh $rootfs_dst/init
cp $CI_COMMON/init*.sh $rootfs_dst/
date +'%F %T'
# Make JWT token available as file in the bare-metal storage to enable access
# to MinIO
cp "${S3_JWT_FILE}" "${rootfs_dst}${S3_JWT_FILE}"
date +'%F %T'
cp $CI_COMMON/capture-devcoredump.sh $rootfs_dst/
cp $CI_COMMON/intel-gpu-freq.sh $rootfs_dst/
cp $CI_COMMON/kdl.sh $rootfs_dst/
cp "$SCRIPTS_DIR/setup-test-env.sh" "$rootfs_dst/"
set +x
# Pass through relevant env vars from the gitlab job to the baremetal init script
echo "Variables passed through:"
"$CI_COMMON"/generate-env.sh | tee $rootfs_dst/set-job-env-vars.sh
set -x
# Add the Mesa drivers we built, and make a consistent symlink to them.
mkdir -p $rootfs_dst/$CI_PROJECT_DIR
rsync -aH --delete $CI_PROJECT_DIR/install/ $rootfs_dst/$CI_PROJECT_DIR/install/
date +'%F %T'

View File

@@ -1,185 +0,0 @@
#!/usr/bin/env python3
#
# Copyright © 2020 Google LLC
#
# 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.
import argparse
from datetime import datetime, timezone
import queue
import serial
import threading
import time
class SerialBuffer:
def __init__(self, dev, filename, prefix, timeout=None, line_queue=None):
self.filename = filename
self.dev = dev
if dev:
self.f = open(filename, "wb+")
self.serial = serial.Serial(dev, 115200, timeout=timeout)
else:
self.f = open(filename, "rb")
self.serial = None
self.byte_queue = queue.Queue()
# allow multiple SerialBuffers to share a line queue so you can merge
# servo's CPU and EC streams into one thing to watch the boot/test
# progress on.
if line_queue:
self.line_queue = line_queue
else:
self.line_queue = queue.Queue()
self.prefix = prefix
self.timeout = timeout
self.sentinel = object()
self.closing = False
if self.dev:
self.read_thread = threading.Thread(
target=self.serial_read_thread_loop, daemon=True)
else:
self.read_thread = threading.Thread(
target=self.serial_file_read_thread_loop, daemon=True)
self.read_thread.start()
self.lines_thread = threading.Thread(
target=self.serial_lines_thread_loop, daemon=True)
self.lines_thread.start()
def close(self):
self.closing = True
if self.serial:
self.serial.cancel_read()
self.read_thread.join()
self.lines_thread.join()
if self.serial:
self.serial.close()
# Thread that just reads the bytes from the serial device to try to keep from
# buffer overflowing it. If nothing is received in 1 minute, it finalizes.
def serial_read_thread_loop(self):
greet = "Serial thread reading from %s\n" % self.dev
self.byte_queue.put(greet.encode())
while not self.closing:
try:
b = self.serial.read()
if len(b) == 0:
break
self.byte_queue.put(b)
except Exception as err:
print(self.prefix + str(err))
break
self.byte_queue.put(self.sentinel)
# Thread that just reads the bytes from the file of serial output that some
# other process is appending to.
def serial_file_read_thread_loop(self):
greet = "Serial thread reading from %s\n" % self.filename
self.byte_queue.put(greet.encode())
while not self.closing:
line = self.f.readline()
if line:
self.byte_queue.put(line)
else:
time.sleep(0.1)
self.byte_queue.put(self.sentinel)
# Thread that processes the stream of bytes to 1) log to stdout, 2) log to
# file, 3) add to the queue of lines to be read by program logic
def serial_lines_thread_loop(self):
line = bytearray()
while True:
bytes = self.byte_queue.get(block=True)
if bytes == self.sentinel:
self.read_thread.join()
self.line_queue.put(self.sentinel)
break
# Write our data to the output file if we're the ones reading from
# the serial device
if self.dev:
self.f.write(bytes)
self.f.flush()
for b in bytes:
line.append(b)
if b == b'\n'[0]:
line = line.decode(errors="replace")
time = datetime.now().strftime('%y-%m-%d %H:%M:%S')
print("{endc}{time} {prefix}{line}".format(
time=time, prefix=self.prefix, line=line, endc='\033[0m'), flush=True, end='')
self.line_queue.put(line)
line = bytearray()
def lines(self, timeout=None, phase=None):
start_time = time.monotonic()
while True:
read_timeout = None
if timeout:
read_timeout = timeout - (time.monotonic() - start_time)
if read_timeout <= 0:
print("read timeout waiting for serial during {}".format(phase))
self.close()
break
try:
line = self.line_queue.get(timeout=read_timeout)
except queue.Empty:
print("read timeout waiting for serial during {}".format(phase))
self.close()
break
if line == self.sentinel:
print("End of serial output")
self.lines_thread.join()
break
yield line
def main():
parser = argparse.ArgumentParser()
parser.add_argument('--dev', type=str, help='Serial device')
parser.add_argument('--file', type=str,
help='Filename for serial output', required=True)
parser.add_argument('--prefix', type=str,
help='Prefix for logging serial to stdout', nargs='?')
args = parser.parse_args()
ser = SerialBuffer(args.dev, args.file, args.prefix or "")
for line in ser.lines():
# We're just using this as a logger, so eat the produced lines and drop
# them
pass
if __name__ == '__main__':
main()

View File

@@ -1,41 +0,0 @@
#!/usr/bin/python3
# Copyright © 2020 Christian Gmeiner
#
# 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.
#
# Tiny script to read bytes from telnet, and write the output to stdout, with a
# buffer in between so we don't lose serial output from its buffer.
#
import sys
import telnetlib
host = sys.argv[1]
port = sys.argv[2]
tn = telnetlib.Telnet(host, port, 1000000)
while True:
bytes = tn.read_some()
sys.stdout.buffer.write(bytes)
sys.stdout.flush()
tn.close()

View File

@@ -1 +0,0 @@
../bin/ci

View File

@@ -1,7 +0,0 @@
#!/bin/sh
# shellcheck disable=SC1091
set -e
_COMPILER=clang++-15
. compiler-wrapper.sh

View File

@@ -1,7 +0,0 @@
#!/bin/sh
# shellcheck disable=SC1091
set -e
_COMPILER=clang++
. compiler-wrapper.sh

View File

@@ -1,7 +0,0 @@
#!/bin/sh
# shellcheck disable=SC1091
set -e
_COMPILER=clang-15
. compiler-wrapper.sh

View File

@@ -1,7 +0,0 @@
#!/bin/sh
# shellcheck disable=SC1091
set -e
_COMPILER=clang
. compiler-wrapper.sh

View File

@@ -1,7 +0,0 @@
#!/bin/sh
# shellcheck disable=SC1091
set -e
_COMPILER=g++
. compiler-wrapper.sh

View File

@@ -1,7 +0,0 @@
#!/bin/sh
# shellcheck disable=SC1091
set -e
_COMPILER=gcc
. compiler-wrapper.sh

View File

@@ -1,21 +0,0 @@
# shellcheck disable=SC1091
# shellcheck disable=SC2086 # we want word splitting
if command -V ccache >/dev/null 2>/dev/null; then
CCACHE=ccache
else
CCACHE=
fi
if echo "$@" | grep -E 'meson-private/tmp[^ /]*/testfile.c' >/dev/null; then
# Invoked for meson feature check
exec $CCACHE $_COMPILER "$@"
fi
if [ "$(eval printf "'%s'" "\"\${$(($#-1))}\"")" = "-c" ]; then
# Not invoked for linking
exec $CCACHE $_COMPILER "$@"
fi
# Compiler invoked by ninja for linking. Add -Werror to turn compiler warnings into errors
# with LTO. (meson's werror should arguably do this, but meanwhile we need to)
exec $CCACHE $_COMPILER "$@" -Werror

View File

@@ -1,714 +0,0 @@
# Shared between windows and Linux
.build-common:
extends: .container+build-rules
# Cancel job if a newer commit is pushed to the same branch
interruptible: true
# Build jobs don't take more than 1-3 minutes. 5-8 min max on a fresh runner
# without a populated ccache.
# These jobs are never slow, either they finish within reasonable time or
# something has gone wrong and the job will never terminate, so we should
# instead timeout so that the retry mechanism can kick in.
# A few exception are made, see `timeout:` overrides in the rest of this
# file.
timeout: 30m
artifacts:
name: "mesa_${CI_JOB_NAME}"
when: always
paths:
- _build/meson-logs/*.txt
- _build/meson-logs/strace
- artifacts
# Just Linux
.build-linux:
extends: .build-common
variables:
CCACHE_COMPILERCHECK: "content"
CCACHE_COMPRESS: "true"
CCACHE_DIR: /cache/mesa/ccache
# Use ccache transparently, and print stats before/after
before_script:
- !reference [default, before_script]
- |
export PATH="/usr/lib/ccache:$PATH"
export CCACHE_BASEDIR="$PWD"
if test -x /usr/bin/ccache; then
section_start ccache_before "ccache stats before build"
ccache --show-stats
section_end ccache_before
fi
after_script:
- if test -x /usr/bin/ccache; then ccache --show-stats | grep "Hits:"; fi
- !reference [default, after_script]
.build-windows:
extends:
- .build-common
- .windows-docker-tags
cache:
key: ${CI_JOB_NAME}
paths:
- subprojects/packagecache
.meson-build:
extends:
- .build-linux
- .use-debian/x86_64_build
stage: build-x86_64
variables:
LLVM_VERSION: 15
script:
- .gitlab-ci/meson/build.sh
debian-testing:
extends:
- .meson-build
- .ci-deqp-artifacts
variables:
UNWIND: "enabled"
DRI_LOADERS: >
-D glx=dri
-D gbm=enabled
-D egl=enabled
-D glvnd=disabled
-D platforms=x11,wayland
GALLIUM_ST: >
-D dri3=enabled
-D gallium-nine=true
-D gallium-va=enabled
-D gallium-rusticl=true
GALLIUM_DRIVERS: "swrast,virgl,radeonsi,zink,crocus,iris,i915,r300,svga"
VULKAN_DRIVERS: "swrast,amd,intel,intel_hasvk,virtio,nouveau"
BUILDTYPE: "debugoptimized"
EXTRA_OPTION: >
-D spirv-to-dxil=true
-D valgrind=disabled
-D perfetto=true
-D tools=drm-shim
S3_ARTIFACT_NAME: mesa-x86_64-default-${BUILDTYPE}
LLVM_VERSION: 15
script:
- .gitlab-ci/meson/build.sh
- .gitlab-ci/prepare-artifacts.sh
artifacts:
reports:
junit: artifacts/ci_scripts_report.xml
debian-testing-asan:
extends:
- debian-testing
variables:
C_ARGS: >
-Wno-error=stringop-truncation
EXTRA_OPTION: >
-D b_sanitize=address
-D valgrind=disabled
-D tools=dlclose-skip
-D intel-clc=system
S3_ARTIFACT_NAME: ""
ARTIFACTS_DEBUG_SYMBOLS: 1
# Do a host build for intel-clc (asan complains not being loaded
# as the first library)
HOST_BUILD_OPTIONS: >
-D build-tests=false
-D enable-glcpp-tests=false
-D gallium-opencl=disabled
-D gallium-drivers=
-D vulkan-drivers=
-D video-codecs=
-D glx=disabled
-D platforms=
-D intel-clc=enabled
-D install-intel-clc=true
debian-testing-msan:
# https://github.com/google/sanitizers/wiki/MemorySanitizerLibcxxHowTo
# msan cannot fully work until it's used together with msan libc
extends:
- debian-clang
variables:
# l_undef is incompatible with msan
EXTRA_OPTION:
-D b_sanitize=memory
-D b_lundef=false
-D intel-clc=system
S3_ARTIFACT_NAME: ""
ARTIFACTS_DEBUG_SYMBOLS: 1
# Don't run all the tests yet:
# GLSL has some issues in sexpression reading.
# gtest has issues in its test initialization.
MESON_TEST_ARGS: "--suite glcpp --suite format"
GALLIUM_DRIVERS: "freedreno,iris,nouveau,kmsro,r300,r600,swrast,svga,v3d,vc4,virgl,etnaviv,panfrost,lima,zink,radeonsi,tegra,d3d12,crocus"
VULKAN_DRIVERS: intel,amd,broadcom,virtio
# Do a host build for intel-clc (msan complains about
# uninitialized values in the LLVM libs)
HOST_BUILD_OPTIONS: >
-D build-tests=false
-D enable-glcpp-tests=false
-D gallium-opencl=disabled
-D gallium-drivers=
-D vulkan-drivers=
-D video-codecs=
-D glx=disabled
-D platforms=
-D intel-clc=enabled
-D install-intel-clc=true
debian-build-testing:
extends: .meson-build
variables:
BUILDTYPE: debug
UNWIND: "enabled"
DRI_LOADERS: >
-D glx=dri
-D gbm=enabled
-D egl=enabled
-D glvnd=disabled
-D platforms=x11,wayland
GALLIUM_ST: >
-D dri3=enabled
-D gallium-extra-hud=true
-D gallium-vdpau=enabled
-D gallium-omx=bellagio
-D gallium-va=enabled
-D gallium-xa=enabled
-D gallium-nine=true
-D gallium-rusticl=false
GALLIUM_DRIVERS: "iris,nouveau,kmsro,r300,r600,freedreno,swrast,svga,v3d,vc4,virgl,etnaviv,panfrost,lima,zink,d3d12,asahi,crocus"
VULKAN_DRIVERS: swrast
EXTRA_OPTION: >
-D spirv-to-dxil=true
-D osmesa=true
-D tools=drm-shim,etnaviv,freedreno,glsl,intel,intel-ui,nir,nouveau,lima,panfrost,asahi
-D b_lto=true
LLVM_VERSION: 15
S3_ARTIFACT_NAME: debian-build-testing
script: |
section_start lava-pytest "lava-pytest"
.gitlab-ci/lava/lava-pytest.sh
section_switch shellcheck "shellcheck"
.gitlab-ci/run-shellcheck.sh
section_switch yamllint "yamllint"
.gitlab-ci/run-yamllint.sh
section_switch meson "meson"
.gitlab-ci/meson/build.sh
.gitlab-ci/prepare-artifacts.sh
timeout: 15m
shader-db:
stage: code-validation
extends:
- .use-debian/x86_64_build
- .container+build-rules
needs:
- debian-build-testing
variables:
S3_ARTIFACT_NAME: debian-build-testing
before_script:
- !reference [.download_s3, before_script]
script: |
section_switch shader-db "shader-db"
.gitlab-ci/run-shader-db.sh
artifacts:
paths:
- shader-db
timeout: 15m
# Test a release build with -Werror so new warnings don't sneak in.
debian-release:
extends: .meson-build
variables:
LLVM_VERSION: 15
UNWIND: "enabled"
C_ARGS: >
-Wno-error=stringop-overread
DRI_LOADERS: >
-D glx=dri
-D gbm=enabled
-D egl=enabled
-D glvnd=disabled
-D platforms=x11,wayland
GALLIUM_ST: >
-D dri3=enabled
-D gallium-extra-hud=true
-D gallium-vdpau=enabled
-D gallium-omx=disabled
-D gallium-va=enabled
-D gallium-xa=enabled
-D gallium-nine=false
-D gallium-rusticl=false
-D llvm=enabled
GALLIUM_DRIVERS: "i915,iris,nouveau,kmsro,freedreno,r300,svga,swrast,v3d,vc4,virgl,etnaviv,panfrost,lima,zink,d3d12,crocus"
VULKAN_DRIVERS: "amd,imagination-experimental,microsoft-experimental"
EXTRA_OPTION: >
-D spirv-to-dxil=true
-D osmesa=true
-D tools=all
-D intel-clc=enabled
-D intel-rt=enabled
-D imagination-srv=true
BUILDTYPE: "release"
S3_ARTIFACT_NAME: "mesa-x86_64-default-${BUILDTYPE}"
script:
- .gitlab-ci/meson/build.sh
- 'if [ -n "$MESA_CI_PERFORMANCE_ENABLED" ]; then .gitlab-ci/prepare-artifacts.sh; fi'
alpine-build-testing:
extends:
- .meson-build
- .use-alpine/x86_64_build
stage: build-x86_64
variables:
BUILDTYPE: "release"
C_ARGS: >
-Wno-error=cpp
-Wno-error=array-bounds
-Wno-error=stringop-overread
DRI_LOADERS: >
-D glx=disabled
-D gbm=enabled
-D egl=enabled
-D glvnd=disabled
-D platforms=wayland
LLVM_VERSION: "16"
GALLIUM_DRIVERS: "crocus,etnaviv,freedreno,iris,kmsro,lima,nouveau,panfrost,r300,r600,radeonsi,svga,swrast,tegra,v3d,vc4,virgl,zink"
GALLIUM_ST: >
-D dri3=enabled
-D gallium-extra-hud=true
-D gallium-vdpau=disabled
-D gallium-omx=disabled
-D gallium-va=enabled
-D gallium-xa=disabled
-D gallium-nine=true
-D gallium-rusticl=false
-D gles1=disabled
-D gles2=enabled
-D llvm=enabled
-D microsoft-clc=disabled
-D shared-llvm=enabled
UNWIND: "disabled"
VULKAN_DRIVERS: "amd,broadcom,freedreno,intel,imagination-experimental"
fedora-release:
extends:
- .meson-build
- .use-fedora/x86_64_build
variables:
BUILDTYPE: "release"
C_LINK_ARGS: >
-Wno-error=stringop-overflow
-Wno-error=stringop-overread
CPP_ARGS: >
-Wno-error=dangling-reference
-Wno-error=overloaded-virtual
CPP_LINK_ARGS: >
-Wno-error=stringop-overflow
-Wno-error=stringop-overread
DRI_LOADERS: >
-D glx=dri
-D gbm=enabled
-D egl=enabled
-D glvnd=enabled
-D platforms=x11,wayland
EXTRA_OPTION: >
-D b_lto=true
-D osmesa=true
-D selinux=true
-D tools=drm-shim,etnaviv,freedreno,glsl,intel,nir,nouveau,lima,panfrost,imagination
-D vulkan-layers=device-select,overlay
-D intel-rt=enabled
-D imagination-srv=true
-D teflon=true
GALLIUM_DRIVERS: "crocus,etnaviv,freedreno,i915,iris,kmsro,lima,nouveau,panfrost,r300,r600,radeonsi,svga,swrast,tegra,v3d,vc4,virgl,zink"
GALLIUM_ST: >
-D dri3=enabled
-D gallium-extra-hud=true
-D gallium-vdpau=enabled
-D gallium-omx=disabled
-D gallium-va=enabled
-D gallium-xa=enabled
-D gallium-nine=false
-D gallium-rusticl=true
-D gles1=disabled
-D gles2=enabled
-D llvm=enabled
-D microsoft-clc=disabled
-D shared-llvm=enabled
LLVM_VERSION: ""
UNWIND: "disabled"
VULKAN_DRIVERS: "amd,broadcom,freedreno,imagination-experimental,intel,intel_hasvk"
debian-android:
extends:
- .meson-cross
- .use-debian/android_build
- .ci-deqp-artifacts
variables:
BUILDTYPE: debug
UNWIND: "disabled"
C_ARGS: >
-Wno-error=asm-operand-widths
-Wno-error=constant-conversion
-Wno-error=enum-conversion
-Wno-error=initializer-overrides
-Wno-error=sometimes-uninitialized
CPP_ARGS: >
-Wno-error=c99-designator
-Wno-error=unused-variable
-Wno-error=unused-but-set-variable
-Wno-error=self-assign
DRI_LOADERS: >
-D glx=disabled
-D gbm=disabled
-D egl=enabled
-D glvnd=disabled
-D platforms=android
EXTRA_OPTION: >
-D android-stub=true
-D llvm=disabled
-D platform-sdk-version=33
-D valgrind=disabled
-D android-libbacktrace=disabled
-D intel-clc=system
GALLIUM_ST: >
-D dri3=disabled
-D gallium-vdpau=disabled
-D gallium-omx=disabled
-D gallium-va=disabled
-D gallium-xa=disabled
-D gallium-nine=false
-D gallium-rusticl=false
LLVM_VERSION: "15"
PKG_CONFIG_LIBDIR: "/disable/non/android/system/pc/files"
HOST_BUILD_OPTIONS: >
-D build-tests=false
-D enable-glcpp-tests=false
-D gallium-opencl=disabled
-D gallium-drivers=
-D vulkan-drivers=
-D video-codecs=
-D glx=disabled
-D platforms=
-D intel-clc=enabled
-D install-intel-clc=true
ARTIFACTS_DEBUG_SYMBOLS: 1
S3_ARTIFACT_NAME: mesa-x86_64-android-${BUILDTYPE}
script:
- CROSS=aarch64-linux-android GALLIUM_DRIVERS=etnaviv,freedreno,lima,panfrost,vc4,v3d VULKAN_DRIVERS=freedreno,broadcom,virtio .gitlab-ci/meson/build.sh
# x86_64 build:
# Can't do Intel because gen_decoder.c currently requires libexpat, which
# is not a dependency that AOSP wants to accept. Can't do Radeon Gallium
# drivers because they requires LLVM, which we don't have an Android build
# of.
- CROSS=x86_64-linux-android GALLIUM_DRIVERS=iris,virgl VULKAN_DRIVERS=amd,intel .gitlab-ci/meson/build.sh
- .gitlab-ci/prepare-artifacts.sh
.meson-cross:
extends:
- .meson-build
stage: build-misc
variables:
UNWIND: "disabled"
DRI_LOADERS: >
-D glx=dri
-D gbm=enabled
-D egl=enabled
-D platforms=x11,wayland
-D osmesa=false
GALLIUM_ST: >
-D dri3=enabled
-D gallium-vdpau=disabled
-D gallium-omx=disabled
-D gallium-va=disabled
-D gallium-xa=disabled
-D gallium-nine=false
.meson-arm:
extends:
- .meson-cross
- .use-debian/arm64_build
needs:
- debian/arm64_build
variables:
VULKAN_DRIVERS: freedreno,broadcom
GALLIUM_DRIVERS: "etnaviv,freedreno,kmsro,lima,nouveau,panfrost,swrast,tegra,v3d,vc4,zink"
BUILDTYPE: "debugoptimized"
tags:
- aarch64
debian-arm32:
extends:
- .meson-arm
- .ci-deqp-artifacts
variables:
CROSS: armhf
DRI_LOADERS:
-D glvnd=disabled
EXTRA_OPTION: >
-D llvm=disabled
-D valgrind=disabled
S3_ARTIFACT_NAME: mesa-arm32-default-${BUILDTYPE}
# The strip command segfaults, failing to strip the binary and leaving
# tempfiles in our artifacts.
ARTIFACTS_DEBUG_SYMBOLS: 1
script:
- .gitlab-ci/meson/build.sh
- .gitlab-ci/prepare-artifacts.sh
debian-arm32-asan:
extends:
- debian-arm32
variables:
DRI_LOADERS:
-D glvnd=disabled
EXTRA_OPTION: >
-D llvm=disabled
-D b_sanitize=address
-D valgrind=disabled
-D tools=dlclose-skip
ARTIFACTS_DEBUG_SYMBOLS: 1
S3_ARTIFACT_NAME: mesa-arm32-asan-${BUILDTYPE}
MESON_TEST_ARGS: "--no-suite mesa:compiler --no-suite mesa:util"
debian-arm64:
extends:
- .meson-arm
- .ci-deqp-artifacts
variables:
C_ARGS: >
-Wno-error=array-bounds
-Wno-error=stringop-truncation
VULKAN_DRIVERS: "freedreno,broadcom,panfrost,imagination-experimental"
DRI_LOADERS:
-D glvnd=disabled
EXTRA_OPTION: >
-D llvm=disabled
-D valgrind=disabled
-D imagination-srv=true
-D perfetto=true
-D freedreno-kmds=msm,virtio
-D teflon=true
S3_ARTIFACT_NAME: mesa-arm64-default-${BUILDTYPE}
script:
- .gitlab-ci/meson/build.sh
- .gitlab-ci/prepare-artifacts.sh
debian-arm64-asan:
extends:
- debian-arm64
variables:
DRI_LOADERS:
-D glvnd=disabled
EXTRA_OPTION: >
-D llvm=disabled
-D b_sanitize=address
-D valgrind=disabled
-D tools=dlclose-skip
ARTIFACTS_DEBUG_SYMBOLS: 1
S3_ARTIFACT_NAME: mesa-arm64-asan-${BUILDTYPE}
MESON_TEST_ARGS: "--no-suite mesa:compiler"
debian-arm64-build-test:
extends:
- .meson-arm
- .ci-deqp-artifacts
variables:
VULKAN_DRIVERS: "amd"
DRI_LOADERS:
-D glvnd=disabled
EXTRA_OPTION: >
-Dtools=panfrost,imagination
debian-arm64-release:
extends:
- debian-arm64
variables:
BUILDTYPE: release
S3_ARTIFACT_NAME: mesa-arm64-default-${BUILDTYPE}
C_ARGS: >
-Wno-error=array-bounds
-Wno-error=stringop-truncation
-Wno-error=stringop-overread
script:
- .gitlab-ci/meson/build.sh
- 'if [ -n "$MESA_CI_PERFORMANCE_ENABLED" ]; then .gitlab-ci/prepare-artifacts.sh; fi'
debian-clang:
extends: .meson-build
variables:
BUILDTYPE: debug
LLVM_VERSION: 15
UNWIND: "enabled"
C_ARGS: >
-Wno-error=constant-conversion
-Wno-error=enum-conversion
-Wno-error=initializer-overrides
-Wno-error=sometimes-uninitialized
-Werror=misleading-indentation
CPP_ARGS: >
-Wno-error=c99-designator
-Wno-error=overloaded-virtual
-Wno-error=tautological-constant-out-of-range-compare
-Wno-error=unused-private-field
DRI_LOADERS: >
-D glx=dri
-D gbm=enabled
-D egl=enabled
-D glvnd=enabled
-D platforms=x11,wayland
GALLIUM_ST: >
-D dri3=enabled
-D gallium-extra-hud=true
-D gallium-vdpau=enabled
-D gallium-omx=bellagio
-D gallium-va=enabled
-D gallium-xa=enabled
-D gallium-nine=true
-D gles1=enabled
-D gles2=enabled
-D llvm=enabled
-D microsoft-clc=disabled
-D shared-llvm=enabled
-D opencl-spirv=true
-D shared-glapi=enabled
GALLIUM_DRIVERS: "iris,nouveau,kmsro,r300,r600,freedreno,swrast,svga,v3d,vc4,virgl,etnaviv,panfrost,lima,zink,radeonsi,tegra,d3d12,crocus,i915,asahi"
VULKAN_DRIVERS: intel,amd,freedreno,broadcom,virtio,swrast,panfrost,imagination-experimental,microsoft-experimental,nouveau
EXTRA_OPTION:
-D spirv-to-dxil=true
-D osmesa=true
-D imagination-srv=true
-D tools=drm-shim,etnaviv,freedreno,glsl,intel,intel-ui,nir,nouveau,lima,panfrost,asahi,imagination
-D vulkan-layers=device-select,overlay
-D build-aco-tests=true
-D intel-clc=enabled
-D intel-rt=enabled
-D imagination-srv=true
-D teflon=true
CC: clang-${LLVM_VERSION}
CXX: clang++-${LLVM_VERSION}
debian-clang-release:
extends: debian-clang
variables:
BUILDTYPE: "release"
DRI_LOADERS: >
-D glx=xlib
-D platforms=x11,wayland
GALLIUM_ST: >
-D dri3=enabled
-D gallium-extra-hud=true
-D gallium-vdpau=enabled
-D gallium-omx=bellagio
-D gallium-va=enabled
-D gallium-xa=enabled
-D gallium-nine=true
-D gles1=disabled
-D gles2=disabled
-D llvm=enabled
-D microsoft-clc=disabled
-D shared-llvm=enabled
-D opencl-spirv=true
-D shared-glapi=disabled
windows-msvc:
extends:
- .build-windows
- .use-windows_build_msvc
- .windows-build-rules
stage: build-misc
script:
- pwsh -ExecutionPolicy RemoteSigned .\.gitlab-ci\windows\mesa_build.ps1
artifacts:
paths:
- _build/meson-logs/*.txt
- _install/
debian-vulkan:
extends: .meson-build
variables:
BUILDTYPE: debug
LLVM_VERSION: 15
UNWIND: "disabled"
DRI_LOADERS: >
-D glx=disabled
-D gbm=disabled
-D egl=disabled
-D opengl=false
-D gles1=disabled
-D gles2=disabled
-D glvnd=disabled
-D platforms=x11,wayland
-D osmesa=false
GALLIUM_ST: >
-D dri3=enabled
-D gallium-vdpau=disabled
-D gallium-omx=disabled
-D gallium-va=disabled
-D gallium-xa=disabled
-D gallium-nine=false
-D gallium-rusticl=false
-D b_sanitize=undefined
-D c_args=-fno-sanitize-recover=all
-D cpp_args=-fno-sanitize-recover=all
UBSAN_OPTIONS: "print_stacktrace=1"
VULKAN_DRIVERS: amd,broadcom,freedreno,intel,intel_hasvk,panfrost,virtio,imagination-experimental,microsoft-experimental,nouveau
EXTRA_OPTION: >
-D vulkan-layers=device-select,overlay
-D build-aco-tests=true
-D intel-rt=disabled
-D imagination-srv=true
debian-x86_32:
extends:
- .meson-cross
- .use-debian/x86_32_build
variables:
BUILDTYPE: debug
CROSS: i386
VULKAN_DRIVERS: intel,amd,swrast,virtio
GALLIUM_DRIVERS: "iris,nouveau,r300,r600,radeonsi,swrast,virgl,zink,crocus,d3d12"
LLVM_VERSION: 15
DRI_LOADERS:
-D glvnd=disabled
EXTRA_OPTION: >
-D vulkan-layers=device-select,overlay
-D intel-clc=system
HOST_BUILD_OPTIONS: >
-D build-tests=false
-D enable-glcpp-tests=false
-D gallium-opencl=disabled
-D gallium-drivers=
-D vulkan-drivers=
-D video-codecs=
-D glx=disabled
-D platforms=
-D intel-clc=enabled
-D install-intel-clc=true
debian-s390x:
extends:
- debian-ppc64el
- .use-debian/s390x_build
- .s390x-rules
tags:
- kvm
variables:
CROSS: s390x
GALLIUM_DRIVERS: "swrast,zink"
LLVM_VERSION: 15
VULKAN_DRIVERS: "swrast"
DRI_LOADERS:
-D glvnd=disabled
debian-ppc64el:
extends:
- .meson-cross
- .use-debian/ppc64el_build
- .ppc64el-rules
variables:
BUILDTYPE: debug
CROSS: ppc64el
GALLIUM_DRIVERS: "nouveau,radeonsi,swrast,virgl,zink"
VULKAN_DRIVERS: "amd,swrast"
DRI_LOADERS:
-D glvnd=disabled

View File

@@ -1,35 +0,0 @@
#!/usr/bin/env bash
# shellcheck disable=SC2035
# shellcheck disable=SC2061
# shellcheck disable=SC2086 # we want word splitting
while true; do
devcds=$(find /sys/devices/virtual/devcoredump/ -name data 2>/dev/null)
for i in $devcds; do
echo "Found a devcoredump at $i."
if cp $i /results/first.devcore; then
echo 1 > $i
echo "Saved to the job artifacts at /first.devcore"
exit 0
fi
done
i915_error_states=$(find /sys/devices/ -path */drm/card*/error)
for i in $i915_error_states; do
tmpfile=$(mktemp)
cp "$i" "$tmpfile"
filesize=$(stat --printf="%s" "$tmpfile")
# Does the file contain "No error state collected" ?
if [ "$filesize" = 25 ]; then
rm "$tmpfile"
else
echo "Found an i915 error state at $i size=$filesize."
if cp "$tmpfile" /results/first.i915_error_state; then
rm "$tmpfile"
echo 1 > "$i"
echo "Saved to the job artifacts at /first.i915_error_state"
exit 0
fi
fi
done
sleep 10
done

View File

@@ -1,137 +0,0 @@
#!/bin/bash
VARS=(
ACO_DEBUG
ARTIFACTS_BASE_URL
ASAN_OPTIONS
BASE_SYSTEM_FORK_HOST_PREFIX
BASE_SYSTEM_MAINLINE_HOST_PREFIX
CI_COMMIT_BRANCH
CI_COMMIT_REF_NAME
CI_COMMIT_TITLE
CI_JOB_ID
S3_JWT_FILE
CI_JOB_STARTED_AT
CI_JOB_NAME
CI_JOB_URL
CI_MERGE_REQUEST_SOURCE_BRANCH_NAME
CI_MERGE_REQUEST_TITLE
CI_NODE_INDEX
CI_NODE_TOTAL
CI_PAGES_DOMAIN
CI_PIPELINE_ID
CI_PIPELINE_URL
CI_PROJECT_DIR
CI_PROJECT_NAME
CI_PROJECT_PATH
CI_PROJECT_ROOT_NAMESPACE
CI_RUNNER_DESCRIPTION
CI_SERVER_URL
CROSVM_GALLIUM_DRIVER
CROSVM_GPU_ARGS
CURRENT_SECTION
DEQP_BIN_DIR
DEQP_CONFIG
DEQP_EXPECTED_RENDERER
DEQP_FRACTION
DEQP_HEIGHT
DEQP_RESULTS_DIR
DEQP_RUNNER_OPTIONS
DEQP_SUITE
DEQP_TEMP_DIR
DEQP_VER
DEQP_WIDTH
DEVICE_NAME
DRIVER_NAME
EGL_PLATFORM
ETNA_MESA_DEBUG
FDO_CI_CONCURRENT
FDO_UPSTREAM_REPO
FD_MESA_DEBUG
FLAKES_CHANNEL
FREEDRENO_HANGCHECK_MS
GALLIUM_DRIVER
GALLIVM_PERF
GPU_VERSION
GTEST
GTEST_FAILS
GTEST_FRACTION
GTEST_RESULTS_DIR
GTEST_RUNNER_OPTIONS
GTEST_SKIPS
HWCI_FREQ_MAX
HWCI_KERNEL_MODULES
HWCI_KVM
HWCI_START_WESTON
HWCI_START_XORG
HWCI_TEST_SCRIPT
IR3_SHADER_DEBUG
JOB_ARTIFACTS_BASE
JOB_RESULTS_PATH
JOB_ROOTFS_OVERLAY_PATH
KERNEL_IMAGE_BASE
KERNEL_IMAGE_NAME
LD_LIBRARY_PATH
LIBGL_ALWAYS_SOFTWARE
LP_NUM_THREADS
MESA_BASE_TAG
MESA_BUILD_PATH
MESA_DEBUG
MESA_GLES_VERSION_OVERRIDE
MESA_GLSL_VERSION_OVERRIDE
MESA_GL_VERSION_OVERRIDE
MESA_IMAGE
MESA_IMAGE_PATH
MESA_IMAGE_TAG
MESA_LOADER_DRIVER_OVERRIDE
MESA_TEMPLATES_COMMIT
MESA_VK_ABORT_ON_DEVICE_LOSS
MESA_VK_IGNORE_CONFORMANCE_WARNING
S3_HOST
S3_RESULTS_UPLOAD
NIR_DEBUG
PAN_I_WANT_A_BROKEN_VULKAN_DRIVER
PAN_MESA_DEBUG
PANVK_DEBUG
PIGLIT_FRACTION
PIGLIT_NO_WINDOW
PIGLIT_OPTIONS
PIGLIT_PLATFORM
PIGLIT_PROFILES
PIGLIT_REPLAY_ANGLE_TAG
PIGLIT_REPLAY_ARTIFACTS_BASE_URL
PIGLIT_REPLAY_DEVICE_NAME
PIGLIT_REPLAY_EXTRA_ARGS
PIGLIT_REPLAY_LOOP_TIMES
PIGLIT_REPLAY_REFERENCE_IMAGES_BASE
PIGLIT_REPLAY_SUBCOMMAND
PIGLIT_RESULTS
PIGLIT_TESTS
PIGLIT_TRACES_FILE
PIPELINE_ARTIFACTS_BASE
RADEON_DEBUG
RADV_DEBUG
RADV_PERFTEST
SKQP_ASSETS_DIR
SKQP_BACKENDS
TU_DEBUG
USE_ANGLE
VIRGL_HOST_API
WAFFLE_PLATFORM
VK_CPU
VK_DRIVER
# required by virglrender CI
VK_DRIVER_FILES
VKD3D_PROTON_RESULTS
VKD3D_CONFIG
VKD3D_TEST_EXCLUDE
ZINK_DESCRIPTORS
ZINK_DEBUG
LVP_POISON_MEMORY
)
for var in "${VARS[@]}"; do
if [ -n "${!var+x}" ]; then
echo "export $var=${!var@Q}"
fi
done

View File

@@ -1,25 +0,0 @@
#!/bin/sh
# Very early init, used to make sure devices and network are set up and
# reachable.
set -ex
cd /
findmnt --mountpoint /proc || mount -t proc none /proc
findmnt --mountpoint /sys || mount -t sysfs none /sys
mount -t debugfs none /sys/kernel/debug
findmnt --mountpoint /dev || mount -t devtmpfs none /dev
mkdir -p /dev/pts
mount -t devpts devpts /dev/pts
mkdir /dev/shm
mount -t tmpfs -o noexec,nodev,nosuid tmpfs /dev/shm
mount -t tmpfs tmpfs /tmp
echo "nameserver 8.8.8.8" > /etc/resolv.conf
[ -z "$NFS_SERVER_IP" ] || echo "$NFS_SERVER_IP caching-proxy" >> /etc/hosts
# Set the time so we can validate certificates before we fetch anything;
# however as not all DUTs have network, make this non-fatal.
for _ in 1 2 3; do sntp -sS pool.ntp.org && break || sleep 2; done || true

View File

@@ -1,235 +0,0 @@
#!/bin/bash
# shellcheck disable=SC1090
# shellcheck disable=SC1091
# shellcheck disable=SC2086 # we want word splitting
# shellcheck disable=SC2155
# Second-stage init, used to set up devices and our job environment before
# running tests.
shopt -s extglob
# Make sure to kill itself and all the children process from this script on
# exiting, since any console output may interfere with LAVA signals handling,
# which based on the log console.
cleanup() {
if [ "$BACKGROUND_PIDS" = "" ]; then
return 0
fi
set +x
echo "Killing all child processes"
for pid in $BACKGROUND_PIDS
do
kill "$pid" 2>/dev/null || true
done
# Sleep just a little to give enough time for subprocesses to be gracefully
# killed. Then apply a SIGKILL if necessary.
sleep 5
for pid in $BACKGROUND_PIDS
do
kill -9 "$pid" 2>/dev/null || true
done
BACKGROUND_PIDS=
set -x
}
trap cleanup INT TERM EXIT
# Space separated values with the PIDS of the processes started in the
# background by this script
BACKGROUND_PIDS=
for path in '/dut-env-vars.sh' '/set-job-env-vars.sh' './set-job-env-vars.sh'; do
[ -f "$path" ] && source "$path"
done
. "$SCRIPTS_DIR"/setup-test-env.sh
set -ex
# Set up any devices required by the jobs
[ -z "$HWCI_KERNEL_MODULES" ] || {
echo -n $HWCI_KERNEL_MODULES | xargs -d, -n1 /usr/sbin/modprobe
}
# Set up ZRAM
HWCI_ZRAM_SIZE=2G
if /sbin/zramctl --find --size $HWCI_ZRAM_SIZE -a zstd; then
mkswap /dev/zram0
swapon /dev/zram0
echo "zram: $HWCI_ZRAM_SIZE activated"
else
echo "zram: skipping, not supported"
fi
#
# Load the KVM module specific to the detected CPU virtualization extensions:
# - vmx for Intel VT
# - svm for AMD-V
#
# Additionally, download the kernel image to boot the VM via HWCI_TEST_SCRIPT.
#
if [ "$HWCI_KVM" = "true" ]; then
unset KVM_KERNEL_MODULE
{
grep -qs '\bvmx\b' /proc/cpuinfo && KVM_KERNEL_MODULE=kvm_intel
} || {
grep -qs '\bsvm\b' /proc/cpuinfo && KVM_KERNEL_MODULE=kvm_amd
}
{
[ -z "${KVM_KERNEL_MODULE}" ] && \
echo "WARNING: Failed to detect CPU virtualization extensions"
} || \
modprobe ${KVM_KERNEL_MODULE}
mkdir -p /lava-files
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
-o "/lava-files/${KERNEL_IMAGE_NAME}" \
"${KERNEL_IMAGE_BASE}/amd64/${KERNEL_IMAGE_NAME}"
fi
# Fix prefix confusion: the build installs to $CI_PROJECT_DIR, but we expect
# it in /install
ln -sf $CI_PROJECT_DIR/install /install
export LD_LIBRARY_PATH=/install/lib
export LIBGL_DRIVERS_PATH=/install/lib/dri
# https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22495#note_1876691
# The navi21 boards seem to have trouble with ld.so.cache, so try explicitly
# telling it to look in /usr/local/lib.
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/lib
# Store Mesa's disk cache under /tmp, rather than sending it out over NFS.
export XDG_CACHE_HOME=/tmp
# Make sure Python can find all our imports
export PYTHONPATH=$(python3 -c "import sys;print(\":\".join(sys.path))")
# If we need to specify a driver, it means several drivers could pick up this gpu;
# ensure that the other driver can't accidentally be used
if [ -n "$MESA_LOADER_DRIVER_OVERRIDE" ]; then
rm /install/lib/dri/!($MESA_LOADER_DRIVER_OVERRIDE)_dri.so
fi
ls -1 /install/lib/dri/*_dri.so || true
if [ "$HWCI_FREQ_MAX" = "true" ]; then
# Ensure initialization of the DRM device (needed by MSM)
head -0 /dev/dri/renderD128
# Disable GPU frequency scaling
DEVFREQ_GOVERNOR=$(find /sys/devices -name governor | grep gpu || true)
test -z "$DEVFREQ_GOVERNOR" || echo performance > $DEVFREQ_GOVERNOR || true
# Disable CPU frequency scaling
echo performance | tee -a /sys/devices/system/cpu/cpufreq/policy*/scaling_governor || true
# Disable GPU runtime power management
GPU_AUTOSUSPEND=$(find /sys/devices -name autosuspend_delay_ms | grep gpu | head -1)
test -z "$GPU_AUTOSUSPEND" || echo -1 > $GPU_AUTOSUSPEND || true
# Lock Intel GPU frequency to 70% of the maximum allowed by hardware
# and enable throttling detection & reporting.
# Additionally, set the upper limit for CPU scaling frequency to 65% of the
# maximum permitted, as an additional measure to mitigate thermal throttling.
/intel-gpu-freq.sh -s 70% --cpu-set-max 65% -g all -d
fi
# Start a little daemon to capture sysfs records and produce a JSON file
if [ -x /kdl.sh ]; then
echo "launch kdl.sh!"
/kdl.sh &
BACKGROUND_PIDS="$! $BACKGROUND_PIDS"
else
echo "kdl.sh not found!"
fi
# Increase freedreno hangcheck timer because it's right at the edge of the
# spilling tests timing out (and some traces, too)
if [ -n "$FREEDRENO_HANGCHECK_MS" ]; then
echo $FREEDRENO_HANGCHECK_MS | tee -a /sys/kernel/debug/dri/128/hangcheck_period_ms
fi
# Start a little daemon to capture the first devcoredump we encounter. (They
# expire after 5 minutes, so we poll for them).
if [ -x /capture-devcoredump.sh ]; then
/capture-devcoredump.sh &
BACKGROUND_PIDS="$! $BACKGROUND_PIDS"
fi
# If we want Xorg to be running for the test, then we start it up before the
# HWCI_TEST_SCRIPT because we need to use xinit to start X (otherwise
# without using -displayfd you can race with Xorg's startup), but xinit will eat
# your client's return code
if [ -n "$HWCI_START_XORG" ]; then
echo "touch /xorg-started; sleep 100000" > /xorg-script
env \
VK_DRIVER_FILES="/install/share/vulkan/icd.d/${VK_DRIVER}_icd.$(uname -m).json" \
xinit /bin/sh /xorg-script -- /usr/bin/Xorg -noreset -s 0 -dpms -logfile /Xorg.0.log &
BACKGROUND_PIDS="$! $BACKGROUND_PIDS"
# Wait for xorg to be ready for connections.
for _ in 1 2 3 4 5; do
if [ -e /xorg-started ]; then
break
fi
sleep 5
done
export DISPLAY=:0
fi
if [ -n "$HWCI_START_WESTON" ]; then
WESTON_X11_SOCK="/tmp/.X11-unix/X0"
if [ -n "$HWCI_START_XORG" ]; then
echo "Please consider dropping HWCI_START_XORG and instead using Weston XWayland for testing."
WESTON_X11_SOCK="/tmp/.X11-unix/X1"
fi
export WAYLAND_DISPLAY=wayland-0
# Display server is Weston Xwayland when HWCI_START_XORG is not set or Xorg when it's
export DISPLAY=:0
mkdir -p /tmp/.X11-unix
env \
VK_DRIVER_FILES="/install/share/vulkan/icd.d/${VK_DRIVER}_icd.$(uname -m).json" \
weston -Bheadless-backend.so --use-gl -Swayland-0 --xwayland --idle-time=0 &
BACKGROUND_PIDS="$! $BACKGROUND_PIDS"
while [ ! -S "$WESTON_X11_SOCK" ]; do sleep 1; done
fi
set +e
bash -c ". $SCRIPTS_DIR/setup-test-env.sh && $HWCI_TEST_SCRIPT"
EXIT_CODE=$?
set -e
# Let's make sure the results are always stored in current working directory
mv -f ${CI_PROJECT_DIR}/results ./ 2>/dev/null || true
[ ${EXIT_CODE} -ne 0 ] || rm -rf results/trace/"$PIGLIT_REPLAY_DEVICE_NAME"
# Make sure that capture-devcoredump is done before we start trying to tar up
# artifacts -- if it's writing while tar is reading, tar will throw an error and
# kill the job.
cleanup
# upload artifacts
if [ -n "$S3_RESULTS_UPLOAD" ]; then
tar --zstd -cf results.tar.zst results/;
ci-fairy s3cp --token-file "${S3_JWT_FILE}" results.tar.zst https://"$S3_RESULTS_UPLOAD"/results.tar.zst;
fi
# We still need to echo the hwci: mesa message, as some scripts rely on it, such
# as the python ones inside the bare-metal folder
[ ${EXIT_CODE} -eq 0 ] && RESULT=pass || RESULT=fail
set +x
# Print the final result; both bare-metal and LAVA look for this string to get
# the result of our run, so try really hard to get it out rather than losing
# the run. The device gets shut down right at this point, and a630 seems to
# enjoy corrupting the last line of serial output before shutdown.
for _ in $(seq 0 3); do echo "hwci: mesa: $RESULT"; sleep 1; echo; done
exit $EXIT_CODE

View File

@@ -1,768 +0,0 @@
#!/usr/bin/env bash
# shellcheck disable=SC2013
# shellcheck disable=SC2015
# shellcheck disable=SC2034
# shellcheck disable=SC2046
# shellcheck disable=SC2059
# shellcheck disable=SC2086 # we want word splitting
# shellcheck disable=SC2154
# shellcheck disable=SC2155
# shellcheck disable=SC2162
# shellcheck disable=SC2229
#
# This is an utility script to manage Intel GPU frequencies.
# It can be used for debugging performance problems or trying to obtain a stable
# frequency while benchmarking.
#
# Note the Intel i915 GPU driver allows to change the minimum, maximum and boost
# frequencies in steps of 50 MHz via:
#
# /sys/class/drm/card<n>/<freq_info>
#
# Where <n> is the DRM card index and <freq_info> one of the following:
#
# - gt_max_freq_mhz (enforced maximum freq)
# - gt_min_freq_mhz (enforced minimum freq)
# - gt_boost_freq_mhz (enforced boost freq)
#
# The hardware capabilities can be accessed via:
#
# - gt_RP0_freq_mhz (supported maximum freq)
# - gt_RPn_freq_mhz (supported minimum freq)
# - gt_RP1_freq_mhz (most efficient freq)
#
# The current frequency can be read from:
# - gt_act_freq_mhz (the actual GPU freq)
# - gt_cur_freq_mhz (the last requested freq)
#
# Also note that in addition to GPU management, the script offers the
# possibility to adjust CPU operating frequencies. However, this is currently
# limited to just setting the maximum scaling frequency as percentage of the
# maximum frequency allowed by the hardware.
#
# Copyright (C) 2022 Collabora Ltd.
# Author: Cristian Ciocaltea <cristian.ciocaltea@collabora.com>
#
# SPDX-License-Identifier: MIT
#
#
# Constants
#
# GPU
DRM_FREQ_SYSFS_PATTERN="/sys/class/drm/card%d/gt_%s_freq_mhz"
ENF_FREQ_INFO="max min boost"
CAP_FREQ_INFO="RP0 RPn RP1"
ACT_FREQ_INFO="act cur"
THROTT_DETECT_SLEEP_SEC=2
THROTT_DETECT_PID_FILE_PATH=/tmp/thrott-detect.pid
# CPU
CPU_SYSFS_PREFIX=/sys/devices/system/cpu
CPU_PSTATE_SYSFS_PATTERN="${CPU_SYSFS_PREFIX}/intel_pstate/%s"
CPU_FREQ_SYSFS_PATTERN="${CPU_SYSFS_PREFIX}/cpu%s/cpufreq/%s_freq"
CAP_CPU_FREQ_INFO="cpuinfo_max cpuinfo_min"
ENF_CPU_FREQ_INFO="scaling_max scaling_min"
ACT_CPU_FREQ_INFO="scaling_cur"
#
# Global variables.
#
unset INTEL_DRM_CARD_INDEX
unset GET_ACT_FREQ GET_ENF_FREQ GET_CAP_FREQ
unset SET_MIN_FREQ SET_MAX_FREQ
unset MONITOR_FREQ
unset CPU_SET_MAX_FREQ
unset DETECT_THROTT
unset DRY_RUN
#
# Simple printf based stderr logger.
#
log() {
local msg_type=$1
shift
printf "%s: %s: " "${msg_type}" "${0##*/}" >&2
printf "$@" >&2
printf "\n" >&2
}
#
# Helper to print sysfs path for the given card index and freq info.
#
# arg1: Frequency info sysfs name, one of *_FREQ_INFO constants above
# arg2: Video card index, defaults to INTEL_DRM_CARD_INDEX
#
print_freq_sysfs_path() {
printf ${DRM_FREQ_SYSFS_PATTERN} "${2:-${INTEL_DRM_CARD_INDEX}}" "$1"
}
#
# Helper to set INTEL_DRM_CARD_INDEX for the first identified Intel video card.
#
identify_intel_gpu() {
local i=0 vendor path
while [ ${i} -lt 16 ]; do
[ -c "/dev/dri/card$i" ] || {
i=$((i + 1))
continue
}
path=$(print_freq_sysfs_path "" ${i})
path=${path%/*}/device/vendor
[ -r "${path}" ] && read vendor < "${path}" && \
[ "${vendor}" = "0x8086" ] && INTEL_DRM_CARD_INDEX=$i && return 0
i=$((i + 1))
done
return 1
}
#
# Read the specified freq info from sysfs.
#
# arg1: Flag (y/n) to also enable printing the freq info.
# arg2...: Frequency info sysfs name(s), see *_FREQ_INFO constants above
# return: Global variable(s) FREQ_${arg} containing the requested information
#
read_freq_info() {
local var val info path print=0 ret=0
[ "$1" = "y" ] && print=1
shift
while [ $# -gt 0 ]; do
info=$1
shift
var=FREQ_${info}
path=$(print_freq_sysfs_path "${info}")
[ -r ${path} ] && read ${var} < ${path} || {
log ERROR "Failed to read freq info from: %s" "${path}"
ret=1
continue
}
[ -n "${var}" ] || {
log ERROR "Got empty freq info from: %s" "${path}"
ret=1
continue
}
[ ${print} -eq 1 ] && {
eval val=\$${var}
printf "%6s: %4s MHz\n" "${info}" "${val}"
}
done
return ${ret}
}
#
# Display requested info.
#
print_freq_info() {
local req_freq
[ -n "${GET_CAP_FREQ}" ] && {
printf "* Hardware capabilities\n"
read_freq_info y ${CAP_FREQ_INFO}
printf "\n"
}
[ -n "${GET_ENF_FREQ}" ] && {
printf "* Enforcements\n"
read_freq_info y ${ENF_FREQ_INFO}
printf "\n"
}
[ -n "${GET_ACT_FREQ}" ] && {
printf "* Actual\n"
read_freq_info y ${ACT_FREQ_INFO}
printf "\n"
}
}
#
# Helper to print frequency value as requested by user via '-s, --set' option.
# arg1: user requested freq value
#
compute_freq_set() {
local val
case "$1" in
+)
val=${FREQ_RP0}
;;
-)
val=${FREQ_RPn}
;;
*%)
val=$((${1%?} * FREQ_RP0 / 100))
# Adjust freq to comply with 50 MHz increments
val=$((val / 50 * 50))
;;
*[!0-9]*)
log ERROR "Cannot set freq to invalid value: %s" "$1"
return 1
;;
"")
log ERROR "Cannot set freq to unspecified value"
return 1
;;
*)
# Adjust freq to comply with 50 MHz increments
val=$(($1 / 50 * 50))
;;
esac
printf "%s" "${val}"
}
#
# Helper for set_freq().
#
set_freq_max() {
log INFO "Setting GPU max freq to %s MHz" "${SET_MAX_FREQ}"
read_freq_info n min || return $?
[ ${SET_MAX_FREQ} -gt ${FREQ_RP0} ] && {
log ERROR "Cannot set GPU max freq (%s) to be greater than hw max freq (%s)" \
"${SET_MAX_FREQ}" "${FREQ_RP0}"
return 1
}
[ ${SET_MAX_FREQ} -lt ${FREQ_RPn} ] && {
log ERROR "Cannot set GPU max freq (%s) to be less than hw min freq (%s)" \
"${SET_MIN_FREQ}" "${FREQ_RPn}"
return 1
}
[ ${SET_MAX_FREQ} -lt ${FREQ_min} ] && {
log ERROR "Cannot set GPU max freq (%s) to be less than min freq (%s)" \
"${SET_MAX_FREQ}" "${FREQ_min}"
return 1
}
[ -z "${DRY_RUN}" ] || return 0
if ! printf "%s" ${SET_MAX_FREQ} | tee $(print_freq_sysfs_path max) \
$(print_freq_sysfs_path boost) > /dev/null;
then
log ERROR "Failed to set GPU max frequency"
return 1
fi
}
#
# Helper for set_freq().
#
set_freq_min() {
log INFO "Setting GPU min freq to %s MHz" "${SET_MIN_FREQ}"
read_freq_info n max || return $?
[ ${SET_MIN_FREQ} -gt ${FREQ_max} ] && {
log ERROR "Cannot set GPU min freq (%s) to be greater than max freq (%s)" \
"${SET_MIN_FREQ}" "${FREQ_max}"
return 1
}
[ ${SET_MIN_FREQ} -lt ${FREQ_RPn} ] && {
log ERROR "Cannot set GPU min freq (%s) to be less than hw min freq (%s)" \
"${SET_MIN_FREQ}" "${FREQ_RPn}"
return 1
}
[ -z "${DRY_RUN}" ] || return 0
if ! printf "%s" ${SET_MIN_FREQ} > $(print_freq_sysfs_path min);
then
log ERROR "Failed to set GPU min frequency"
return 1
fi
}
#
# Set min or max or both GPU frequencies to the user indicated values.
#
set_freq() {
# Get hw max & min frequencies
read_freq_info n RP0 RPn || return $?
[ -z "${SET_MAX_FREQ}" ] || {
SET_MAX_FREQ=$(compute_freq_set "${SET_MAX_FREQ}")
[ -z "${SET_MAX_FREQ}" ] && return 1
}
[ -z "${SET_MIN_FREQ}" ] || {
SET_MIN_FREQ=$(compute_freq_set "${SET_MIN_FREQ}")
[ -z "${SET_MIN_FREQ}" ] && return 1
}
#
# Ensure correct operation order, to avoid setting min freq
# to a value which is larger than max freq.
#
# E.g.:
# crt_min=crt_max=600; new_min=new_max=700
# > operation order: max=700; min=700
#
# crt_min=crt_max=600; new_min=new_max=500
# > operation order: min=500; max=500
#
if [ -n "${SET_MAX_FREQ}" ] && [ -n "${SET_MIN_FREQ}" ]; then
[ ${SET_MAX_FREQ} -lt ${SET_MIN_FREQ} ] && {
log ERROR "Cannot set GPU max freq to be less than min freq"
return 1
}
read_freq_info n min || return $?
if [ ${SET_MAX_FREQ} -lt ${FREQ_min} ]; then
set_freq_min || return $?
set_freq_max
else
set_freq_max || return $?
set_freq_min
fi
elif [ -n "${SET_MAX_FREQ}" ]; then
set_freq_max
elif [ -n "${SET_MIN_FREQ}" ]; then
set_freq_min
else
log "Unexpected call to set_freq()"
return 1
fi
}
#
# Helper for detect_throttling().
#
get_thrott_detect_pid() {
[ -e ${THROTT_DETECT_PID_FILE_PATH} ] || return 0
local pid
read pid < ${THROTT_DETECT_PID_FILE_PATH} || {
log ERROR "Failed to read pid from: %s" "${THROTT_DETECT_PID_FILE_PATH}"
return 1
}
local proc_path=/proc/${pid:-invalid}/cmdline
[ -r ${proc_path} ] && grep -qs "${0##*/}" ${proc_path} && {
printf "%s" "${pid}"
return 0
}
# Remove orphaned PID file
rm -rf ${THROTT_DETECT_PID_FILE_PATH}
return 1
}
#
# Control detection and reporting of GPU throttling events.
# arg1: start - run throttle detector in background
# stop - stop throttle detector process, if any
# status - verify if throttle detector is running
#
detect_throttling() {
local pid
pid=$(get_thrott_detect_pid)
case "$1" in
status)
printf "Throttling detector is "
[ -z "${pid}" ] && printf "not running\n" && return 0
printf "running (pid=%s)\n" ${pid}
;;
stop)
[ -z "${pid}" ] && return 0
log INFO "Stopping throttling detector (pid=%s)" "${pid}"
kill ${pid}; sleep 1; kill -0 ${pid} 2>/dev/null && kill -9 ${pid}
rm -rf ${THROTT_DETECT_PID_FILE_PATH}
;;
start)
[ -n "${pid}" ] && {
log WARN "Throttling detector is already running (pid=%s)" ${pid}
return 0
}
(
read_freq_info n RPn || exit $?
while true; do
sleep ${THROTT_DETECT_SLEEP_SEC}
read_freq_info n act min cur || exit $?
#
# The throttling seems to occur when act freq goes below min.
# However, it's necessary to exclude the idle states, where
# act freq normally reaches RPn and cur goes below min.
#
[ ${FREQ_act} -lt ${FREQ_min} ] && \
[ ${FREQ_act} -gt ${FREQ_RPn} ] && \
[ ${FREQ_cur} -ge ${FREQ_min} ] && \
printf "GPU throttling detected: act=%s min=%s cur=%s RPn=%s\n" \
${FREQ_act} ${FREQ_min} ${FREQ_cur} ${FREQ_RPn}
done
) &
pid=$!
log INFO "Started GPU throttling detector (pid=%s)" ${pid}
printf "%s\n" ${pid} > ${THROTT_DETECT_PID_FILE_PATH} || \
log WARN "Failed to write throttle detector PID file"
;;
esac
}
#
# Retrieve the list of online CPUs.
#
get_online_cpus() {
local path cpu_index
printf "0"
for path in $(grep 1 ${CPU_SYSFS_PREFIX}/cpu*/online); do
cpu_index=${path##*/cpu}
printf " %s" ${cpu_index%%/*}
done
}
#
# Helper to print sysfs path for the given CPU index and freq info.
#
# arg1: Frequency info sysfs name, one of *_CPU_FREQ_INFO constants above
# arg2: CPU index
#
print_cpu_freq_sysfs_path() {
printf ${CPU_FREQ_SYSFS_PATTERN} "$2" "$1"
}
#
# Read the specified CPU freq info from sysfs.
#
# arg1: CPU index
# arg2: Flag (y/n) to also enable printing the freq info.
# arg3...: Frequency info sysfs name(s), see *_CPU_FREQ_INFO constants above
# return: Global variable(s) CPU_FREQ_${arg} containing the requested information
#
read_cpu_freq_info() {
local var val info path cpu_index print=0 ret=0
cpu_index=$1
[ "$2" = "y" ] && print=1
shift 2
while [ $# -gt 0 ]; do
info=$1
shift
var=CPU_FREQ_${info}
path=$(print_cpu_freq_sysfs_path "${info}" ${cpu_index})
[ -r ${path} ] && read ${var} < ${path} || {
log ERROR "Failed to read CPU freq info from: %s" "${path}"
ret=1
continue
}
[ -n "${var}" ] || {
log ERROR "Got empty CPU freq info from: %s" "${path}"
ret=1
continue
}
[ ${print} -eq 1 ] && {
eval val=\$${var}
printf "%6s: %4s Hz\n" "${info}" "${val}"
}
done
return ${ret}
}
#
# Helper to print freq. value as requested by user via '--cpu-set-max' option.
# arg1: user requested freq value
#
compute_cpu_freq_set() {
local val
case "$1" in
+)
val=${CPU_FREQ_cpuinfo_max}
;;
-)
val=${CPU_FREQ_cpuinfo_min}
;;
*%)
val=$((${1%?} * CPU_FREQ_cpuinfo_max / 100))
;;
*[!0-9]*)
log ERROR "Cannot set CPU freq to invalid value: %s" "$1"
return 1
;;
"")
log ERROR "Cannot set CPU freq to unspecified value"
return 1
;;
*)
log ERROR "Cannot set CPU freq to custom value; use +, -, or % instead"
return 1
;;
esac
printf "%s" "${val}"
}
#
# Adjust CPU max scaling frequency.
#
set_cpu_freq_max() {
local target_freq res=0
case "${CPU_SET_MAX_FREQ}" in
+)
target_freq=100
;;
-)
target_freq=1
;;
*%)
target_freq=${CPU_SET_MAX_FREQ%?}
;;
*)
log ERROR "Invalid CPU freq"
return 1
;;
esac
local pstate_info=$(printf "${CPU_PSTATE_SYSFS_PATTERN}" max_perf_pct)
[ -e "${pstate_info}" ] && {
log INFO "Setting intel_pstate max perf to %s" "${target_freq}%"
if ! printf "%s" "${target_freq}" > "${pstate_info}";
then
log ERROR "Failed to set intel_pstate max perf"
res=1
fi
}
local cpu_index
for cpu_index in $(get_online_cpus); do
read_cpu_freq_info ${cpu_index} n ${CAP_CPU_FREQ_INFO} || { res=$?; continue; }
target_freq=$(compute_cpu_freq_set "${CPU_SET_MAX_FREQ}")
[ -z "${target_freq}" ] && { res=$?; continue; }
log INFO "Setting CPU%s max scaling freq to %s Hz" ${cpu_index} "${target_freq}"
[ -n "${DRY_RUN}" ] && continue
if ! printf "%s" ${target_freq} > $(print_cpu_freq_sysfs_path scaling_max ${cpu_index});
then
res=1
log ERROR "Failed to set CPU%s max scaling frequency" ${cpu_index}
fi
done
return ${res}
}
#
# Show help message.
#
print_usage() {
cat <<EOF
Usage: ${0##*/} [OPTION]...
A script to manage Intel GPU frequencies. Can be used for debugging performance
problems or trying to obtain a stable frequency while benchmarking.
Note Intel GPUs only accept specific frequencies, usually multiples of 50 MHz.
Options:
-g, --get [act|enf|cap|all]
Get frequency information: active (default), enforced,
hardware capabilities or all of them.
-s, --set [{min|max}=]{FREQUENCY[%]|+|-}
Set min or max frequency to the given value (MHz).
Append '%' to interpret FREQUENCY as % of hw max.
Use '+' or '-' to set frequency to hardware max or min.
Omit min/max prefix to set both frequencies.
-r, --reset Reset frequencies to hardware defaults.
-m, --monitor [act|enf|cap|all]
Monitor the indicated frequencies via 'watch' utility.
See '-g, --get' option for more details.
-d|--detect-thrott [start|stop|status]
Start (default operation) the throttling detector
as a background process. Use 'stop' or 'status' to
terminate the detector process or verify its status.
--cpu-set-max [FREQUENCY%|+|-}
Set CPU max scaling frequency as % of hw max.
Use '+' or '-' to set frequency to hardware max or min.
-r, --reset Reset frequencies to hardware defaults.
--dry-run See what the script will do without applying any
frequency changes.
-h, --help Display this help text and exit.
EOF
}
#
# Parse user input for '-g, --get' option.
# Returns 0 if a value has been provided, otherwise 1.
#
parse_option_get() {
local ret=0
case "$1" in
act) GET_ACT_FREQ=1;;
enf) GET_ENF_FREQ=1;;
cap) GET_CAP_FREQ=1;;
all) GET_ACT_FREQ=1; GET_ENF_FREQ=1; GET_CAP_FREQ=1;;
-*|"")
# No value provided, using default.
GET_ACT_FREQ=1
ret=1
;;
*)
print_usage
exit 1
;;
esac
return ${ret}
}
#
# Validate user input for '-s, --set' option.
# arg1: input value to be validated
# arg2: optional flag indicating input is restricted to %
#
validate_option_set() {
case "$1" in
+|-|[0-9]%|[0-9][0-9]%)
return 0
;;
*[!0-9]*|"")
print_usage
exit 1
;;
esac
[ -z "$2" ] || { print_usage; exit 1; }
}
#
# Parse script arguments.
#
[ $# -eq 0 ] && { print_usage; exit 1; }
while [ $# -gt 0 ]; do
case "$1" in
-g|--get)
parse_option_get "$2" && shift
;;
-s|--set)
shift
case "$1" in
min=*)
SET_MIN_FREQ=${1#min=}
validate_option_set "${SET_MIN_FREQ}"
;;
max=*)
SET_MAX_FREQ=${1#max=}
validate_option_set "${SET_MAX_FREQ}"
;;
*)
SET_MIN_FREQ=$1
validate_option_set "${SET_MIN_FREQ}"
SET_MAX_FREQ=${SET_MIN_FREQ}
;;
esac
;;
-r|--reset)
RESET_FREQ=1
SET_MIN_FREQ="-"
SET_MAX_FREQ="+"
;;
-m|--monitor)
MONITOR_FREQ=act
parse_option_get "$2" && MONITOR_FREQ=$2 && shift
;;
-d|--detect-thrott)
DETECT_THROTT=start
case "$2" in
start|stop|status)
DETECT_THROTT=$2
shift
;;
esac
;;
--cpu-set-max)
shift
CPU_SET_MAX_FREQ=$1
validate_option_set "${CPU_SET_MAX_FREQ}" restricted
;;
--dry-run)
DRY_RUN=1
;;
-h|--help)
print_usage
exit 0
;;
*)
print_usage
exit 1
;;
esac
shift
done
#
# Main
#
RET=0
identify_intel_gpu || {
log INFO "No Intel GPU detected"
exit 0
}
[ -n "${SET_MIN_FREQ}${SET_MAX_FREQ}" ] && { set_freq || RET=$?; }
print_freq_info
[ -n "${DETECT_THROTT}" ] && detect_throttling ${DETECT_THROTT}
[ -n "${CPU_SET_MAX_FREQ}" ] && { set_cpu_freq_max || RET=$?; }
[ -n "${MONITOR_FREQ}" ] && {
log INFO "Entering frequency monitoring mode"
sleep 2
exec watch -d -n 1 "$0" -g "${MONITOR_FREQ}"
}
exit ${RET}

View File

@@ -1,24 +0,0 @@
#!/usr/bin/env bash
# shellcheck disable=SC1091 # the path is created in build-kdl and
# here is check if exist
terminate() {
echo "ci-kdl.sh caught SIGTERM signal! propagating to child processes"
for job in $(jobs -p)
do
kill -15 "$job"
done
}
trap terminate SIGTERM
if [ -f /ci-kdl.venv/bin/activate ]; then
source /ci-kdl.venv/bin/activate
/ci-kdl.venv/bin/python /ci-kdl.venv/bin/ci-kdl | tee -a /results/kdl.log &
child=$!
wait $child
mv kdl_*.json /results/kdl.json
else
echo -e "Not possible to activate ci-kdl virtual environment"
fi

View File

@@ -1,21 +0,0 @@
#!/bin/sh
set -ex
_XORG_SCRIPT="/xorg-script"
_FLAG_FILE="/xorg-started"
echo "touch ${_FLAG_FILE}; sleep 100000" > "${_XORG_SCRIPT}"
if [ "x$1" != "x" ]; then
export LD_LIBRARY_PATH="${1}/lib"
export LIBGL_DRIVERS_PATH="${1}/lib/dri"
fi
xinit /bin/sh "${_XORG_SCRIPT}" -- /usr/bin/Xorg vt45 -noreset -s 0 -dpms -logfile /Xorg.0.log &
# Wait for xorg to be ready for connections.
for _ in 1 2 3 4 5; do
if [ -e "${_FLAG_FILE}" ]; then
break
fi
sleep 5
done

View File

@@ -1,67 +0,0 @@
#!/usr/bin/env bash
# shellcheck disable=SC1091
# When changing this file, you need to bump the following
# .gitlab-ci/image-tags.yml tags:
# ALPINE_X86_64_BUILD_TAG
set -e
set -o xtrace
export LLVM_VERSION="${LLVM_VERSION:=16}"
EPHEMERAL=(
)
DEPS=(
bash
bison
ccache
clang16-dev
cmake
clang-dev
coreutils
curl
flex
gcc
g++
git
gettext
glslang
linux-headers
llvm16-static
llvm16-dev
meson
expat-dev
elfutils-dev
libdrm-dev
libselinux-dev
libva-dev
libpciaccess-dev
zlib-dev
python3-dev
py3-cparser
py3-mako
py3-ply
vulkan-headers
spirv-tools-dev
util-macros
wayland-dev
wayland-protocols
)
apk --no-cache add "${DEPS[@]}" "${EPHEMERAL[@]}"
. .gitlab-ci/container/build-llvm-spirv.sh
. .gitlab-ci/container/build-libclc.sh
. .gitlab-ci/container/container_pre_build.sh
############### Uninstall the build software
apk del "${EPHEMERAL[@]}"
. .gitlab-ci/container/container_post_build.sh

View File

@@ -1,29 +0,0 @@
#!/usr/bin/env bash
# This is a ci-templates build script to generate a container for LAVA SSH client.
# shellcheck disable=SC1091
set -e
set -o xtrace
EPHEMERAL=(
)
# We only need these very basic packages to run the tests.
DEPS=(
openssh-client # for ssh
iputils # for ping
bash
curl
)
apk --no-cache add "${DEPS[@]}" "${EPHEMERAL[@]}"
. .gitlab-ci/container/container_pre_build.sh
############### Uninstall the build software
apk del "${EPHEMERAL[@]}"
. .gitlab-ci/container/container_post_build.sh

View File

@@ -1,62 +0,0 @@
#!/usr/bin/env bash
set -e
set -o xtrace
# Fetch the arm-built rootfs image and unpack it in our x86_64 container (saves
# network transfer, disk usage, and runtime on test jobs)
# shellcheck disable=SC2154 # arch is assigned in previous scripts
if curl -X HEAD -s "${ARTIFACTS_PREFIX}/${FDO_UPSTREAM_REPO}/${ARTIFACTS_SUFFIX}/${arch}/done"; then
ARTIFACTS_URL="${ARTIFACTS_PREFIX}/${FDO_UPSTREAM_REPO}/${ARTIFACTS_SUFFIX}/${arch}"
else
ARTIFACTS_URL="${ARTIFACTS_PREFIX}/${CI_PROJECT_PATH}/${ARTIFACTS_SUFFIX}/${arch}"
fi
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
"${ARTIFACTS_URL}"/lava-rootfs.tar.zst -o rootfs.tar.zst
mkdir -p /rootfs-"$arch"
tar -C /rootfs-"$arch" '--exclude=./dev/*' --zstd -xf rootfs.tar.zst
rm rootfs.tar.zst
if [[ $arch == "arm64" ]]; then
mkdir -p /baremetal-files
pushd /baremetal-files
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
-O "${KERNEL_IMAGE_BASE}"/arm64/Image
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
-O "${KERNEL_IMAGE_BASE}"/arm64/Image.gz
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
-O "${KERNEL_IMAGE_BASE}"/arm64/cheza-kernel
DEVICE_TREES=""
DEVICE_TREES="$DEVICE_TREES apq8016-sbc.dtb"
DEVICE_TREES="$DEVICE_TREES apq8096-db820c.dtb"
DEVICE_TREES="$DEVICE_TREES tegra210-p3450-0000.dtb"
DEVICE_TREES="$DEVICE_TREES imx8mq-nitrogen.dtb"
for DTB in $DEVICE_TREES; do
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
-O "${KERNEL_IMAGE_BASE}/arm64/$DTB"
done
popd
elif [[ $arch == "armhf" ]]; then
mkdir -p /baremetal-files
pushd /baremetal-files
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
-O "${KERNEL_IMAGE_BASE}"/armhf/zImage
DEVICE_TREES=""
DEVICE_TREES="$DEVICE_TREES imx6q-cubox-i.dtb"
DEVICE_TREES="$DEVICE_TREES tegra124-jetson-tk1.dtb"
for DTB in $DEVICE_TREES; do
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
-O "${KERNEL_IMAGE_BASE}/armhf/$DTB"
done
popd
fi

View File

@@ -1,58 +0,0 @@
#!/usr/bin/env bash
set -ex
ANGLE_REV="0518a3ff4d4e7e5b2ce8203358f719613a31c118"
# DEPOT tools
git clone --depth 1 https://chromium.googlesource.com/chromium/tools/depot_tools.git
PWD=$(pwd)
export PATH=$PWD/depot_tools:$PATH
export DEPOT_TOOLS_UPDATE=0
mkdir /angle-build
pushd /angle-build
git init
git remote add origin https://chromium.googlesource.com/angle/angle.git
git fetch --depth 1 origin "$ANGLE_REV"
git checkout FETCH_HEAD
# source preparation
python3 scripts/bootstrap.py
mkdir -p build/config
gclient sync
sed -i "/catapult/d" testing/BUILD.gn
mkdir -p out/Release
echo '
is_debug = false
angle_enable_swiftshader = false
angle_enable_null = false
angle_enable_gl = false
angle_enable_vulkan = true
angle_has_histograms = false
build_angle_trace_perf_tests = false
build_angle_deqp_tests = false
angle_use_custom_libvulkan = false
dcheck_always_on=true
' > out/Release/args.gn
if [[ "$DEBIAN_ARCH" = "arm64" ]]; then
build/linux/sysroot_scripts/install-sysroot.py --arch=arm64
fi
gn gen out/Release
# depot_tools overrides ninja with a version that doesn't work. We want
# ninja with FDO_CI_CONCURRENT anyway.
/usr/local/bin/ninja -C out/Release/
mkdir /angle
cp out/Release/lib*GL*.so /angle/
ln -s libEGL.so /angle/libEGL.so.1
ln -s libGLESv2.so /angle/libGLESv2.so.2
rm -rf out
popd
rm -rf ./depot_tools

View File

@@ -1,25 +0,0 @@
#!/usr/bin/env bash
# shellcheck disable=SC2086 # we want word splitting
# When changing this file, you need to bump the following
# .gitlab-ci/image-tags.yml tags:
# DEBIAN_X86_64_TEST_GL_TAG
# DEBIAN_X86_64_TEST_VK_TAG
# KERNEL_ROOTFS_TAG
set -ex
APITRACE_VERSION="0a6506433e1f9f7b69757b4e5730326970c4321a"
git clone https://github.com/apitrace/apitrace.git --single-branch --no-checkout /apitrace
pushd /apitrace
git checkout "$APITRACE_VERSION"
git submodule update --init --depth 1 --recursive
cmake -S . -B _build -G Ninja -DCMAKE_BUILD_TYPE=Release -DENABLE_GUI=False -DENABLE_WAFFLE=on $EXTRA_CMAKE_ARGS
cmake --build _build --parallel --target apitrace eglretrace
mkdir build
cp _build/apitrace build
cp _build/eglretrace build
${STRIP_CMD:-strip} build/*
find . -not -path './build' -not -path './build/*' -delete
popd

View File

@@ -1,44 +0,0 @@
#!/usr/bin/env bash
# shellcheck disable=SC2086 # we want word splitting
set -ex
git config --global user.email "mesa@example.com"
git config --global user.name "Mesa CI"
CROSVM_VERSION=1641c55bcc922588e24de73e9cca7b5e4005bd6d
git clone --single-branch -b main --no-checkout https://chromium.googlesource.com/crosvm/crosvm /platform/crosvm
pushd /platform/crosvm
git checkout "$CROSVM_VERSION"
git submodule update --init
VIRGLRENDERER_VERSION=d9c002fac153b834a2c17731f2b85c36e333e102
rm -rf third_party/virglrenderer
git clone --single-branch -b main --no-checkout https://gitlab.freedesktop.org/virgl/virglrenderer.git third_party/virglrenderer
pushd third_party/virglrenderer
git checkout "$VIRGLRENDERER_VERSION"
meson setup build/ -D libdir=lib -D render-server-worker=process -D venus=true $EXTRA_MESON_ARGS
meson install -C build
popd
cargo update -p pkg-config@0.3.26 --precise 0.3.27
RUSTFLAGS='-L native=/usr/local/lib' cargo install \
bindgen-cli \
--locked \
-j ${FDO_CI_CONCURRENT:-4} \
--root /usr/local \
--version 0.65.1 \
$EXTRA_CARGO_ARGS
CROSVM_USE_SYSTEM_MINIGBM=1 CROSVM_USE_SYSTEM_VIRGLRENDERER=1 RUSTFLAGS='-L native=/usr/local/lib' cargo install \
-j ${FDO_CI_CONCURRENT:-4} \
--locked \
--features 'default-no-sandbox gpu x virgl_renderer' \
--path . \
--root /usr/local \
$EXTRA_CARGO_ARGS
popd
rm -rf /platform/crosvm

View File

@@ -1,70 +0,0 @@
#!/usr/bin/env bash
# shellcheck disable=SC2086 # we want word splitting
# When changing this file, you need to bump the following
# .gitlab-ci/image-tags.yml tags:
# DEBIAN_X86_64_TEST_ANDROID_TAG
# DEBIAN_X86_64_TEST_GL_TAG
# DEBIAN_X86_64_TEST_VK_TAG
# KERNEL_ROOTFS_TAG
set -ex
DEQP_RUNNER_VERSION=0.18.0
DEQP_RUNNER_GIT_URL="${DEQP_RUNNER_GIT_URL:-https://gitlab.freedesktop.org/mesa/deqp-runner.git}"
if [ -n "${DEQP_RUNNER_GIT_TAG}${DEQP_RUNNER_GIT_REV}" ]; then
# Build and install from source
DEQP_RUNNER_CARGO_ARGS="--git $DEQP_RUNNER_GIT_URL"
if [ -n "${DEQP_RUNNER_GIT_TAG}" ]; then
DEQP_RUNNER_CARGO_ARGS="--tag ${DEQP_RUNNER_GIT_TAG} ${DEQP_RUNNER_CARGO_ARGS}"
DEQP_RUNNER_GIT_CHECKOUT="$DEQP_RUNNER_GIT_TAG"
else
DEQP_RUNNER_CARGO_ARGS="--rev ${DEQP_RUNNER_GIT_REV} ${DEQP_RUNNER_CARGO_ARGS}"
DEQP_RUNNER_GIT_CHECKOUT="$DEQP_RUNNER_GIT_REV"
fi
DEQP_RUNNER_CARGO_ARGS="${DEQP_RUNNER_CARGO_ARGS} ${EXTRA_CARGO_ARGS}"
else
# Install from package registry
DEQP_RUNNER_CARGO_ARGS="--version ${DEQP_RUNNER_VERSION} ${EXTRA_CARGO_ARGS} -- deqp-runner"
DEQP_RUNNER_GIT_CHECKOUT="v$DEQP_RUNNER_VERSION"
fi
if [[ "$RUST_TARGET" != *-android ]]; then
cargo install --locked \
-j ${FDO_CI_CONCURRENT:-4} \
--root /usr/local \
${DEQP_RUNNER_CARGO_ARGS}
else
mkdir -p /deqp-runner
pushd /deqp-runner
git clone --branch "$DEQP_RUNNER_GIT_CHECKOUT" --depth 1 "$DEQP_RUNNER_GIT_URL" deqp-runner-git
pushd deqp-runner-git
cargo install --locked \
-j ${FDO_CI_CONCURRENT:-4} \
--root /usr/local --version 2.10.0 \
cargo-ndk
rustup target add $RUST_TARGET
RUSTFLAGS='-C target-feature=+crt-static' cargo ndk --target $RUST_TARGET build --release
mv target/$RUST_TARGET/release/deqp-runner /deqp-runner
cargo uninstall --locked \
--root /usr/local \
cargo-ndk
popd
rm -rf deqp-runner-git
popd
fi
# remove unused test runners to shrink images for the Mesa CI build (not kernel,
# which chooses its own deqp branch)
if [ -z "${DEQP_RUNNER_GIT_TAG}${DEQP_RUNNER_GIT_REV}" ]; then
rm -f /usr/local/bin/igt-runner
fi

View File

@@ -1,264 +0,0 @@
#!/usr/bin/env bash
# shellcheck disable=SC2086 # we want word splitting
# When changing this file, you need to bump the following
# .gitlab-ci/image-tags.yml tags:
# DEBIAN_X86_64_TEST_ANDROID_TAG
# DEBIAN_X86_64_TEST_GL_TAG
# DEBIAN_X86_64_TEST_VK_TAG
# KERNEL_ROOTFS_TAG
set -ex -o pipefail
# See `deqp_build_targets` below for which release is used to produce which
# binary. Unless this comment has bitrotten:
# - the VK release produces `deqp-vk`,
# - the GL release produces `glcts`, and
# - the GLES release produces `deqp-gles*` and `deqp-egl`
DEQP_VK_VERSION=1.3.8.2
DEQP_GL_VERSION=4.6.4.0
DEQP_GLES_VERSION=3.2.10.0
# Patches to VulkanCTS may come from commits in their repo (listed in
# cts_commits_to_backport) or patch files stored in our repo (in the patch
# directory `$OLDPWD/.gitlab-ci/container/patches/` listed in cts_patch_files).
# Both list variables would have comments explaining the reasons behind the
# patches.
# shellcheck disable=SC2034
vk_cts_commits_to_backport=(
# Fix more ASAN errors due to missing virtual destructors
dd40bcfef1b4035ea55480b6fd4d884447120768
# Remove "unused shader stages" tests
7dac86c6bbd15dec91d7d9a98cd6dd57c11092a7
)
# shellcheck disable=SC2034
vk_cts_patch_files=(
)
if [ "${DEQP_TARGET}" = 'android' ]; then
vk_cts_patch_files+=(
build-deqp-vk_Allow-running-on-Android-from-the-command-line.patch
build-deqp-vk_Android-prints-to-stdout-instead-of-logcat.patch
)
fi
# shellcheck disable=SC2034
gl_cts_commits_to_backport=(
)
# shellcheck disable=SC2034
gl_cts_patch_files=(
)
if [ "${DEQP_TARGET}" = 'android' ]; then
gl_cts_patch_files+=(
build-deqp-gl_Allow-running-on-Android-from-the-command-line.patch
build-deqp-gl_Android-prints-to-stdout-instead-of-logcat.patch
)
fi
# shellcheck disable=SC2034
# GLES builds also EGL
gles_cts_commits_to_backport=(
# Implement support for the EGL_EXT_config_select_group extension
88ba9ac270db5be600b1ecacbc6d9db0c55d5be4
)
# shellcheck disable=SC2034
gles_cts_patch_files=(
# Correct detection mechanism for EGL_EXT_config_select_group extension
build-deqp-egl_Correct-EGL_EXT_config_select_group-extension-query.patch
)
if [ "${DEQP_TARGET}" = 'android' ]; then
gles_cts_patch_files+=(
build-deqp-gles_Allow-running-on-Android-from-the-command-line.patch
build-deqp-gles_Android-prints-to-stdout-instead-of-logcat.patch
)
fi
### Careful editing anything below this line
git config --global user.email "mesa@example.com"
git config --global user.name "Mesa CI"
# shellcheck disable=SC2153
case "${DEQP_API}" in
VK) DEQP_VERSION="vulkan-cts-$DEQP_VK_VERSION";;
GL) DEQP_VERSION="opengl-cts-$DEQP_GL_VERSION";;
GLES) DEQP_VERSION="opengl-es-cts-$DEQP_GLES_VERSION";;
esac
git clone \
https://github.com/KhronosGroup/VK-GL-CTS.git \
-b $DEQP_VERSION \
--depth 1 \
/VK-GL-CTS
pushd /VK-GL-CTS
mkdir -p /deqp
# shellcheck disable=SC2153
deqp_api=${DEQP_API,,}
cts_commits_to_backport="${deqp_api}_cts_commits_to_backport[@]"
for commit in "${!cts_commits_to_backport}"
do
PATCH_URL="https://github.com/KhronosGroup/VK-GL-CTS/commit/$commit.patch"
echo "Apply patch to ${DEQP_API} CTS from $PATCH_URL"
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 $PATCH_URL | \
git am -
done
cts_patch_files="${deqp_api}_cts_patch_files[@]"
for patch in "${!cts_patch_files}"
do
echo "Apply patch to ${DEQP_API} CTS from $patch"
git am < $OLDPWD/.gitlab-ci/container/patches/$patch
done
{
echo "dEQP base version $DEQP_VERSION"
echo "The following local patches are applied on top:"
git log --reverse --oneline $DEQP_VERSION.. --format=%s | sed 's/^/- /'
} > /deqp/version-$deqp_api
# --insecure is due to SSL cert failures hitting sourceforge for zlib and
# libpng (sigh). The archives get their checksums checked anyway, and git
# always goes through ssh or https.
python3 external/fetch_sources.py --insecure
# Save the testlog stylesheets:
cp doc/testlog-stylesheet/testlog.{css,xsl} /deqp
popd
pushd /deqp
if [ "${DEQP_API}" = 'GLES' ]; then
if [ "${DEQP_TARGET}" = 'android' ]; then
cmake -S /VK-GL-CTS -B . -G Ninja \
-DDEQP_TARGET=android \
-DCMAKE_BUILD_TYPE=Release \
$EXTRA_CMAKE_ARGS
mold --run ninja modules/egl/deqp-egl
mv /deqp/modules/egl/deqp-egl /deqp/modules/egl/deqp-egl-android
else
# When including EGL/X11 testing, do that build first and save off its
# deqp-egl binary.
cmake -S /VK-GL-CTS -B . -G Ninja \
-DDEQP_TARGET=x11_egl_glx \
-DCMAKE_BUILD_TYPE=Release \
$EXTRA_CMAKE_ARGS
mold --run ninja modules/egl/deqp-egl
mv /deqp/modules/egl/deqp-egl /deqp/modules/egl/deqp-egl-x11
cmake -S /VK-GL-CTS -B . -G Ninja \
-DDEQP_TARGET=wayland \
-DCMAKE_BUILD_TYPE=Release \
$EXTRA_CMAKE_ARGS
mold --run ninja modules/egl/deqp-egl
mv /deqp/modules/egl/deqp-egl /deqp/modules/egl/deqp-egl-wayland
fi
fi
cmake -S /VK-GL-CTS -B . -G Ninja \
-DDEQP_TARGET=${DEQP_TARGET} \
-DCMAKE_BUILD_TYPE=Release \
$EXTRA_CMAKE_ARGS
# Make sure `default` doesn't silently stop detecting one of the platforms we care about
if [ "${DEQP_TARGET}" = 'default' ]; then
grep -q DEQP_SUPPORT_WAYLAND=1 build.ninja
grep -q DEQP_SUPPORT_X11=1 build.ninja
grep -q DEQP_SUPPORT_XCB=1 build.ninja
fi
deqp_build_targets=()
case "${DEQP_API}" in
VK)
deqp_build_targets+=(deqp-vk)
;;
GL)
deqp_build_targets+=(glcts)
;;
GLES)
deqp_build_targets+=(deqp-gles{2,3,31})
# deqp-egl also comes from this build, but it is handled separately above.
;;
esac
if [ "${DEQP_TARGET}" != 'android' ]; then
deqp_build_targets+=(testlog-to-xml)
deqp_build_targets+=(testlog-to-csv)
deqp_build_targets+=(testlog-to-junit)
fi
mold --run ninja "${deqp_build_targets[@]}"
if [ "${DEQP_TARGET}" != 'android' ]; then
# Copy out the mustpass lists we want.
mkdir -p /deqp/mustpass
if [ "${DEQP_API}" = 'VK' ]; then
for mustpass in $(< /VK-GL-CTS/external/vulkancts/mustpass/main/vk-default.txt) ; do
cat /VK-GL-CTS/external/vulkancts/mustpass/main/$mustpass \
>> /deqp/mustpass/vk-main.txt
done
fi
if [ "${DEQP_API}" = 'GL' ]; then
cp \
/VK-GL-CTS/external/openglcts/data/mustpass/gl/khronos_mustpass/4.6.1.x/*-main.txt \
/deqp/mustpass/
cp \
/VK-GL-CTS/external/openglcts/data/mustpass/gl/khronos_mustpass_single/4.6.1.x/*-single.txt \
/deqp/mustpass/
fi
if [ "${DEQP_API}" = 'GLES' ]; then
cp \
/VK-GL-CTS/external/openglcts/data/mustpass/gles/aosp_mustpass/3.2.6.x/*.txt \
/deqp/mustpass/
cp \
/VK-GL-CTS/external/openglcts/data/mustpass/egl/aosp_mustpass/3.2.6.x/egl-main.txt \
/deqp/mustpass/
cp \
/VK-GL-CTS/external/openglcts/data/mustpass/gles/khronos_mustpass/3.2.6.x/*-main.txt \
/deqp/mustpass/
fi
# Save *some* executor utils, but otherwise strip things down
# to reduct deqp build size:
mkdir /deqp/executor.save
cp /deqp/executor/testlog-to-* /deqp/executor.save
rm -rf /deqp/executor
mv /deqp/executor.save /deqp/executor
fi
# Remove other mustpass files, since we saved off the ones we wanted to conventient locations above.
rm -rf /deqp/external/**/mustpass/
rm -rf /deqp/external/vulkancts/modules/vulkan/vk-main*
rm -rf /deqp/external/vulkancts/modules/vulkan/vk-default
rm -rf /deqp/external/openglcts/modules/cts-runner
rm -rf /deqp/modules/internal
rm -rf /deqp/execserver
rm -rf /deqp/framework
find . -depth \( -iname '*cmake*' -o -name '*ninja*' -o -name '*.o' -o -name '*.a' \) -exec rm -rf {} \;
if [ "${DEQP_API}" = 'VK' ]; then
${STRIP_CMD:-strip} external/vulkancts/modules/vulkan/deqp-vk
fi
if [ "${DEQP_API}" = 'GL' ]; then
${STRIP_CMD:-strip} external/openglcts/modules/glcts
fi
if [ "${DEQP_API}" = 'GLES' ]; then
${STRIP_CMD:-strip} modules/*/deqp-*
fi
du -sh ./*
rm -rf /VK-GL-CTS
popd

View File

@@ -1,15 +0,0 @@
#!/usr/bin/env bash
# shellcheck disable=SC2086 # we want word splitting
# When changing this file, you need to bump the following
# .gitlab-ci/image-tags.yml tags:
# DEBIAN_BUILD_TAG
set -ex
git clone https://github.com/microsoft/DirectX-Headers -b v1.613.1 --depth 1
pushd DirectX-Headers
meson setup build --backend=ninja --buildtype=release -Dbuild-test=false $EXTRA_MESON_ARGS
meson install -C build
popd
rm -rf DirectX-Headers

View File

@@ -1,19 +0,0 @@
#!/bin/bash
# When changing this file, you need to bump the following
# .gitlab-ci/image-tags.yml tags:
# DEBIAN_X86_64_TEST_VK_TAG
# KERNEL_ROOTFS_TAG
set -ex
git clone https://github.com/ValveSoftware/Fossilize.git
cd Fossilize
git checkout b43ee42bbd5631ea21fe9a2dee4190d5d875c327
git submodule update --init
mkdir build
cd build
cmake -S .. -B . -G Ninja -DCMAKE_BUILD_TYPE=Release
ninja -C . install
cd ../..
rm -rf Fossilize

View File

@@ -1,19 +0,0 @@
#!/usr/bin/env bash
set -ex
GFXRECONSTRUCT_VERSION=761837794a1e57f918a85af7000b12e531b178ae
git clone https://github.com/LunarG/gfxreconstruct.git \
--single-branch \
-b master \
--no-checkout \
/gfxreconstruct
pushd /gfxreconstruct
git checkout "$GFXRECONSTRUCT_VERSION"
git submodule update --init
git submodule update
cmake -S . -B _build -G Ninja -DCMAKE_BUILD_TYPE=Release -DCMAKE_INSTALL_PREFIX:PATH=/gfxreconstruct/build -DBUILD_WERROR=OFF
cmake --build _build --parallel --target tools/{replay,info}/install/strip
find . -not -path './build' -not -path './build/*' -delete
popd

View File

@@ -1,16 +0,0 @@
#!/bin/bash
set -ex
PARALLEL_DEQP_RUNNER_VERSION=fe557794b5dadd8dbf0eae403296625e03bda18a
git clone https://gitlab.freedesktop.org/mesa/parallel-deqp-runner --single-branch -b master --no-checkout /parallel-deqp-runner
pushd /parallel-deqp-runner
git checkout "$PARALLEL_DEQP_RUNNER_VERSION"
meson . _build
ninja -C _build hang-detection
mkdir -p build/bin
install _build/hang-detection build/bin
strip build/bin/*
find . -not -path './build' -not -path './build/*' -delete
popd

View File

@@ -1,23 +0,0 @@
#!/usr/bin/env bash
# shellcheck disable=SC1091 # the path is created by the script
set -ex
KDL_REVISION="5056f71b100a68b72b285c6fc845a66a2ed25985"
mkdir ci-kdl.git
pushd ci-kdl.git
git init
git remote add origin https://gitlab.freedesktop.org/gfx-ci/ci-kdl.git
git fetch --depth 1 origin ${KDL_REVISION}
git checkout FETCH_HEAD
popd
python3 -m venv ci-kdl.venv
source ci-kdl.venv/bin/activate
pushd ci-kdl.git
pip install -r requirements.txt
pip install .
popd
rm -rf ci-kdl.git

View File

@@ -1,31 +0,0 @@
#!/usr/bin/env bash
# shellcheck disable=SC2086 # we want word splitting
# shellcheck disable=SC2153
set -ex
mkdir -p kernel
pushd kernel
if [[ ${DEBIAN_ARCH} = "arm64" ]]; then
KERNEL_IMAGE_NAME+=" cheza-kernel"
fi
for image in ${KERNEL_IMAGE_NAME}; do
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
-o "/lava-files/${image}" "${KERNEL_IMAGE_BASE}/${DEBIAN_ARCH}/${image}"
done
for dtb in ${DEVICE_TREES}; do
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
-o "/lava-files/${dtb}" "${KERNEL_IMAGE_BASE}/${DEBIAN_ARCH}/${dtb}"
done
mkdir -p "/lava-files/rootfs-${DEBIAN_ARCH}"
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
-O "${KERNEL_IMAGE_BASE}/${DEBIAN_ARCH}/modules.tar.zst"
tar --keep-directory-symlink --zstd -xf modules.tar.zst -C "/lava-files/rootfs-${DEBIAN_ARCH}/"
popd
rm -rf kernel

View File

@@ -1,31 +0,0 @@
#!/usr/bin/env bash
set -ex
export LLVM_CONFIG="llvm-config-${LLVM_VERSION:?"llvm unset!"}"
LLVM_TAG="llvmorg-15.0.7"
$LLVM_CONFIG --version
git config --global user.email "mesa@example.com"
git config --global user.name "Mesa CI"
git clone \
https://github.com/llvm/llvm-project \
--depth 1 \
-b "${LLVM_TAG}" \
/llvm-project
mkdir /libclc
pushd /libclc
cmake -S /llvm-project/libclc -B . -G Ninja -DLLVM_CONFIG="$LLVM_CONFIG" -DLIBCLC_TARGETS_TO_BUILD="spirv-mesa3d-;spirv64-mesa3d-" -DCMAKE_BUILD_TYPE=Release -DCMAKE_INSTALL_PREFIX=/usr -DLLVM_SPIRV=/usr/bin/llvm-spirv
ninja
ninja install
popd
# workaroud cmake vs debian packaging.
mkdir -p /usr/lib/clc
ln -s /usr/share/clc/spirv64-mesa3d-.spv /usr/lib/clc/
ln -s /usr/share/clc/spirv-mesa3d-.spv /usr/lib/clc/
du -sh ./*
rm -rf /libclc /llvm-project

View File

@@ -1,16 +0,0 @@
#!/usr/bin/env bash
# Script used for Android and Fedora builds
# shellcheck disable=SC2086 # we want word splitting
set -ex
export LIBDRM_VERSION=libdrm-2.4.119
curl -L -O --retry 4 -f --retry-all-errors --retry-delay 60 \
https://dri.freedesktop.org/libdrm/"$LIBDRM_VERSION".tar.xz
tar -xvf "$LIBDRM_VERSION".tar.xz && rm "$LIBDRM_VERSION".tar.xz
cd "$LIBDRM_VERSION"
meson setup build -D vc4=disabled -D freedreno=disabled -D etnaviv=disabled $EXTRA_MESON_ARGS
meson install -C build
cd ..
rm -rf "$LIBDRM_VERSION"

View File

@@ -1,22 +0,0 @@
#!/usr/bin/env bash
set -ex
VER="${LLVM_VERSION:?llvm not set}.0.0"
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
-O "https://github.com/KhronosGroup/SPIRV-LLVM-Translator/archive/refs/tags/v${VER}.tar.gz"
tar -xvf "v${VER}.tar.gz" && rm "v${VER}.tar.gz"
mkdir "SPIRV-LLVM-Translator-${VER}/build"
pushd "SPIRV-LLVM-Translator-${VER}/build"
cmake .. -G Ninja -DCMAKE_BUILD_TYPE=Release -DCMAKE_INSTALL_PREFIX=/usr
ninja
ninja install
# For some reason llvm-spirv is not installed by default
ninja llvm-spirv
cp tools/llvm-spirv/llvm-spirv /usr/bin/
popd
du -sh "SPIRV-LLVM-Translator-${VER}"
rm -rf "SPIRV-LLVM-Translator-${VER}"

View File

@@ -1,22 +0,0 @@
#!/usr/bin/env bash
set -ex
# When changing this file, you need to bump the following
# .gitlab-ci/image-tags.yml tags:
# DEBIAN_BASE_TAG
# DEBIAN_BUILD_TAG
# FEDORA_X86_64_BUILD_TAG
# KERNEL_ROOTFS_TAG
MOLD_VERSION="2.4.1"
git clone -b v"$MOLD_VERSION" --single-branch --depth 1 https://github.com/rui314/mold.git
pushd mold
cmake -DCMAKE_BUILD_TYPE=Release -D BUILD_TESTING=OFF -D MOLD_LTO=ON
cmake --build . --parallel
cmake --install .
popd
rm -rf mold

View File

@@ -1,25 +0,0 @@
#!/usr/bin/env bash
# When changing this file, you need to bump the following
# .gitlab-ci/image-tags.yml tags:
# DEBIAN_X86_64_TEST_GL_TAG
set -ex -o pipefail
### Careful editing anything below this line
git config --global user.email "mesa@example.com"
git config --global user.name "Mesa CI"
git clone https://github.com/axeldavy/Xnine.git /Xnine
mkdir /Xnine/build
pushd /Xnine/build
git checkout c64753d224c08006bcdcfa7880ada826f27164b1
cmake .. -DBUILD_TESTS=1 -DWITH_DRI3=1 -DD3DADAPTER9_LOCATION=/install/lib/d3d/d3dadapter9.so
make
mkdir -p /NineTests/
mv NineTests/NineTests /NineTests/
popd
rm -rf /Xnine

View File

@@ -1,33 +0,0 @@
#!/bin/bash
# shellcheck disable=SC2086 # we want word splitting
set -ex
# When changing this file, you need to bump the following
# .gitlab-ci/image-tags.yml tags:
# DEBIAN_X86_64_TEST_GL_TAG
# DEBIAN_X86_64_TEST_VK_TAG
# KERNEL_ROOTFS_TAG
REV="f7ece74a107a2f99b2f494d978c84f8d51faa703"
git clone https://gitlab.freedesktop.org/mesa/piglit.git --single-branch --no-checkout /piglit
pushd /piglit
git checkout "$REV"
patch -p1 <$OLDPWD/.gitlab-ci/piglit/disable-vs_in.diff
cmake -S . -B . -G Ninja -DCMAKE_BUILD_TYPE=Release $PIGLIT_OPTS $EXTRA_CMAKE_ARGS
ninja $PIGLIT_BUILD_TARGETS
find . -depth \( -name .git -o -name '*ninja*' -o -iname '*cmake*' -o -name '*.[chao]' \) -exec rm -rf {} \;
rm -rf target_api
if [ "$PIGLIT_BUILD_TARGETS" = "piglit_replayer" ]; then
find . -depth \
! -regex "^\.$" \
! -regex "^\.\/piglit.*" \
! -regex "^\.\/framework.*" \
! -regex "^\.\/bin$" \
! -regex "^\.\/bin\/replayer\.py" \
! -regex "^\.\/templates.*" \
! -regex "^\.\/tests$" \
! -regex "^\.\/tests\/replay\.py" \
-exec rm -rf {} \; 2>/dev/null
fi
popd

View File

@@ -1,39 +0,0 @@
#!/bin/bash
# Note that this script is not actually "building" rust, but build- is the
# convention for the shared helpers for putting stuff in our containers.
set -ex
# cargo (and rustup) wants to store stuff in $HOME/.cargo, and binaries in
# $HOME/.cargo/bin. Make bin a link to a public bin directory so the commands
# are just available to all build jobs.
mkdir -p "$HOME"/.cargo
ln -s /usr/local/bin "$HOME"/.cargo/bin
# Rusticl requires at least Rust 1.66.0 and NAK requires 1.73.0
#
# Also, pick a specific snapshot from rustup so the compiler doesn't drift on
# us.
RUST_VERSION=1.73.0-2023-10-05
# For rust in Mesa, we use rustup to install. This lets us pick an arbitrary
# version of the compiler, rather than whatever the container's Debian comes
# with.
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
--proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- \
--default-toolchain $RUST_VERSION \
--profile minimal \
-y
rustup component add clippy rustfmt
# Set up a config script for cross compiling -- cargo needs your system cc for
# linking in cross builds, but doesn't know what you want to use for system cc.
cat > /root/.cargo/config <<EOF
[target.armv7-unknown-linux-gnueabihf]
linker = "arm-linux-gnueabihf-gcc"
[target.aarch64-unknown-linux-gnu]
linker = "aarch64-linux-gnu-gcc"
EOF

View File

@@ -1,14 +0,0 @@
#!/usr/bin/env bash
# When changing this file, you need to bump the following
# .gitlab-ci/image-tags.yml tags:
# DEBIAN_BUILD_TAG
set -ex
pushd /usr/local
git clone https://gitlab.freedesktop.org/mesa/shader-db.git --depth 1
rm -rf shader-db/.git
cd shader-db
make
popd

View File

@@ -1,89 +0,0 @@
#!/usr/bin/env bash
# SPDX-License-Identifier: MIT
#
# Copyright © 2022 Collabora Limited
# Author: Guilherme Gallo <guilherme.gallo@collabora.com>
#
# When changing this file, you need to bump the following
# .gitlab-ci/image-tags.yml tags:
# KERNEL_ROOTFS_TAG
SKQP_BRANCH=android-cts-12.1_r5
# hack for skqp see the clang
pushd /usr/bin/
ln -s ../lib/llvm-15/bin/clang clang
ln -s ../lib/llvm-15/bin/clang++ clang++
popd
create_gn_args() {
# gn can be configured to cross-compile skia and its tools
# It is important to set the target_cpu to guarantee the intended target
# machine
cp "${BASE_ARGS_GN_FILE}" "${SKQP_OUT_DIR}"/args.gn
echo "target_cpu = \"${SKQP_ARCH}\"" >> "${SKQP_OUT_DIR}"/args.gn
}
download_skia_source() {
if [ -z ${SKIA_DIR+x} ]
then
return 1
fi
# Skia cloned from https://android.googlesource.com/platform/external/skqp
# has all needed assets tracked on git-fs
SKQP_REPO=https://android.googlesource.com/platform/external/skqp
git clone --branch "${SKQP_BRANCH}" --depth 1 "${SKQP_REPO}" "${SKIA_DIR}"
}
set -ex
SCRIPT_DIR=$(realpath "$(dirname "$0")")
SKQP_PATCH_DIR="${SCRIPT_DIR}/patches"
BASE_ARGS_GN_FILE="${SCRIPT_DIR}/build-skqp_base.gn"
SKQP_ARCH=${SKQP_ARCH:-x64}
SKIA_DIR=${SKIA_DIR:-$(mktemp -d)}
SKQP_OUT_DIR=${SKIA_DIR}/out/${SKQP_ARCH}
SKQP_INSTALL_DIR=${SKQP_INSTALL_DIR:-/skqp}
SKQP_ASSETS_DIR="${SKQP_INSTALL_DIR}/assets"
SKQP_BINARIES=(skqp list_gpu_unit_tests list_gms)
download_skia_source
pushd "${SKIA_DIR}"
# Apply all skqp patches for Mesa CI
cat "${SKQP_PATCH_DIR}"/build-skqp_*.patch |
patch -p1
# Fetch some needed build tools needed to build skia/skqp.
# Basically, it clones repositories with commits SHAs from ${SKIA_DIR}/DEPS
# directory.
python tools/git-sync-deps
mkdir -p "${SKQP_OUT_DIR}"
mkdir -p "${SKQP_INSTALL_DIR}"
create_gn_args
# Build and install skqp binaries
bin/gn gen "${SKQP_OUT_DIR}"
for BINARY in "${SKQP_BINARIES[@]}"
do
/usr/bin/ninja -C "${SKQP_OUT_DIR}" "${BINARY}"
# Strip binary, since gn is not stripping it even when `is_debug == false`
${STRIP_CMD:-strip} "${SKQP_OUT_DIR}/${BINARY}"
install -m 0755 "${SKQP_OUT_DIR}/${BINARY}" "${SKQP_INSTALL_DIR}"
done
# Move assets to the target directory, which will reside in rootfs.
mv platform_tools/android/apps/skqp/src/main/assets/ "${SKQP_ASSETS_DIR}"
popd
rm -Rf "${SKIA_DIR}"
set +ex

View File

@@ -1,59 +0,0 @@
cc = "clang"
cxx = "clang++"
extra_cflags = [
"-Wno-error",
"-DSK_ENABLE_DUMP_GPU",
"-DSK_BUILD_FOR_SKQP"
]
extra_cflags_cc = [
"-Wno-error",
# skqp build process produces a lot of compilation warnings, silencing
# most of them to remove clutter and avoid the CI job log to exceed the
# maximum size
# GCC flags
"-Wno-redundant-move",
"-Wno-suggest-override",
"-Wno-class-memaccess",
"-Wno-deprecated-copy",
"-Wno-uninitialized",
# Clang flags
"-Wno-macro-redefined",
"-Wno-anon-enum-enum-conversion",
"-Wno-suggest-destructor-override",
"-Wno-return-std-move-in-c++11",
"-Wno-extra-semi-stmt",
"-Wno-reserved-identifier",
"-Wno-bitwise-instead-of-logical",
"-Wno-reserved-identifier",
"-Wno-psabi",
"-Wno-unused-but-set-variable",
"-Wno-sizeof-array-div",
"-Wno-string-concatenation",
]
cc_wrapper = "ccache"
is_debug = false
skia_enable_fontmgr_android = false
skia_enable_fontmgr_empty = true
skia_enable_pdf = false
skia_enable_skottie = false
skia_skqp_global_error_tolerance = 8
skia_tools_require_resources = true
skia_use_dng_sdk = false
skia_use_expat = true
skia_use_icu = false
skia_use_libheif = false
skia_use_lua = false
skia_use_piex = false
skia_use_vulkan = true
target_os = "linux"

View File

@@ -1,25 +0,0 @@
#!/usr/bin/env bash
# shellcheck disable=SC2086 # we want word splitting
# When changing this file, you need to bump the following
# .gitlab-ci/image-tags.yml tags:
# KERNEL_ROOTFS_TAG
set -ex
git config --global user.email "mesa@example.com"
git config --global user.name "Mesa CI"
git clone \
https://github.com/intel/libva-utils.git \
-b 2.18.1 \
--depth 1 \
/va-utils
pushd /va-utils
# Too old libva in Debian 11. TODO: when this PR gets in, refer to the patch.
curl -L https://github.com/intel/libva-utils/pull/329.patch | git am
meson setup build -D tests=true -Dprefix=/va $EXTRA_MESON_ARGS
meson install -C build
popd
rm -rf /va-utils

View File

@@ -1,43 +0,0 @@
#!/bin/bash
# When changing this file, you need to bump the following
# .gitlab-ci/image-tags.yml tags:
# DEBIAN_X86_64_TEST_VK_TAG
# KERNEL_ROOTFS_TAG
set -ex
VKD3D_PROTON_COMMIT="c3b385606a93baed42482d822805e0d9c2f3f603"
VKD3D_PROTON_DST_DIR="/vkd3d-proton-tests"
VKD3D_PROTON_SRC_DIR="/vkd3d-proton-src"
VKD3D_PROTON_BUILD_DIR="/vkd3d-proton-$VKD3D_PROTON_VERSION"
function build_arch {
local arch="$1"
shift
meson "$@" \
-Denable_tests=true \
--buildtype release \
--prefix "$VKD3D_PROTON_DST_DIR" \
--strip \
--bindir "x${arch}" \
--libdir "x${arch}" \
"$VKD3D_PROTON_BUILD_DIR/build.${arch}"
ninja -C "$VKD3D_PROTON_BUILD_DIR/build.${arch}" install
install -D -m755 -t "${VKD3D_PROTON_DST_DIR}/x${arch}/bin" "$VKD3D_PROTON_BUILD_DIR/build.${arch}/tests/d3d12"
}
git clone https://github.com/HansKristian-Work/vkd3d-proton.git --single-branch -b master --no-checkout "$VKD3D_PROTON_SRC_DIR"
pushd "$VKD3D_PROTON_SRC_DIR"
git checkout "$VKD3D_PROTON_COMMIT"
git submodule update --init --recursive
git submodule update --recursive
build_arch 64
build_arch 86
popd
rm -rf "$VKD3D_PROTON_BUILD_DIR"
rm -rf "$VKD3D_PROTON_SRC_DIR"

View File

@@ -1,18 +0,0 @@
#!/usr/bin/env bash
# When changing this file, you need to bump the following
# .gitlab-ci/image-tags.yml tags:
# DEBIAN_X86_64_TEST_GL_TAG
# KERNEL_ROOTFS_TAG:
set -ex
VALIDATION_TAG="v1.3.281"
git clone -b "$VALIDATION_TAG" --single-branch --depth 1 https://github.com/KhronosGroup/Vulkan-ValidationLayers.git
pushd Vulkan-ValidationLayers
python3 scripts/update_deps.py --dir external --config debug
cmake -G Ninja -DCMAKE_BUILD_TYPE=Release -DCMAKE_INSTALL_PREFIX=/usr -DBUILD_TESTS=OFF -DBUILD_WERROR=OFF -C external/helper.cmake -S . -B build
ninja -C build install
popd
rm -rf Vulkan-ValidationLayers

View File

@@ -1,32 +0,0 @@
#!/usr/bin/env bash
# shellcheck disable=SC2086 # we want word splitting
set -ex
# When changing this file, you need to bump the following
# .gitlab-ci/image-tags.yml tags:
# DEBIAN_BUILD_TAG
# DEBIAN_X86_64_TEST_ANDROID_TAG
# DEBIAN_X86_64_TEST_GL_TAG
# DEBIAN_X86_64_TEST_VK_TAG
# FEDORA_X86_64_BUILD_TAG
# KERNEL_ROOTFS_TAG
export LIBWAYLAND_VERSION="1.21.0"
export WAYLAND_PROTOCOLS_VERSION="1.34"
git clone https://gitlab.freedesktop.org/wayland/wayland
cd wayland
git checkout "$LIBWAYLAND_VERSION"
meson setup -Ddocumentation=false -Ddtd_validation=false -Dlibraries=true _build $EXTRA_MESON_ARGS
meson install -C _build
cd ..
rm -rf wayland
git clone https://gitlab.freedesktop.org/wayland/wayland-protocols
cd wayland-protocols
git checkout "$WAYLAND_PROTOCOLS_VERSION"
meson setup _build $EXTRA_MESON_ARGS
meson install -C _build
cd ..
rm -rf wayland-protocols

View File

@@ -1,14 +0,0 @@
#!/usr/bin/env bash
if test -f /etc/debian_version; then
apt-get autoremove -y --purge
fi
# Clean up any build cache
rm -rf /root/.cache
rm -rf /root/.cargo
rm -rf /.cargo
if test -x /usr/bin/ccache; then
ccache --show-stats
fi

View File

@@ -1,52 +0,0 @@
#!/bin/sh
if test -x /usr/bin/ccache; then
if test -f /etc/debian_version; then
CCACHE_PATH=/usr/lib/ccache
elif test -f /etc/alpine-release; then
CCACHE_PATH=/usr/lib/ccache/bin
else
CCACHE_PATH=/usr/lib64/ccache
fi
# Common setup among container builds before we get to building code.
export CCACHE_COMPILERCHECK=content
export CCACHE_COMPRESS=true
export CCACHE_DIR=/cache/$CI_PROJECT_NAME/ccache
export PATH=$CCACHE_PATH:$PATH
# CMake ignores $PATH, so we have to force CC/GCC to the ccache versions.
export CC="${CCACHE_PATH}/gcc"
export CXX="${CCACHE_PATH}/g++"
ccache --show-stats
fi
# When not using the mold linker (e.g. unsupported architecture), force
# linkers to gold, since it's so much faster for building. We can't use
# lld because we're on old debian and it's buggy. mingw fails meson builds
# with it with "meson.build:21:0: ERROR: Unable to determine dynamic linker"
find /usr/bin -name \*-ld -o -name ld | \
grep -v mingw | \
xargs -n 1 -I '{}' ln -sf '{}.gold' '{}'
# Make a wrapper script for ninja to always include the -j flags
{
echo '#!/bin/sh -x'
# shellcheck disable=SC2016
echo '/usr/bin/ninja -j${FDO_CI_CONCURRENT:-4} "$@"'
} > /usr/local/bin/ninja
chmod +x /usr/local/bin/ninja
# Set MAKEFLAGS so that all make invocations in container builds include the
# flags (doesn't apply to non-container builds, but we don't run make there)
export MAKEFLAGS="-j${FDO_CI_CONCURRENT:-4}"
# make wget to try more than once, when download fails or timeout
echo -e "retry_connrefused = on\n" \
"read_timeout = 300\n" \
"tries = 4\n" \
"retry_on_host_error = on\n" \
"retry_on_http_error = 429,500,502,503,504\n" \
"wait_retry = 32" >> /etc/wgetrc

View File

@@ -1,37 +0,0 @@
#!/bin/bash
ndk=$1
arch=$2
cpu_family=$3
cpu=$4
cross_file="/cross_file-$arch.txt"
sdk_version=$5
# armv7 has the toolchain split between two names.
arch2=${6:-$2}
# Note that we disable C++ exceptions, because Mesa doesn't use exceptions,
# and allowing it in code generation means we get unwind symbols that break
# the libEGL and driver symbol tests.
cat > "$cross_file" <<EOF
[binaries]
ar = '$ndk/toolchains/llvm/prebuilt/linux-x86_64/bin/llvm-ar'
c = ['ccache', '$ndk/toolchains/llvm/prebuilt/linux-x86_64/bin/${arch2}${sdk_version}-clang', '-fno-exceptions', '-fno-unwind-tables', '-fno-asynchronous-unwind-tables']
cpp = ['ccache', '$ndk/toolchains/llvm/prebuilt/linux-x86_64/bin/${arch2}${sdk_version}-clang++', '-fno-exceptions', '-fno-unwind-tables', '-fno-asynchronous-unwind-tables', '-static-libstdc++']
c_ld = 'lld'
cpp_ld = 'lld'
strip = '$ndk/toolchains/llvm/prebuilt/linux-x86_64/bin/llvm-strip'
pkgconfig = ['/usr/bin/pkgconf']
[host_machine]
system = 'android'
cpu_family = '$cpu_family'
cpu = '$cpu'
endian = 'little'
[properties]
needs_exe_wrapper = true
pkg_config_libdir = '/usr/local/lib/${arch2}/pkgconfig/:/${ndk}/toolchains/llvm/prebuilt/linux-x86_64/sysroot/usr/lib/${arch2}/pkgconfig/'
EOF

View File

@@ -1,40 +0,0 @@
#!/bin/sh
# shellcheck disable=SC2086 # we want word splitting
# Makes a .pc file in the Android NDK for meson to find its libraries.
set -ex
ndk="$1"
pc="$2"
cflags="$3"
libs="$4"
version="$5"
sdk_version="$6"
sysroot=$ndk/toolchains/llvm/prebuilt/linux-x86_64/sysroot
for arch in \
x86_64-linux-android \
i686-linux-android \
aarch64-linux-android \
arm-linux-androideabi; do
pcdir=$sysroot/usr/lib/$arch/pkgconfig
mkdir -p $pcdir
cat >$pcdir/$pc <<EOF
prefix=$sysroot
exec_prefix=$sysroot
libdir=$sysroot/usr/lib/$arch/$sdk_version
sharedlibdir=$sysroot/usr/lib/$arch
includedir=$sysroot/usr/include
Name: zlib
Description: zlib compression library
Version: $version
Requires:
Libs: -L$sysroot/usr/lib/$arch/$sdk_version $libs
Cflags: -I$sysroot/usr/include $cflags
EOF
done

View File

@@ -1,54 +0,0 @@
#!/bin/bash
arch=$1
cross_file="/cross_file-$arch.txt"
meson env2mfile --cross --debarch "$arch" -o "$cross_file"
# Explicitly set ccache path for cross compilers
sed -i "s|/usr/bin/\([^-]*\)-linux-gnu\([^-]*\)-g|/usr/lib/ccache/\\1-linux-gnu\\2-g|g" "$cross_file"
# Rely on qemu-user being configured in binfmt_misc on the host
# shellcheck disable=SC1003 # how this sed doesn't seems to work for me locally
sed -i -e '/\[properties\]/a\' -e "needs_exe_wrapper = False" "$cross_file"
# Add a line for rustc, which meson env2mfile is missing.
cc=$(sed -n "s|^c\s*=\s*\[?'\(.*\)'\]?|\1|p" < "$cross_file")
if [[ "$arch" = "arm64" ]]; then
rust_target=aarch64-unknown-linux-gnu
elif [[ "$arch" = "armhf" ]]; then
rust_target=armv7-unknown-linux-gnueabihf
elif [[ "$arch" = "i386" ]]; then
rust_target=i686-unknown-linux-gnu
elif [[ "$arch" = "ppc64el" ]]; then
rust_target=powerpc64le-unknown-linux-gnu
elif [[ "$arch" = "s390x" ]]; then
rust_target=s390x-unknown-linux-gnu
else
echo "Needs rustc target mapping"
fi
# shellcheck disable=SC1003 # how this sed doesn't seems to work for me locally
sed -i -e '/\[binaries\]/a\' -e "rust = ['rustc', '--target=$rust_target', '-C', 'linker=$cc']" "$cross_file"
# Set up cmake cross compile toolchain file for dEQP builds
toolchain_file="/toolchain-$arch.cmake"
if [[ "$arch" = "arm64" ]]; then
GCC_ARCH="aarch64-linux-gnu"
DE_CPU="DE_CPU_ARM_64"
elif [[ "$arch" = "armhf" ]]; then
GCC_ARCH="arm-linux-gnueabihf"
DE_CPU="DE_CPU_ARM"
fi
if [[ -n "$GCC_ARCH" ]]; then
{
echo "set(CMAKE_SYSTEM_NAME Linux)";
echo "set(CMAKE_SYSTEM_PROCESSOR arm)";
echo "set(CMAKE_C_COMPILER /usr/lib/ccache/$GCC_ARCH-gcc)";
echo "set(CMAKE_CXX_COMPILER /usr/lib/ccache/$GCC_ARCH-g++)";
echo "set(CMAKE_CXX_FLAGS_INIT \"-Wno-psabi\")"; # makes ABI warnings quiet for ARMv7
echo "set(ENV{PKG_CONFIG} \"/usr/bin/$GCC_ARCH-pkgconf\")";
echo "set(DE_CPU $DE_CPU)";
} > "$toolchain_file"
fi

View File

@@ -1,92 +0,0 @@
#!/usr/bin/env bash
# shellcheck disable=SC2086 # we want word splitting
set -e
set -o xtrace
export DEBIAN_FRONTEND=noninteractive
export LLVM_VERSION="${LLVM_VERSION:=15}"
# Ephemeral packages (installed for this script and removed again at the end)
EPHEMERAL=(
)
DEPS=(
"crossbuild-essential-$arch"
"pkgconf:$arch"
"libasan8:$arch"
"libdrm-dev:$arch"
"libelf-dev:$arch"
"libexpat1-dev:$arch"
"libffi-dev:$arch"
"libpciaccess-dev:$arch"
"libstdc++6:$arch"
"libvulkan-dev:$arch"
"libx11-dev:$arch"
"libx11-xcb-dev:$arch"
"libxcb-dri2-0-dev:$arch"
"libxcb-dri3-dev:$arch"
"libxcb-glx0-dev:$arch"
"libxcb-present-dev:$arch"
"libxcb-randr0-dev:$arch"
"libxcb-shm0-dev:$arch"
"libxcb-xfixes0-dev:$arch"
"libxdamage-dev:$arch"
"libxext-dev:$arch"
"libxrandr-dev:$arch"
"libxshmfence-dev:$arch"
"libxxf86vm-dev:$arch"
"libwayland-dev:$arch"
)
dpkg --add-architecture $arch
echo "deb [trusted=yes] https://gitlab.freedesktop.org/gfx-ci/ci-deb-repo/-/raw/${PKG_REPO_REV}/ ${FDO_DISTRIBUTION_VERSION%-*} main" | tee /etc/apt/sources.list.d/gfx-ci_.list
apt-get update
apt-get install -y --no-remove "${DEPS[@]}" "${EPHEMERAL[@]}" \
$EXTRA_LOCAL_PACKAGES
if [[ $arch != "armhf" ]]; then
# We don't need clang-format for the crossbuilds, but the installed amd64
# package will conflict with libclang. Uninstall clang-format (and its
# problematic dependency) to fix.
apt-get remove -y "clang-format-${LLVM_VERSION}" "libclang-cpp${LLVM_VERSION}" \
"llvm-${LLVM_VERSION}-runtime" "llvm-${LLVM_VERSION}-linker-tools"
# llvm-*-tools:$arch conflicts with python3:amd64. Install dependencies only
# with apt-get, then force-install llvm-*-{dev,tools}:$arch with dpkg to get
# around this.
apt-get install -y --no-remove --no-install-recommends \
"libclang-cpp${LLVM_VERSION}:$arch" \
"libgcc-s1:$arch" \
"libtinfo-dev:$arch" \
"libz3-dev:$arch" \
"llvm-${LLVM_VERSION}:$arch" \
zlib1g
fi
. .gitlab-ci/container/create-cross-file.sh $arch
. .gitlab-ci/container/container_pre_build.sh
# dependencies where we want a specific version
MULTIARCH_PATH=$(dpkg-architecture -A $arch -qDEB_TARGET_MULTIARCH)
export EXTRA_MESON_ARGS="--cross-file=/cross_file-${arch}.txt -D libdir=lib/${MULTIARCH_PATH}"
. .gitlab-ci/container/build-wayland.sh
. .gitlab-ci/container/build-directx-headers.sh
apt-get purge -y "${EPHEMERAL[@]}"
. .gitlab-ci/container/container_post_build.sh
# This needs to be done after container_post_build.sh, or apt-get breaks in there
if [[ $arch != "armhf" ]]; then
apt-get download llvm-"${LLVM_VERSION}"-{dev,tools}:"$arch"
dpkg -i --force-depends llvm-"${LLVM_VERSION}"-*_"${arch}".deb
rm llvm-"${LLVM_VERSION}"-*_"${arch}".deb
fi

View File

@@ -1,93 +0,0 @@
#!/usr/bin/env bash
# shellcheck disable=SC2086 # we want word splitting
# When changing this file, you need to bump the following
# .gitlab-ci/image-tags.yml tags:
# DEBIAN_BUILD_TAG
set -ex
EPHEMERAL=(
autoconf
rdfind
unzip
)
apt-get install -y --no-remove "${EPHEMERAL[@]}"
# Fetch the NDK and extract just the toolchain we want.
ndk=$ANDROID_NDK
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
-o $ndk.zip https://dl.google.com/android/repository/$ndk-linux.zip
unzip -d / $ndk.zip "$ndk/toolchains/llvm/*"
rm $ndk.zip
# Since it was packed as a zip file, symlinks/hardlinks got turned into
# duplicate files. Turn them into hardlinks to save on container space.
rdfind -makehardlinks true -makeresultsfile false /${ndk}/
# Drop some large tools we won't use in this build.
find /${ndk}/ -type f \( -iname '*clang-check*' -o -iname '*clang-tidy*' -o -iname '*lldb*' \) -exec rm -f {} \;
sh .gitlab-ci/container/create-android-ndk-pc.sh /$ndk zlib.pc "" "-lz" "1.2.3" $ANDROID_SDK_VERSION
sh .gitlab-ci/container/create-android-cross-file.sh /$ndk x86_64-linux-android x86_64 x86_64 $ANDROID_SDK_VERSION
sh .gitlab-ci/container/create-android-cross-file.sh /$ndk i686-linux-android x86 x86 $ANDROID_SDK_VERSION
sh .gitlab-ci/container/create-android-cross-file.sh /$ndk aarch64-linux-android aarch64 armv8 $ANDROID_SDK_VERSION
sh .gitlab-ci/container/create-android-cross-file.sh /$ndk arm-linux-androideabi arm armv7hl $ANDROID_SDK_VERSION armv7a-linux-androideabi
for arch in \
x86_64-linux-android \
i686-linux-android \
aarch64-linux-android \
arm-linux-androideabi ; do
EXTRA_MESON_ARGS="--cross-file=/cross_file-$arch.txt --libdir=lib/$arch -Dnouveau=disabled -Dintel=disabled" \
. .gitlab-ci/container/build-libdrm.sh
done
rm -rf $LIBDRM_VERSION
export LIBELF_VERSION=libelf-0.8.13
curl -L --retry 4 -f --retry-all-errors --retry-delay 60 \
-O https://fossies.org/linux/misc/old/$LIBELF_VERSION.tar.gz
# Not 100% sure who runs the mirror above so be extra careful
if ! echo "4136d7b4c04df68b686570afa26988ac ${LIBELF_VERSION}.tar.gz" | md5sum -c -; then
echo "Checksum failed"
exit 1
fi
tar -xf ${LIBELF_VERSION}.tar.gz
cd $LIBELF_VERSION
# Work around a bug in the original configure not enabling __LIBELF64.
autoreconf
for arch in \
x86_64-linux-android \
i686-linux-android \
aarch64-linux-android \
arm-linux-androideabi ; do
ccarch=${arch}
if [ "${arch}" == 'arm-linux-androideabi' ]
then
ccarch=armv7a-linux-androideabi
fi
export CC=/${ndk}/toolchains/llvm/prebuilt/linux-x86_64/bin/llvm-ar
export CC=/${ndk}/toolchains/llvm/prebuilt/linux-x86_64/bin/${ccarch}${ANDROID_SDK_VERSION}-clang
export CXX=/${ndk}/toolchains/llvm/prebuilt/linux-x86_64/bin/${ccarch}${ANDROID_SDK_VERSION}-clang++
export LD=/${ndk}/toolchains/llvm/prebuilt/linux-x86_64/bin/${arch}-ld
export RANLIB=/${ndk}/toolchains/llvm/prebuilt/linux-x86_64/bin/llvm-ranlib
# The configure script doesn't know about android, but doesn't really use the host anyway it
# seems
./configure --host=x86_64-linux-gnu --disable-nls --disable-shared \
--libdir=/usr/local/lib/${arch}
make install
make distclean
done
cd ..
rm -rf $LIBELF_VERSION
apt-get purge -y "${EPHEMERAL[@]}"

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