Compare commits

..

1572 Commits

Author SHA1 Message Date
Julien Cristau
1cfffc29f6 Prepare changelog for upload 2010-05-20 17:34:45 +02:00
Julien Cristau
27da77b49a Bump changelogs 2010-05-20 17:33:16 +02:00
Julien Cristau
95b8885da8 Merge branch 'mesa_7_7_branch' of git://anongit.freedesktop.org/git/mesa/mesa into debian-unstable 2010-05-20 17:32:13 +02:00
Julien Cristau
7a8e3614f7 debian/rules: use DEB_HOST_ARCH_CPU instead of DEB_HOST_GNU_CPU.
Prompted by Ubuntu which changed the latter to i686, breaking their
packages on i386.
2010-05-20 11:00:01 +02:00
José Fonseca
8ba378d969 mesa/st: Fill in native program limits.
In the lack of more fine grained capabilities in Gallium, assume that if
the pipe driver supports GLSL then native limits match Mesa software
limits.
(cherry picked from commit 40a90cd11234a09c2477f5c9984dd6d9fac3f52c)
2010-05-04 15:36:30 +01:00
Ian Romanick
942ec8722d Fix __glXInitializeVisualConfigFromTags's handling of unrecognized fbconfig
__glXInitializeVisualConfigFromTags doesn't skip the payload of
unrecognized tags.  Instead, it treats the value as if it were the
next tag, which can happen if the server's GLX extension is not
Mesa's.  For example, this falls down when NVIDIA sends a
GLX_FLOAT_COMPONENTS_NV = 0 pair, causing
__glXInitializeVisualConfigFromTags to bail out early.

Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
2010-04-23 11:53:50 -07:00
Tormod Volden
45affda4bf radeon: 9800 SE has only one quadpipe
Although these cards have 2 pipelines on the silicon only
the first passed the QA and the other should be disabled.

http://www.digital-daily.com/video/ati-radeon9800se/
http://www.rojakpot.com/showarticle.aspx?artno=101&pgno=1

Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2010-04-22 17:17:47 -04:00
Alex Deucher
593f7cc315 r300: fix vertex unit setup
RV3xx is 2, RV560,RV570 is 8

Noticed by Tormod Volden.
2010-04-22 02:16:41 -04:00
Brian Paul
6d18fd6756 st/mesa: invert scissor rect depending on FB orientation
Fixes fd.o bug 27715

(cherry picked from commit c060265bdb)
2010-04-19 10:13:40 -06:00
Alex Deucher
51d4d0a12e r600: add new r7xx pci ids 2010-04-09 20:22:06 -04:00
Maciej Cencora
a548964959 r300: set proper vertex index limits also in non indexed mode
Fixes #27521, broken menus in UT2004 and broken water refraction in Sauerbraten.
2010-04-09 21:26:01 +02:00
Brian Paul
08b98fac62 mesa: fix instruction indexing bugs
We were looping over instructions but only looking at the 0th instruction's
opcode.

Fixes fd.o bug 27566.
(cherry picked from commit b22a00bff4)
2010-04-09 10:12:02 -06:00
Jeremy Huddleston
65c337e29b Revert accidental commits from the xquartz tree
This reverts commit 5726f2f1b8.
This reverts commit 2d5c596ca4.
This reverts commit 48ea93eb88.
This reverts commit 3e6d5aedf0.
This reverts commit cfb4f95bd5.
This reverts commit d57aa45dac.
2010-04-01 18:15:46 -07:00
Jeremy Huddleston
5726f2f1b8 apple: Use mesa gl.h rather than generating one.
Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
2010-04-01 10:05:57 -07:00
Jeremy Huddleston
2d5c596ca4 apple: Integrate our libGL into the existing build system better
Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
2010-04-01 10:05:56 -07:00
Jeremy Huddleston
48ea93eb88 apple: Change ifdefs for DRI to be DRI && !APPLE
Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
2010-04-01 10:05:56 -07:00
Jeremy Huddleston
3e6d5aedf0 apple: Remove duplicate headers that already exist in mesa.
Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
2010-04-01 10:05:56 -07:00
Jeremy Huddleston
cfb4f95bd5 apple: Adding .gitignore for src/glx/x11/apple
Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
2010-04-01 10:05:56 -07:00
Jeremy Huddleston
d57aa45dac apple: Initial import of libGL for OSX from AppleSGLX svn repository.
Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
2010-04-01 10:05:55 -07:00
Brice Goglin
8d98d78320 Prepare changelog for upload 2010-03-30 17:15:15 +02:00
Brice Goglin
ce01e79d7b Bump Standards-Version to 3.8.4 2010-03-30 17:14:41 +02:00
Brian Paul
e6f5ca0fa3 docs: fix 7.7.1 release date 2010-03-29 10:57:48 -06:00
Timo Aaltonen
44083f0cd6 Update the changelogs. 2010-03-29 19:52:23 +03:00
Timo Aaltonen
9f4072889f Merge commit 'mesa_7_7_1' into debian-unstable 2010-03-29 19:51:04 +03:00
Ian Romanick
0c88e34049 mesa: Add 7.7.1 release MD5 sums 2010-03-28 16:25:47 -07:00
Ian Romanick
663642b435 mesa: set version string to 7.7.1
Also set the correct release date.
2010-03-28 16:22:44 -07:00
Brian Paul
535742d75f mesa: move/update hash function comments 2010-03-27 08:58:02 -06:00
Török Edwin
8fe3b3f66a mesa: fix deadlock in _mesa_HashFindFreeKeyBlock()
Fixes fd.o bug 27340.
2010-03-27 08:55:47 -06:00
Michel Dänzer
4cf14fa80b dri/swrast: Fix frontbuffer rendering.
Was broken since the endianness fixes.
2010-03-26 18:29:18 +01:00
Michel Dänzer
6ec259eb17 dri/swrast: Fix missed conversion of one pixel pointer increment.
This probably broke the swrast DRI driver when running X in depth 16.
2010-03-26 16:29:59 +01:00
Brian Paul
ca1c8a183d docs: added news item for 7.7.1 release 2010-03-23 09:19:44 -06:00
Brian Paul
3f3b1094bf docs: insert expected 7.7.1 release date 2010-03-23 09:19:26 -06:00
Ian Romanick
e33121b2d8 mesa: set version string to 7.7.1-rc2 2010-03-22 18:15:53 -07:00
Pauli Nieminen
415d0326bb r200: Fix emit size prediction to account elt splitting.
Emit sizes prediction didn't account for render splitting in
hwtnl path.
2010-03-21 23:32:14 +02:00
Pauli Nieminen
7e24ce2d9b r200: Don't flush when closing elts in KMS.
Flush in middle of rendering in KMS is not allowed because
buffers are discarded in flush.

Fixes crash when emiting split indices with RADEON_DEBUG=all.
2010-03-21 23:30:23 +02:00
Pauli Nieminen
2ae754b7b9 r200: Fix swtnl fallback to flush pending rendering before transition.
Flush after transition would emit wrong state that could cause
wrong state emited for pending rendering operation.

Fixes wan once from extrement tuxracer that is using per vertex
materials.
2010-03-21 20:23:01 +02:00
Pauli Nieminen
501156b36b r200: Fix mixed indetion in r200TclFallback. 2010-03-21 20:22:56 +02:00
Pauli Nieminen
4ff3244457 vbo: Fix vbo_split_copy to pass correct max_index to draw.
vbo_split_copy was passing one past the max_index to draw function
which caused _tnl_draw_prims function to read uninitialized values
from copied array.

Bug was spoted in valgrind report of progs/tests/cva_huge.
2010-03-21 12:42:10 +02:00
Michel Dänzer
3889556d70 Revert "Revert "mesa/st: Make ST_SURFACE_DEPTH index consistent with mesa's BUFFER_DEPTH.""
This reverts commit 52d83efdbc.

The fix it reverted should be fine with the two previous fixes.
2010-03-19 17:32:54 +01:00
Brian Paul
6412046f65 st/mesa: s/BUFFER_FRONT_LEFT/surfIndex/
(cherry picked from commit 5d5c524cbe)
2010-03-19 17:20:26 +01:00
Brian Paul
6e96cea6e2 st/mesa: fix st_set_framebuffer_surface() state validation
Set the _NEW_BUFFERS flag and remove the code which updated the
parent framebuffer size.  Normal Mesa state validation will do that.

Fixes issues with Warsow on r300g and possibly other bugs.
(cherry picked from commit c472ce5302)
2010-03-19 17:20:19 +01:00
Ian Romanick
3a3ef3d6c9 intel: Use bit-wise not instead of logical not (i830 path)
The assertion is checking that the low-order bits of offset are not
set.  It does this by anding the inverted offset mask with the
offset.  This is clearly intended to be a bit-wise "invert".

Fixes bug #25984.
(cherry picked from commit 062a208814)
2010-03-18 18:23:21 -07:00
Ian Romanick
98aed6dc69 intel: Correct value of S0_VB_OFFSET_MASK to match hardware docs.
(cherry picked from commit 689e4b5541)
2010-03-18 17:31:22 -07:00
Ian Romanick
d3a607f889 Use bit-wise not instead of logical not.
The assertion is checking that the low-order bits of offset are not
set.  It does this by anding the inverted offset mask with the
offset.  This is clearly intended to be a bit-wise "invert".

Fixes bug #25984.
(cherry picked from commit fda5078324)
2010-03-18 15:39:00 -07:00
Timo Aaltonen
65a728348a Update the changelogs. 2010-03-18 13:51:45 +02:00
Timo Aaltonen
99d174f882 Merge remote branch 'upstream/mesa_7_7_branch' into debian-unstable 2010-03-18 13:30:34 +02:00
Thomas Hellstrom
00e41e007e st/xorg: Fix Xv cliprect scaling.
Due to a quantization error, different cliprects of scaled video windows may
not have identical x / y scale.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-03-13 20:32:00 +01:00
Ian Romanick
6624845a5d mesa: set version string to 7.7.1-rc1 2010-03-16 10:26:52 -07:00
Matthew W. S. Bell
c9c54180e4 Correct GL_EQUIV code in r67/7xx.
From 247e121106e8d3e389f2e5a6edf13ea70ac18df7 Mon Sep 17 00:00:00 2001

These seem to be documented in
<http://www.svgopen.org/2003/papers/RasterOperationsUsingFilterElements/index.html>.
2010-03-16 11:38:32 -04:00
José Fonseca
93e77b0028 mesa: List Quake3 extensions first.
Quake3 truncates the extension string, and GL_EXT_compiled_vertex_array
wasn't being detected, making it very slow.

This is a quick fix. The IMHO best way to address this in a more general
fashion is to sort by year.
2010-03-13 10:41:32 +00:00
Keith Whitwell
0c9e8e6c6e mesa: Fix memory leak in decompress-with-blit.
(cherry picked from commit f05a4ee6f2840590c90da4be2fe5c6295410a5af)
2010-03-13 10:40:56 +00:00
Michel Dänzer
c50477c255 st/dri: Always try to set up R5G6B5 configs.
Allows compiz to work in depth 16.

The DRI2 getBuffersWithFormat hook is only required for 16/32 bit depth
buffers, for colour buffers the only requirement is that the format matches
the drawable depth, which we can't check here.
2010-03-12 18:24:34 +01:00
Brice Goglin
bf8957e7b9 Remove Thierry Reding from Uploaders 2010-03-04 20:52:25 +01:00
Brice Goglin
b5f7f02399 Prepare Changelog for upload 2010-02-24 22:44:25 +01:00
Brice Goglin
cfbe87746f New upstream snapshot, to avoid fd.o bug #26734 2010-02-24 22:44:03 +01:00
Brice Goglin
09dfea4380 Merge branch 'upstream-experimental' into debian-unstable 2010-02-24 22:42:33 +01:00
Brian Paul
293f4d51b4 dri: remove old assertion (see bug 26734) 2010-02-24 14:04:26 -07:00
Brice Goglin
d6f8c95d32 New upstream snapshot 2010-02-24 20:53:58 +01:00
Brice Goglin
fec3df9cfb Merge branch 'upstream-experimental' into debian-experimental
Conflicts:
	scons/crossmingw.py
	scons/gallium.py
2010-02-24 20:51:54 +01:00
Alex Deucher
f0e99179bc r600: update state count for CB_COLOR0_FRAG & CB_COLOR0_TILE relocs 2010-02-24 12:03:10 -05:00
Jerome Glisse
69a94e1452 r6xx/r7xx: emit relocation for FRAG & TILE buffer
FRAG & TILE buffer are unused but still they need
to be associated with a valid relocation so that
userspace can't try to abuse them to overwritte
GART and then try to write anywhere in system
memory.
2010-02-24 12:03:00 -05:00
Dave Airlie
bc7e12e5e3 radeon/r200: fix bad state emission causes kernel to do bad depth clear
The kernel lets you clear depth without getting a depth offset
from userspace, mesa used to emit state before clear, but that got
lost in the refactoring, which made the kernel bug show up. Fix
mesa driver to emit the state properly now.

cherry-pick + squash master commits.

Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-02-24 19:16:13 +10:00
Jakob Bornecrantz
7accf8ced6 vmware/xorg: Bump to match vmware driver 2010-02-22 20:07:06 +01:00
Jakob Bornecrantz
52d83efdbc Revert "mesa/st: Make ST_SURFACE_DEPTH index consistent with mesa's BUFFER_DEPTH."
This reverts commit 9d17ad2891.

Fun stuff so fixing exposes another bug which I'm having trouble
tracking down. So for now I'm just going to revert this untill
I can fix the real bug. Sorry about this.
2010-02-22 19:37:09 +01:00
Jakob Bornecrantz
1702db3a35 st/xorg: Make resize fail gracefully 2010-02-22 19:34:56 +01:00
Jakob Bornecrantz
fa6eee135e st/xorg: Obey max {width|height} from kernel and Gallium 2010-02-22 19:34:56 +01:00
José Fonseca
b5fa760972 svga: Upload user buffers only once. 2010-02-22 19:38:10 +00:00
José Fonseca
69cf45cdae svga: Fix the guest offset of piecewise buffer DMAs.
It was being erroneously set equal to the host offset, but it should be
zero.
2010-02-22 19:38:08 +00:00
José Fonseca
f5ffbe0bc3 svga: Emit a scalar rcp in shadow maps.
Small improvement in Lightsmark 2008.
2010-02-22 19:38:05 +00:00
José Fonseca
46d8ca023d svga: Don't emit zero writemasks.
This fixes a regression with Lightsmark, where more compact TGSI from Mesa
was causing a zero mask MOV to be emitted for shadow map compare, causing
problems in some backends.

Add a few more assertions to catch cases like this.
2010-02-22 19:38:04 +00:00
José Fonseca
9bef69782d tgsi: Report an error when a destination has an empty writemask. 2010-02-22 19:38:01 +00:00
Xavier Chantry
b0e5dcb859 glxinfo: Fix User error in glGetProgramivARB
glxinfo needed fixing after commit 4bccd69.

Move fragment program only parameters into their own list
so that they are not queried for a vertex program.

Signed-off-by: Xavier Chantry <chantry.xavier@gmail.com>
Signed-off-by: Brian Paul <brianp@vmware.com>
2010-02-22 08:53:03 -07:00
Brian Paul
bcd561c667 st/mesa: change viewport Z scale/bias for glBitmap/glDrawPixels
This fixes incorrect Z position of glBitmap, glDraw/CopyPixels for the
svga driver.  Now we use 0.5, 0.5 as is typical for ordinary 3D rendering.
2010-02-22 08:01:28 -07:00
José Fonseca
c98eced9ae scons: Control caching via new SCONS_CACHE_DIR environment variable.
This serves several purposes:
- disable caching in situations were is it useless or undesired
- share caches among all trees
- simplify purging the cache (when it's a single location)
- move the cache out of the tree, since that slows downs IDEs considerably

To retain previous behavior just define do

  export SCONS_CACHE_DIR=$PWD/build/cache

before invoking scons.
2010-02-22 11:28:55 +00:00
Brian Paul
13cbb5fff6 st/mesa: Add checks for ST_SURFACE_x vs MESA_BUFFER_x
The ST_SURFACE_x values should match the Mesa BUFFER_x values.
Added some assertions to prevent future mix-ups.
2010-02-21 12:38:46 +01:00
Thomas Hellstrom
dd8d78c908 st/xorg: Encapsulate all customizable stuff in a separate class.
This avoids exposing the ms driver structure to the winsys,
and nicely encapsulates driver customizable stuff.
In the future more things might be customizable by the winsys, like
throttling, 3D readback etc.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-02-21 12:25:33 +01:00
Thomas Hellstrom
9d17ad2891 mesa/st: Make ST_SURFACE_DEPTH index consistent with mesa's BUFFER_DEPTH.
Some st functions assume that they are identical.
2010-02-21 10:39:38 +01:00
Brian Paul
d437d905e6 softpipe: fix depth testing problems
The optimized Z-test functions assumed that the array of incoming quads
are adjacent, but that's not always true.  The fragment shader can cull
intermediate quads, for example.

Now these Z-test functions can cope with non-adjacent quads.  A little bit
of performance is probably lost, but it's probably not worth worring about.

This fixes broken glBitmap() Z testing, among other things.
2010-02-18 16:47:27 -07:00
Jon TURNEY
551c96979e mklib: Teach mklib to fail build if link fails on cygwin
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Signed-off-by: Brian Paul <brianp@vmware.com>
2010-02-18 12:41:30 -07:00
Jon TURNEY
c1a4f249f1 configure: PTHREADS should be defined on cygwin, as well
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Signed-off-by: Brian Paul <brianp@vmware.com>
2010-02-18 12:41:08 -07:00
José Fonseca
18ff85e91d svga: Upload the dirty ranges instead of the whole buffer. 2010-02-18 18:22:31 +00:00
José Fonseca
12617c7e30 svga: Break uploads of big buffers into smaller pieces.
Fixes crash in Homeworld2 which tries to create a 14MB buffer, because we
now avoid creating GMR buffers larger than 8MB to ensure progress given
we have only a 16MB pool.
2010-02-18 18:22:31 +00:00
José Fonseca
5dbf44953c svga: Note ranges even when there is no hardware buffer around.
Not having a hardware buffer around doesn't change the fact that the
range is dirty and needs to be uploaded eventually.
2010-02-18 18:22:31 +00:00
José Fonseca
a51d638ff3 svga: Further cleanup/comment svga buffer code.
Just cosmetic changes -- no behavior change.
2010-02-18 18:22:30 +00:00
José Fonseca
770945cff4 svga: Temporarily create a sw vertex buf when failed to create a hw buf.
Many apps don't check the return of map buffer so it is better not to
fail.
2010-02-18 18:22:30 +00:00
Jakob Bornecrantz
eaa4066bfc vmware/xorg: Export modinfo just as the other vmware X drivers does 2010-02-18 17:07:32 +01:00
Jakob Bornecrantz
46cf606cd3 st/xorg: Fixup configure
Make sure we always test for XEXT version.
Make sure that the user has a recent version of libkms and libdrm installed.

Not that the st/xorg code requires so new versions to work but these are
needed to make a proper vmwgfx_drv.so driver which is the only real user.
2010-02-18 17:07:32 +01:00
Jakob Bornecrantz
a69a7b9688 st/xorg: A bit more debugging info 2010-02-18 16:13:18 +01:00
Jakob Bornecrantz
eb7590a0d9 st/xorg: Make default of debugging fallbacks smart
Set it to the same as 2d acceleration
2010-02-18 16:13:18 +01:00
Jakob Bornecrantz
86870a691c drm/vmware: Latest interface changes are backwards incopatible
Since the execbuffer change actually changed size off the ioctl
struct and not just a reuse of padded bits, we can't support
old kernels as easily as the scanout change was.
2010-02-18 16:13:18 +01:00
Brian Paul
7c34c237a2 swrast: fix left side clipping
Fixes bug 26623.  Original patch was submitted by Mathias Frohlich
and modified by Brian.
2010-02-18 09:37:58 -07:00
José Fonseca
13cd4298fb mesa: Take back some assertions.
It seems there are still some places where draw can happen with mapped
buffers... Remove the assertions since there is not much more than can
be done at this moment.

This partially reverts commit af2023e31c.
2010-02-17 20:02:32 +00:00
José Fonseca
3b724f91c5 svga: Remove unused buffer download code.
Unnecessary now that we never destroy buffer storage.
2010-02-16 21:24:38 +00:00
José Fonseca
2077f375c7 pipebuffer: Don't synchronize when checking for buffer overflows.
To avoid masking synchronization issues in debug builds.
2010-02-16 21:00:15 +00:00
Michal Krol
055265b0a3 svga: Translate point_sprite rasterizer state. 2010-02-16 19:53:13 +01:00
Mike Stroyan
3094adb3ca mesa: Lock mutex around _mesa_HashLookup linked list chase.
Remove const qualifier from _mesa_HashLookup() table parameter to
avoid LOCK/UNLOCK warnings in the function body.

Signed-off-by: Brian Paul <brianp@vmware.com>
2010-02-16 08:18:12 -07:00
Mike Stroyan
7c7247ddbf mesa: Test for failed malloc in _mesa_HashInsert.
Signed-off-by: Brian Paul <brianp@vmware.com>
2010-02-16 08:17:06 -07:00
José Fonseca
af2023e31c vbo: Ensure vertices are always unmapped before the context is flushed.
Purpose is two fold:

- when doing user-space memory management mapping a buffer stored in the
  graphics aperture effectively pins it there, increasing the likelyhood
  of failure of validating other buffers when flushing

- certain hardware/platform combinations do not allow a buffer to be
  simultaneously mapped and validated

This fixes assertion failures in HL Uplink with the svga driver, where
vbo was holding a map to a buffer which was also referred in the command
stream.

Note: this a non-invasive fix and shouldn't be just cherry-picked into
master as-is -- a cleaner fix for this problem should be searched.
2010-02-16 13:24:01 +00:00
Michal Krol
2eedbc94c2 slang: Fix handling of if/elif/else cases.
Once if/elif evalutes to true, all subsequent conditions are always false.
2010-02-16 10:34:07 +01:00
José Fonseca
672f6bb545 wgl: Be lenient when sharing contexts.
My first reading of MS docs was wrong. It says:

  All rendering contexts of a shared display list must use an identical
  pixel format. Otherwise the results depend on the implementation of
  OpenGL used.

That is, it is OK to share contexts with different pixel formats.

Adobe Premiere Pro tries to do that: share lists between a rgbx8 and a
rgba8 pixel format.
2010-02-15 19:34:38 +00:00
Brian Paul
ab6bcef99a docs: document glMultiDrawElements() fix 2010-02-15 08:56:39 -07:00
Brian Paul
c5a4cfb03f vbo: fix broken glMultiDrawElements()
Seems to be a regression from commit 60b08eb1fd.
2010-02-15 08:52:48 -07:00
Thomas Hellstrom
df944efdbf st/xorg, vmware/xorg: Fix xnfcalloc arguments.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-02-15 14:26:25 +01:00
Brian Paul
1ff9cd5079 softpipe: fix broken cubemap / mipmap selection code
This is a quick-fix for the time being...

The per-face mipmap LOD computation was invalid at cube edges.  In
mip_filter_nearest/linear() we were trying to compute LOD using
texcoords that were sometimes indexes into different cube faces.
The subtraction used to compute the partial derivatives basically
gave random values, so the LOD was unpredictable.  This fix simply
uses the same cube face for all four pixels in the quad.  The per-
face texcoords all reference the same cube face so the partial
deriviates are computed properly.

A more elaborate fix would involve computing the LOD at the same
time as we choose the cube faces.  But for now, this solution works
well and allows the piglit/cubemap test to pass.
2010-02-13 17:45:05 -07:00
José Fonseca
4a2b54cbdb mesa: Export GL_EXT_texture_cube_map.
Still used by some applications.
2010-02-13 15:10:57 +00:00
José Fonseca
6bd6a15ab3 wgl: Implement wglSwapMultipleBuffers. 2010-02-13 15:10:57 +00:00
José Fonseca
c1a5c9bb4c svga: Fix texture border color.
Also opposite ordering.
2010-02-13 15:10:57 +00:00
Brian Paul
b3c7dc6ff2 st/mesa: restore draw rasterization stage after rasterpos
This fixes invalid calls to rastpos_point/line/tri() that can occur
when glRasterPos() is called while in feedback or selection mode.
2010-02-12 17:18:21 -07:00
José Fonseca
a1025ec041 svga: A few more tweaks to blend color state emission. 2010-02-12 23:17:04 +00:00
José Fonseca
3a4068474c svga: Clamp max cube texture levels to supported max 2d levels. 2010-02-12 21:31:18 +00:00
José Fonseca
1ae976be4a svga: Report the 2d/3d texture sizes supported by the host. 2010-02-12 21:31:14 +00:00
Roland Scheidegger
b685927156 svga: emit blend color 2010-02-12 22:12:55 +01:00
Zack Rusin
df0c8d029d st/dri: don't expose visuals we'll have trouble supporting
without getBuffersWithFormat some visuals will just cause headaches (crashes),
so if we're running on an older system simply don't advertise them.
2010-02-12 15:49:03 -05:00
Zack Rusin
3477dc4c48 st/dri, st/xorg: fix buffers that have attachements of different depth
we actually need to specify the formats for different attachements, otherwise
if the color buffer is 24bpp and the app asks for 16bpp depth buffer than
we end up fetching the depth from the drawable which is 24bpp and end up
creating the wrong depth buffer. use the new getBuffersWithFormat extension
to pass the depth correctly.
2010-02-12 15:48:52 -05:00
Andre Maasikas
0426bccadd r600: be more exact in vb size calculation
to make kernel cs checker happier, last attribs don't need full stride of
space. Calculate as count-1*stride + size of attrib
2010-02-12 10:21:44 -05:00
Andre Maasikas
841333cd21 r600: support GL_SHORT attributes
normalized seems to work with this setting.
2010-02-12 10:21:23 -05:00
José Fonseca
e541dceb67 wgl: Go into dormant state when DLL is unloaded unclealy.
When our DLL is unloaded, even if we leave the data structures in memory
for sake of future calls, the MS CRT will destroy the heap. Instead we
make all calls no-ops by setting stw_dev to NULL.
2010-02-12 13:59:16 +00:00
José Fonseca
fea7a70a1a scons: Target Windows7. 2010-02-12 13:22:59 +00:00
José Fonseca
426f607aaa wgl: Do not reach out and destroy contexts on cleanup.
Simply skip cleanup when contexts are still active.

This addresses two issues:

- in some situations the ICD DLL may be unloaded before the DLL that is
  using GL contexts is, so we may receive GL calls after stw_cleanup.

- when aborting (exception, or control-c) the contexts may have been left
  in an inconsistent state and attempting to destroy can cause
  unpredictable results.
2010-02-12 11:03:33 +00:00
Brian Paul
408f32dc16 svga: check min_index, max_index before assertions
It's possible for min_index and max_index to be ~0 if the min/max
values were not actually set or computed in the state tracker.
Skip some assertions in that case.

This only effects the debug build error checking.
2010-02-11 16:47:34 -07:00
José Fonseca
94028edfc5 gdi: Never fail at DLL load time.
Windows doesn't really expect things to fail at this point -- it
will try many times until it finally gives up, worse, something bad
happens.

The WGL state tracker will gracefully decline to do work even when it
is loaded.
2010-02-11 16:38:16 +00:00
José Fonseca
7fec5f88a5 wgl: Fail gracefully whenever 3D could not be enabled for some reason. 2010-02-11 16:38:13 +00:00
Michal Krol
67007670bb mesa: Enable true refcounting for NullBufferObj.
This object can be shared with another context, so we cannot just
delete it when the owning context is being destroyed.

Ensuring that buffer objects are properly refcounted guarantees
NullBufferObj is destroyed when all references to it are removed.
2010-02-10 14:12:11 +01:00
Michal Krol
e8865f199d mesa: Protect buffer objects reference counting with a mutex. 2010-02-10 14:12:10 +01:00
Zack Rusin
2d3262d47d st/xorg: fix a silly offset bug in xv
we were multiplying the difference in the wrong direction.
2010-02-09 21:15:22 -05:00
Thomas Hellstrom
3470d821ba vmware: Update vmwgfx_drm.h to kernel version.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-02-09 22:54:15 +01:00
Brian Paul
0371956d66 docs: fix typo: s/osdemo/osdemos/ 2010-02-09 11:21:15 -07:00
Jakob Bornecrantz
3bf13656d3 st/xorg: Include cursorstr.h 2010-02-06 16:58:32 +00:00
Brian Paul
fa4083d38b docs: updates to 7.7.1 release notes 2010-03-10 14:14:04 -07:00
Brian Paul
d311ded31d gallium/util: added surface dump code for unix 2010-03-10 10:39:46 -07:00
Brian Paul
34f0207161 st/mesa: fix incorrect glCopyPixels between window/FBO
There was a DrawBuffer/ReadBuffer typo and we were neglecting to invert
the texture coords when copying from a window to an FBO.

Plus, add some surface dump/debug code (disabled).
2010-03-10 10:39:45 -07:00
Brian Paul
8d3f629a13 st/mesa: fix incorrect glDrawPixels into FBO
We weren't inverting the textured quad when drawing into an fbo.
2010-03-10 10:39:45 -07:00
Thomas Hellstrom
b98ef495d4 st/xorg Avoid advertizing rotation / reflection support.
We don't support it yet since we don't implement the shadow allocate
functions.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-03-06 15:52:31 +01:00
Michel Dänzer
21c91b410a st/xorg: Work around cursor reference counting bugs in older X servers.
Could result in use of freed memory and consequently random crashes, e.g. on
screen resize.
2010-03-10 13:16:00 +01:00
Thomas Hellstrom
a8f3b3f88a st/xorg, vmware: Make throttling configurable.
The xorg state tracker gets two new options to let the user choose
whether to enable / disable dirty throttling and swapbuffer throttling.
The default value of these options are enabled, unless the winsys
supplies a customizer with other values. The customizer record has been
extended to allow this, and also to set winsys-based throttling on a per-
context basis.

The vmware part of this patch disables the dirty throttling if the kernel
supports command submission throttling, and also in that case sets kernel
based throttling for everything but swapbuffers. The vmware winsys does not
set throttling per context, even if it theoretically could, but instead
sets throttling per screen. This should perhaps be changed, should the
xorg state tracker start to use multiple rendering contexts. Kernel throttling
is off by default for all new screens/contexts, so the dri state tracker
is not affected.

This significantly improves interactivity of the vmware xorg driver.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-03-06 11:03:55 +01:00
Thomas Hellstrom
e8a8c5e339 svga: Add a winsys callback to get the svga_winsys_context
The winsys may need to extract the svga_winsys_context from a
pipe_context. Add a function to enable that functionality.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-03-06 11:03:54 +01:00
Keith Whitwell
7941d31ee6 svga: Don't do culling while the software pipeline is active.
It does it for us, and additionally introduces potentially
back-facing triangles.
2010-03-09 19:19:56 +01:00
Brian Paul
842351dd76 softpipe: zero-out entire clear_flags array in sp_tile_cache_flush_clear()
Before, we only cleared the flags for the active tiles (the ones inside
the framebuffer bound).  The problem is if we later bound a different,
larger surface to the tile cache we'd have some stale clear-flags still
set (and mistakenly clear some tiles in the new surface).

Fixes fd.o bug 26932.

(cherry picked from commit e1762fb870)
2010-03-08 11:40:40 -07:00
Brian Paul
d74929702f st/mesa: don't detach renderbuffer, surface in st_finish_render_texture()
There's no reason to release the renderbuffer from the framebuffer object
or release the gallium surface in this function (they're reference counted).
In fact, we don't want to do this because we may later use the texture as a
pixel source (ex: glBlitFramebuffer) and need the surface.

Fixes fd.o bug 26923 and is part of the fix for bug 26932.

(cherry picked from commit 80dc54e308)
2010-03-08 11:40:25 -07:00
Brian Paul
1e431f0454 mesa: s/GL_DEPTH_STENCIL/GL_DEPTH_COMPONENT/ for MESA_FORMAT_Z16 renderbuffer
MESA_FORMAT_Z16 has no stencil bits.

(cherry picked from commit 614f490ca9)
2010-03-08 10:00:17 -07:00
Alex Deucher
4cc8d1d79f r600: recalculate point size, if point min/max size changes 2010-03-08 11:31:06 -05:00
Alex Deucher
3d2bc6848a r600: no need to flush on context init 2010-03-08 11:30:54 -05:00
Maciej Cencora
d5327538e7 r300: recalculate point size, if point min/max size changes
Fixes two wine d3d9 unit tests
2010-03-08 11:29:12 -05:00
Maciej Cencora
86ac140937 r300: no need to flush on context init 2010-03-08 11:28:57 -05:00
Maciej Cencora
b584e780ab r300: remove unnecessary code
_tnl_UpdateFixedFunctionProgram is already called in r300_draw.c
2010-03-08 11:28:39 -05:00
Alex Deucher
981e8a2087 r600: add new rs880 pci id 2010-03-03 14:11:10 -05:00
Brian Paul
72d380b363 glx: replace assertion with conditional
See fd.o bug 26832.
2010-03-02 07:38:12 -07:00
Dave Airlie
cf8af9bcf1 r600: fix blender picking.
This fixes the sw fallback for GL_SELECT picking modes.

Fixes object picking blender + depthpick test

http://bugs.freedesktop.org/show_bug.cgi?id=26419

Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-03-02 14:14:47 +10:00
Alex Deucher
7123f3d77a r600: enable OQ on rv740 on drms with working pipe config 2010-02-26 15:39:40 -05:00
Klaus Schnass
2edb1b9534 r600: fail to validate unsupported texture formats 2010-02-26 15:39:17 -05:00
Markus Fleschutz
69334d6784 glx: fix incorrect array stack memory allocation
The array stack space wasn't allocated to the proper size.  Fixes out of
bounds memory writes when the client/array stack depth exceeds one.

See fd.o bug 26768.
2010-02-26 10:38:21 -07:00
Brian Paul
61482ddc1c mesa: fix _BaseFormat assignment in _mesa_soft_renderbuffer_storage()
The rb->InternalFormat field will be set by the caller if the allocation
succeeds.  Until then, this field's value can't be used.  Fixes a failed
assertion with FlightGear.

(cherry picked from commit fe25476c04)
2010-02-26 09:14:12 -07:00
Brian Paul
b0e84e22d5 mesa: use simplified _BaseFormat value in render-to-texture code
Fixes fd.o bug 26762.

(cherry picked from commit c9e8ff1976)
2010-02-26 09:10:27 -07:00
Jakob Bornecrantz
c0e8d443fe st/xorg: Re-enable crtc on resize 2010-02-25 16:52:02 +01:00
Jakob Bornecrantz
b95d4cd680 st/xorg: Fix copy-topy 2010-02-25 16:52:02 +01:00
Jakob Bornecrantz
45c4addea7 vmware/core: Support drm interface version 1.0.0 2010-02-06 01:30:29 +01:00
Jakob Bornecrantz
1eba0eb37d vmware/core: Use correct flags in scanout code 2010-02-06 01:29:47 +01:00
Dave Airlie
d87fb5e003 radeon: fix UMS since stp addition.
UMS doesn't need the STP atom.

Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-02-08 12:07:04 +02:00
Pauli Nieminen
4be7922a8e r200: Fix UMS notto emit stp.
Polgon stipples are handled by kernel in UMS. Mark the
state as never to be emited for UMS.
2010-02-08 12:07:04 +02:00
Eric Anholt
a1cac0732b intel: Check aperture size when doing a blit glClear.
Fixes failure in cairo-gl firefox-planet-gnome.
(cherry picked from commit 67e5fd7397)

Conflicts:

	src/mesa/drivers/dri/intel/intel_blit.c
2010-02-06 17:45:34 +01:00
Eric Anholt
fb32e0fcc5 intel: Respect texture tiling when doing a PBO blit teximage upload.
Bug #26008.  Fixes piglit pbo-teximage-tiling-2.
(cherry picked from commit 22bbc97994)
2010-02-06 17:45:34 +01:00
Eric Anholt
9564a6fa13 intel: Set the region's tiling to none when attaching a PBO to a region.
Note that when detaching the PBO from the region and making a new BO
for the region, we don't make it tiled even if the region originally
was.

Fixes piglit pbo-teximage-tiling.
(cherry picked from commit f20e83210e)
2010-02-06 17:45:34 +01:00
Eric Anholt
0b9990b2fd intel: Fix PBO blit ReadPixels from an FBO.
Bug #25921 -- clutter PBO usage gave unreliable results.
(cherry picked from commit 63b10e8fe7)

Conflicts:

	src/mesa/drivers/dri/intel/intel_pixel_read.c
2010-02-06 17:16:51 +01:00
Eric Anholt
8ce99c85e7 i965: Fix loads of non-relative-addr constants after a reladdr load.
Fixes piglit vp-arl-constant-array-huge-overwritten.
(cherry picked from commit d5b964234c)
2010-02-06 17:08:53 +01:00
Jakob Bornecrantz
a2c402ba53 vmware/core: Handle new scanout field 2010-02-06 01:34:12 +01:00
Jakob Bornecrantz
1ceb906c12 vmware/core: Update header to latest version 2010-02-06 01:34:12 +01:00
Zack Rusin
fb1fe8e76f gallium/tgsi: fix soa dependencies in mul and cmp instructions
master has this done in a lot nicer way, this is a simple patch to
stop the bleeding in a minimally intrusive manner.
2010-02-05 19:32:01 -05:00
Zack Rusin
f8b05566aa gallium/python: support two more formats in retrace dumps
dump PIPE_FORMAT_R8G8B8A8_UNORM and PIPE_FORMAT_R16G16B16_SNORM formats.
2010-02-05 19:25:01 -05:00
Zack Rusin
a6148b8eba gallium/draw: fix overflowing element indices
we piggy back flags in the top four bits of the element indices, so if
the maximum index stored in any of the elements is greater than
sizeof(ushort) - sizeof(flags we piggy back) then we overflow. fix it by
simply falling back to the slow path if we notice the overflow.
2010-02-05 19:25:01 -05:00
Brian Paul
64be837b0b st/mesa: allow negative index for PROGRAM_STATE_VAR
The piglit vp-address-01 test uses negative address-relative
offsets.  In this test we're indexing into PROGRAM_STATE_VAR
which, in turn, contains references to ENV vars.

We previously fixed this issue for PROGRAM_CONSTANT.

piglit/vp-address-01 (the version from Feb 5) passes now.
2010-02-05 14:54:00 -07:00
Brian Paul
e070c1d183 tgsi: disable inf/nan checking
Inf/NaN can happen normally.  Only check for those values
when we need to debug something.
2010-02-05 14:48:42 -07:00
Tom Fogal
e6ee4b49c4 docs: Fix a typo in mesa3d.org HTML.
Signed-off-by: Brian Paul <brianp@vmware.com>
2010-02-05 13:17:37 -07:00
Tom Fogal
e65029e9b3 osmesa: Add OSMesaColorClamp and OSMesaGetProcAddress to symbol defs.
Without this patch, the two symbols get an underscore prepended
and an "@4" appended when compiling with VC8.

Signed-off-by: Brian Paul <brianp@vmware.com>
2010-02-05 13:15:33 -07:00
Brian Paul
3ca6cb3440 softpipe: fix bug in perspective corrected interpolants for lines
Thanks to Michal for finding this.

This will get fixed on the gallium-cylindrical-wrap branch and
then merged to master.
2010-02-05 12:57:54 -07:00
Pauli Nieminen
34b36277b7 r100: Add the polygon stipple state to the state list.
This fixes glean paths test case.
2010-02-05 11:28:01 +02:00
Pauli Nieminen
c8ea0212fe r200: Add the polygon stipple state to the state list.
This fixes glean paths test case.
2010-02-04 23:56:37 +02:00
Pauli Nieminen
7d6cbcdd9a swrast/sse: Fix _mesa_sse_transform_pointsX_3d_no_rot
PXOR user in code were causing the lowest SP float register to have NaN
values which made all math operations in that slot fail. Correct istruction
to clear float registers is XORPS which handles single precission floats
correctly.

Fixes progs/tests/fog in swrast SSE mode.

Now the correct commit instead of 9d9c1f17dc which is not even close
of correct fix for the bug.
2010-02-04 13:07:04 +02:00
Pauli Nieminen
9d9c1f17dc Revert "mesa: fix transform_points_3d_no_rot using undefined values in %xmm0"
This reverts commit 4c31632817.
2010-02-04 05:00:02 +02:00
Brian Paul
f1afb352da scons: add another prefix for mingw32 on Fedora 2010-02-03 11:46:36 -07:00
Brian Paul
e4c3abbf55 tgsi: fix XPD when dst reg is a src reg
Store XPD results in temporaries before writing to the dest register.
This issue has already been fixed in Mesa/master for XPD (and all
other instructions?).

Fixes progs/demos/bump.c when using TGSI interpreter (GALLIUM_NOSSE=1).

A new glean/glsl1 test has been added to test this case.
2010-02-03 08:17:37 -07:00
Keith Whitwell
141b5775c0 svga: fix TXD and TXL opcode translation 2010-02-02 18:01:52 +00:00
Keith Whitwell
c0d5f1d3ad svga: deriv insns not valid in dynamic flow control either 2010-02-02 17:45:32 +00:00
Keith Whitwell
70947e531e svga: texture from lod zero inside dynamic branching
Texture derivatives are potentially undefined inside dynamic branches,
so hardwire lod zero in this case.  Treating all if/endif and loop
constructs as dynamic branches.
2010-02-02 15:34:01 +00:00
Pauli Nieminen
7b92cb45b2 r200: Fix EXT_fogcoord rendering.
The fogcoord calue was not pushed to GPU because of implicit float to int conversion.
Fix is to use float pointer to buffer object so no conversion is done in assigment.
2010-02-02 06:09:36 +02:00
Brian Paul
43e4b58422 st/mesa: fix texture deallocation bug
This fixes a bug reported by Christoph Bumiller on mesa3d-dev.

When a texture is first created as RGBA, then re-defined with
glTexImage(internalFormat=GL_DEPTH_COMPONENT) we failed to deallocate
the original texture.  When this texture was bound as a FBO surface,
the depth/Z surface format was RGBA instead of Z.  Depending on the
driver this led to a failed assertion or FBO validation failure.

This patch does three things:

1. Remove ancient code that mysteriously tested if we were replacing
the smallest mipmap level and tested if the texture was not a cube map
texture.  I can't see any reason for those tests.

2. Move the width=height=depth=0 test to after the code which frees
texture data.  Calling glTexImage with width=height=depth=0 and data=NULL
is a way to free a single mipmap level.

3. Update the code comments.

There are no apparent conform, glean or piglit regressions from this change.
2010-02-01 18:55:54 -07:00
Brian Paul
e0d01c9d7f mesa: change _mesa_find_free_register() to find multiple free regs
Before, _mesa_find_free_register() would scan the given shader to
find a free/unused register of the given type.  But subsequent calls
would return the same register again.  This caused a failure in the
_mesa_remove_output_reads() function which sometimes needs several
free temps.

Now use a new function which build a vector of 'used' flags and another
function which searches that vector for an unused register starting at
a position that's incremented for each call.

Fixes fd.o bug 26317.  Note that a regression test for this has been
added to the glean/glsl1 test.
2010-02-01 17:31:24 -07:00
Brice Goglin
eef7b39a36 Prepare changelog for upload 2010-02-01 22:55:49 +01:00
José Fonseca
b90b3667a1 mesa: Ensure object refcount is null when destroying the buffer.
Lets see if this is not too pedantic. Obj pointers are never exposed to
GL apps so it should be possible to get this right.

Furthermore apps with GL widgets and test suits create and destroy many
contexts and objects, so bad reference counting is not really an option.
2010-02-01 21:41:04 +00:00
José Fonseca
0123a2d042 mesa: Fix null buffer object reference counting.
Always use _mesa_reference_buffer_object, and never call
ctx->Driver.DeleteBuffer() directly to prevent dangling pointers to the
null buffer object.

This fixes crash/assertions in sharedtex_mt and Autodesk Mudbox.
2010-02-01 21:41:04 +00:00
José Fonseca
51a2cc5499 mesa: Always do proper ref counting of shared state. 2010-02-01 21:41:04 +00:00
Brice Goglin
04f4a8921b New upstream snapshot 2010-02-01 22:38:35 +01:00
Brice Goglin
7078992fea Merge branch 'upstream-experimental' into debian-experimental 2010-02-01 22:37:11 +01:00
Jakob Bornecrantz
f5145a6ec3 vmware/core: Remove old ioctl structs not used
Leftover that was never used in this driver
2010-01-29 17:14:53 +01:00
José Fonseca
ddedfe12d4 svga: Actually call fence_reference in texture downloads. 2010-01-29 16:23:46 +00:00
Brian Paul
cb5447f79c glsl: fix preprocessor bug involving //-style comments in macro definitions
See comments for details.
2010-01-29 08:26:23 -07:00
Brian Paul
9fd3c74724 tnl: check that state is validated before drawing 2010-01-29 08:24:36 -07:00
Brian Paul
4d1234e222 st/mesa: check that state is validated before drawing 2010-01-29 08:24:36 -07:00
Brian Paul
3cba779e16 vbo: fix missing state validation bugs
Commit 2708ddfb06 caused a few regressions.
We need to check/validate state after calling bind_arrays() because
it might set the _NEW_ARRAYS flag if the varying VP inputs change.

The symptom of this problem was some attribute arrays being ignored
(or interpreted as constant-valued) in glDrawRangeElements or
glMultiDrawElements.

A follow-on patch will add some additional asserts to try to catch
this kind of thing in the future.
2010-01-29 08:24:36 -07:00
Brian Paul
23eda89ec8 mesa: do state validation in _mesa_valid_to_render()
...rather than checking/validating before all the calls to
_mesa_valid_to_render() and valid_to_render().

The next patch will actually fix some bugs...
2010-01-29 08:24:36 -07:00
José Fonseca
6e68898b05 softpipe: Flush draw module when fragment pipeline state changes. 2010-01-28 20:22:44 +00:00
José Fonseca
1befcd5a2a softpipe: Flush draw module before switching framebuffer.
Otherwise geometry might end up in the wrong rendertarget.
2010-01-28 20:22:44 +00:00
José Fonseca
12ba355978 llvmpipe: Use assert instead of abort. Only verify functions on debug builds. 2010-01-28 20:22:43 +00:00
José Fonseca
a0907a645f llvmpipe: Fix memory leak. 2010-01-28 20:22:43 +00:00
José Fonseca
1acf7a09e7 llvmpipe: Use the generic conversion routine for depths.
This allows for z32f depth format to work correctly.
2010-01-28 20:22:43 +00:00
José Fonseca
14dc02a1b2 llvmpipe: Be more conservative with the supported formats.
We'll likely support much more formats, but doing this allows to run
more testsuites without immediately hit assertion failures.
2010-01-28 20:22:43 +00:00
José Fonseca
88cf87bd56 llvmpipe: Unmapping vertex/index buffers does NOT flush draw module anymore.
Not since 6094e79f4e.

Drivers now need to flush draw module explicitely (which explains why
all those previous commits adding draw_flushes calls were necessary).

This is a good thing, but it's tricky to get this right in face of user buffers
(it's not even clear who has the responsibility to flush when a user buffer
is seen -- statetracker or pipe driver), so just force flush (temporarily)
since it's not a bottleneck now.
2010-01-28 20:22:43 +00:00
José Fonseca
31b3420688 llvmpipe: Treat state changes systematically.
That is:
- check for no op
- update/flush draw module
- update bound state and mark it as dirty

In particular flushing the draw module is important since it may contain
unflushed primitives which would otherwise be draw with wrong state.
2010-01-28 20:22:43 +00:00
José Fonseca
4e95983fa8 llvmpipe: Flush draw module before switching framebuffer.
Otherwise geometry will end up in the wrong rendertarget.
2010-01-28 20:22:43 +00:00
José Fonseca
4e506eac8f llvmpipe: add LP_DEBUG env var
Cherry-picked from dec35d04ae.
2010-01-28 20:22:43 +00:00
Brian Paul
46167149ce llvmpipe: fix broken TGSI_OPCODE_FRC codegen 2010-01-28 20:22:42 +00:00
Brian Paul
96ec4eb755 llvmpipe: fix broken lp_build_abs() 2010-01-28 20:22:42 +00:00
José Fonseca
e20547042c llvmpipe: Fix yet another copynpaste typo in lp_build_log2_approx.
Now fslight looks perfect.
2010-01-28 20:22:42 +00:00
José Fonseca
0451d0fd01 llvmpipe: Fix lp_build_polynomial comment. 2010-01-28 20:22:42 +00:00
José Fonseca
aa8b23e077 llvmpipe: Fix typo in lp_build_log constant. 2010-01-28 20:22:42 +00:00
José Fonseca
4eb48a3af7 llvmpipe: Fix bad SI -> FP conversion into lp_build_log2_approx.
It should be a bitcast  as the integer value is actually an encoded FP
already.
2010-01-28 20:22:42 +00:00
Jakob Bornecrantz
8db8adfd01 st/xorg: Update to new libkms interface but be backwards compatible 2010-01-28 18:02:10 +01:00
Keith Whitwell
8e240d7e0e svga: remove sleep in DEBUG_FLUSH path 2010-01-28 17:13:50 +00:00
José Fonseca
40298bf272 python/retrace: Dump all vertices for verbosity level 3. 2010-01-28 17:13:49 +00:00
Brian Paul
a0518e66b2 docs: Fixed Windows build for 7.7.1 2010-01-27 17:08:03 -07:00
Brian Paul
82c76cd16f windows: assorted build fixes
These changes were originally submitted by Karl Schultz for Mesa 7.7-rc2
but weren't applied.
2010-01-27 17:04:49 -07:00
Rodolfo Ribeiro Gomes
50e890bc51 glx: fix possible memory leaks in dri2CreateScreen()
Original patch fixed up by Brian Paul.
2010-01-27 17:04:49 -07:00
Alex Deucher
bba9557019 r600: fix warning 2010-01-27 12:20:04 -05:00
Alex Deucher
2041d3e4b7 r600: rv670 support 8 tex instructions just like other r6xx
also clarify some other const values.
2010-01-27 12:19:54 -05:00
Andre Maasikas
77b7b3a1ab r600: increase max texture units to 16 2010-01-27 11:38:29 -05:00
Andre Maasikas
0dab80fbfb r600: fix XPD with writemask
same variable used for 2 different temp registers
fixes e.g. glsl/bump
2010-01-27 11:38:16 -05:00
José Fonseca
e3257912e0 mesa: Warn when indices are out of bounds, but do not skip the draw in debug builds.
The driver (or preferably the hardware) should handle out of bounds
indices.

If there are problems then it's better to detect those in the debug
builds.
2010-01-27 16:25:31 +00:00
José Fonseca
15fe491822 svga: Prevent buffer overflow in buffer ranges.
Do this by extending the nearest range to cover the new range.

This fixes an access fault in Call of Duty which was doing many disjoint
glBufferSubData calls.
2010-01-27 15:41:25 +00:00
José Fonseca
ac597f5acc svga: Fix PIPE_LOGICOP_INVERT.
Actually the current XOR implementation is an INVERT.

This fixes rectangle selection in Maya.
2010-01-27 14:45:56 +00:00
Keith Whitwell
5cb255f0d7 svga: better path for generating white pixels for xor logicop workaround 2010-01-27 14:26:48 +00:00
Brice Goglin
adcbe9af69 Prepare changelog for upload 2010-01-27 09:14:43 +01:00
Brice Goglin
527f2433fc New upstream snapshot 2010-01-27 09:14:16 +01:00
Brice Goglin
81f129113a Merge branch 'mesa_7_7_branch' of git://anongit.freedesktop.org/mesa/mesa into debian-experimental
Conflicts:
	progs/tests/getprocaddress.c
	progs/tests/vparray.c
	progs/trivial/tri-fbo-tex-mip.c
	progs/trivial/tri-fbo-tex.c
	scons/gallium.py
2010-01-27 09:12:04 +01:00
Brice Goglin
9abcffc1c6 Merge branch 'debian-unstable' into debian-experimental
Conflicts:
	debian/changelog
2010-01-27 09:09:36 +01:00
Dave Airlie
2f28ca0a27 radeon/r200/r300: don't clean non-emitted state.
So if we don't actually emit an atom to the hw because we don't
need it in the current state (e.g. lighting related atoms when
lighting is off) then don't mark it as clean, because when
lighting gets switched on we won't emit it at all.

This fixes funky gears colors.

Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-01-27 12:32:46 +10:00
Dave Airlie
0580e488da radeon: remove unused file 2010-01-27 12:32:44 +10:00
José Fonseca
5435f790fd python: Add missing library on debug builds. 2010-01-26 20:37:35 +00:00
Julien Cristau
5aee9e344b Remove myself from Uploaders 2010-01-23 01:46:41 +01:00
Julien Cristau
b1a8810204 Rename the build directory to not include DEB_BUILD_GNU_TYPE for no good reason. Thanks, Colin Watson! 2010-01-16 16:48:55 +00:00
Brice Goglin
2b5f0b8669 Prepare changelog for upload 2010-01-11 17:52:44 +01:00
Brice Goglin
5be25443dd Merge branch 'debian-unstable' of git+ssh://git.debian.org/git/pkg-xorg/lib/mesa into debian-experimental
Conflicts:
	ChangeLog
	debian/changelog
2010-01-11 17:52:13 +01:00
Brice Goglin
eb1cc8273b New upstream release 2010-01-11 17:49:26 +01:00
Brice Goglin
5144f580d0 Merge branch 'upstream-experimental' into debian-experimental
Conflicts:
	progs/egl/eglgears.c
	progs/egl/eglscreen.c
	progs/fp/Makefile
	progs/fp/SConscript
	progs/rbug/simple_server.c
	progs/tests/getprocaddress.c
	progs/tests/interleave.c
	progs/tests/texwrap.c
	progs/tests/vparray.c
	progs/vp/vp-tris.c
	src/glut/glx/win32_menu.c
2010-01-11 17:47:52 +01:00
Brice Goglin
4afc359cc3 Revert "Bump the libdrm build-dep to 2.4.17 because of the changed radeon API."
This reverts commit ed7cd35ba0.

Only libdrm 2.4.15 is uploaded in Debian for now.
2010-01-11 17:44:11 +01:00
Timo Aaltonen
ed7cd35ba0 Bump the libdrm build-dep to 2.4.17 because of the changed radeon API. 2010-01-09 02:24:49 +02:00
Julien Cristau
53d395336d Prepare changelog for upload 2009-12-29 10:42:29 +00:00
Julien Cristau
bc94d48259 Update patch stack
Drop hunk from 05_hurd-ftbfs.diff that was applied upstream.
Refresh other patches.
2009-12-29 10:40:05 +00:00
Julien Cristau
fabfd3eaca Bump linux-libc-dev build-dep to 2.6.31 for the r600 dri driver
fixes ftbfs on mips
2009-12-29 10:19:48 +00:00
Julien Cristau
089e2e656c Bump changelogs 2009-12-29 10:17:41 +00:00
Julien Cristau
7030b61d99 Merge branch 'mesa_7_6_branch' of git://anongit.freedesktop.org/mesa/mesa into debian-unstable
Conflicts:
	progs/tests/bug_texstore_i8.c
	progs/tests/crossbar.c
	progs/tests/cva.c
	progs/tests/fptest1.c
	progs/tests/invert.c
	progs/tests/packedpixels.c
	progs/tests/quads.c
	progs/tests/scissor-viewport.c
	progs/tests/scissor.c
	progs/tests/stencilwrap.c
	progs/tests/tex1d.c
	progs/tests/texcomp_image.h
	progs/tests/unfilledclip.c
	progs/tests/vpeval.c
	progs/trivial/clear-fbo-tex.c
	progs/trivial/createwin.c
	progs/trivial/dlist-begin-call-end.c
	progs/trivial/draw2arrays.c
	progs/trivial/drawarrays.c
	progs/trivial/tri-blend-max.c
	progs/trivial/tri-blend-min.c
	progs/trivial/tri-blend-revsub.c
	progs/trivial/tri-blend-sub.c
	progs/trivial/tri-blend.c
	progs/trivial/tri-fbo-tex.c
	progs/trivial/tri-fp.c
	progs/trivial/tri-logicop-none.c
	progs/trivial/tri-logicop-xor.c
	progs/trivial/tri-z.c
	progs/trivial/vbo-drawarrays.c
	progs/trivial/vbo-noninterleaved.c
	scons/gallium.py
	scons/generic.py
2009-12-29 10:16:21 +00:00
Julien Cristau
36461f0d66 Delete some files not present in tarballs 2009-12-25 19:48:46 +01:00
Julien Cristau
5e20a66515 Bump changelogs 2009-12-25 17:06:27 +01:00
Julien Cristau
12251b4915 Merge remote branch 'upstream/mesa_7_7_branch' into debian-experimental
Conflicts:
	progs/tests/bug_texstore_i8.c
	progs/tests/crossbar.c
	progs/tests/cva.c
	progs/tests/fptest1.c
	progs/tests/invert.c
	progs/tests/packedpixels.c
	progs/tests/quads.c
	progs/tests/scissor-viewport.c
	progs/tests/scissor.c
	progs/tests/stencilwrap.c
	progs/tests/tex1d.c
	progs/tests/unfilledclip.c
	progs/tests/vpeval.c
	progs/trivial/clear-fbo-tex.c
	progs/trivial/createwin.c
	progs/trivial/dlist-begin-call-end.c
	progs/trivial/draw2arrays.c
	progs/trivial/drawarrays.c
	progs/trivial/tri-blend-max.c
	progs/trivial/tri-blend-min.c
	progs/trivial/tri-blend-revsub.c
	progs/trivial/tri-blend-sub.c
	progs/trivial/tri-blend.c
	progs/trivial/tri-fbo-tex.c
	progs/trivial/tri-fp.c
	progs/trivial/tri-logicop-none.c
	progs/trivial/tri-logicop-xor.c
	progs/trivial/tri-z.c
	progs/trivial/vbo-drawarrays.c
	progs/trivial/vbo-noninterleaved.c
	src/glut/glx/glut_menu.c
2009-12-25 16:25:56 +01:00
Julien Cristau
addd04c255 Don't include GLUT sources since we don't use them. 2009-12-25 16:19:26 +01:00
Julien Cristau
2596bd88d7 Add freedesktop.org ftp to watch file since that's where newer upstream tarballs are. 2009-12-25 16:13:38 +01:00
Brice Goglin
35994294ea Bump libdrm build dependency to 2.4.15 2009-12-14 09:20:38 +01:00
Brice Goglin
bb1193ca83 Prepare Changelog for upload 2009-12-12 13:03:07 +01:00
Brice Goglin
14cc33508c Do not build s3v and trident anymore 2009-12-12 13:02:44 +01:00
Brice Goglin
890a0f155f New upstream release candidate 2009-12-12 12:35:13 +01:00
Brice Goglin
6640b46db9 Merge branch 'upstream-experimental' into debian-experimental
Conflicts:
	doxygen/gallium.doc
	progs/egl/demo1.c
	progs/egl/demo2.c
	progs/egl/demo3.c
	progs/egl/eglgears.c
	progs/egl/eglinfo.c
	progs/slang/cltest.c
	progs/slang/vstest.c
	progs/tests/arbgpuprog.c
	progs/tests/getprocaddress.c
	progs/tests/getteximage.c
	progs/tests/mipmap_limits.c
	progs/tests/packedpixels.c
	progs/tests/sharedtex.c
	progs/tests/texcmp.c
	progs/tests/texcomp_image.h
	progs/tests/texcompsub.c
	progs/tests/zreaddraw.c
	scons/crossmingw.py
	scons/dxsdk.py
	scons/gallium.py
	scons/generic.py
	scons/llvm.py
	scons/winddk.py
	src/gallium/state_trackers/xorg/xvmc/tests/.gitignore
	src/gallium/state_trackers/xorg/xvmc/tests/test_context.c
	src/gallium/state_trackers/xorg/xvmc/tests/test_rendering.c
	src/gallium/state_trackers/xorg/xvmc/tests/testlib.c
	src/gallium/state_trackers/xorg/xvmc/tests/xvmc_bench.c
	src/mesa/drivers/windows/gldirect/mesasw/gld_wgl_mesasw.c
	windows/VC7/mesa/mesa/mesa.vcproj
2009-12-12 12:31:27 +01:00
Brice Goglin
7f8952d242 Prepare Changelog for upload 2009-12-11 18:36:40 +01:00
Brice Goglin
adae195cc1 Drop 07_sparc64_xform_asm.patch, applied upstream 2009-12-11 18:36:25 +01:00
Brice Goglin
3fa885911b New upstream release candidate 2009-12-11 18:33:34 +01:00
Brice Goglin
ebf57c6202 Merge branch 'mesa_7_6_branch' of git://anongit.freedesktop.org/mesa/mesa into debian-unstable
Conflicts:
	progs/tests/Makefile
	progs/tests/prog_parameter.c
2009-12-11 18:31:38 +01:00
Brice Goglin
1f47135544 Build r600_dri.so 2009-12-11 15:54:32 +01:00
Brice Goglin
ac180fa6a7 fix sparc64 FTBFS 2009-12-11 07:52:39 +01:00
Brice Goglin
a7109f3ae5 Prepare Changelog for upload 2009-12-06 00:15:21 +01:00
Brice Goglin
e5e54a9a6e Refresh patches 2009-12-06 00:15:21 +01:00
Brice Goglin
c2ad82a90b New upstream release candidate 2009-12-06 00:15:07 +01:00
Brice Goglin
9347983556 Merge branch 'mesa_7_6_branch' of git://anongit.freedesktop.org/mesa/mesa into debian-unstable
Conflicts:
	progs/fp/Makefile
	progs/fp/fp-tri.c
	progs/fp/point-position.c
	progs/fp/tri-depth.c
	progs/fp/tri-depth2.c
	progs/fp/tri-depthwrite.c
	progs/fp/tri-depthwrite2.c
	progs/fp/tri-inv.c
	progs/fp/tri-param.c
	progs/fp/tri-tex.c
	progs/tests/Makefile
	progs/tests/SConscript
	progs/tests/arraytexture.c
	progs/tests/copypixrate.c
	progs/tests/crossbar.c
	progs/tests/drawbuffers.c
	progs/tests/invert.c
	progs/tests/minmag.c
	progs/tests/mipmap_comp.c
	progs/tests/prog_parameter.c
	progs/tests/random.c
	progs/tests/shader_api.c
	progs/tests/stencil_twoside.c
	progs/tests/texcmp.c
	progs/tests/texdown.c
	progs/tests/vao-01.c
	progs/tests/vao-02.c
	progs/trivial/clear-fbo-tex.c
	progs/trivial/clear-fbo.c
	progs/trivial/clear-random.c
	progs/trivial/clear-scissor.c
	progs/trivial/clear.c
	progs/trivial/createwin.c
	progs/trivial/dlist-begin-call-end.c
	progs/trivial/dlist-dangling.c
	progs/trivial/dlist-edgeflag-dangling.c
	progs/trivial/dlist-edgeflag.c
	progs/trivial/dlist-flat-tri.c
	progs/trivial/dlist-mat-tri.c
	progs/trivial/dlist-recursive-call.c
	progs/trivial/dlist-tri-flat-tri.c
	progs/trivial/dlist-tri-mat-tri.c
	progs/trivial/line-clip.c
	progs/trivial/line-cull.c
	progs/trivial/line-flat.c
	progs/trivial/line-stipple-wide.c
	progs/trivial/line-userclip-clip.c
	progs/trivial/line-userclip-nop-clip.c
	progs/trivial/line-userclip-nop.c
	progs/trivial/line-userclip.c
	progs/trivial/line-wide.c
	progs/trivial/line.c
	progs/trivial/lineloop-clip.c
	progs/trivial/lineloop.c
	progs/trivial/linestrip-clip.c
	progs/trivial/linestrip-flat-stipple.c
	progs/trivial/linestrip-stipple-wide.c
	progs/trivial/linestrip-stipple.c
	progs/trivial/linestrip.c
	progs/trivial/point-clip.c
	progs/trivial/point-param.c
	progs/trivial/point-sprite.c
	progs/trivial/point-wide-smooth.c
	progs/trivial/point-wide.c
	progs/trivial/point.c
	progs/trivial/poly-flat-clip.c
	progs/trivial/poly-flat-unfilled-clip.c
	progs/trivial/poly-flat.c
	progs/trivial/poly-unfilled.c
	progs/trivial/poly.c
	progs/trivial/quad-clip-all-vertices.c
	progs/trivial/quad-clip.c
	progs/trivial/quad-degenerate.c
	progs/trivial/quad-flat.c
	progs/trivial/quad-offset-factor.c
	progs/trivial/quad-offset-unfilled.c
	progs/trivial/quad-offset-units.c
	progs/trivial/quad-tex-alpha.c
	progs/trivial/quad-tex-pbo.c
	progs/trivial/quad-unfilled-clip.c
	progs/trivial/quad-unfilled-stipple.c
	progs/trivial/quad-unfilled.c
	progs/trivial/quad.c
	progs/trivial/quads.c
	progs/trivial/quadstrip-cont.c
	progs/trivial/quadstrip-flat.c
	progs/trivial/quadstrip.c
	progs/trivial/readpixels.c
	progs/trivial/tri-alpha-tex.c
	progs/trivial/tri-alpha.c
	progs/trivial/tri-blend-color.c
	progs/trivial/tri-clear.c
	progs/trivial/tri-clip.c
	progs/trivial/tri-cull-both.c
	progs/trivial/tri-dlist.c
	progs/trivial/tri-fbo.c
	progs/trivial/tri-flat-clip.c
	progs/trivial/tri-flat.c
	progs/trivial/tri-fog.c
	progs/trivial/tri-fp-const-imm.c
	progs/trivial/tri-fp.c
	progs/trivial/tri-lit-material.c
	progs/trivial/tri-lit.c
	progs/trivial/tri-multitex-vbo.c
	progs/trivial/tri-orig.c
	progs/trivial/tri-query.c
	progs/trivial/tri-scissor-tri.c
	progs/trivial/tri-square.c
	progs/trivial/tri-stipple.c
	progs/trivial/tri-tex-3d.c
	progs/trivial/tri-tex.c
	progs/trivial/tri-tri.c
	progs/trivial/tri-unfilled-clip.c
	progs/trivial/tri-unfilled-edgeflag.c
	progs/trivial/tri-unfilled-point.c
	progs/trivial/tri-unfilled-smooth.c
	progs/trivial/tri-unfilled-tri-lit.c
	progs/trivial/tri-unfilled-tri.c
	progs/trivial/tri-unfilled-userclip-stip.c
	progs/trivial/tri-unfilled-userclip.c
	progs/trivial/tri-unfilled.c
	progs/trivial/tri-userclip.c
	progs/trivial/tri-z-9.c
	progs/trivial/tri-z-eq.c
	progs/trivial/trifan-flat-clip.c
	progs/trivial/trifan-flat-unfilled-clip.c
	progs/trivial/trifan-flat.c
	progs/trivial/trifan-unfilled.c
	progs/trivial/trifan.c
	progs/trivial/tristrip-clip.c
	progs/trivial/tristrip-flat.c
	progs/trivial/tristrip.c
	progs/trivial/vp-tri-cb-pos.c
	progs/trivial/vp-tri-cb-tex.c
	progs/trivial/vp-tri-invariant.c
	progs/vp/vp-tris.c
	progs/vpglsl/psiz-mul.glsl
	progs/vpglsl/vp-tris.c
2009-12-05 23:36:06 +01:00
Brice Goglin
ee9811de78 Prepare changelog for upload 2009-09-29 11:52:04 +02:00
Brice Goglin
adaeac03dd Add bug closers 2009-09-29 11:51:51 +02:00
Brice Goglin
e47955de4e Refresh patches 2009-09-29 11:45:58 +02:00
Brice Goglin
58c42a5b1d New upstream release 2009-09-29 11:43:15 +02:00
Brice Goglin
753604e607 Merge tag 'mesa_7_6' into debian-unstable
Conflicts:
	progs/egl/Makefile
	progs/egl/demo3.c
	progs/slang/vstest.c
	progs/tests/Makefile
	progs/tests/SConscript
	progs/tests/bufferobj.c
	progs/tests/floattex.c
	progs/tests/getprocaddress.c
	progs/tests/getprocaddress.py
	progs/tests/mapbufrange.c
	progs/tests/mapvbo.c
	progs/tests/texcmp.c
	progs/tests/texcompsub.c
	progs/tests/zreaddraw.c
	progs/trivial/.gitignore
	progs/vp/arl-static.txt
	progs/vp/arl-unused.txt
	progs/vp/vp-tris.c
	scons/gallium.py
2009-09-29 11:39:19 +02:00
Julien Cristau
f8f07cb999 Remove unused INSTALL_TARGETS variable
The install targets were removed in the previous commit.
2009-09-04 17:29:21 +02:00
Julien Cristau
c14446ab00 Don't run install from the various configs in parallel
Hopefully this fixes a bug in the previous debian/rules that could lead
to FTBFS.  Thanks to Bernhard R. Link for the suggestions.
2009-09-04 17:22:31 +02:00
Brice Goglin
70fd9421ac Prepare changelog for upload 2009-09-04 11:39:18 +02:00
Brice Goglin
1810757c29 Bump Standards-Version to 3.8.3 to 3.8.3 now that we have README.source 2009-09-04 11:35:10 +02:00
Brice Goglin
4c79e26a9d Add a Mesa-specific README.source inspired from the one in xsfbs 2009-09-04 11:33:17 +02:00
Brice Goglin
1783d52ace Prepare changelog for upload 2009-09-04 10:58:58 +02:00
Brice Goglin
c5bf7f94c0 New upstream release 2009-09-04 10:55:25 +02:00
Brice Goglin
8222bfc1d1 Merge tag 'mesa_7_5_1' into debian-unstable
Conflicts:
	progs/tests/fbotest1.c
	progs/tests/texwrap.c
	progs/trivial/.gitignore
	progs/trivial/Makefile
2009-09-04 10:53:45 +02:00
Julien Cristau
3f970a6d4e Bump debhelper build-dep for dh_lintian 2009-08-22 21:30:15 +02:00
Julien Cristau
a151efc6cc Use dh_lintian. 2009-07-27 00:29:45 +02:00
Julien Cristau
05c0e10a9e Override 'package-name-doesnt-match-sonames' lintian warnings 2009-07-27 00:29:44 +02:00
Brice Goglin
3c4727a4e9 Prepare changelog for upload 2009-07-24 00:29:35 +02:00
Brice Goglin
40f217aea7 Enable i686 optimized libraries on hurd-i386 2009-07-24 00:29:25 +02:00
Brice Goglin
d3403389cd Only install the huge upstream changelog in mesa-common-dev 2009-07-23 23:41:19 +02:00
Brice Goglin
a70d1eb1ee Pull from upstream mesa_7_5_branch 2009-07-23 21:11:22 +02:00
Brice Goglin
d7aa5d5ca0 Merge branch 'mesa_7_5_branch' of git://anongit.freedesktop.org/mesa/mesa into debian-unstable 2009-07-23 21:09:47 +02:00
Brice Goglin
04493de67b Prepare changelog for upload 2009-07-21 22:22:07 +02:00
Brice Goglin
e06e2b001f Install the upstream ChangeLog 2009-07-21 22:21:47 +02:00
Brice Goglin
d740456ee5 Install the upstream ChangeLog 2009-07-21 20:42:50 +02:00
Brice Goglin
45c5ed4bb8 Pull from upstream mesa_7_5_branch 2009-07-21 20:38:30 +02:00
Brice Goglin
19a8ce6f70 Merge branch 'mesa_7_5_branch' of git://anongit.freedesktop.org/mesa/mesa into debian-unstable 2009-07-21 20:05:24 +02:00
Brice Goglin
767c0b16f3 Prepare changelog for upload 2009-07-19 12:54:22 +02:00
Brice Goglin
3041819a4e New upstream release 2009-07-19 12:51:08 +02:00
Brice Goglin
4a9beaf2e9 Merge tag 'mesa_7_5' into debian-experimental
Conflicts:
	progs/tests/.gitignore
	progs/tests/Makefile
	progs/tests/SConscript
	progs/trivial/Makefile
	scons/dxsdk.py
	scons/gallium.py
2009-07-19 12:50:28 +02:00
Brice Goglin
fe9b9fec39 Enable i686 optimized libraries on kfreebsd-i386 2009-07-17 18:21:27 +02:00
Timo Aaltonen
082f547492 Move dri.pc to mesa-common-dev (closes: #521667) 2009-06-29 15:22:21 +03:00
Julien Cristau
9032d6b40b Prepare changelog for upload 2009-06-28 20:22:05 +02:00
Julien Cristau
489ca73db1 Merge commit 'mesa-7.4.4-1' into debian-experimental
Conflicts:
	debian/changelog
2009-06-28 20:20:56 +02:00
Julien Cristau
2c8046b1bc Prepare changelog for upload 2009-06-28 18:58:41 +02:00
Julien Cristau
18bb120fce Refresh patches 2009-06-28 18:41:21 +02:00
Julien Cristau
d2c8712fa4 Remove some directories absent from the tarballs, and add some missing ones 2009-06-28 18:09:58 +02:00
Julien Cristau
b59211a602 Update changelog 2009-06-28 17:46:10 +02:00
Julien Cristau
adafe6d607 Merge commit 'mesa_7_5_rc4' into debian-experimental
Conflicts:
	progs/fp/point-position.c
	progs/fp/tri-depth.c
	progs/fp/tri-depth2.c
	progs/fp/tri-depthwrite.c
	progs/fp/tri-depthwrite2.c
	progs/fp/tri-inv.c
	progs/fp/tri-param.c
	progs/fp/tri-tex.c
	progs/tests/.gitignore
	progs/tests/Makefile
	progs/tests/zreaddraw.c
2009-06-28 17:41:50 +02:00
Julien Cristau
c38bc21cf5 add bug closer 2009-06-24 19:54:26 +02:00
Julien Cristau
1df36638c0 Don't build gallium for now. 2009-06-24 17:30:00 +02:00
Julien Cristau
dd89f8e3ea Merge branch 'debian-unstable' into debian-experimental
Conflicts:
	debian/changelog
	debian/patches/06_kfreebsd-ftbfs.diff
2009-06-24 17:26:42 +02:00
Julien Cristau
2ca61d4474 Merge commit 'mesa_7_4_4' into debian-experimental 2009-06-24 17:23:34 +02:00
Julien Cristau
22f1c449ae update changelog 2009-06-24 17:20:23 +02:00
Julien Cristau
2be8cd65cb Merge tag 'mesa_7_5_rc3' into debian-experimental 2009-06-24 17:19:16 +02:00
Julien Cristau
ec4c889266 Update 04_osmesa_version.diff, refresh 06_kfreebsd-ftbfs.diff. 2009-06-24 12:37:17 +02:00
Julien Cristau
c84a384fdf Update changelog 2009-06-24 12:32:20 +02:00
Julien Cristau
f56634caea Merge tag 'mesa_7_4_4' into debian-unstable 2009-06-24 12:27:10 +02:00
Brian Paul
d0c391b6a2 docs: set 7.4.4 release date 2009-06-23 18:53:31 -06:00
Brian Paul
eaed8dc609 mesa: bump version to 7.4.4 2009-06-22 15:27:03 -06:00
Brian Paul
b5609e6444 docs: release notes for 7.4.4 2009-06-22 15:26:19 -06:00
Brian Paul
0b870bbb2a docs: 7.4.3 MD5 checksums 2009-06-22 11:24:09 -06:00
Brian Paul
1dbbc39f48 intel: added a null ptr check
This was hit during screen tear-down.
2009-06-22 11:23:44 -06:00
Brian Paul
4b8cd0b0ad intel: call _mesa_unreference_framebuffer() in intelDestroyBuffer()
This was a call to _mesa_reference_framebuffer(ptr, NULL) but the NULL
pointer is incorrect in Mesa 7.4.x.

Fixes a failed assertion during screen tear-down.
2009-06-22 11:23:13 -06:00
Brian Paul
5d387a35ff docs: set 7.4.3 release date 2009-06-19 15:16:01 -06:00
Brian Paul
2ae6859215 docs: bump version to 7.4.3 2009-06-19 15:14:49 -06:00
Brian Paul
afe0ca1f3f mesa: rework viewport/scissor initialization code
The first time a context is bound to a drawable, the viewport and scissor
bounds are initialized to the buffer's size.  This is actually a bit tricky.

A new _mesa_check_init_viewport() function is called in several places
to check if the viewport has been initialized.  We also use a new
ctx->ViewportInitialized flag instead of the overloaded
ctx->FirstTimeCurrent flag.

Hand-picked from mesa_7_5_branch, commit 3f856c6b6b
2009-06-17 08:43:17 -06:00
Brian Paul
6b782231d0 mesa: added null ptr check in Fake_glXCreatePixmap()
Fixes segfault in progs/xdemos/glxgears_pixmap.c

(cherry picked from master, commit d18c57aaea)
2009-06-17 08:40:16 -06:00
Brian Paul
2b890c9818 docs: document GLX/glean makeCurrent fix 2009-06-17 08:06:50 -06:00
Brian Paul
e0b13e15f1 GLX: attempt to fix glean makeCurrent test cases.
Two parts to this:

One we don't keep pointers to possibly freed memory anymore once we unbind the
drawables from the context. Brian I need to figure out what the comment
you made there, can we get a glean/piglit test so we can fix it properly?

If the new gc is the same as the oldGC, we call the unbind even though
we just bound it in that function. doh.

(cherry picked from master, commit 77506dac8e)
2009-06-17 08:05:54 -06:00
Brian Paul
420c74b870 docs: document viewport clamping bug 2009-06-16 09:40:22 -06:00
Brian Paul
a64f994a3a mesa: fix incorrect viewport clamping in _mesa_set_viewport()
A 0 by 0 viewport size is legal.  Don't clamp against lower bound of one.
The error checking earlier in the function prevents negative values.
2009-06-16 09:39:33 -06:00
Brian Paul
c568446dc6 docs: fixed bug 21872 2009-06-16 08:48:26 -06:00
Brian Paul
acbf9c6f3c mesa: fix REMAINDER() macro
The results were incorrect for some negative values of A.
See bug 21872.

(cherry picked from mesa_7_5_branch, commit ed7f4b4230)
2009-06-16 08:48:12 -06:00
Brian Paul
3b842feead docs: recent fixes for 7.4.3 2009-06-15 16:24:03 -06:00
Brian Paul
6d62b8d5c1 intel: Release fb backing regions in intelDestroyBuffer()
Fixes memory leak when destroying framebuffers.

(cherry picked from mesa_7_5_branch, commit d027e8feff)
2009-06-15 16:22:54 -06:00
Brian Paul
02659868e8 glsl: fix warnings, update comments, s/TRUE/GL_TRUE/
(cherry picked from master, commit 7fdd64ab29)
2009-06-10 12:26:12 -06:00
Brian Paul
457f4efab0 glsl: Handle continuation characters in preprocessor.
(cherry picked from master, commit cc22620e4b)
2009-06-10 12:25:35 -06:00
Brian Paul
0269eb5e3f glsl: Fix symbol replacement handling in preprocessor.
(cherry picked from master, commit d9617deb00)
2009-06-10 12:25:11 -06:00
Brian Paul
b7d9be69b2 glsl: Expand nested preprocessor macros.
(cherry picked from master, commit ef8caec29a)
2009-06-10 12:24:52 -06:00
Brian Paul
7b8b856c97 glsl: Fix preprocessor define argument parsing for dead sections.
(cherry picked from master, commit 19a54d9f10)
2009-06-10 12:24:32 -06:00
Brian Paul
24bffc4cc2 glsl: preprocessor debug code (disabled)
(cherry picked from master, commit 854151ba62)
2009-06-10 12:24:11 -06:00
Brian Paul
586341afd9 docs: document recent bug fixes for 7.4.x 2009-06-05 17:30:51 -06:00
Dan Nicholson
1e4334f623 osmesa: Link with -ldl for dlopen code
Now that the dlopen wrappers are built into libmesa.a, we need to link
standalone libOSMesa with libdl to resolve dlopen and friends on
platforms that need it.
(cherry picked from commit 4795dd5950)
2009-06-04 19:45:58 -07:00
Dan Nicholson
8861577fd1 osmesa: Allow building standalone in all three channel widths
autoconf had been designating the 8 bit libOSMesa as the default
standalone osmesa, but the Makefile expected it to be linked to libGL.
Fix up the osmesa Makefile so that it allows any of the combinations of
standalone and channel width to be built.

Fixes bug #21980.
(cherry picked from commit 7441dcd90b)
2009-06-04 19:05:48 -07:00
Brian Paul
cf997e0d2e mesa: fix loop over generic attribs in update_arrays()
(cherry picked from commit 1045481dd9)
2009-06-01 15:04:35 -06:00
Timo Aaltonen
f3496a3cbd mesa-common-dev.install: Don't install glxew.h, conflicts with libglew. 2009-06-01 22:26:27 +03:00
Julien Cristau
4bcb9cc14f Update patches
+ 02_use-ieee-fp-on-s390-and-m68k.patch moved from imports.h to compiler.h
+ 03_optional-progs-and-install.patch refreshed
+ 05_hurd-ftbfs.diff partly applied upstream
+ 06_kfreebsd-ftbfs.diff refreshed
2009-05-31 13:15:00 +02:00
Julien Cristau
537f3e7a1e Install dri.pc, which will be needed to build xorg-server 1.7. 2009-05-31 12:52:47 +02:00
Julien Cristau
2096234053 clarify changelog: this is only a release candidate 2009-05-30 21:08:54 +02:00
Timo Aaltonen
99b4daf5e1 Disable EGL.
Update the changelog too.
2009-05-25 12:47:02 +03:00
Brian Paul
506d340363 demos: fix multitex.c VertCoord attribute mapping
If the multitex.vert shader uses the VertCoord generic vertex attribute
instead of the pre-defined gl_Vertex attribute, we need to make sure that
VertCoord gets bound to generic vertex attribute zero.

That's because we need to call glVertexAttrib2fv(0, xy) after all the other
vertex attributes have been set since setting generic attribute 0 triggers
vertex submission.  Before, we wound up issuing the vertex attributes in
the order 0, 1, 2 which caused the first vertex to be submitted before all
the attributes were set.  Now, the attributes are set in 1, 2, 0 order.

(cherry picked from commit 58fadc6242)
2009-05-22 13:20:48 -06:00
Brian Paul
9dbe32f5eb vbo: fix crash in vbo_exec_bind_arrays()
When a vertex shader uses generic vertex attribute 0, but not gl_Vertex,
we need to set attribute[16] to point to attribute[0].  We were setting the
attribute size, but not the pointer.

Fixes crash in glsl/multitex.c when using the VertCoord attribute instead
of gl_Vertex.

(cherry picked from commit c3538969e1)
2009-05-22 13:19:42 -06:00
Brian Paul
9adca8814d mesa: allow GL_BITMAP type in _mesa_image_image_stride()
It's possible to hand a GL_COLOR_INDEX/GL_BITMAP image to glTexImage3D()
which gets converted to RGBA via the glPixelMap tables.

This fixes a failure with piglit/fdo10370 with Gallium.

(cherry picked from commit 995456f930)
2009-05-22 09:38:18 -06:00
Brian Paul
8125c5ce92 docs: updates for version 7.4.3 2009-05-22 07:13:47 -06:00
Brian Paul
ba3f8dc0aa mesa: bump version to 7.4.3 2009-05-22 07:13:34 -06:00
Brian Paul
85cf1675d4 mesa: bump version to 7.4.3 2009-05-22 07:13:30 -06:00
Brian Paul
3aeace2928 vbo: fix incorrect loop limit in bind_array_obj()
The generic_array[] is 16 elements in size, but the loop was doing 32
iterations.  The out of bounds array write was clobbering the following
inputs[] array but as luck would have it, that didn't matter.

(cherry picked from commit 8da09e6924)
2009-05-22 07:10:47 -06:00
Brian Paul
0edb61afef docs: add glGet case for GL_VERTEX_ARRAY_BINDING_APPLE 2009-05-22 07:08:38 -06:00
Brian Paul
accffe2b73 mesa: add missing glGet*() case for GL_VERTEX_ARRAY_BINDING_APPLE
(cherry picked from commit 70c4b81e88)
2009-05-22 07:07:38 -06:00
Brian Paul
35f5222ae0 mesa: remove texUnit->_Current = NULL assignment
A follow-on to commit 46a2aad2cf which
fixes a texture object memory leak.

See bug 21756.
2009-05-20 08:14:45 -06:00
Brian Paul
1096a6c579 docs: 7.4.3 relnotes edits 2009-05-19 10:06:58 -06:00
Brian Paul
1eb59f49a4 docs: start 7.4.3 release notes 2009-05-19 10:06:09 -06:00
Brian Paul
63d1e89138 mesa: allow depth/stencil textures to be attached to GL_STENCIL_ATTACHMENT
See sourceforge bug #2793846.

(cherry picked from commit 042d9a5132)
2009-05-19 10:04:17 -06:00
Brian Paul
3ed7e9bfcc mesa: assign trb->Base.StencilBits in update_wrapper().
When we render to a depth/stencil texture there are stencil bits.

(cherry picked from commit c99a60c40d)
2009-05-19 10:02:55 -06:00
Brian Paul
46a2aad2cf mesa: fix texture object reference counting in texture_override()
Fixes bug 21756: r200: Mesa-7.4.2 release regression: "Quit on exit".
2009-05-19 09:03:19 -06:00
Timo Aaltonen
73670d56e8 Merge branch 'debian-experimental' into test
Conflicts:
	Makefile
	configs/default
	docs/download.html
	docs/news.html
	docs/relnotes-7.4.2.html
	doxygen/.gitignore
	include/GL/internal/dri_interface.h
	include/GLES/egl.h
	progs/demos/Windows/all.dsp
	progs/demos/Windows/bounce.dsp
	progs/demos/Windows/clearspd.dsp
	progs/demos/Windows/cubemap.dsp
	progs/demos/Windows/demos.dsw
	progs/demos/Windows/drawpix.dsp
	progs/demos/Windows/fire.dsp
	progs/demos/Windows/gears.dsp
	progs/demos/Windows/geartrain.dsp
	progs/demos/Windows/glinfo.dsp
	progs/demos/Windows/gloss.dsp
	progs/demos/Windows/gltestperf.dsp
	progs/demos/Windows/glutfx.dsp
	progs/demos/Windows/ipers.dsp
	progs/demos/Windows/isosurf.dsp
	progs/demos/Windows/loadbias.dsp
	progs/demos/Windows/morph3d.dsp
	progs/demos/Windows/multiarb.dsp
	progs/demos/Windows/occlude.dsp
	progs/demos/Windows/osdemo.dsp
	progs/demos/Windows/paltex.dsp
	progs/demos/Windows/pixeltex.dsp
	progs/demos/Windows/pointblast.dsp
	progs/demos/Windows/rain.dsp
	progs/demos/Windows/ray.dsp
	progs/demos/Windows/readpix.dsp
	progs/demos/Windows/reflect.dsp
	progs/demos/Windows/renormal.dsp
	progs/demos/Windows/shadowtex.dsp
	progs/demos/Windows/spectex.dsp
	progs/demos/Windows/spriteblast.dsp
	progs/demos/Windows/stex3d.dsp
	progs/demos/Windows/teapot.dsp
	progs/demos/Windows/terrain.dsp
	progs/demos/Windows/tessdemo.dsp
	progs/demos/Windows/texcyl.dsp
	progs/demos/Windows/texdown.dsp
	progs/demos/Windows/texenv.dsp
	progs/demos/Windows/texobj.dsp
	progs/demos/Windows/trispd.dsp
	progs/demos/Windows/tunnel.dsp
	progs/demos/Windows/tunnel2.dsp
	progs/demos/Windows/winpos.dsp
	progs/demos/dinoshade.c
	progs/demos/projtex.c
	progs/egl/Makefile
	progs/egl/demo1.c
	progs/egl/demo2.c
	progs/egl/demo3.c
	progs/egl/eglgears.c
	progs/egl/eglinfo.c
	progs/fp/.gitignore
	progs/fp/Makefile
	progs/miniglx/Makefile
	progs/redbook/Windows/redbook.dsw
	progs/redbook/Windows/teapots.dsp
	progs/slang/windows/vc60/cltest.dsp
	progs/slang/windows/vc60/framework.dsp
	progs/slang/windows/vc60/slang.dsw
	progs/slang/windows/vc60/sotest.dsp
	progs/slang/windows/vc60/vstest.dsp
	progs/tests/.gitignore
	progs/tests/Makefile
	progs/tests/afsmultiarb.c
	progs/tests/antialias.c
	progs/tests/api_speed.c
	progs/tests/arbfpspec.c
	progs/tests/arbfptest1.c
	progs/tests/arbfptexture.c
	progs/tests/arbfptrig.c
	progs/tests/arbnpot-mipmap.c
	progs/tests/arbnpot.c
	progs/tests/arbvptest1.c
	progs/tests/arbvptest3.c
	progs/tests/arbvptorus.c
	progs/tests/arbvpwarpmesh.c
	progs/tests/arraytexture.c
	progs/tests/blendminmax.c
	progs/tests/blendsquare.c
	progs/tests/blendxor.c
	progs/tests/bufferobj.c
	progs/tests/bug_3050.c
	progs/tests/bug_3101.c
	progs/tests/bug_3195.c
	progs/tests/copypixrate.c
	progs/tests/crossbar.c
	progs/tests/cva.c
	progs/tests/debugger.c
	progs/tests/drawbuffers.c
	progs/tests/exactrast.c
	progs/tests/ext422square.c
	progs/tests/fbotest1.c
	progs/tests/fbotest2.c
	progs/tests/floattex.c
	progs/tests/fog.c
	progs/tests/fogcoord.c
	progs/tests/fptest1.c
	progs/tests/fptexture.c
	progs/tests/interleave.c
	progs/tests/invert.c
	progs/tests/jkrahntest.c
	progs/tests/lineclip.c
	progs/tests/manytex.c
	progs/tests/minmag.c
	progs/tests/mipmap_limits.c
	progs/tests/multipal.c
	progs/tests/multitexarray.c
	progs/tests/multiwindow.c
	progs/tests/no_s3tc.c
	progs/tests/packedpixels.c
	progs/tests/pbo.c
	progs/tests/prog_parameter.c
	progs/tests/random.c
	progs/tests/readrate.c
	progs/tests/rubberband.c
	progs/tests/seccolor.c
	progs/tests/shader_api.c
	progs/tests/stencil_twoside.c
	progs/tests/stencil_wrap.c
	progs/tests/stencilwrap.c
	progs/tests/subtexrate.c
	progs/tests/tex1d.c
	progs/tests/texcmp.c
	progs/tests/texcompress2.c
	progs/tests/texfilt.c
	progs/tests/texgenmix.c
	progs/tests/texline.c
	progs/tests/texrect.c
	progs/tests/texwrap.c
	progs/tests/unfilledclip.c
	progs/tests/vao-01.c
	progs/tests/vao-02.c
	progs/tests/vparray.c
	progs/tests/vpeval.c
	progs/tests/vptest1.c
	progs/tests/vptest2.c
	progs/tests/vptest3.c
	progs/tests/vptorus.c
	progs/tests/vpwarpmesh.c
	progs/tests/yuvrect.c
	progs/tests/yuvsquare.c
	progs/tests/zreaddraw.c
	progs/trivial/.gitignore
	progs/trivial/Makefile
	progs/trivial/clear.c
	progs/trivial/dlist-dangling.c
	progs/trivial/dlist-edgeflag-dangling.c
	progs/trivial/dlist-edgeflag.c
	progs/trivial/drawarrays.c
	progs/trivial/drawelements.c
	progs/trivial/drawrange.c
	progs/trivial/line-clip.c
	progs/trivial/line-cull.c
	progs/trivial/line-userclip-clip.c
	progs/trivial/line-userclip-nop-clip.c
	progs/trivial/line-userclip-nop.c
	progs/trivial/line-userclip.c
	progs/trivial/line.c
	progs/trivial/lineloop-clip.c
	progs/trivial/lineloop.c
	progs/trivial/point-clip.c
	progs/trivial/point-param.c
	progs/trivial/point-wide.c
	progs/trivial/point.c
	progs/trivial/poly-flat.c
	progs/trivial/poly-unfilled.c
	progs/trivial/poly.c
	progs/trivial/quad-clip-all-vertices.c
	progs/trivial/quad-clip-nearplane.c
	progs/trivial/quad-clip.c
	progs/trivial/quad-degenerate.c
	progs/trivial/quad-flat.c
	progs/trivial/quad-offset-factor.c
	progs/trivial/quad-offset-unfilled.c
	progs/trivial/quad-offset-units.c
	progs/trivial/quad-tex-2d.c
	progs/trivial/quad-tex-3d.c
	progs/trivial/quad-tex-pbo.c
	progs/trivial/quad-unfilled.c
	progs/trivial/quad.c
	progs/trivial/quads.c
	progs/trivial/quadstrip-flat.c
	progs/trivial/quadstrip.c
	progs/trivial/tri-blend.c
	progs/trivial/tri-clip.c
	progs/trivial/tri-cull.c
	progs/trivial/tri-dlist.c
	progs/trivial/tri-edgeflag.c
	progs/trivial/tri-flat-clip.c
	progs/trivial/tri-flat.c
	progs/trivial/tri-tex-3d.c
	progs/trivial/tri-unfilled-clip.c
	progs/trivial/tri-unfilled-smooth.c
	progs/trivial/tri-unfilled-userclip.c
	progs/trivial/tri-unfilled.c
	progs/trivial/tri-userclip.c
	progs/trivial/tri.c
	progs/trivial/tristrip-clip.c
	progs/trivial/tristrip.c
	progs/trivial/vbo-drawarrays.c
	progs/trivial/vbo-drawelements.c
	progs/trivial/vbo-drawrange.c
	progs/trivial/vp-array.c
	progs/trivial/vp-clip.c
	progs/trivial/vp-line-clip.c
	progs/trivial/vp-tri.c
	progs/trivial/vp-unfilled.c
	progs/vp/Makefile
	progs/vp/vp-tris.c
	src/egl/Makefile
	src/egl/drivers/demo/Makefile
	src/egl/drivers/demo/demo.c
	src/egl/drivers/dri/Makefile
	src/egl/drivers/dri/egldri.c
	src/egl/drivers/dri/egldri.h
	src/egl/main/Makefile
	src/egl/main/eglapi.c
	src/egl/main/eglapi.h
	src/egl/main/eglconfig.c
	src/egl/main/eglconfig.h
	src/egl/main/eglcontext.c
	src/egl/main/eglcontext.h
	src/egl/main/egldisplay.c
	src/egl/main/egldisplay.h
	src/egl/main/egldriver.c
	src/egl/main/egldriver.h
	src/egl/main/eglglobals.c
	src/egl/main/eglglobals.h
	src/egl/main/egllog.c
	src/egl/main/eglmode.h
	src/egl/main/eglsurface.c
	src/egl/main/eglsurface.h
	src/egl/main/egltypedefs.h
	src/glx/mini/Makefile
	src/glx/x11/dri2_glx.c
	src/mesa/drivers/allegro/amesa.c
	src/mesa/drivers/dri/intel/intel_buffers.c
	src/mesa/drivers/dri/intel/intel_clear.c
	src/mesa/drivers/dri/intel/intel_context.c
	src/mesa/drivers/dri/intel/intel_pixel.c
	src/mesa/drivers/dri/r300/r300_texstate.c
	src/mesa/drivers/dri/radeon/radeon_screen.c
	src/mesa/drivers/windows/gldirect/dx7/gld_wgl_dx7.c
	src/mesa/drivers/windows/gldirect/dx8/gld_wgl_dx8.c
	src/mesa/drivers/windows/gldirect/dx9/gld_wgl_dx9.c
	src/mesa/main/glheader.h
	src/mesa/main/light.c
	src/mesa/main/teximage.c
	src/mesa/main/texparam.c
	src/mesa/main/version.h
	src/mesa/swrast/s_texfilter.c
2009-05-18 14:05:53 +03:00
Aidan Thornton
b5eafa2d51 Initialize psp->waitX/waitGL for swrast_dri.so.
Fixes http://bugs.freedesktop.org/show_bug.cgi?id=21053 .
(cherry picked from commit f2445dfd85)
2009-05-18 12:51:12 +02:00
Timo Aaltonen
ac00f6a47a Merge commit 'mesa_7_4' into test 2009-05-18 13:21:01 +03:00
Brian Paul
d1fcb4a263 docs 7.4.2 tarball md5 sums 2009-05-15 06:32:24 -06:00
Brian Paul
da709f163f docs: set 7.4.2 release date 2009-05-15 06:21:56 -06:00
Brian Paul
3a1bcbe579 mesa: bump version to 7.4.2 2009-05-14 13:20:13 -06:00
Brian Paul
ec863602f2 i965: fix render to FBO/texture orientation bugs
Anytime we're not rendering to the default/window FBO, need to invert
rendering, not just when rendering to a texture.  Otherwise, if a FBO
consists of a mix of textures and renderbuffers the up/down orientation
was inconsistant.

Fixes shadowtex.c bad rendering.
(cherry picked from commit ba367f68cc)
2009-05-13 14:13:46 -07:00
Brian Paul
2ff47b80f5 docs: updated bug fix list for 7.4.2 2009-05-11 09:49:46 -06:00
Brian Paul
63cde0ea0e mesa: Fixed a texture memory leak
The current texture for any particular texture unit is given an additional
reference in update_texture_state(); but if the context is closed before
that texture can be released (which is quite frequent in normal use, unless
a program unbinds and deletes the texture and renders without it to force
a call to update_texture_state(), the memory is lost.

This affects general Mesa; but the i965 is particularly affected because
it allocates a considerable amount of additional memory for each allocated
texture.

(cherry picked from master, commit c230767d69)
2009-05-11 09:47:17 -06:00
Brice Goglin
ab885d7088 Enable motif in GLw 2009-05-09 20:49:15 +02:00
Eric Anholt
c521d4629b Flag GLSL programs as depending on constant state when they use uniforms.
This is not a cherry-pick, as it got fixed in master with the
_NEW_PROGRAM_CONSTANTS change.  Fixes bad rendering with various GLSL programs
idr and I have been writing.
2009-05-05 12:14:29 -07:00
Brian Paul
e9d5569bf3 glx: replace Xmalloc() calls with Xcalloc()
Fixes a bug where psp->WaitX was uninitialized.  Reported by Chris Clayton.

(cherry picked from commit dd4c142e90)
2009-05-04 08:34:20 -06:00
Julien Cristau
84ae506072 Don't build hardware dri drivers on s390. 2009-05-03 19:30:02 +02:00
Julien Cristau
4842cde28f Prepare changelog for upload 2009-05-03 16:05:30 +02:00
Julien Cristau
b651c02dca Fix Architecture field for libgl1-mesa-dri-dbg 2009-05-03 16:04:39 +02:00
Julien Cristau
ec44d955bb Move -dbg packages to new 'debug' section. 2009-05-03 16:03:58 +02:00
Julien Cristau
914edc46fa Update changelog 2009-05-03 14:31:05 +02:00
Julien Cristau
64cb29deca Merge branch 'mesa_7_4_branch' of git.freedesktop.org:/git/mesa/mesa into debian-unstable 2009-05-03 14:26:30 +02:00
Brian Paul
6337525497 mesa: fix state validation bug for glCopyTex[Sub]Image()
We need to make sure the framebuffer state is up to date to make sure we
read pixels from the right buffer when doing a texture image copy.

(cherry-picked from master, commit 63f0130980)
2009-05-01 20:24:39 -06:00
Brian Paul
7541c3806a docs: fix typos 2009-04-30 17:08:37 -06:00
Michel Dänzer
a1ce4efefb r300: Increase reference count of texture objects referenced by current state.
Fixes a use-after-free reported in
http://bugs.freedesktop.org/show_bug.cgi?id=20539, so this possibly fixes that
bug. It has been confirmed to fix
http://bugs.freedesktop.org/show_bug.cgi?id=17895 .
(cherry picked from commit c28707b507)
2009-04-30 13:50:36 +02:00
Brian Paul
656cd707db mesa: Fix buffer overflow when parsing generic vertex attributes.
(cherry picked from master, commit fa92756400)
2009-04-28 10:58:13 -06:00
Alex Deucher
ce7f78694d R300: add quadpipe overrides
RV410 SE chips only have 1 quadpipe.
Also, handle other R300 chip with quadpipe override
2009-04-28 12:40:04 -04:00
Brian Paul
fd95e8ee33 docs: fix swrast texrect bug 21461 2009-04-28 09:11:05 -06:00
Brian Paul
d2b8c41b7a swrast: add missing break in clamp_rect_coord_linear()
See bug 21461.

(cherry picked from master, commit dcf571aff9)
2009-04-28 09:09:41 -06:00
Alan Hourihane
bfd6bb31a6 demos: ensure display lists are destroyed for next generation 2009-04-24 17:22:52 +01:00
Eric Anholt
c3e7fc394f i965: Support drawing to FBO cube faces other than positive X.
Also fixes drawing to 3D texture depth levels.
(cherry picked from commit 8374379572)
2009-04-23 14:05:58 -07:00
Eric Anholt
dd56897b27 intel: Fix commented-out glViewport in intel_meta_set_passthrough_transform.
Too much commit -a while debugging.
(cherry picked from commit 8910da5b7a)
2009-04-23 14:00:56 -07:00
Brian Paul
e8807a14a6 fixed intel front-buffer rendering segfault 2009-04-20 16:13:04 -06:00
Brian Paul
c38397866d intel: added null screen->dri2.loader pointer check
Fixes front-buffer rendering with DRI1.
2009-04-20 16:12:16 -06:00
Brice Goglin
d7b9a6c9ef Add patch by Aurelien Jarno to fix FTBPS on kfreebsd-i386 2009-04-20 12:36:44 +02:00
Julien Cristau
4f2bc058f3 Add patch by Samuel Thibault to fix FTBFS on hurd-i386. 2009-04-19 00:46:32 +02:00
Julien Cristau
909afc5304 Adjust libgl1-mesa-glx and libgl1-mesa-dri package relationships
Drop the dependency on libgl1-mesa-glx from -dri, and make -glx
recommend -dri instead.  The dri drivers are also loaded by the X
server, which doesn't need libGL.  On the other hand, libGL needs one of
the dri drivers for direct rendering (either software or hardware).
Mark libgl1-mesa-dri as breaking old xserver-xorg-core and
libgl1-mesa-glx, to avoid incompatibilities.
2009-04-18 22:28:35 +02:00
Julien Cristau
4deb679b42 Make libgl1-mesa-dri 'Architecture: any'
swrast_dri.so should get built everywhere.
2009-04-18 22:27:51 +02:00
Brian Paul
dce84002c2 docs: create 7.4.2 release notes page 2009-04-18 11:12:41 -06:00
Brian Paul
c490349825 docs: remove unused sections 2009-04-18 11:12:33 -06:00
Brian Paul
f62d45fd0d docs: MD5 sums for 7.4.1 release 2009-04-18 11:03:25 -06:00
Brian Paul
b2dfc29a82 intel: update driver date, remove RC3 designation 2009-04-18 10:51:41 -06:00
Brian Paul
7c1e61f41c mesa: bump version to 7.4.1 2009-04-18 10:49:35 -06:00
Brian Paul
5654197d12 docs: prep for 7.4.1 release 2009-04-18 10:49:08 -06:00
Brian Paul
87cb55e9c6 mesa: bump version to 7.4.1 2009-04-18 10:48:57 -06:00
Brian Paul
e7d499f935 mesa: add switch case for GL_VERTEX_STATE_PROGRAM_NV in _mesa_new_program()
Fixes bug seen in progs/tests/vptest1.c

(cherry picked from master, commit c5af2ed60f)
2009-04-18 10:21:00 -06:00
Michel Dänzer
51e7600d77 intel: Handle ARB_vertex_buffer_object state in intel_clear_tris().
Fixes gearsvbo app by Michael Clark.
(cherry picked from commit 54fb6f0053)
2009-04-18 15:53:08 +02:00
Brian Paul
69b17abd6b intel: make sure polygon mode is set properly in intel_clear_tris()
Fixes progs/glsl/skinning.c demo.

(cherry picked from master, commit 794d488e6d)
2009-04-17 16:14:39 -06:00
Brian Paul
a975da7aca docs: fixed user clip plane restore bug in glPopAttrib() 2009-04-16 10:16:38 -06:00
Brian Paul
ecf47b5046 mesa: fix bad mask bit in clip plane restore code for glPopAttrib()
(cherry picked from master commit d82876e850)
2009-04-16 10:16:26 -06:00
Brian Paul
fed8dc53ad mesa: add distclean target to top-level Makefile
(cherry picked from master, commit 666702baec)
2009-04-16 09:34:06 -06:00
Brian Paul
1f462e26c0 demos: use larger buffer for snprintf() call, see bug 21220
(cherry picked from master, commit 0d0028e6df)
2009-04-16 09:26:54 -06:00
Ian Romanick
85396d8ef4 intel: Bump driver date 2009-04-15 15:12:48 -07:00
Ian Romanick
f11b84998d docs: more bug fixes for Mesa 7.4.1 2009-04-15 15:10:33 -07:00
Julien Cristau
5d4659e7c7 Install pkgconfig files for libGLU, libOSMesa and libGLw. 2009-04-15 21:31:24 +02:00
Julien Cristau
c422347bee Make libgl1-mesa-dev and mesa-common-dev 'Architecture: any'.
This gets rid of uninstallability when a new upstream version
isn't built on all architectures, and allows us to ship potentially
arch-specific .pc files.
2009-04-15 21:25:17 +02:00
Brian Paul
d2f6791062 glx: added null pointer check in glXGetFBConfigs()
Fixes segfault seen with glxinfo with NVIDIA OpenGL.

(cherry picked from master, commit 05471828dc)
2009-04-15 08:02:38 -06:00
Ian Romanick
49e0c74ddd DRI2: Don't fault on NULL DrawBuffer
It is possible for ctx->DrawBuffer to be NULL, so don't fault when
that happens.  This change is not being committed to master because it
doesn't appear to be necessary there.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
2009-04-11 20:24:15 -07:00
Eric Anholt
83e14773c5 dri2: Don't crash if the server returns more buffers than expected.
(cherry picked from commit f967e8b507)
2009-04-09 15:15:33 -07:00
Ian Romanick
4605937843 intel / DRI2: Accept fake front-buffer from loader
Handle the loader returning a fake front-buffer.  Since the driver
never specifically requests a fake front-buffer, the driver assumes
that it will never receive both a fake and a real front-buffer.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kristian Høgsberg <krh@redhat.com>
(cherry picked from commit 6a6e478e55)
2009-04-09 14:44:53 -07:00
Ian Romanick
b8ef30c45c DRI2: Assume that there is always a front buffer
Assume that the front-buffer exists even if the server didn't tell the
client that it exists.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kristian Høgsberg <krh@redhat.com>
(cherry picked from commit e6386e0d9a)
2009-04-09 14:44:44 -07:00
Ian Romanick
4f17040cb1 intel / DRI2: Track and flush front-buffer rendering
Track two flags:  whether or not front-buffer rendering is currently
enabled and whether or not front-buffer rendering has been enabled
since the last glFlush.  If the second flag is set, the front-buffer
is flushed via a loader call back.  If the first flag is cleared, the
second flag is cleared at this time.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kristian Høgsberg <krh@redhat.com>
(cherry picked from commit 43cf0d1eeb)
2009-04-09 14:44:34 -07:00
Ian Romanick
98c2c9ef8c DRI2: Provide an interface for drivers to flush front-buffer rendering
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kristian Høgsberg <krh@redhat.com>
(cherry picked from commit 82634ee8df)
2009-04-09 14:44:17 -07:00
Brian Paul
d805c82068 mesa: fix potential recursive locking deadlock in _mesa_HashWalk()
If the walk callback called _mesa_HashRemove() we'd deadlock.

(cherry picked from master, commit deff099215)
2009-04-09 10:55:06 -06:00
Julien Cristau
3d81e934e5 Upload to unstable. 2009-04-08 23:53:54 +01:00
Brian Paul
6c3bec4387 swrast: fix point rendering function selection
Need to clamp default point size to min/max range before checking if it's one.
Fixes glean pointAtten test.

(cherry picked from master, commit 228f20e324)
2009-04-07 13:44:50 -06:00
Brian Paul
7d3561c871 docs: more bug fixes for Mesa 7.4.1 2009-04-07 08:37:22 -06:00
Brian Paul
5c47d53924 mesa: in mesa_add_named_constant(), avoid adding duplicate constants
(cherry picked from master, commit 80197a0c1b)
2009-04-07 08:35:01 -06:00
Brian Paul
a18216308a mesa: fix parameter counting in ARB vertex/fragment program parsing
Duplicated unnamed constants were getting counted more than once.

(cherry picked from master, commit 866bdd0509)
2009-04-07 08:34:32 -06:00
Brian Paul
14f13fbcfa mesa: replace >= with > when testing if we've exceeded max local params
Now a program that uses 256 locals works as it should.

(cherry picked from master, commit a4173956eb)
2009-04-07 08:34:04 -06:00
Brian Paul
ca24095c1e mesa: only clear matrix MAT_DIRTY_INVERSE flag when we actually compute the inverse
If _math_matrix_analyse() got called before we allocated the inverse
matrix array we could lose the flag indicating that we needed to compute
the inverse.  This could happen with certain vertex shader cases.

(cherry picked from master, commit ce461ffc5a)
2009-04-07 08:33:43 -06:00
Brian Paul
d5e0e03d43 mesa: for OPCODE_LIT, use _mesa_pow() instead of exp() and log()
Also, s/pow/_mesa_pow/

(cherry picked from master, commit b8a200ac9d)
2009-04-07 08:32:42 -06:00
Michel Dänzer
775ca8e3fa radeon: Expose a 32 bit RGBA fbconfig even when the screen depth is 16.
Otherwise current xserver / libGL no longer expose a 32 bit RGBA GLX visual,
and compiz fails.

Fixes http://bugs.freedesktop.org/show_bug.cgi?id=20479 .

(Cherry picked from commit e798bf8053)
2009-04-06 09:27:17 +02:00
Brian Paul
39345a4928 mesa: don't normalize spot light direction until validation time
In glLight() we're only supposed to transform the direction by the modelview
matrix, not normalized it too.

(cherry picked from commit 650d147289)
2009-04-03 10:16:31 -06:00
Roland Scheidegger
7be149cfd1 mesa: fix TexParameter functions
premature return in TexParameterf caused mesa to never call Driver.TexParameter
breaking drivers relying on this (fix bug #20966).
While here, also fix using ctx->ErrorValue when deciding to call
Driver.TexParameter. Errors are sticky and uncleared errors thus would cause
this to no longer get called. Since we thus need return value of
set_tex_parameter[if] can also optimize this to only call when value changed.
2009-04-03 00:00:40 +02:00
Brian Paul
8c9d7b2474 docs: fix TXB texture unit bias error 2009-04-02 13:11:56 -06:00
Brian Paul
6ef0951b17 mesa: don't call ctx->Driver.ReadPixels() if width or height is zero
(cherry picked from commit master, 7b9bf39543)
2009-04-02 13:11:33 -06:00
Brian Paul
2b33b77337 mesa: don't call ctx->Driver.Draw/CopyPixels() if width or height is zero
(cherry picked from master, commit f6a3f1f52a)
2009-04-02 13:10:41 -06:00
Brian Paul
0590edeea0 mesa: use correct tex unit lod bias for TXB instruction
(cherry picked from master, commit 1ab225017e)
2009-04-02 13:09:32 -06:00
Brian Paul
7511d76bd7 swrast: fix glDrawBuffer(GL_FRONT_AND_BACK)
We weren't putting the right colors into the back buffer in this mode.

(cherry picked from master, commit 9cc79fc2dc)
2009-04-02 13:07:53 -06:00
Julien Cristau
99f159339e Prepare changelog for upload 2009-04-01 20:25:25 +02:00
Brian Paul
7f30aba920 docs: fix bug 20986 2009-04-01 07:39:43 -06:00
Brian Paul
4fa700260e glsl: fix texgen state variable tokens in emit_statevars()
This fixes broken variable indexing into the gl_Eye/ObjectPlaneS/T/R/Q arrays.
See bug 20986.

(cherry picked from master, commit f8dd6594bf)
2009-04-01 07:39:26 -06:00
Brian Paul
781fb79c59 docs: first 7.4.1 bug fixes 2009-03-31 16:29:23 -06:00
Brian Paul
5b7e9f2f3a mesa: fix bug in GPU codegen for fixed-function two-sided lighting
The 'dots' register wasn't getting properly un-negated and un-swizzled
after emitting the code for back-face lighting.  So, if more than one
light source was enabled, the specular exponent for the next light source
was wrong.

During execution we were evaluating pow(x, y) where y was negative instead
of positive.  This led to the outcome being zero or NaN.

This fixes the occasional black triangles seen in isosurf when hacked to
enable two-sided lighting.

(cherry picked from master, commit 919f57078b)
2009-03-31 16:28:17 -06:00
Brian Paul
d4d4b63d7f docs: prep for Mesa 7.4.1 release notes 2009-03-31 16:26:24 -06:00
Jeremy Huddleston
9e9fe51acd Updated CPU_TO_LE32 to work on darwin 2009-03-31 14:59:28 -07:00
Jeremy Huddleston
b65bc1b6cb Fix compiling indirect.c when GLX_DIRECT_RENDERING is not defined 2009-03-31 14:48:19 -07:00
Jeremy Huddleston
6cb796f6fc Updated darwin config for when X11 is not in the same location as we're installing to
(cherry picked from commit 7817fea0b6)
2009-03-31 14:37:27 -07:00
Timo Aaltonen
4830fab603 update the changelog. 2009-03-30 10:41:39 +03:00
Timo Aaltonen
1e142bf63d Merge commit 'mesa_7_4' into debian-experimental 2009-03-30 10:38:45 +03:00
Brian Paul
de197cf991 docs: 7.4 release md5 sums 2009-03-27 19:12:16 -06:00
Brian Paul
bf1e6581a5 docs: set 7.4 release date 2009-03-27 18:57:39 -06:00
Brian Paul
80ec7843bc mesa: set VERSION=7.4 2009-03-27 18:57:23 -06:00
Brian Paul
366d6e8cd1 mesa: set version to 7.4 2009-03-27 18:57:07 -06:00
Ian Romanick
63ed288396 intel: Add extra, stronger flushes around CopyTexSubImage blits 2009-03-27 15:22:27 -07:00
Eric Anholt
88ce34f400 swrast: Add support for x8r8g8b8 fbconfig.
This lets swrast produce an fbconfig suitable for the root visual now that
the server's not allowing mismatched fbconfigs.
(cherry picked from commit 529d1d720e)
2009-03-27 15:20:09 -07:00
Ian Romanick
f535de8721 Mark current bits as 2009Q1-RC2 for Intel driver. 2009-03-26 15:32:29 -07:00
Eric Anholt
1db913f687 i965: Fix glFrontFacing in twoside GLSL demo.
This also cuts instructions by just using the existing bit in the payload
rather than computing it from the determinant in the SF unit and passing it
as a varying down to the WM.  Something still goes wrong with getting the
backface color right, but a simpler shader appears to get the right result.
(cherry picked from commit 699db6d842)
2009-03-26 15:15:51 -07:00
Eric Anholt
0e5d1f43af i965: Fix fog coordinate g,b,a values when glFrontFacing isn't used.
Previously, we would sample (f,glFrontFacing,undef,undef) instead of the
(f,0,0,1) that fragment.fogcoord is supposed to return.  Due to
glFrontFacing's presence in FOGC.y, we'll still give bad results there when
glFrontFacing is used.

Bug #19122, piglit testcase fp-fog.
(cherry picked from commit 411d913cce)
2009-03-26 15:15:27 -07:00
Eric Anholt
09c587ae55 i965: Clean up a bit of mess with unneeded variables in emit_interp.
(cherry picked from commit b013f945d8)
2009-03-26 15:15:14 -07:00
Brian Paul
775d8431e9 i965: separate emit_op() and emit_tex_op() functions
(cherry picked from commit a79186e29e)
2009-03-26 15:14:56 -07:00
Brian Paul
3217ea8a1d i965: minor clean-up, comments, etc.
(cherry picked from commit 9e7903e492)
2009-03-26 15:14:22 -07:00
Brian Paul
00aea5853d docs: mention Solaris updates in release notes 2009-03-25 21:08:21 -06:00
Brian Paul
20d34b50b5 mesa: bump version to 7.4 rc2 2009-03-25 21:06:41 -06:00
Alan Coopersmith
9cff0acb36 define __builtin_expect for non-gcc compilers in two more glx files
Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
2009-03-25 16:26:03 -07:00
Alan Coopersmith
791b59493f mklib improvements for Solaris
Move flags for linking standard C/C++ libraries from configure.ac to mklib
Use -norunpath flag when linking with Sun C++ compiler
Convert mklib -exports list into a linker mapfile
Set FINAL_LIBS correctly when -noprefix is used

Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
2009-03-25 16:26:02 -07:00
Michel Dänzer
7dff7029fa Revert "dri2: Avoid round-tripping on DRI2GetBuffers for the same set of buffers."
This scheme breaks when the display connection doesn't receive ConfigureNotify
events. This caused reporoducible problems (cropped / misplaced output) when
starting a 3D application in a guest operating system in VMware Workstation.

This reverts commit dd1c68f151.

Conflicts:

	src/glx/x11/dri2_glx.c
(cherry picked from commit 11da7e02aa)
2009-03-25 15:55:20 +01:00
Brian Paul
f8fd9bc46b docs: update glext.h to version 48 2009-03-25 08:35:15 -06:00
Brian Paul
53f0e141c0 gl: update include/GL/glext.h to version 48 2009-03-25 08:35:02 -06:00
Julien Cristau
af3c568475 Prepare changelog for upload 2009-03-25 11:34:58 +01:00
Julien Cristau
7d01429512 Build-depend on linux-libc-dev >= 2.6.29 on linux archs.
The 2.6.28 kernel headers miss some compat defines.
2009-03-25 01:51:02 +01:00
Brian Paul
14a794300e demos: add LDFLAGS for corender target, bug 20844
(cherry picked from commit c4c4358e08)
2009-03-24 16:09:23 -06:00
Eric Anholt
912ec292c1 i965: Fix occlusion query when no other WM state updates occur.
Turns out that XXX comment was important.  We weren't flagging the WM to
re-update with the statistics enable, so we got zeroes out of our query.

Bug #20740, fixes piglit occlusion_query test.

Signed-off-by: Eric Anholt <eric@anholt.net>
(cherry picked from commit c0d6e07909)
2009-03-23 12:49:31 -07:00
Tom Fogal
e814b5ae2f mesa: fix _glxapi_get_proc_address() for mangled names
(cherry picked from commit 7399d56ec6)
2009-03-23 12:49:13 -07:00
Brian Paul
0fea05915a glapi: update find_entry() for mangled names
(cherry picked from commit 4447fddc82)
2009-03-23 12:49:05 -07:00
Brian Paul
dfa44f0ff9 mesa: regenerated gl_mange.h file
(cherry picked from commit fe0ccf323d)
2009-03-23 12:48:56 -07:00
Julien Cristau
8d088cd34f update changelog 2009-03-23 17:25:04 +01:00
Julien Cristau
094d627885 Merge tag 'mesa_7_4_rc1' into debian-experimental 2009-03-23 17:21:26 +01:00
Brian Paul
f850ff2bcc mesa: remove unused 'out' label 2009-03-23 08:27:00 -06:00
Brian Paul
66f972ab55 Add Solaris to OS'es using PROT_EXEC mmap() to get executable heap space
Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>

(cherry picked from master, commit a06dd4de87)
2009-03-23 08:18:48 -06:00
Brian Paul
17db2db9dc docs: prep for 7.4 release 2009-03-20 17:34:54 -06:00
Brian Paul
dba79af9bc mesa: prep for 7.4-rc1 release 2009-03-20 17:26:04 -06:00
Brian Paul
b009a32bf4 r300: don't crash on sw tcl hw if point size vertex attrib is sent
(cherry picked from master, commit 005ad1a71d)
2009-03-20 08:49:39 -06:00
Brian Paul
7122490982 mesa: avoid setting texObj->_Complete = GL_FALSE when there's no state change
Avoid a little bit of unneeded state validation and fixes a bug where the
texture complete flags was set to false, but we didn't signal _NEW_TEXTURE.

Fixes piglit tex1d-2dborder failure.
(cherry picked from commit aad3f546a0)
2009-03-18 14:47:47 -07:00
Brian Paul
6ada1d47d9 mesa: add no-change testing for a few more texture parameters
(cherry picked from commit 37c768b36a)
2009-03-18 10:54:09 -07:00
Robert Ellison
55865335f4 i965: fix polygon stipple when rendering to FBO
The polygon stipple pattern, like the viewport and the
polygon face orientation, must be inverted on the i965
when rendering to a FBO (which itself has an inverted pixel
coordinate system compared to raw Mesa).

In addition, the polygon stipple offset, which orients
the stipple to the window system, disappears when rendering
to an FBO (because the window system offset doesn't apply,
and there's no associated FBO offset).

With these fixes, the conform triangle and polygon stipple
tests pass when rendering to texture.
(cherry picked from commit 29309b45b0)
2009-03-18 10:47:24 -07:00
Robert Ellison
918e5221ef i965: fix polygon face orientation when rendering to FBO
In the i965, the FBO coordinate system is inverted from the standard
OpenGL/Mesa coordinate system; that means that the viewport and the
polygon face orientation have to be inverted if rendering to a FBO.

The viewport was already being handled correctly; but polygon face
was not.  This caused a conform failure when rendering to texture with
two-sided lighting enabled.

This fixes the problem in the i965 driver, and adds to the comment about
the gl_framebuffer "Name" field so that this isn't a surprise to other
driver writers.
(cherry picked from commit 6dceeb2eb8)
2009-03-18 10:47:24 -07:00
Brian Paul
9feb26584a swrast: use better _swrast_compute_lambda() function
The MAX-based function can produce values that are non-monotonic for a span
which causes glitches in texture filtering.  The sqrt-based one avoids that.

This is perhaps slightly slower than before, but the difference
probably isn't noticable given we're doing software mipmap filtering.

Issue reported by Nir Radian <nirr@horizonsemi.com>

(cherry picked from master, commit c334ce273e)
2009-03-17 10:34:45 -06:00
Ian Romanick
a8528a2e86 Mark current bits as 2009Q1-RC1 for Intel driver. 2009-03-16 13:39:32 -07:00
Brian Paul
119360cccd i965: init dest reg CondMask = COND_TR (the proper default)
Plus fix up a debug printf.
(cherry picked from commit 20f49252e1)
2009-03-13 17:38:40 -07:00
Brian Paul
ac2216542d glsl: fix vec4_texp_rect IR code (need projective version) (cherry picked from commit ad2cfa4199) 2009-03-13 17:38:40 -07:00
Brian Paul
37e5c057f1 i965: check if we run out of GRF/temp registers
Before this change we would up emitting instructions with invalid register
numbers.  This typically (but not always) hung the GPU.  For now, just
prevent emitting bad instructions to avoid hangs.  Still need to do some
kind of proper error recovery.
(cherry picked from commit e60b3067d8)
2009-03-13 17:38:40 -07:00
Brian Paul
4d9b8e0f93 mesa: added _mesa_fprintf() wrapper (cherry picked from commit 596b8fbbbf) 2009-03-13 17:38:40 -07:00
Brian Paul
c1b71f46b5 i965: fix emit_math1() function used for scalar instructions
Instructions such as RCP, RSQ, LOG must smear the result of the function
across the dest register's X, Y, Z and W channels (subject to write masking).
Before this change, only the X component was getting written.

Among other things, this fixes cube map texture sampling in GLSL shaders
(since cube lookups involve normalizing the texcoord).
(cherry picked from commit 3485801978)
2009-03-13 17:38:39 -07:00
Brian Paul
4a25ac9564 i965: rewrite the code for handling shader subroutine calls
Previously, the prog_instruction::Data field was used to map original Mesa
instructions to brw instructions in order to resolve subroutine calls.  This
was a rather tangled mess.  Plus it's an obstacle to implementing dynamic
allocation/growing of the instruction buffer (it's still a fixed size).

Mesa's GLSL compiler emits a label for each subroutine and CAL instruction.
Now we use those labels to patch the subroutine calls after code generation
has been done.  We just keep a list of all CAL instructions that needs patching
and a list of all subroutine labels.  It's a simple matter to resolve them.

This also consolidates some redundant post-emit code between brw_vs_emit.c and
brw_wm_glsl.c and removes some loops that cleared the prog_instruction::Data
fields at the end.

Plus, a bunch of new comments.
(cherry picked from commit c51c822ee0)
2009-03-13 17:38:39 -07:00
Eric Anholt
35d965b39b intel: Fix bpp setting of blits to 8bpp targets.
This was causing hangs in cairogears, as we would blit to the 8bpp target
(A8 texture) as 16bpp, and stomp over state objects.
(cherry picked from commit 19e134051c)
2009-03-13 17:38:39 -07:00
Robert Ellison
d0aab00609 i965: add software fallback for conformant 3D textures and GL_CLAMP
The i965 hardware cannot do GL_CLAMP behavior on textures; an earlier
commit forced a software fallback if strict conformance was required
(i.e. the INTEL_STRICT_CONFORMANCE environment variable was set) and
2D textures were used, but it was somewhat flawed - it could trigger
the software fallback even if 2D textures weren't enabled, as long
as one texture unit was enabled.

This fixes that, and adds software fallback for GL_CLAMP behavior with
1D and 3D textures.

It also adds support for a particular setting of the INTEL_STRICT_CONFORMANCE
environment variable, which forces software fallbacks to be taken *all*
the time.  This is helpful with debugging.  The value is:
   export INTEL_STRICT_CONFORMANCE=2
(cherry picked from commit 3468315087)
2009-03-13 17:38:39 -07:00
Brian Paul
ef33d0f103 mesa: don't draw arrays if vertex position array is not enabled
For regular GL, we must have vertex positions in order to draw.  But ES2
doesn't have that requirement (positions can be computed from any array
of data).

See bug 19911.
(cherry picked from commit 97dd2ddbd9)
2009-03-13 17:38:39 -07:00
Robert Ellison
349b819768 i965: texture fixes: bordered textures, fallback rendering
i965 doesn't natively support GL_CLAMP; it treats it like
GL_CLAMP_TO_EDGE, which fails conformance tests.

This fix adds a clause to the check_fallbacks() test to check
whether GL_CLAMP is in use on any enabled 2D texture.  If so,
and if strict conformance is required (via INTEL_STRICT_CONFORMANCE),
a software fallback is mandated.

In addition, validate textures *before* checking for fallbacks,
rather than after; otherwise, the texture state is never validated
and can't be trusted.  (In particular, if texturing is enabled and
the sampler would access any level beyond level 0 of a texture, the
sampler will segfault, because texture validation sets the firstLevel
and lastLevel fields of a texture object so that the valid levels
will be mapped and accessed correctly.  If texture validation doesn't
occur, only level 0 is accessed correctly, and that only because
firstLevel and lastLevel happen to be set to 0.)
(cherry picked from commit 17c7852bf9)
2009-03-13 17:38:39 -07:00
Dave Airlie
50443db882 texmem: fix typo from brianp's changes.
Reported by cjb via tinderbox on irc
(cherry picked from commit 487a55af78)
2009-03-13 17:38:39 -07:00
Brian Paul
c503cacb53 mesa: use an array for current texture objects
Use loops to consolidate lots of texture object code.
(cherry picked from commit 9818734e01)
2009-03-13 17:38:39 -07:00
Brian Paul
e862b4e38f mesa: use an array for default texture objects
Replace Default1D/2D/3D/Cube/etc with DefaultTex[TEXTURE_x_INDEX].
The same should be done with the Current1D/2D/3D/etc pointers...
(cherry picked from commit 4d24b639d1)
2009-03-13 17:38:39 -07:00
Robert Ellison
e9dfc858fb Fix an i965 assertion failure on glClear()
While running conform with render-to-texture:

	conform -d 33 -v 2 -t -direct

the i965 driver failed this assertion:

intel_clear.c:77: intel_clear_tris: Assertion `(mask & ~((1 << BUFFER_BACK_LEFT) | (1 << BUFFER_FRONT_LEFT) | (1 << BUFFER_DEPTH) | (1 << BUFFER_STENCIL))) == 0' failed.

The problem is that intel_clear_tris() is called by intelClear() to
clear any and all of the available color buffers, but intel_clear_tris()
actually only handles the back left and front left color buffers; so
the assertion fails as soon as you try to clear a non-standard color
buffer.

The fix is to have intelClear() only call intel_clear_tris() with
buffers that intel_clear_tris() can support.  intelClear() already backs
down to _swrast_Clear() for all buffers that aren't handled explicitly.
(cherry picked from commit 0ccbc3c905)
2009-03-13 17:38:39 -07:00
Brian Paul
5efbca1e21 intel: fix datatype typo, s/GLboolean/GLuint/
Fixes mysterious failures in glean glsl1 test.
(cherry picked from commit da2b661ee4)
2009-03-13 17:38:39 -07:00
Kristian Høgsberg
d6d8663a88 intel: Fix intelSetTexBuffer miptree leak.
The intelImage also holds a reference to the miptree, so unref that as well.
(cherry picked from commit 5b354d39d4)
2009-03-13 17:38:39 -07:00
Eric Anholt
13990bd98a intel: tell libdrm whether we want a cpu-ready or gpu-ready BO for regions.
This lets us avoid allocing new buffers for renderbuffers, finalized miptrees,
and PBO-uploaded textures when there's an unreferenced but still active one
cached, while also avoiding CPU waits for batchbuffers and CPU-uploaded
textures.  The size of BOs allocated for a desktop running current GL
cairogears on i915 is cut in half with this.

Note that this means we require libdrm 2.4.5.
(cherry picked from commit 40dd024be6)
2009-03-13 17:38:39 -07:00
Eric Anholt
def630d0a2 i965: Fix render target read domains.
We were asking for something illegal (write_domain != 0 && read_domains !=
write_domain) because at the time of writing the region surfaces were used
for texturing occasionally as well, and we weren't really clear on the model
GEM was going to use.

This reliably triggered a kernel bug with domain handling, resulting in
oglconform mustpass.c failure.  Of course, it only became visible after
01bc4d441f cleaned up some gratuitous flushing.
(cherry picked from commit 078e8a61b2)
2009-03-13 17:38:39 -07:00
Eric Anholt
ca2564f1e6 intel: Don't do the extra MI_FLUSH in flushing except when doing glFlush().
Everything other than "make sure the last rendering ends up visible on the
screen" doesn't need that behavior.
(cherry picked from commit 01bc4d441f)
2009-03-13 17:38:38 -07:00
Eric Anholt
711a57f7c7 intel: don't crash when dri2 tells us about buffers we don't care about. (cherry picked from commit f82f1ffba9) 2009-03-13 17:38:38 -07:00
Eric Anholt
058cf2fff1 dri2: Initialize variables for the getbuffers round-trip reduction.
Missed setting the initial values which usually didn't hurt at runtime.
(cherry picked from commit 680c708dee)
2009-03-13 17:38:38 -07:00
Eric Anholt
715f509c5d intel: Fix tri clear to do FBO color attachments as well.
This is a 2% win in fbo_firecube, and would avoid a sw fallback for
masked clears.
(cherry picked from commit fd51cf1531)
2009-03-13 17:38:38 -07:00
Eric Anholt
e704e6c6f3 i965: Fix fallback on stencil drawing to fbo when the visual lacks stencil.
Noticed this with the fbotexture demo.
(cherry picked from commit c06f4e2a37)
2009-03-13 17:38:38 -07:00
Brian Paul
01f34e667a i965: need to disable current shader, if any, in intel_clear_tris()
Fixes bad background in all the progs/glsl/ tests.
(cherry picked from commit 60b3fe6c19)
2009-03-13 17:38:38 -07:00
Eric Anholt
4d0a7b33c5 intel: Speed up glDrawPixels(GL_ALPHA) by using an alpha texture format. (cherry picked from commit 0b63f6449e) 2009-03-13 17:38:38 -07:00
Eric Anholt
fb1d2ed34c intel: Fix some state leakage of {Client,}ActiveTexture in metaops.
Found while debugging cairo-gl.
(cherry picked from commit d11981e0d7)
2009-03-13 17:38:38 -07:00
Eric Anholt
9395fedf98 i965: Remove brw->attribs now that we can just always look in the GLcontext. (cherry picked from commit 052c1d66a1) 2009-03-13 17:38:38 -07:00
Eric Anholt
9abc0d80af i965: Delete old metaops code now that there are no remaining consumers. (cherry picked from commit 14321fcfde) 2009-03-13 17:38:38 -07:00
Brian Paul
90e89c01df i965: init array->Format fields (see bug 19708) (cherry picked from commit 628b52241b) 2009-03-13 17:38:38 -07:00
Eric Anholt
790d93e362 intel: If we're doing a depth clear with tris, do color with it.
This is a 10% win on the ever-important glxgears not-a-benchmark.
(cherry picked from commit 67ee22c89f)
2009-03-13 17:38:38 -07:00
Eric Anholt
e9f3783169 i915: Only call CalcViewport from DrawBuffers instead of Viewport.
This saves an inadvertent round-trip to the X Server on DrawBuffers, which was
hurting some metaops.
(cherry picked from commit ac0dfbdf0f)
2009-03-13 17:38:38 -07:00
Brian Paul
e41780fedc intel: move some driver functions around
A step toward consolidating i915/intel_state.c and i965/intel_state.c
(cherry picked from commit 84c8b5bbf9)
2009-03-13 17:38:38 -07:00
Eric Anholt
d0edbbb3f3 intel: replace custom metaops clear with generic.
No real-world impact on performance seen.  Even glxgears seems to be, if
anything, happier.
(cherry picked from commit c96bac0950)
2009-03-13 17:38:38 -07:00
Brian Paul
e10119c23e intel: make intelUpdateScreenFromSAREA() static (cherry picked from commit aae2729aeb) 2009-03-13 17:38:38 -07:00
Brian Paul
208c087f39 intel: remove unused var (cherry picked from commit 66c7f06413) 2009-03-13 17:38:38 -07:00
Shaohua Li
0c8b40b42b i915: Add support for a new G33-like chipset.
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
(cherry picked from commit 40290745ea)
2009-03-13 17:38:37 -07:00
Xiang, Haihao
5199f451bb i965: fix for RHW workaround
It is possible that an object whose vertices all are outside of a
view plane is passed to clip thread due to the RHW workaround. This
object should be rejected by clip thread. Fix bug #19879
(cherry picked from commit 68915fd6fa)
2009-03-13 17:38:37 -07:00
Brian Paul
dd7cde5230 i965: tell GLSL compiler to emit code using condition codes
The default for EmitCondCodes got flipped when gallium-0.2 was merged.
This fixes GLSL if/else/endif regressions.
Drivers that use GLSL should always explicitly set the flag to be safe.
(cherry picked from commit a9e753c84c)
2009-03-13 17:38:37 -07:00
Eric Anholt
ab5746e5d8 tdfx: Fix begin/endquery for current API. (cherry picked from commit 0cb295584f) 2009-03-13 17:38:37 -07:00
Brian Paul
3bd446c0ab i965: add missing break for OPCODE_RET case
This doesn't effect correctness, but we were emitting an extraneous ADD.
(cherry picked from commit 74b6d55864)
2009-03-13 17:38:37 -07:00
Brian Paul
9b78402ee4 mesa: more info in error messages (cherry picked from commit 1ca05a066b) 2009-03-13 17:38:37 -07:00
Brian Paul
cf1ef0f308 mesa: minor error msg improvement (cherry picked from commit 803504e69f) 2009-03-13 17:38:37 -07:00
Chris Wilson
fd1f65ba61 intel: Decode MI operands using specific length masks
The MI opcodes have different variable length masks, so use an operand
specific mask to decode the length.
(cherry picked from commit e92d97d75b)
2009-03-13 17:38:37 -07:00
Chris Wilson
967345b497 intel: Correct decoding of 3DSTATE_PIXEL_SHADER_CONSTANTS
A couple of minor typos that proclaimed an error in the wrong command, and
failed to offset the mask.
(cherry picked from commit 05d130a35a)
2009-03-13 17:38:37 -07:00
Ian Romanick
41ad853a6a glxgears: No, really. Fix the dyslexia. (cherry picked from commit 5b5ddfb89c) 2009-03-13 17:38:37 -07:00
Ian Romanick
842baa29cb Fix dyslexia. (cherry picked from commit 9d6880ec8d) 2009-03-13 17:38:37 -07:00
Ian Romanick
9556d870a1 glxgears: Log a message if synched to vblank
Tries to use either GLX_MESA_swap_control or GLX_SGI_video_sync to
detect whether the display is synchronized to the vertical blank.  If
it detects this, a message will be printed.  HOPEFULLY this will
prevent some of the bug reports such as "glxgears only gets 59.7fps.
What's wrong with my driver?"
(cherry picked from commit 58b9cd411f)
2009-03-13 17:38:37 -07:00
Brian Paul
42ce790086 mesa: refactor glTexParameter code (cherry picked from commit 318e53a4bf) 2009-03-13 17:38:37 -07:00
Robert Ellison
b17769af42 mesa: add missing texture_put_row_rgb() function in texrender.c
The wrap_texture() function doesn't set the renderbuffer PutRowRGB() method,
which is used to implement DrawPixels().  This fix adds an implementation
of this method.
(cherry picked from commit 523febe12e)
2009-03-13 17:38:37 -07:00
Brian Paul
9030277b0d intel: move glClear-related code into new intel_clear.c file (cherry picked from commit 4451eb2e75) 2009-03-13 17:38:37 -07:00
Brian Paul
a37a9bf769 intel: Move swap-related functions from intel_buffers.c to new intel_swapbuffers.c (cherry picked from commit 6fcebbe719) 2009-03-13 17:38:37 -07:00
Brian Paul
8afb8a8145 i965: scissor rect was inverted when rendering to texture (cherry picked from commit 3b23a8e07d) 2009-03-13 17:38:37 -07:00
Brian Paul
801f933694 mesa: consolidate glGetTexEnvi/f() code with new get_texenvi() helper (cherry picked from commit ad338c14c2) 2009-03-13 17:38:36 -07:00
Jesse Barnes
75ecd6ebbf intel: fix the mismerge of the vblank pipe enable sanity check
Fix the last merge fix, had the blocks ordered incorrectly.
(cherry picked from commit e57e398960)
2009-03-13 17:38:36 -07:00
Jesse Barnes
1cd7317fdb intel: move pipe enable sanity check to where it belongs
Bah, applied the patches in the wrong order, not Owain's fault...
(cherry picked from commit 216bff5fd4)
2009-03-13 17:38:36 -07:00
Owain Ainsworth
1865291701 intel: fix vblank crtc selection with DRI1 when only one pipe is enabled.
On Mobile chipsets, we often enable PipeB instead of PipeA, but the test
in here was insufficient, falling back to pipe A if the area
intersection returned zero. Therefore, in the case where a window went
off to the top of the left of the screen, it would freeze, waiting on
the wrong vblank.

Fix this mess by checking the sarea for a crtc being zero sized, and in
that case always default to the other one.
(cherry picked from commit 0b5266ff64)
2009-03-13 17:38:36 -07:00
Owain G. Ainsworth
3edbcce7d5 intel: Prevent an "irq is not working" printf when only pipe B is enabled.
intelMakeCurrent is called before intelWindowMoved (in fact, it calls
it), so calculation of the correct vblank crtc has not happened yet.
Fix this by making a function that fixes up a set of vblank flags and
call if from both functions.
(cherry picked from commit 39b4061bb9)
2009-03-13 17:38:36 -07:00
Robert Ellison
b1fb34d0ac i965: fix line stipple fallback for GL_LINE_STRIP primitives
When doing line stipple, the stipple count resets on each line segment,
unless the primitive is a GL_LINE_LOOP or a GL_LINE_STRIP.

The existing code correctly identifies the need for a software fallback
to handle conformant line stipple on GL_LINE_LOOP primitives, but
neglects to make the same assessment on GL_LINE_STRIP primitives.
This fixes it so they match.
(cherry picked from commit 73658ff04f)
2009-03-13 17:38:36 -07:00
Ian Romanick
606099f545 swrast: Enable GL_EXT_stencil_two_side (cherry picked from commit b5fbdef7ec) 2009-03-13 17:38:36 -07:00
Brian Paul
32e3142071 i965: minor improvements in brw_wm_populate_key() (cherry picked from commit 89fddf978c) 2009-03-13 17:38:36 -07:00
Brian Paul
66eea4e104 mesa: fix/update/restore comments related to two-sided stencil (cherry picked from commit a304cc6cca) 2009-03-13 17:38:36 -07:00
Brian Paul
e428cf3afd mesa: initialize ctx->Stencil._BackFace = 1
Back-face stencil operations didn't work correctly because this value was
zero.  It needs to be 1 or 2.  The only place it's set otherwise is in
glEnable/Disable(GL_STENCIL_TEST_TWO_SIDE_EXT).
(cherry picked from commit 2a968113a9)
2009-03-13 17:38:36 -07:00
Ian Romanick
842b5164d2 Track two sets of back-face stencil state
Track separate back-face stencil state for OpenGL 2.0 /
GL_ATI_separate_stencil and GL_EXT_stencil_two_side.  This allows all
three to be enabled in a driver.  One set of state is set via the 2.0
or ATI functions and is used when STENCIL_TEST_TWO_SIDE_EXT is
disabled.  The other is set by StencilFunc and StencilOp when the
active stencil face is set to BACK.  The GL_EXT_stencil_two_side spec has
more details.

http://opengl.org/registry/specs/EXT/stencil_two_side.txt
(cherry picked from commit dde7cb9628)
2009-03-13 17:38:36 -07:00
Brian Paul
a6f7e909a7 mesa: fix transposed red/blue in store_texel_rgb888/bgr888() functions
(cherry picked from master, commit 862dccd560)
2009-03-12 09:32:55 -06:00
Brian Paul
ca87e5a7bd mesa: add missing _glthread_INIT_MUTEX in _mesa_new_framebuffer()
(cherry picked from master, commit 81569c2f69)
2009-03-12 08:12:54 -06:00
Alan Hourihane
45c4b4dfbd xdemos: On termination with esc close the contexts correctly. 2009-03-11 13:36:26 +00:00
Brian Paul
2adaec1226 xmesa: set back-buffer's drawable field
Fixes back-buffer rendering when MESA_BACK_BUFFER=pixmap

(cherry picked from master, commit 22bac2a1a0)
2009-03-09 16:28:50 -06:00
Brian Paul
549586c319 i965: fix cube map lock-up / corruption
If we're using anything but GL_NEAREST sampling of a cube map, we need to
use the BRW_TEXCOORDMODE_CUBE texcoord wrap mode.  Before this, the GPU
would either lock up or subsequent texture filtering would be corrupted.

(cherry picked from master, commit 6f915b10d5)
2009-03-09 11:57:16 -06:00
Dan Nicholson
463ac421a5 autoconf: Restore _GNU_SOURCE for all Linux systems
This catches the linux-uclibc case and any others that were being set
prior to 98fcdf3f. Fixes bug 20345.

Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
(cherry picked from commit ac55db1d7d)
2009-03-06 07:59:00 -08:00
Brian Paul
6801240205 mesa: Reads must also be done with lock held.
Otherwise two threads might think each made the refcount go zero.

(cherry picked from master, commit 8bbb6b352a)
2009-03-06 07:51:18 -07:00
Brian Paul
a22088b7d5 mesa: Fix typo.
Windows threads block if one over-unlocks them.

(cherry picked from master, commit f6159ba4d5)
2009-03-06 07:50:55 -07:00
Brian Paul
e89c49147f docs: update glext.h to version 46 2009-03-05 08:25:17 -07:00
Brian Paul
67634240d1 gl: update glext.h to version 46 2009-03-05 08:25:09 -07:00
Brian Paul
819b028a8e mesa: fix sw fallback state validation bug
When a hw driver fell back to swrast, swrast wasn't always getting informed
of program changes.  When fixed function is translated into shaders, flags
like _NEW_LIGHT, _NEW_TEXTURE, etc. should really signal _NEW_PROGRAM.
In this case, swrast wasn't seeing _NEW_PROGRAM when new fragment shaders
were generated.

(cherry picked from master, commit de1caa5507)
2009-03-03 14:25:28 -07:00
Brian Paul
a43c30ef41 mesa: update fragResults array in arb_output_attrib_string()
Plus add some comments.

(cherry picked from master, commit a070937c00)
2009-02-27 13:49:58 -07:00
Brian Paul
b65bfde84d mesa: set bufObj->Pointer = NULL after unmapping
Also, ctx->Driver.UnmapBuffer can never be null, so remove conditional.

(cherry picked from master, commit 67025f7893)
2009-02-27 13:13:31 -07:00
Brian Paul
c39e6917ac mesa: if a buffer object is mapped when glDeleteBuffers() is called, unmap it
(cherry picked from master, commit a7f434b486)
2009-02-27 13:05:51 -07:00
Brian Paul
c79079895f mesa: fix incorrect error handling in glBufferDataARB()
If glBufferDataARB() is called while a buffer object is currently mapped
we're supposed to unmap the current buffer, then replace it.  Don't generate
an error.

(cherry picked from master, commit 75e3ccf6a5)
2009-02-27 12:58:07 -07:00
Alex Deucher
4480e631cd R300: Add support for RS600 chips 2009-02-25 17:48:45 -05:00
Brian Paul
c32661c3ce mesa: Build DRI by default on Linux/sparc
Signed-off-by: David S. Miller <davem@davemloft.net>

(cherry picked from master, commit 32dc28ac7a)
2009-02-24 20:08:00 -07:00
Brian Paul
d5c06bb812 glsl: yet another swizzled expression fix
This fixes swizzled conditional expressions such "(b ? p : q).x"

(cherry picked from master, commit 00f0b05d5f)
2009-02-24 08:32:01 -07:00
Brian Paul
9d5aa36239 glsl: fix another swizzle-related bug
This fixes the case of "infinitely" nested swizzles such as EXPR.wzyx.yxwz.xxyz
This doesn't appear in typical shaders but with function inlining and the
compiler's internal use of swizzles it can happen.
New glean glsl1 test case added for this.

(cherry picked from master, commit d9881356a6)
2009-02-24 08:31:02 -07:00
Brian Paul
e3050c1777 docs: Haiku fixes 2009-02-23 08:16:09 -07:00
Brian Paul
06295ddbdd mesa: fixes for building on Haiku
(cherry picked from master, commit 41172c0430)

Conflicts:
	src/mesa/main/dlopen.c
2009-02-23 08:15:43 -07:00
Brian Paul
e2092bb23c demos: update multitex.c GLSL demo to use vertex arrays or glVertex-mode
Press 'a' to toggle drawing mode.

(cherry picked from master, commit 8e8b25c26a)
2009-02-19 17:05:14 -07:00
Brian Paul
528836d0ad util: added more functions to extfuncs.h
(cherry picked from master, commit 4c5f390328)
2009-02-19 17:04:42 -07:00
Brian Paul
3511442e43 mesa: fix GLSL issue preventing use of all 16 generic vertex attributes
Only 15 actually worked before since we always reserved generic[0] as an
alias for vertex position.

The case of vertex attribute 0 is tricky.  The spec says that there is no
aliasing between generic vertex attributes 0..MAX_VERTEX_ATTRIBS-1 and the
conventional attributes.  But it also says that calls to glVertexAttrib(0, v)
are equivalent to glVertex(v).  The distinction seems to be in glVertex-mode
versus vertex array mode.

So update the VBO code so that if the shader uses generic[0] but not gl_Vertex,
route the attribute data set with glVertex() to go to shader input generic[0].

No change needed for the glDrawArrays/Elements() path.

This is a potentially risky change so regressions are possible.  All the usual
tests seem OK though.

(cherry picked from mesa, commit dea0d4d563)

Follow-up: This patch doesn't seem to be as risky as noted above.
There have been no known regressions on Mesa/master because of it.
2009-02-19 17:01:17 -07:00
Brian Paul
d20c53b511 glsl: update program->InputsRead when referencing input attributes
This info will be used in the linker for allocating generic vertex attribs.

(cherry picked from master, commit 1cb7cd1292)
2009-02-19 17:00:06 -07:00
Brian Paul
091cc122ad glsl: asst improvements, clean-ups in set_program_uniform()
Move the is_boolean/integer_type() calls out of the loops.
Move the is_sampler_type() function near the bool/int functions.
Add a bunch of comments.

(cherry picked from master, commit c4ffbf009e)
2009-02-18 17:53:08 -07:00
Brian Paul
c983abbb9e glsl: fix inequality in set_program_uniform()
We were off by one when checking for too many uniform values.

(cherry picked from master, commit b9d8f717d2)
2009-02-18 17:52:47 -07:00
Brian Paul
a0a81204c2 mesa: bump gl.h version comment to 7.4 (bug 20170) 2009-02-18 15:37:44 -07:00
Brian Paul
409d4f839c glsl: fix link failure for variable-indexed varying output arrays
If the vertex shader writes to a varying array with a variable index,
mark all the elements of that array as being written.

For example, if the vertex shader does:

   for (i = 0; i < 4; i++)
      gl_TexCoord[i] = expr;

Mark all texcoord outputs as being written, not just the first.

Linking will fail if a fragment shader tries to read an input that's not
written by the vertex shader.  Before this fix, this linker test could fail.

(cherry picked from master, commit dac19f17f3)
2009-02-18 14:29:46 -07:00
Brian Paul
0ab36958ea mesa: improved error msg
(cherry picked from master, commit 621c999d82)
2009-02-18 13:41:12 -07:00
Brian Paul
c2445f43a3 mesa: increase MAX_UNIFORMS to 1024 (of vec4 type)
Old limit was 256.  Note that no arrays are declared to this size.
The only place we have to be careful about raising this limit is the
prog_src/dst_register Index bitfields.  These have been bumped up too.

Added assertions to check we don't exceed the bitfield in the future too.

(cherry picked from master, commit 5b2f8dc013)
2009-02-18 13:40:57 -07:00
Brian Paul
f3d1c136b7 glsl: fix a swizzle-related regression
This new issue was exposed by commit 6eabfc27f1

(cherry picked from master, commit 212f41b80f)
2009-02-18 13:37:31 -07:00
Brian Paul
dd10a8b09d docs: 7.4 relnotes: report GLSL 1.20, fixed GLSL array index bug 2009-02-17 16:39:45 -07:00
Brian Paul
dd312366da glsl: silence some uninit var warnings
(cherry picked from master, commit 9d49802b7a)
2009-02-17 16:38:30 -07:00
Brian Paul
d2fe466ef3 mesa: turn on reporting of GLSL version 1.20
The new array features, precision/invariant/centroid qualifiers, etc. were
done a while back.  The glGetString(GL_SHADING_LANGUAGE_VERSION) query returns
"1.20" now (for drivers that support it anyway).

(cherry picked from master, commit f59719c6c7)
2009-02-17 16:37:41 -07:00
Brian Paul
fe5328bfad glsl: fix mistake in a comment
(cherry picked from master, commit be8dd01678)
2009-02-17 16:36:15 -07:00
Brian Paul
c0891dde07 glsl: fix an array indexing bug
This fixes a bug found with swizzled array indexes such as in "array[index.z]"
where "index" is an ivec4.

(cherry picked from master, commit 6eabfc27f1)
2009-02-17 16:35:59 -07:00
Brian Paul
0e6d9c1a55 docs: recent 7.4 bug fixes 2009-02-16 20:35:12 -07:00
Brian Paul
5f74a66132 glsl: allow setting arrays of samplers in set_program_uniform()
Arrays of sampler vars haven't been tested much and might actually be broken.
Will need to be revisited someday.

Another fix for bug 20056.

(cherry picked from master, commit 2b4f0216bf)
2009-02-16 20:30:49 -07:00
Brian Paul
46f8b62d5f glsl: raise GL_INVALID_OPERATION for glUniform(location < -1)
location = -1 is silently ignored, but other negative values should raise
an error.

Another fix for bug 20056.

(cherry picked from master, commit 234f03e90a)
2009-02-16 20:30:06 -07:00
Brian Paul
3e3e80c431 glsl: rework _mesa_get_uniform[fi]v() to avoid using a fixed size intermediate array
(cherry picked from master, commit 4ef7a93296)
2009-02-16 20:29:36 -07:00
Brian Paul
0486d117e4 glsl: fix glUniform() array bounds error checking
If too many array elements are specified, they're to be silently ignored (don't
raise a GL error).

Fixes another issue in bug 20056.

(cherry picked from master, commit 2c1ea0720d)
2009-02-16 20:28:57 -07:00
Brian Paul
29981c14da glsl: fix incorrect size returned by glGetActiveUniform() for array elements.
Fixes one of the issues in bug 20056.

(cherry picked from master, commit 369d1859d7)
2009-02-16 20:28:14 -07:00
Brian Paul
98d5efb2c5 glsl: use _slang_var_swizzle() in a few places to simplify the code.
(cherry picked from master, commit 285b500bb7)
2009-02-16 20:26:58 -07:00
Brian Paul
77a44aafaf mesa: another fix for program/texture state validation
This fixes a regression introduced in 46ae1abbac

Break program validation into two steps, do part before texture state
validation and do the rest after:
1. Determine Vertex/Fragment _Enabled state.
2. Update texture state.
3. Determine pointers to current Vertex/Fragment programs (which may involve
generating new "fixed-function" programs).

See comments in the code for more details of the dependencies.

(cherry picked from master, commit 537d3ed6f3)
2009-02-16 20:23:17 -07:00
Brian Paul
8278c70716 mesa: fix/change state validation order for program/texture state
Program state needs to be updated before texture state since the later depends
on the former.

Fixes piglit texgen failure.  The second time through the modes (press 't'
three times) we disable the fragment program and return to conventional texture
mode.  State validation failed here because update_texture() saw stale fragment
program state.

(cherry picked from master, commit 46ae1abbac)
2009-02-16 20:22:46 -07:00
Alan Hourihane
ac0955f5e2 demos: fix glxpixmap and call glXWaitGL before calling XCopyArea. 2009-02-16 11:55:47 +00:00
Alan Hourihane
4130c35ca8 dri2: support glXWaitX & glXWaitGL by using fake front buffer. 2009-02-16 11:48:24 +00:00
Brian Paul
d440647eb8 autoconf: Use include-fixed directory with makedepend on newer GCC
On newer GCC releases, the compiler's headers have been split between
the include and include-fixed directories. Add both if the directories
exist.

Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
Signed-off-by: Julien Cristau <jcristau@debian.org>

(cherry picked from master, commit a3d223f0d2)
2009-02-13 09:10:09 -07:00
Brian Paul
7e8f2c56c0 mesa: fix logic error in computing enableBits in update_texture_state()
If we had a vertex shader but no fragment shader (i.e. fixed function) we
didn't get the right enabled texture targets.

Fixes blank/white texture problem.

(cherry picked from master, commit b46611633c)
2009-02-07 12:04:15 -07:00
Brian Paul
a0880a152f mesa: fix tnl->render_inputs_bitset setup for fragment program texcoords
Handle the case where there's no per-vertex texcoords but the fragment shader
needs texcoords.

Fixes piglit shaders/fp-generic/dph test.

(cherry picked from master, commit f6d23943cd)
2009-02-07 12:03:37 -07:00
Brian Paul
4608a9172f swrast: return (0,0,0,1) when sampling incomplete textures, not (0,0,0,0)
Fixes piglit shaders/fp-incomplete-tex test.

(cherry picked from master, commit 1df62651b2)

Conflicts:
	src/mesa/swrast/s_fragprog.c

Also updated the fetch_texel_lod(), fetch_texel_deriv() functions to match
those in git/master.
2009-02-07 12:02:22 -07:00
Brian Paul
d0e38f7ec0 glsl: replace assertion with conditional in _slang_pop_var_table()
We were hitting the assertion when we ran out of registers, which can happen.

Also, add some additional assertions and freshen up some comments.

(cherry picked from master, commit 0744805d58)
2009-02-07 11:57:43 -07:00
Brian Paul
a69bdbbbdd docs: fixed Windows build error, added MESA_GLX_FORCE_DIRECT env var 2009-02-03 15:40:16 -07:00
Brian Paul
88c97a2662 docs: document new MESA_GLX_FORCE_DIRECT env var for the Xlib driver
(cherry picked from master, commit 2f51be75c5)
2009-02-03 15:39:13 -07:00
Brian Paul
ce9c69ddbc xlib: use MESA_GLX_FORCE_DIRECT to make glXIsDirect() always return True
Some apps won't run w/ indirect rendering contexts.
Also, consolidate some context-init code in new init_glx_context() function.

(cherry-picked from master, commit 49e80bf6b1)
2009-02-03 15:38:44 -07:00
Eric Anholt
e430fcb716 dri2: Avoid round-tripping on DRI2GetBuffers for the same set of buffers.
We only wanted to request when asked for the same set of buffers when a resize
has happened.  We can just watch the protocol stream for a ConfigureNotify
and flag to do it then.

This is about a 5% win from doing two glViewport()s per frame in openarena.
2009-02-03 21:36:38 +00:00
Brian Paul
11a363e632 mesa: move code after decls. Fixes Window build failure.
(cherry picked from master, commit 92ced46eaf)
2009-02-02 07:50:31 -07:00
Julien Cristau
9a3a20b2b9 debian/rules: on lpia, only build the i915 and i965 dri drivers
Based on Ubuntu changes, modified to also build swrast.
2009-01-31 23:04:06 +01:00
Julien Cristau
4672811541 debian/control: build the dri drivers on lpia
add lpia to the Architecture field for libgl1-mesa-dri{,-dbg} to match Ubuntu.
2009-01-31 22:30:15 +01:00
Julien Cristau
3cad3d28a5 mangle upstream version in debian/watch
Fix watch file to make uscan not consider release candidates as newer than
actual releases.
2009-01-31 22:07:51 +01:00
Brian Paul
41e63fbdb2 docs: recent bug fixes for 7.4 2009-01-30 16:09:58 -07:00
Brian Paul
b543401a5c mesa: fix incorrect call to clear_teximage_fields() in _mesa_TexImage2D()
Fixes failed assertion / segfault for particular proxy texture tests.

(cherry picked from master, commit ea4b183b8c)
2009-01-30 16:06:52 -07:00
Brian Paul
8c4f08fbad mesa: add missing _mesa_reference_texobj() calls for texture array targets
(cherry picked from master, commit 7c48719a5f)
2009-01-30 16:06:21 -07:00
Brian Paul
f3081a34d2 mesa: remove incorrect refcounting adjustment in adjust_buffer_object_ref_counts()
Fixes bug 19835.  However, a more elaborate fix should be implemented someday
which uses proper reference counting for gl_array_object.

(cherry picked from master, commit 88e0b92a74)
2009-01-30 16:05:51 -07:00
Julien Cristau
cff0971f87 Prepare changelog for upload 2009-01-30 20:00:37 +01:00
Julien Cristau
5ac4c69d83 More superfluous directories 2009-01-30 19:56:29 +01:00
Julien Cristau
a314c40cc4 Document my changes in debian/changelog 2009-01-30 19:42:38 +01:00
Julien Cristau
959fd6b8f6 Add back progs/fbdev
It was added to the tarballs in 22b0b2f47e
2009-01-30 19:34:47 +01:00
Julien Cristau
937f0da04a Delete more directories that aren't in the tarball 2009-01-30 19:25:50 +01:00
Julien Cristau
73bc5967e2 Merge commit '5a458977113e4a0d3c389b03801cf6ce580d3935' into debian-experimental 2009-01-30 19:06:19 +01:00
Brian Paul
7b1d3cf392 docs: i965 fixes 2009-01-28 17:02:39 -07:00
Brian Paul
25515b557e i965: fix bug in pass0_precalc_mov()
Previously, "in-place" swizzles such as:
   MOV t, t.xxyx;
were handled incorrectly.  Fixed by splitting the one loop into two loops so we
get all the refs before assigning them (to avoid potential clobbering).

(cherry picked from master/commit faa48915d2)
2009-01-28 17:01:34 -07:00
Brian Paul
51625ff86e i965: widen per-texture bitfields for 16 texture image units
(cherry picked from master/commit f78c388b6c)
2009-01-28 17:01:03 -07:00
Brian Paul
4d8f661726 intel: check if stencil test is enabled in intel_stencil_drawpixels()
(cherry picked from master/commit 72ee0e247d)
2009-01-28 17:00:23 -07:00
Brian Paul
efa69a2f66 intel: save/restore GL matrix mode in intel_meta_set_passthrough_transform(), intel_meta_restore_transform()
(cherry picked from master, commit 723648f2ee)
2009-01-28 16:55:16 -07:00
Timo Aaltonen
a2dce49158 New changelog entry for 7.3 2009-01-25 17:51:16 +02:00
Timo Aaltonen
e4b28de803 Merge branch 'upstream-experimental' into debian-experimental 2009-01-25 17:30:47 +02:00
Brian Paul
30a6dd78d5 docs: fix typo 2009-01-22 10:39:05 -07:00
Brian Paul
9f75836308 mesa: set version to 7.4 for mesa_7_4_branch 2009-01-22 10:19:39 -07:00
Julien Cristau
f5c47105db Prepare changelog for upload 2009-01-21 19:01:24 +01:00
Julien Cristau
f1939c8685 Refresh patches 03 and 04. 2009-01-21 19:01:17 +01:00
Julien Cristau
3ac4d678ef Delete some more files which aren't in the tarballs 2009-01-21 18:51:07 +01:00
Julien Cristau
fcfec66865 update changelog 2009-01-20 18:50:10 +00:00
Julien Cristau
53426d7113 Merge tag 'mesa_7_3_rc3' into debian-experimental
Conflicts:

	progs/tests/Makefile
2009-01-20 18:49:21 +00:00
Timo Aaltonen
1d4756de06 Update the changelog. 2009-01-16 23:12:14 +02:00
Timo Aaltonen
c1abb821f9 Merge commit 'mesa_7_3_rc2' into debian-experimental 2009-01-16 23:05:20 +02:00
Julien Cristau
88083e9bab Prepare changelog for upload 2009-01-10 22:34:36 +01:00
Julien Cristau
47e54545ed Delete some now unused code from debian/rules. 2009-01-10 22:13:17 +01:00
Julien Cristau
54885ff0ad Delete unused configs/debian-*, and install-source.sh script.
We've switched to using autoconf, and mesa-swx11-source is gone.
2009-01-10 21:55:19 +01:00
Julien Cristau
502fbe7cb3 Merge branch 'debian-unstable' into debian-experimental
Conflicts:

	debian/changelog
	debian/rules
2009-01-10 21:42:57 +01:00
Julien Cristau
08fdb497fb Merge commit 'origin/upstream-unstable' into debian-experimental 2009-01-10 21:38:10 +01:00
Julien Cristau
2a6711c0e5 add bug closers 2009-01-10 21:02:55 +01:00
Julien Cristau
382a8f6e29 Require libdrm-dev 2.4.3. 2009-01-10 20:41:16 +01:00
Julien Cristau
a713ba4399 Update changelog 2009-01-10 20:31:20 +01:00
Julien Cristau
5359febc9b Merge tag 'mesa_7_3_rc1' into debian-experimental
Conflicts:

	progs/tests/Makefile
	progs/tests/floattex.c
2009-01-10 20:30:41 +01:00
Julien Cristau
85d882983f Prepare changelog for upload 2008-12-14 09:59:33 +05:30
Timo Aaltonen
5387880f18 delete progs/trivial/ 2008-12-09 02:16:44 +02:00
Timo Aaltonen
a7a9398411 Build-depend on x11proto-dri2-dev (>= 1.99.3) 2008-12-09 02:01:40 +02:00
Timo Aaltonen
7593023f0b Update the changelog for the snapshot
also drop 01_disable-intel-classic-warn.diff since it's obsolete now.
2008-12-09 01:57:44 +02:00
Timo Aaltonen
92d9a71f60 Merge commit 'origin/debian-experimental' into debian-experimental
Conflicts:
	include/GLES/egltypes.h
	progs/miniglx/glfbdevtest.c
	progs/miniglx/manytex.c
	progs/miniglx/sample_server.c
	progs/miniglx/sample_server2.c
	progs/miniglx/texline.c
	progs/tests/Makefile.win
	progs/tests/antialias.c
	progs/tests/cva.c
	progs/tests/dinoshade.c
	progs/tests/getprocaddress.py
	progs/tests/jkrahntest.c
	progs/tests/manytex.c
	progs/tests/multipal.c
	progs/tests/multiwindow.c
	progs/tests/sharedtex.c
	progs/tests/texline.c
	progs/tests/texrect.c
	progs/tests/texwrap.c
	progs/trivial/Makefile
2008-12-09 01:38:08 +02:00
Timo Aaltonen
661890211e Merge commit 'mesa_7_2' into debian-experimental 2008-12-09 01:36:36 +02:00
Julien Cristau
0206e20240 Also build the x86-specific dri drivers on kfreebsd (closes: #492894). 2008-10-26 15:58:46 +01:00
Julien Cristau
83a4719056 Update changelog 2008-10-26 15:34:37 +01:00
Ian Romanick
6cfe871c75 Use 3Dnow! x86-64 routines only on processors that support 3Dnow!
Added an x86-64 CPUID function and use it to detect 3Dnow!  If 3Dnow!
is available, use _mesa_3dnow_transform_points4_3d_no_rot,
_mesa_3dnow_transform_points4_perspective,
_mesa_3dnow_transform_points4_2d_no_rot, and _mesa_3dnow_transform_points4_2d.

This fixes long standing bug #8724.
(cherry picked from commit 2b8d8989fb)
2008-10-26 15:23:27 +01:00
Julien Cristau
84bb8fc695 Prepare changelog for upload 2008-09-24 14:28:28 +02:00
Julien Cristau
872cb1a0dd intel: don't warn about TTM init failure. 2008-09-24 14:25:30 +02:00
Julien Cristau
ee80a3a623 Remove the build-dep on dri2proto, DRI2 support has been removed. 2008-09-24 12:55:15 +02:00
Timo Aaltonen
e700c5ad99 Update the changelog, refresh patch 04. 2008-09-23 16:09:00 +03:00
Timo Aaltonen
08c0ec7a89 Merge commit 'mesa_7_2' into debian-experimental
Conflicts:

	include/GLES/egl.h
2008-09-23 16:06:40 +03:00
Brian Paul
35fe0e56e6 mesa: docs: remove parenthesis 2008-09-20 09:28:22 -06:00
Brian Paul
69e5311ad3 mesa: set 7.2 version for tarballs 2008-09-20 09:24:17 -06:00
Brian Paul
e884c4db2b mesa: prep for 7.2 release 2008-09-20 09:23:50 -06:00
Adam Jackson
e681bee04e Update to SGI FreeB 2.0.
Under the terms of version 1.1, "once Covered Code has been published
under a particular version of the License, Recipient may, for the
duration of the License, continue to use it under the terms of that
version, or choose to use such Covered Code under the terms of any
subsequent version published by SGI."

FreeB 2.0 license refers to "dates of first publication".  They are here
taken to be 1991-2000, as noted in the original license text:

 ** Original Code. The Original Code is: OpenGL Sample Implementation,
 ** Version 1.2.1, released January 26, 2000, developed by Silicon Graphics,
 ** Inc. The Original Code is Copyright (c) 1991-2000 Silicon Graphics, Inc.
 ** Copyright in any portions created by third parties is as indicated
 ** elsewhere herein. All Rights Reserved.

Official FreeB 2.0 text:

http://oss.sgi.com/projects/FreeB/SGIFreeSWLicB.2.0.pdf

As always, this code has not been tested for conformance with the OpenGL
specification.  OpenGL conformance testing is available from
http://khronos.org/ and is required for use of the OpenGL logo in
product advertising and promotion.
2008-09-20 09:21:34 -06:00
Julien Cristau
07c1fcc604 Prepare changelog for upload 2008-09-20 16:32:02 +02:00
Julien Cristau
7959b95efa Update debian/copyright to the SGI Free Software License B, version 2.0.
It now mirrors the free X11 license used by X.Org (closes: #368560).
http://www.sgi.com/company_info/newsroom/press_releases/2008/september/opengl.html
2008-09-19 18:03:59 +02:00
Brian Paul
3dd48d903f document _tnl_InvalidateState() fix 2008-09-18 15:28:31 -06:00
Brian Paul
e0ea4337fc mesa: fix null ptr deref bug in _tnl_InvalidateState(), bug 15834 2008-09-18 15:27:55 -06:00
Brian Paul
6290d7fd29 mesa: fix bug in previous changes to _slang_resolve_attributes() 2008-09-17 09:13:28 -06:00
Brian Paul
84ad182ab8 document glBindAttribLocation fixes 2008-09-16 16:37:17 -06:00
Brian Paul
a155ead515 mesa: rework GLSL vertex attribute binding
Calls to glBindAttribLocation() should not take effect until the next time
that glLinkProgram() is called.
gl_shader_program::Attributes now just contains user-defined bindings.
gl_shader_program::VertexProgram->Attributes contains the actual/final bindings.
2008-09-16 15:58:05 -06:00
Brian Paul
5b0f652456 mesa: fix display list regression (check if save->prim_count > 0 in vbo_save_EndList()) 2008-09-16 13:32:13 -06:00
Brian Paul
1be50e1638 fix msaa enable/disable bug 2008-09-15 17:15:51 -06:00
Brian Paul
4fefa29d8c mesa: fix MSAA enable state in update_multisample() 2008-09-15 17:14:41 -06:00
Brian Paul
ef479d9f2a document display list bug fix 2008-09-15 09:22:07 -06:00
Brian Paul
fc04604856 mesa: remove some assertions that are invalid during context tear-down 2008-09-15 09:20:30 -06:00
Keith Whitwell
ef2bd7860f mesa: get another class of degenerate dlists working
Primitive begin in one dlist, end in another.
2008-09-15 09:20:23 -06:00
Shane Blackett
dca8b4c964 glut: s/glut_fbc.c/glut_fcb.c/ 2008-09-14 17:51:53 -06:00
Brian Paul
419353dc2f mesa: Makefile: 7.2 release candidate 1 2008-09-13 16:06:12 -06:00
Brian Paul
38d3b2262b mesa: documentation updates for 7.2 release 2008-09-13 14:43:54 -06:00
Brian Paul
4db2a7f36c updated Cell driver info 2008-09-13 14:39:39 -06:00
Brian Paul
1145042fd2 Mesa: version bumps for 7.2 release 2008-09-13 14:36:35 -06:00
Brian Paul
772131280a document 7.2 bug fixes, etc 2008-09-13 14:36:17 -06:00
Jakub Bogusz
ee913c0fcc tdfx: fix crash and lack of visuals bug 2008-09-13 14:34:51 -06:00
Guillaume Melquiond
cd1804c43a i915: fix himask constant init for 64-bit build 2008-09-13 14:26:35 -06:00
Guillaume Melquiond
253784d45e glx: fix 64-bit datatype issue 2008-09-13 14:26:25 -06:00
Guillaume Melquiond
463d19d477 mesa: return after _mesa_problem() calls 2008-09-13 14:26:18 -06:00
Guillaume Melquiond
6e4fd90ca5 mesa: #include <stdio.h> 2008-09-13 14:26:13 -06:00
Guillaume Melquiond
06188c0c5c remove invalid XDestroyWindow() call 2008-09-13 14:26:06 -06:00
Guillaume Melquiond
dec9097647 silence warning 2008-09-13 14:25:59 -06:00
Guillaume Melquiond
e234339dba GLU: fix asst warnings 2008-09-13 14:25:53 -06:00
Ian Romanick
bb09d23e92 Remove DRI2 support
Remove support for the original incarnation of DRI2.  This version has
since been abandoned and will never ship.  This eliminates a build
dependency on a component that only exists in Fedora 9 and in now-dead
up-stream development trees.

I also recommend that we spin a Mesa 7.1.1 release with this patch.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
2008-09-12 13:48:14 -07:00
Xiang, Haihao
1ffbcddb6f i965: Add support for G41 chipset which is another 4 series.
(cherry picked from commit b2e083eba2)
2008-09-12 09:29:00 +08:00
Brice Goglin
5cd892a4e4 Fix grammar and punctuation in glxinfo(1) 2008-09-11 19:13:46 +02:00
Shane Blackett
33cf8f5c06 added 24bpp support 2008-09-11 08:35:37 -06:00
Shane Blackett
7b8d878bed Fixes for Mingw 2008-09-10 08:23:19 -06:00
Brian Paul
86ae34fbb4 mesa: replace MALLOC w/ CALLOC to fix memory error in glPushClientAttrib() 2008-09-05 08:07:07 -06:00
Brian Paul
16f3461fd6 mesa: improved gl_buffer_object reference counting
Use new _mesa_reference_buffer_object() function wherever possible.
Fixes buffer object/display list crash reported in ParaView.
2008-09-04 15:26:07 -06:00
Xiang, Haihao
0ac8886b51 intel: Fix depth_stencil texture.
(cherry picked from commit 0adfd10210)
2008-09-04 11:54:31 +08:00
Xiang, Haihao
1729d37223 mesa: merge stencil values into depth values for MESA_FORMAT_S8_Z24
(cherry picked from commit 218df7f9c5)
2008-09-04 11:54:21 +08:00
Jakob Bornecrantz
81edef6f9f mesa: Support for MESA_FORMAT_S8_Z24 texture
cherry-picked from gallium-0.1
(cherry picked from commit dc44bb8e92)
2008-09-04 11:54:08 +08:00
Brian Paul
e040e18c7b fix BUFFER_DEPTH/BUFFER_ACCUM mix-up 2008-09-02 18:13:53 -06:00
Roland Scheidegger
14d186644b fix no error generated when calling glLight{if}[v] inside begin/end (bug 17408) 2008-09-02 15:49:35 +02:00
Xiang, Haihao
e4535e4d5e i965: force thread switch after IF/ELSE/ENDIF. partial fix for #16882.
A thread switch is implicitly invoked after the issuance of an IF/ELSE/ENDIF
instruction if necessary. Unfortunately it seems sometimes a forced thread
switch is needed.
2008-08-29 10:30:12 +08:00
Xiang, Haihao
a008813890 i965: mask control for BREAK/CONT/DO/WHILE. partial fix fox #16882 2008-08-29 10:30:03 +08:00
Xiang, Haihao
da50dc7bb3 i965: Push/pop instruction state. partial fix for #16882 2008-08-29 10:29:57 +08:00
Brian Paul
7f628d9cbc mesa: bump MAX_INSN to 350 2008-08-28 15:16:33 -06:00
Julien Cristau
f9affb3059 Prepare changelog for upload 2008-08-27 19:52:30 +02:00
Julien Cristau
fd7cc514d2 Some more parallel build updates. 2008-08-27 19:52:14 +02:00
Julien Cristau
f6a927b45f Update changelog 2008-08-27 17:29:28 +02:00
Julien Cristau
8772e02b68 Merge tag 'mesa_7_1' into debian-experimental 2008-08-27 17:28:31 +02:00
Brian Paul
b2ffd0be03 placeholder for upcoming 7.2 release notes 2008-08-26 16:24:37 -06:00
Brian Paul
912b16c689 added 7.1 MD5 sums 2008-08-26 16:11:56 -06:00
Brian Paul
6d9b25e522 mesa: don't check for GLSL 1.2 to advertise GL 2.1
The GLSL 1.2 features are minor...
2008-08-26 15:50:11 -06:00
Corbin Simpson
31bad23f0f note ATI R500 series support 2008-08-26 12:39:27 -06:00
Timo Jyrinki
29b9b25e72 asst. html doc updates 2008-08-26 12:36:23 -06:00
Brian Paul
e794f496c6 mesa: assorted updates/prep for 7.1 release 2008-08-25 14:09:14 -06:00
Brian Paul
68067c66d3 mesa: glsl: regenerated file 2008-08-25 09:16:21 -06:00
Brian Paul
ad5474e995 mesa: glsl: grab latest fixes from gallium-0.1 branch
Includes:
1. Fixes failed asserting about bad swizzles in src reg emit.
2. Tracks uniform var usage.
3. Emit exp() in terms of EXP2 instruction.
2008-08-25 09:15:16 -06:00
Julien Cristau
ac8771ee99 Update changelog 2008-08-24 17:38:03 +02:00
Julien Cristau
16358bb3cc Merge branch 'mesa_7_2_branch' of git.freedesktop.org:/git/mesa/mesa into debian-experimental 2008-08-24 17:34:47 +02:00
Julien Cristau
ef160330e5 Merge tag 'mesa_7_1_rc4' into debian-experimental
Conflicts:

	progs/fp/Makefile
	progs/slang/Makefile
	progs/tools/trace/gltrace_support.cc
	progs/trivial/Makefile
2008-08-16 18:43:53 +02:00
Julien Cristau
332dcfb60b Add parallel build support. 2008-08-04 12:49:03 +02:00
Julien Cristau
a8dde163dc Prepare changelog for upload 2008-07-13 19:42:01 +02:00
Julien Cristau
868d0accef update changelog and patches
01_fix-installdir.patch dropped, applied upstream
04_osmesa_version.diff needed fixing
2008-07-13 19:36:56 +02:00
Julien Cristau
64e57ed257 Merge branch 'master' of git://anongit.freedesktop.org/git/mesa/mesa into debian-experimental
Conflicts:

	progs/tests/antialias.c
	progs/tools/trace/Makefile
2008-07-13 19:20:03 +02:00
Julien Cristau
e325b76444 Prepare changelog for upload 2008-07-12 18:56:22 +02:00
Julien Cristau
34eedc6a98 update changelog 2008-07-12 18:55:55 +02:00
Julien Cristau
8c21c2e07e Merge branch 'mesa_7_0_branch' of git://anongit.freedesktop.org/git/mesa/mesa into debian-unstable 2008-07-12 17:21:50 +02:00
Brian Paul
27425708f5 mesa: return -1, not GL_FALSE if _glapi_add_dispatch() fails name sanity check 2008-07-09 15:06:49 -06:00
Brian Paul
fc07b6cda0 ARB program, GLSL fixes 2008-07-09 08:58:47 -06:00
Brian Paul
562ed27eb5 mesa: remove debug code 2008-07-09 08:52:44 -06:00
Brian Paul
0a0e09078d mesa: implement glGetUniformiv() with new ctx->Driver function
The old implementation could overwrite the caller's param buffer.

cherry-picked from master
2008-07-09 08:51:35 -06:00
Brian Paul
50bc5a75a7 mesa: fix state.clip[n].plane parsing bug (bug 16611) 2008-07-09 08:42:49 -06:00
Brian Paul
6661f53eb3 mesa: check for OpenBSD (bug 15604)
cherry-picked from master
2008-07-09 08:42:28 -06:00
Xiang, Haihao
b8782affc8 i915: fall back to software rendering when shadow comparison is
enabled for 1D texture. fix #12176

Cherry-picked from commit 75e4db1804
2008-07-09 13:44:13 +08:00
Brian Paul
9a9e839b9b added null texObj ptr check (bug 15567) 2008-07-08 15:13:23 -06:00
Julien Cristau
b04ccc9a03 Disable the i915tex driver, it doesn't build against libdrm 2.3.1. 2008-07-08 18:18:56 +02:00
Xiang, Haihao
aa522e14ee i965: official name for GM45 chipset 2008-07-08 14:39:01 +08:00
Julien Cristau
dabd09f5a9 some more debian/rules adjustments
build the software libGL with --disable-gl-osmesa, and try to fix some
rules dependencies
2008-07-06 14:03:58 +02:00
Julien Cristau
46ffb3c0b1 debian/rules cleanup and fixes
* tell configure which dri drivers we want to build, based on the old
  configs/debian-dri-default
* disable direct rendering for the glx libGL on the hurd
* make sure patches are applied before we run autoreconf, and autoreconf
  runs before all builds
2008-07-05 14:03:57 +02:00
Julien Cristau
4b9b01f0a5 Use --with-driver=dri for the dri config
the default is xlib on some architectures and dri on others, so we
need to set that explicitly
2008-07-05 12:59:14 +02:00
Timo Aaltonen
0261355a03 Fix confflags CFLAGS and clean up confflags-dri. 2008-07-04 11:50:09 +03:00
Timo Aaltonen
beb8547c71 Delete two files not available on the tarball.
docs/gears.png
src/glut/os2/src-glut_os2pm.zip
2008-07-04 11:50:08 +03:00
Timo Aaltonen
4aa518a2d7 Add a patch from upstream to set installdir properly.
this also replaces 01_fix-libdir.patch and the need to set --libdir for
every target.
2008-07-04 11:50:08 +03:00
root
08af66d57e Revert "Remove configs/current on clean."
This reverts commit 2380f8bd93.

Conflicts:

	debian/changelog
2008-07-04 11:50:08 +03:00
root
38eb154b39 Use --enable-glx-tls for dri. 2008-07-04 11:50:08 +03:00
Timo Aaltonen
2954e4c3d7 Disable glut for swx11 targets.
It was not built before either and would need new build-deps (xi xmu).
2008-07-04 11:50:08 +03:00
Julien Cristau
213803b4d4 fix libOSMesa versioning, to revert accidental SONAME bump. 2008-07-03 18:30:35 +02:00
Timo Aaltonen
2380f8bd93 Remove configs/current on clean. 2008-07-03 18:37:28 +03:00
Timo Aaltonen
e846ebc1a1 don't chmod configure which does not exist. 2008-07-03 18:14:25 +03:00
Timo Aaltonen
6dedd60350 Further autoconf changes:
add automake & autoconf to build-deps
add some cleanup rules
2008-07-03 15:53:19 +03:00
Timo Aaltonen
063a487fe1 Fix the configure.ac patch somewhat
workaround the problems by setting --libdir for all flavours, otherwise
the path used is /usr/usr/lib.
2008-07-03 13:20:21 +03:00
Timo Aaltonen
99a35fd4e1 01_fix-libdir.patch: libdir handling is broken in configure.ac, fix it. 2008-07-03 09:31:23 +03:00
Timo Aaltonen
cbaac82c2c Further tweaks to the autoconf stuff. Getting closer but still not there 2008-07-03 09:26:46 +03:00
Brian Paul
2c9e332bce mesa: when linking a shader program, make sure all the shaders compiled OK
cherry-picked from master
2008-07-02 17:05:01 -06:00
Brian Paul
e89deed74d mesa: fix error codes in _mesa_shader_source(), _mesa_get_shader_source()
cherry-picked from master
If the 'shader' parameter is wrong, need to either generate GL_INVALID_VALUE
or GL_INVALID_OPERATION.  It depends on whether 'shader' actually names a
'program' or is a totally unknown ID.
There might be other cases to fix...
2008-07-02 17:02:41 -06:00
Brian Paul
5cb6f5515f mesa: regenerated file 2008-07-02 17:02:14 -06:00
Brian Paul
c5024d4a5c mesa: added some missing equal() notEqual() intrinsics
cherry-picked from master
2008-07-02 17:01:26 -06:00
Timo Aaltonen
9d72de1fb2 Run autoreconf before building the various flavours. 2008-07-03 00:20:58 +03:00
Brian Paul
62626467b4 fixes some GLSL bugs 2008-07-02 12:45:16 -06:00
Brian Paul
c55a72ef48 mesa: regenerated files 2008-07-02 12:42:52 -06:00
Brian Paul
8a353c4020 mesa: fix all(bvec2) function typo, add missing bvec2/3/4() constuctors
cherry-picked from master
2008-07-02 12:42:12 -06:00
Brian Paul
8261c40bb9 mesa: init default span fog if fragprog uses fog input register
See bug #16194 for original patch.  Cleaned up a bit here.
2008-07-02 08:58:32 -06:00
Timo Aaltonen
053bf7f5db rules: Replace the old build system with the new autotools-based system. 2008-07-02 17:56:51 +03:00
Timo Aaltonen
e6a5e6512f Add gl.pc to libgl1-mesa-dev 2008-07-02 14:07:23 +03:00
Timo Aaltonen
bc32cd7fc8 Add dri_interface.h to mesa-common-dev. 2008-07-01 18:58:17 +03:00
Timo Aaltonen
24fde3cf34 Drop mesa-swx11-source. 2008-07-01 18:56:38 +03:00
Timo Aaltonen
7ceaa0916a Re-add linux-osmesa-static and linux-osmesa32-static
they were dropped by mistake during the merge.
2008-07-01 18:51:01 +03:00
Timo Aaltonen
4c958fbfb4 Build-depend on x11proto-dri2-dev. 2008-07-01 16:24:46 +03:00
Timo Aaltonen
9795ef4348 Refresh patches, and drop obsolete 00_create-libdir.patch and 01_fix-makefile.patch 2008-07-01 16:07:17 +03:00
Julien Cristau
768d8e0563 update changelog and require libdrm 2.3.1 2008-07-01 13:46:54 +02:00
Julien Cristau
374be37dd9 Merge branch 'master' of git.freedesktop.org:/git/mesa/mesa into debian-experimental
Conflicts:

	.gitignore
	Makefile
	bin/mklib
	configs/.gitignore
	configs/default
	configs/freebsd
	configs/freebsd-dri
	configs/linux-osmesa
	configs/linux-osmesa16
	configs/linux-osmesa16-static
	configs/linux-osmesa32
	docs/download.html
	docs/install.html
	docs/news.html
	docs/relnotes.html
	doxygen/Makefile
	doxygen/doxy.bat
	doxygen/header.html
	doxygen/header_subset.html
	include/GL/internal/dri_interface.h
	progs/demos/.gitignore
	progs/directfb/Makefile
	progs/egl/Makefile
	progs/fbdev/Makefile
	progs/fbdev/glfbdevtest.c
	progs/fp/Makefile
	progs/glsl/.gitignore
	progs/glsl/Makefile
	progs/glsl/bump.c
	progs/glsl/mandelbrot.c
	progs/glsl/noise.c
	progs/glsl/toyball.c
	progs/miniglx/Makefile
	progs/samples/.gitignore
	progs/slang/Makefile
	progs/tests/.gitignore
	progs/tests/Makefile
	progs/tests/arbnpot.c
	progs/tests/cva.c
	progs/tests/dinoshade.c
	progs/tests/drawbuffers.c
	progs/tests/fbotest2.c
	progs/tests/fbotexture.c
	progs/tests/getprocaddress.c
	progs/tests/mipmap_limits.c
	progs/tests/multipal.c
	progs/tests/texline.c
	progs/tools/trace/Makefile
	progs/trivial/Makefile
	progs/trivial/dlist-edgeflag-dangling.c
	progs/trivial/dlist-edgeflag.c
	progs/trivial/point-param.c
	progs/trivial/quad-clip-nearplane.c
	progs/trivial/tri.c
	progs/vp/Makefile
	progs/xdemos/.gitignore
	progs/xdemos/Makefile
	progs/xdemos/glthreads.c
	src/egl/Makefile
	src/egl/drivers/demo/Makefile
	src/egl/drivers/dri/Makefile
	src/egl/main/Makefile
	src/glu/mesa/Makefile
	src/glu/mesa/Makefile.m32
	src/glu/mesa/project.c
	src/glu/mini/project.c
	src/glut/directfb/Makefile
	src/glut/directfb/callback.c
	src/glut/directfb/events.c
	src/glut/directfb/internal.h
	src/glut/directfb/window.c
	src/glut/os2/Makefile
	src/glut/os2/WarpWin.cpp
	src/glut/os2/glutOverlay.cpp
	src/glut/os2/glut_8x13.cpp
	src/glut/os2/glut_9x15.cpp
	src/glut/os2/glut_bitmap.cpp
	src/glut/os2/glut_cindex.cpp
	src/glut/os2/glut_cmap.cpp
	src/glut/os2/glut_cursor.cpp
	src/glut/os2/glut_event.cpp
	src/glut/os2/glut_ext.cpp
	src/glut/os2/glut_fullscrn.cpp
	src/glut/os2/glut_gamemode.cpp
	src/glut/os2/glut_get.cpp
	src/glut/os2/glut_hel10.cpp
	src/glut/os2/glut_hel12.cpp
	src/glut/os2/glut_hel18.cpp
	src/glut/os2/glut_init.cpp
	src/glut/os2/glut_input.cpp
	src/glut/os2/glut_key.cpp
	src/glut/os2/glut_keyctrl.cpp
	src/glut/os2/glut_keyup.cpp
	src/glut/os2/glut_mesa.cpp
	src/glut/os2/glut_modifier.cpp
	src/glut/os2/glut_roman.cpp
	src/glut/os2/glut_shapes.cpp
	src/glut/os2/glut_stroke.cpp
	src/glut/os2/glut_swap.cpp
	src/glut/os2/glut_teapot.cpp
	src/glut/os2/glut_tr24.cpp
	src/glut/os2/glut_util.cpp
	src/glut/os2/glut_vidresize.cpp
	src/glut/os2/glut_warp.cpp
	src/glut/os2/glut_win.cpp
	src/glut/os2/glut_winmisc.cpp
	src/glut/os2/glutbitmap.h
	src/glut/os2/glutstroke.h
	src/glut/os2/layerutil.h
	src/glut/os2/libGlut.DEF
	src/glut/os2/os2_glx.cpp
	src/glut/os2/os2_menu.cpp
	src/glut/os2/os2_winproc.cpp
	src/glw/Makefile
	src/glx/mini/Makefile
	src/glx/mini/miniglx.c
	src/glx/x11/glxext.c
	src/mesa/Makefile
	src/mesa/drivers/allegro/amesa.c
	src/mesa/drivers/allegro/direct.h
	src/mesa/drivers/allegro/generic.h
	src/mesa/drivers/d3d/D3DCAPS.CPP
	src/mesa/drivers/d3d/D3DHAL.H
	src/mesa/drivers/d3d/D3DInit.cpp
	src/mesa/drivers/d3d/D3DMESA.H
	src/mesa/drivers/d3d/D3DRaster.cpp
	src/mesa/drivers/d3d/D3DTEXT.CPP
	src/mesa/drivers/d3d/D3DTextureMgr.cpp
	src/mesa/drivers/d3d/D3DTextureMgr.h
	src/mesa/drivers/d3d/D3DUTILS.CPP
	src/mesa/drivers/d3d/DDrawPROCS.c
	src/mesa/drivers/d3d/DEBUG.C
	src/mesa/drivers/d3d/DEBUG.H
	src/mesa/drivers/d3d/DbgEnv.bat
	src/mesa/drivers/d3d/MAKEFILE
	src/mesa/drivers/d3d/OPENGL32.DEF
	src/mesa/drivers/d3d/WGL.C
	src/mesa/drivers/dri/Makefile.template
	src/mesa/drivers/dri/common/dri_bufmgr.c
	src/mesa/drivers/dri/common/dri_drmpool.c
	src/mesa/drivers/dri/common/dri_util.c
	src/mesa/drivers/dri/common/dri_util.h
	src/mesa/drivers/dri/common/xmlconfig.c
	src/mesa/drivers/dri/glcore/Makefile
	src/mesa/drivers/dri/i915/i915_fragprog.c
	src/mesa/drivers/dri/i915/i915_program.h
	src/mesa/drivers/dri/i915/i915_texprog.c
	src/mesa/drivers/dri/i915/i915_texstate.c
	src/mesa/drivers/dri/i915/i915_vtbl.c
	src/mesa/drivers/dri/i915/intel_context.c
	src/mesa/drivers/dri/i915/intel_context.h
	src/mesa/drivers/dri/i915/intel_pixel.c
	src/mesa/drivers/dri/i915/intel_render.c
	src/mesa/drivers/dri/i915/intel_screen.c
	src/mesa/drivers/dri/i915/intel_state.c
	src/mesa/drivers/dri/i915/intel_tex.c
	src/mesa/drivers/dri/i915/intel_tris.c
	src/mesa/drivers/dri/i915tex/i830_reg.h
	src/mesa/drivers/dri/i915tex/i830_texstate.c
	src/mesa/drivers/dri/i915tex/i830_vtbl.c
	src/mesa/drivers/dri/i915tex/i915_fragprog.c
	src/mesa/drivers/dri/i915tex/i915_metaops.c
	src/mesa/drivers/dri/i915tex/i915_tex_layout.c
	src/mesa/drivers/dri/i915tex/i915_texstate.c
	src/mesa/drivers/dri/i915tex/i915_vtbl.c
	src/mesa/drivers/dri/i915tex/intel_render.c
	src/mesa/drivers/dri/i915tex/intel_screen.c
	src/mesa/drivers/dri/i915tex/intel_tris.c
	src/mesa/drivers/dri/i965/brw_aub.c
	src/mesa/drivers/dri/i965/brw_aub_playback.c
	src/mesa/drivers/dri/i965/brw_cc.c
	src/mesa/drivers/dri/i965/brw_clip.c
	src/mesa/drivers/dri/i965/brw_clip_state.c
	src/mesa/drivers/dri/i965/brw_clip_util.c
	src/mesa/drivers/dri/i965/brw_context.c
	src/mesa/drivers/dri/i965/brw_curbe.c
	src/mesa/drivers/dri/i965/brw_defines.h
	src/mesa/drivers/dri/i965/brw_draw.c
	src/mesa/drivers/dri/i965/brw_draw_upload.c
	src/mesa/drivers/dri/i965/brw_eu.h
	src/mesa/drivers/dri/i965/brw_eu_emit.c
	src/mesa/drivers/dri/i965/brw_gs_state.c
	src/mesa/drivers/dri/i965/brw_misc_state.c
	src/mesa/drivers/dri/i965/brw_sf_emit.c
	src/mesa/drivers/dri/i965/brw_sf_state.c
	src/mesa/drivers/dri/i965/brw_state_cache.c
	src/mesa/drivers/dri/i965/brw_state_pool.c
	src/mesa/drivers/dri/i965/brw_tex.c
	src/mesa/drivers/dri/i965/brw_tex_layout.c
	src/mesa/drivers/dri/i965/brw_vs_state.c
	src/mesa/drivers/dri/i965/brw_vs_tnl.c
	src/mesa/drivers/dri/i965/brw_wm.c
	src/mesa/drivers/dri/i965/brw_wm.h
	src/mesa/drivers/dri/i965/brw_wm_glsl.c
	src/mesa/drivers/dri/i965/brw_wm_sampler_state.c
	src/mesa/drivers/dri/i965/brw_wm_state.c
	src/mesa/drivers/dri/i965/brw_wm_surface_state.c
	src/mesa/drivers/dri/i965/bufmgr_fake.c
	src/mesa/drivers/dri/i965/intel_batchbuffer.c
	src/mesa/drivers/dri/i965/intel_batchbuffer.h
	src/mesa/drivers/dri/i965/intel_blit.c
	src/mesa/drivers/dri/i965/intel_blit.h
	src/mesa/drivers/dri/i965/intel_buffers.c
	src/mesa/drivers/dri/i965/intel_context.c
	src/mesa/drivers/dri/i965/intel_context.h
	src/mesa/drivers/dri/i965/intel_mipmap_tree.c
	src/mesa/drivers/dri/i965/intel_mipmap_tree.h
	src/mesa/drivers/dri/i965/intel_pixel_bitmap.c
	src/mesa/drivers/dri/i965/intel_pixel_copy.c
	src/mesa/drivers/dri/i965/intel_screen.c
	src/mesa/drivers/dri/i965/intel_tex_validate.c
	src/mesa/drivers/dri/i965/server/i830_common.h
	src/mesa/drivers/dri/intel/intel_batchbuffer.c
	src/mesa/drivers/dri/intel/intel_batchbuffer.h
	src/mesa/drivers/dri/intel/intel_blit.c
	src/mesa/drivers/dri/intel/intel_buffers.c
	src/mesa/drivers/dri/intel/intel_context.c
	src/mesa/drivers/dri/intel/intel_context.h
	src/mesa/drivers/dri/intel/intel_mipmap_tree.c
	src/mesa/drivers/dri/intel/intel_mipmap_tree.h
	src/mesa/drivers/dri/intel/intel_screen.h
	src/mesa/drivers/dri/r200/r200_texstate.c
	src/mesa/drivers/dri/r300/r300_cmdbuf.c
	src/mesa/drivers/dri/r300/r300_context.h
	src/mesa/drivers/dri/r300/r300_ioctl.c
	src/mesa/drivers/dri/r300/r300_reg.h
	src/mesa/drivers/dri/r300/r300_state.c
	src/mesa/drivers/dri/r300/r300_tex.h
	src/mesa/drivers/dri/r300/r300_texstate.c
	src/mesa/drivers/dri/radeon/radeon_screen.c
	src/mesa/drivers/windows/gdi/wmesa.c
	src/mesa/drivers/x11/fakeglx.c
	src/mesa/drivers/x11/xm_dd.c
	src/mesa/glapi/Makefile
	src/mesa/glapi/dispatch.h
	src/mesa/glapi/glapioffsets.h
	src/mesa/glapi/glapitable.h
	src/mesa/glapi/glapitemp.h
	src/mesa/glapi/glprocs.h
	src/mesa/main/api_validate.c
	src/mesa/main/attrib.c
	src/mesa/main/config.h
	src/mesa/main/context.c
	src/mesa/main/context.h
	src/mesa/main/dlist.c
	src/mesa/main/drawpix.c
	src/mesa/main/enums.c
	src/mesa/main/fbobject.c
	src/mesa/main/image.c
	src/mesa/main/imports.h
	src/mesa/main/mtypes.h
	src/mesa/main/points.c
	src/mesa/main/state.c
	src/mesa/main/texcompress_fxt1.c
	src/mesa/main/texenvprogram.c
	src/mesa/main/teximage.c
	src/mesa/main/texobj.c
	src/mesa/main/texstate.c
	src/mesa/main/version.h
	src/mesa/shader/prog_execute.c
	src/mesa/shader/program.c
	src/mesa/shader/shader_api.c
	src/mesa/shader/slang/library/slang_core.gc
	src/mesa/shader/slang/library/slang_core_gc.h
	src/mesa/shader/slang/slang_codegen.c
	src/mesa/sources
	src/mesa/sparc/glapi_sparc.S
	src/mesa/swrast/s_aalinetemp.h
	src/mesa/swrast/s_aatriangle.c
	src/mesa/swrast/s_aatritemp.h
	src/mesa/swrast/s_context.c
	src/mesa/swrast/s_copypix.c
	src/mesa/swrast/s_drawpix.c
	src/mesa/swrast/s_fragprog.c
	src/mesa/swrast/s_pointtemp.h
	src/mesa/swrast/s_readpix.c
	src/mesa/swrast/s_span.c
	src/mesa/swrast/s_texfilter.c
	src/mesa/swrast/s_tritemp.h
	src/mesa/swrast/s_zoom.c
	src/mesa/tnl/t_draw.c
	src/mesa/tnl/t_vb_fog.c
	src/mesa/tnl/t_vb_program.c
	src/mesa/tnl/t_vp_build.c
	src/mesa/vbo/vbo_split_copy.c
	src/mesa/x86-64/xform4.S
	src/mesa/x86/common_x86.c
	src/mesa/x86/glapi_x86.S
	windows/VC6/mesa/mesa/mesa.dsp
2008-07-01 13:18:27 +02:00
Xiang, Haihao
9fa552eb0f dri: Take the base image size into account when computing
first level of the mipmap. fix #16421
2008-07-01 11:50:11 +08:00
Brian Paul
5f8ad807be s/GL_INVALID_VALUE/GL_INVALID_OPERATION/ in _mesa_get_uniformfv() 2008-06-28 16:50:07 -06:00
Zou Nan hai
b6e165d661 optimize 965 clip
1. increase clip thread number to 2
  2. do cliptest for -rhw

Cherry-picked from commits b47c9f8c91,
aa88d11e7d.
2008-06-23 15:16:11 +08:00
Roland Scheidegger
afbb645906 r200: fix typo in r200TryDrawPixels parameter validation (bug 16406) 2008-06-21 02:52:18 +02:00
Brice Goglin
2b831e73f5 Prepare changelog for upload 2008-06-18 20:59:24 +02:00
Brice Goglin
6ec170290c Put all configs/ changes into the .diff.gz
Put back our configs/ changes into the .diff.gz since choose-configs
needs them before quilt is invoked. Put 04_cleanup-osmesa-configs.patch
there as well for #485161.
2008-06-18 20:59:09 +02:00
Brice Goglin
7c8ea0899a Revert "Move our configs/ changes from the .diff.gz into our quilt patches"
This reverts commit 03970183fc.

Conflicts:

	debian/changelog
2008-06-18 20:55:42 +02:00
Brice Goglin
f81070dbd5 Pull from mesa_7_0_branch 2008-06-18 20:55:00 +02:00
Brice Goglin
955c2ef48f Merge branch 'mesa_7_0_branch' of git://git.freedesktop.org/git/mesa/mesa into debian-unstable 2008-06-18 20:52:25 +02:00
Xiang, Haihao
2ac4919d24 i965: add support for Intel 4 series chipsets 2008-06-18 15:48:45 +08:00
Xiang, Haihao
3ed89025f3 i915: The pitch passed to intelEmitCopyBlitLocked should be in pixels,
not in bytes. Reported by Christopher Dissauer.
2008-06-18 13:52:04 +08:00
Xiang, Haihao
5b42bbce70 i915: fix data size in intelTryDrawPixels. Reported by Christopher Dissauer 2008-06-18 13:40:22 +08:00
Brian Paul
d2e0a11aab mesa: fix inconsistent use of GL_UNSIGNED_INT vs. GL_UNSIGNED_INT_24_8_EXT for Z unpacking 2008-06-17 16:44:00 -06:00
Brice Goglin
5033e5b36d Prepare changelog for upload 2008-06-17 20:01:01 +02:00
Brice Goglin
6a6a9c1c5a Pull from mesa_7_0_branch 2008-06-17 20:00:37 +02:00
Brice Goglin
3d6aa2e06f Merge branch 'mesa_7_0_branch' of git://git.freedesktop.org/git/mesa/mesa into debian-unstable 2008-06-17 19:54:47 +02:00
Brice Goglin
03970183fc Move our configs/ changes from the .diff.gz into our quilt patches
with 04_cleanup-osmesa-configs.patch renamed into 04_debian-configs.patch
2008-06-17 19:32:21 +02:00
Brian Paul
47d046c93f mesa: make mm.c use unsigned ints for offsets.
If you have a GPU using this code and it has the offsets up in this space,
this fails.

cherry-picked from master
2008-06-17 10:10:53 -06:00
Wilfried Holzke
718724deeb assorted glide driver fixes/updates 2008-06-17 10:03:03 -06:00
Brian Paul
85c325c36c add hyphen to rm command 2008-06-17 10:02:10 -06:00
Brian Paul
7e6d99f5ec glu: silence warnings 2008-06-17 09:01:40 -06:00
Brian Paul
6ce6dc961b bump version to 7.0.4 2008-06-16 10:19:29 -06:00
Brian Paul
186883611e fix glPixelZoom stack over flow on Windows 2008-06-16 10:19:28 -06:00
Brian Paul
48b3c59cb9 mesa: allocate pixel zoom arrays on heap, not stack
Fixes stack overflow on Windows.

cherry-picked from master
2008-06-16 10:19:28 -06:00
Brian Paul
04b9d5bc23 Fix _mesa_new_program() recursive call regression
This was introduced by the "i965 GLSL merge" from master (ce7a9efb09)
2008-06-16 10:19:28 -06:00
Brian Paul
91707e9020 fix GLSL generic vertex attrib linking bug 2008-06-16 10:19:28 -06:00
Xiang, Haihao
82a0e82232 i965: fix intel_batchbuffer_space. (bug#14709) 2008-06-13 13:53:46 +08:00
Brice Goglin
fe4264bf45 Prepare changelog for upload 2008-06-13 06:53:55 +02:00
Brice Goglin
2fa795a213 Pull from mesa_7_0_branch 2008-06-11 18:59:59 +02:00
Brice Goglin
e32aaf5da8 Merge branch 'mesa_7_0_branch' of git://git.freedesktop.org/git/mesa/mesa into debian-unstable 2008-06-11 18:58:59 +02:00
Brian
03447de338 disable ctx->Driver.NewProgram() call in _mesa_new_program()
This was causing infinite recursive calls w/ software drivers.
All vertex/fragment shaders should be allocated by calling
ctx->Driver.NewProgram(), not by calling _mesa_new_program().

(Cherry picked from commit 40133487db,
351a83163a).
2008-06-11 17:03:47 +08:00
Dave Airlie
ee5f4a4caf r300: disable the lowimpact fallbacks by default.
because really we should be able to just fix the driver.
(cherry picked from commit 7013eecf28)

There are actually even better reasons for this, the bottom line being that
enabling these fallbacks makes a lot of apps unusable mostly for no gain
whatsoever.
2008-06-11 10:42:16 +02:00
Xiang, Haihao
c04f3933ab i915: fix fd.o #14966 2008-06-11 11:36:01 +08:00
Xiang, Haihao
8f328c45e5 i915: Keith Whitwell's swizzling TEX patch. fix #8283
Cherry picked from commit 3369cd9a6f
2008-06-11 11:32:12 +08:00
Brice Goglin
9676f0cffc Pull from mesa_7_0_branch 2008-06-10 23:18:28 +02:00
Brice Goglin
2d0ca23319 Merge branch 'mesa_7_0_branch' of git://git.freedesktop.org/git/mesa/mesa into debian-unstable 2008-06-10 22:48:43 +02:00
Xiang, Haihao
6f4c8b5b50 i965: apply commit 6c1a98e97a to glsl
(cherry picked from commit a742bed99a)
2008-06-10 16:46:02 +08:00
Michal Wajdeczko
f8bd9cc30f Add support for ATI_separate_stencil in display lists.
(cherry picked from commit 7f747204ea)
2008-06-10 16:28:06 +08:00
Michal Wajdeczko
d9f9b1cd0b [965] Correctly set read mask for OPCODE_SWZ in pass1.
While OPCODE_SWZ has usually been optimized away in pass0, it may still
exist if a SWZ with dst saturate was emitted in pass_fp.  Fixes an error
in oglconform fpalu.c.
(cherry picked from commit 13a6f73a64)
2008-06-10 16:21:31 +08:00
Michal Wajdeczko
eca283976b [965] Avoid emitting dead code for DPx/math instructions.
The pass1 optimization stage clears out writemasks and registers, but the
instructions themselves are still being processed at this stage, and could
have resulted in them still being emitted.
(cherry picked from commit c60b5dfde8)
2008-06-10 16:18:48 +08:00
Michal Wajdeczko
2176259ca6 [965] Improve pinterp performance by delaying reads of just-written regs.
(cherry picked from commit bb419970ef)
2008-06-10 16:18:09 +08:00
Michal Wajdeczko
8fe6fcb900 [965] Fix negating of unsigned value in emit_wpos_xy.
(cherry picked from commit 6c1a98e97a)
2008-06-10 16:16:52 +08:00
Michal Wajdeczko
76d6edcc38 [965] Add MVP code for position invariant vertex programs.
This fixes the arbvptorus demo.
(cherry picked from commit 5f10438f2d)
2008-06-10 16:13:06 +08:00
Michal Wajdeczko
98d6c671f5 [win32] Use native aligned memory allocation functions.
(cherry picked from commit 31fe7cf5e3)
2008-06-10 16:12:18 +08:00
Andrzej Trznadel
f652811df4 [965] Fix fp temp reg release code to not usually release all temps.
Also, use wrapped ffs() instead of native.
(cherry picked from commit 3105bc1d88)
2008-06-10 15:45:11 +08:00
Andrzej Trznadel
e279f4601d Fix compat implementation of ffs() to return 1-based bit numbers.
(cherry picked from commit e9809a36aa)
2008-06-10 15:44:53 +08:00
Keith Packard
87a30337a1 [i965] short immediate values must be replicated to both halves of the dword
The 32-bit immediate value in the i965 instruction word must contain two
copies of any 16-bit constants. brw_imm_uw and brw_imm_w just needed to
copy the value into both halves of the immediate value instruction field.
(cherry picked from commit ca73488f48)
2008-06-10 15:37:11 +08:00
Eric Anholt
9c2047b275 [965] Don't let the negate flags of src0 affect 1 constants in precalc_dst/lit
This patch is a variant of a submission by Michal Wajdeczko to fix
oglconform fpalu failures.
(cherry picked from commit b4cbf6983e)
2008-06-10 15:32:53 +08:00
Zou Nan hai
1dcb0433a3 [i915] fix fragment.position 2008-06-10 15:28:07 +08:00
Zou Nan hai
5ff27e02b3 [i965] fix wpos height 1 pixel higher
(cherry picked from commit b0f681b458)
2008-06-10 15:25:20 +08:00
Eric Anholt
4beee58e57 [965] Bug #9151: make fragment.position return window coords not screen coords.
(cherry picked from commit 9c8f27ba13)
2008-06-10 15:22:08 +08:00
Eric Anholt
1f9de20719 [915] Fix COS function using same plan as SIN.
The previous COS function failed badly outside of [-pi/2, pi/2].
2008-06-10 14:55:52 +08:00
Eric Anholt
d05a8d9750 [915] Use a quartic term to improve the accuracy of SIN results.
This is described in the link in the comment, and is the same technique that
r300 uses.
2008-06-10 14:46:12 +08:00
Eric Anholt
db5f206c00 [915] Fix fp SIN function, and use a quadratic approximation instead of Taylor.
The Taylor series notably fails at producing sin(pi) == 0, which leads to
discontinuity every 2*pi.  The quadratic gets us sin(pi) == 0 behavior, at the
expense of going from 2.4% THD with working Taylor series to 3.8% THD (easily
seen on comparative graphs of the two).  However, our previous implementation
was producing sin(pi) < -1 and worse, so any reasonable approximation is an
improvement.  This also fixes the repeating behavior, where the previous
implementation would repeat sin(x) for x>pi as sin(x % pi) and the opposite
for x < -pi.
2008-06-10 14:22:36 +08:00
Eric Anholt
9dface8347 [965] Fix potential segfaults from bad realloc.
C has no order of evaluation restrictions on function arguments, so we
attempted to realloc from new-size to new-size.
(cherry picked from commit e747e9a072)
2008-06-10 13:54:01 +08:00
Eric Anholt
32f4940883 [965] Fix inversion of SLT/SGE results in vertex programs.
The WM code had this right, so copy its behavior.  This reverts a flipping
of the arguments to SLT in brw_vs_tnl which came in with the GLSL code that
probably occurred to work around the flipped results, and brings the code back
in line with t_vp_build.c.
(cherry picked from commit 9bae03a583)
2008-06-10 13:44:54 +08:00
Eric Anholt
a7969a9b93 [965] Fix and enable separate stencil.
Note that this does not enable GL_EXT_stencil_two_side, because Mesa's computed
_TestTwoSide ends up respecting only STENCIL_TEST_TWO_SIDE_EXT (defaults to
GL_FALSE), even if the application uses only GL 2.0 / ATI entrypoints.
(cherry picked from commit 9136e1f2c8)
2008-06-10 13:30:25 +08:00
Eric Anholt
ad88130df5 [965] Replace our own depth constants in intel context with GL context ones. 2008-06-10 13:08:09 +08:00
Eric Anholt
d1e71bc08b [965] Remove dead code in upload_wm_surfaces.
(cherry picked from commit 3ecdae82d7)
2008-06-10 11:19:53 +08:00
Roland Scheidegger
fa58fe247c i965: fix OPCODE_TEX when additional ops are needed 2008-06-08 14:04:39 +02:00
Brian Paul
4b71478326 Set the attribute as used.
Cherry-picked from gallium-0.1
2008-06-04 14:42:55 -06:00
Zou Nan hai
0989471fdb [i915] GL_DEPTH_TEXTURE_MODE fix
Cherry picked from commit 7233eabaf0
with manual changes.
2008-06-04 15:57:37 +08:00
Xiang, Haihao
696140bd1d i965: handle source depth to render target for glsl,
(cherry picked from commit d2540e6d4b)
2008-06-04 11:54:25 +08:00
Brian
380d15b7fe replace // comment with /* */ (bug 14916)
(cherry picked from commit eecb3ab7c6)
2008-06-04 11:52:38 +08:00
Xiang, Haihao
41261d64b2 i965: use _Current pointer instead of Current pointer.
Cherry picked from commit de1e9880f8
2008-06-04 11:51:19 +08:00
Zou Nan hai
ce7a9efb09 [i965] Add support for GL shading language in I965 driver.
Cherry picked from commit 6ef27b88e6,
d0ebdca4fa.
2008-06-04 11:47:08 +08:00
Zou Nan hai
8a38ebe328 [i965] This is to fix random crash in some maps of Ut2004 demo.
e.g. bridge of fate.
 	If vs output is big, driver may fall back to use 8 urb entries for vs,
	unfortunally, for some unknown reason, if vs is working at 4x2 mode,
	8 entries is not enough, may lead to gpu hang.

Cherry picked from commmit c9c64a100d
2008-06-03 11:30:41 +08:00
Zou Nan hai
dcc6671b85 [i915] don't use 4x4 filter for 1D shadowmap
Cherry picked from commit d24a5254c2
2008-06-03 11:23:44 +08:00
Zou Nan hai
b53b7581e4 [i965] fix fd.o bug #11471 and #11478
1. Follow EXT_texture_rectangle with YCbCr texture
	2. swap UV component for MESA_FORMAT_YCBCR

Cherry picked from commit 7676980d38.
2008-06-03 11:04:10 +08:00
Zou Nan hai
64a4a03c2a EXT_texture_sRGB support on i965
Cherry picked from commit 6bf81a5edf,
246d1d2522.
2008-06-03 10:09:33 +08:00
Brian Paul
b878c3f518 i915tex: fix fragment fog swizzle (from master) (bug 16195) 2008-06-02 09:29:22 -06:00
Brian Paul
6164163ca8 glDrawElement + VBO bug fix 2008-06-02 09:28:31 -06:00
Zou Nan hai
e92a53cd92 [i965] flip point sprite
Cherry picked from commit 1202c434d9
2008-06-02 17:05:57 +08:00
Zou Nan hai
2467af98b1 ARB sprite point support on i965
Cherry picked from commmit 60179434d1,
505453a04e with manual changes.
2008-06-02 17:01:31 +08:00
Xiang, Haihao
6f851d8875 _generic_read_RGBA_span_BGRA8888_REV_SSE2: It should adjust the source
and target pointers after do the first 2 pixels. fix bug #15850

Cherry-picked from commit 4b7d301c94
2008-06-02 14:28:42 +08:00
Xiang, Haihao
9b99bf89c4 i965: depth offset on glPolygonMode(GL_LINE/GL_POINT)
Cherry picked from 184cf464f4
2008-06-02 14:20:23 +08:00
Xiang, Haihao
7346fca083 965: use RGB565 to render a bitmap if Depth is 16
Cherry-picked from commit 5982d39799.
2008-06-02 14:15:24 +08:00
Xiang, Haihao
7facbb69c6 i965: don't swizzle fogcoord if FogOption is FOG_NONE.
fix #10788 issue on 965.

Cherry picked from commit 83068115e2
2008-06-02 14:08:26 +08:00
Xiang, Haihao
f59267d650 i915: set fogcoord to (f,0,0,1). fix #10788 issue on 915.
Cherry picked from commit 7eef52e975
2008-06-02 14:06:14 +08:00
Xiang, Haihao
71cb014195 mesa: fix a bad cast in put_values_z24.
The values passed to put_values_z24 are GLuint,
not GLubyte. fix #13543

Cherry picked from commit cf46aee14a
2008-06-02 13:59:51 +08:00
Xiang, Haihao
6c0f8db9c2 i965: The jump instruction count is added
to IP pre-increment, and should point to
the first instruction after the do instruction
of the do-while block of code

Cherry picked from commit 46e03d584a
2008-06-02 13:54:45 +08:00
Xiang, Haihao
49f1e2fc4c i965: fix an error in brw_vs_tnl.c
Update the tnl program if the state of TEXMAT is changed.
2008-06-02 13:46:31 +08:00
Eric Anholt
2d26e19535 [965] Clarify a bit of index buffer upload code.
Cherry picked from commit 5a49e84fcd
2008-06-02 13:18:00 +08:00
Xiang, Haihao
5b0c6cd49a i965: align the address of the first element within
the index buffer. (fix#11910)

Cherry picked from ea07a0df9a
2008-06-02 13:16:53 +08:00
Xiang, Haihao
c3ee8e46cc i965: fix projtex_mask
projtex_mask is only an 8bit field, and wm.input_size_masks includes
other attributes' information, therefore right shift is needed.

Cherry picked from 88451b04e9
2008-06-02 13:01:11 +08:00
Xiang, Haihao
46aac24261 i965: fix bad casts in do_blit_bitmap to support WindowPos correctly
Cherry picked from commit e66757c8ba
2008-06-02 11:54:35 +08:00
Xiang, Haihao
e1032ce718 i965: fix DEPTH_TEXTURE_MODE
Cherry picked from commit 6e620162a1 with
manual changes
2008-06-02 11:52:36 +08:00
Brian Paul
ce636f36f2 Fix segfault in _save_OBE_DrawElements() when using VBO and display list (bug 16156)
This was previously fixed in master by commit 982dcb74fd by Haihao Xiang.
2008-05-30 08:46:51 -06:00
Xiang, Haihao
93f2eec6b3 i965: roland's DXTn format texture patch(bug10347)
Cherry picked from commit db928291dc
2008-05-29 15:50:06 +08:00
Eric Anholt
46ef09d787 [965] Replace various alignment code with a shared ALIGN() macro.
In the process, fix some alignment issues:
- Scratch space allocation was aligned into units of 1KB, while the allocation
  wanted units of bytes, so we never allocated enough space for scratch.
- GRF register count was programmed as ALIGN(val - 1, 16) / 16 instead of
  ALIGN(val, 16) / 16 - 1, which overcounted for val != 16n+1.

Cherry picked from commit 77e0523fb7
2008-05-29 15:48:47 +08:00
Xiang, Haihao
7487ec0ff0 i965: align width/height for volume texture
Cherry picked from commit 00b86ecf6f
2008-05-29 15:45:12 +08:00
Xiang, Haihao
a25549866f intel: applying right alignment to compressed texture,
which make small textures(4x4,2x2,1x1) work well.

Cherry picked from commit 8ea66fa2ec
2008-05-29 15:43:31 +08:00
Xiang, Haihao
558cc6e38f i965: set mt->cpp differently with compressed texture
Cherry picked from commit 2cafd749b8
2008-05-29 15:40:44 +08:00
Xiang, Haihao
feb1fa1e83 i965: use BRW_TEXCOORDMODE_CLAMP instead of BRW_TEXCOORDMODE_CLAMP_BORDER
to implement GL_CLAMP

Cherry picked from commit ab99960858. fix #16005
2008-05-28 16:29:17 +08:00
Brian Paul
f32462343d allow GLX_SAMPLES_ARB==0 (bug 16073) 2008-05-27 09:49:44 -06:00
Brian Paul
08ef1b379a added sampler types to sizeof_glsl_type() 2008-05-27 08:50:15 -06:00
Karl Schultz
1e79831b56 dependency fixes (bug 13544) 2008-05-27 08:50:14 -06:00
Tormod Volden
7a29164f70 dri: vblank_mode warning
From what I can see the environment variables LIBGL_THROTTLE_REFRESH
and LIBGL_SYNC_REFRESH were taken out like 3 years ago, but this
warning was never updated.
2008-05-24 18:34:27 +02:00
Brian Paul
f2533e787e AA tri and glMaterial fixes 2008-05-16 15:35:19 -06:00
Brian Paul
b2ccd5c1ae fix memory access error in vbo_bind_vertex_list
Picked from master (commit 8fc1a6808d)
2008-05-16 15:35:19 -06:00
Brian Paul
6f63543dd7 fix an attr/src mix-up when setting-up/binding vertex arrays
This fixes problems with incorrect material coefficients when glMaterial
is called per-vertex.
2008-05-16 15:35:19 -06:00
Brian Paul
c966f1629b fix segfault in AA triangle code when using certain shaders 2008-05-16 15:35:19 -06:00
Xiang, Haihao
709f24adbb intel: Set right cliprects for the current draw region. fix #15057 2008-05-14 15:01:44 +08:00
Brian Paul
52fe7ea3d1 mesa: free shader program data before deleting shader objects.
Picked from master.
Fixes mem corruption seen when glean/api2 test exits.
2008-05-07 16:10:32 +01:00
Xiang, Haihao
44f6a6f9c4 i915: Add E7221 variant to i915.
Cherry picked from commit 39bcbe0921
2008-05-07 14:09:28 +08:00
Brian Paul
ac88b3dd16 Add support for GL_REPLACE_EXT texture env mode.
GL_REPLACE_EXT comes from the ancient GL_EXT_texture extension.  Found an old demo that
actually uses it.
The values of the GL_REPLACE and GL_REPLACE_EXT tokens is different, unfortunately.
2008-04-30 16:08:28 -06:00
Brian Paul
27b6fa5615 mesa: adjust glBitmap coords by a small epsilon
Fixes problem with bitmaps jumping around by one pixel depending on window
size.  The rasterpos is often X.9999 instead of X+1.
Run progs/redbook/drawf and resize window to check.

Cherry picked from gallium-0.1 branch
2008-04-29 18:38:26 -06:00
Brian Paul
68eb76dfb7 Enabled GL_EXT_multi_draw_arrays extension in R200/R300 drivers 2008-04-29 18:38:26 -06:00
Ove Kaaven
392760ee0c r200: fix state submission issue causing bogus textures (bug 15730) 2008-04-29 22:11:23 +02:00
Michel Dänzer
063b60a51f Change default of driconf "allow_large_textures" to announce hardware limits.
The previous default these days served mostly to cause artifical problems with
GLX compositing managers like compiz (see e.g.
http://bugs.freedesktop.org/show_bug.cgi?id=10501).

(cherry picked from commit acba9c1771)
2008-04-29 19:03:44 +02:00
Brian Paul
7b676192e9 enable GL_EXT_multi_draw_arrays (see bug 15670) 2008-04-24 16:31:28 -06:00
Brian Paul
138e0010eb added FreeBSD static config 2008-04-23 08:23:18 -06:00
Anatolij Shkodin
36bad2b478 added freebsd-static 2008-04-23 08:23:18 -06:00
Alan Hourihane
f6d6fc603e revert 2008-04-22 23:08:35 +01:00
Alan Hourihane
5545b46571 small cleanups 2008-04-22 20:29:42 +01:00
Alan Hourihane
06b0a7acce correct the return value 2008-04-22 20:29:00 +01:00
Alan Hourihane
534f30064d Fix error string 2008-04-22 20:28:35 +01:00
Brian Paul
6c9e1b2337 fix GL_ARB_texture_rectangle breakage 2008-04-14 13:44:05 -06:00
Brian Paul
62c67576e0 fixed WIN32 compile problem in libGLU 2008-04-14 13:41:17 -06:00
Brian Paul
91d59e4cb9 check for _WIN32 and __WIN32__ 2008-04-14 13:40:39 -06:00
Brian Paul
8de268bb3c add -Wl case (part of prev DragonFly patch) 2008-04-14 12:58:51 -06:00
Brian Paul
4fd7f6047c define #extension GL_ARB_texture_rectangle 2008-04-14 12:58:51 -06:00
David Flynn
ff3033e190 define #extension GL_ARB_texture_rectangle 2008-04-14 12:58:51 -06:00
Roland Scheidegger
18404076e3 r200: fix XPD vertex program instruction when using temps as inputs
due to the two read ports limit into temp memory may need the MAD_2 instruction
for the second instruction of the decomposed XPD.
While here, also try to avoid MAD_2 for MAD if all 3 inputs are temps but the
temps aren't actually distinct.
2008-04-12 02:40:44 +02:00
Brice Goglin
59c9c40c01 Prepare changelog for upload 2008-04-11 08:42:55 +02:00
Brice Goglin
666e0635da New upstream release 2008-04-11 08:41:32 +02:00
Brice Goglin
f9968a0346 Merge tag 'mesa_7_0_3' into debian-unstable 2008-04-11 08:38:26 +02:00
Hasso Tepper
39dca05d1d New dragonfly configs
A re-do commit, this time with the intended commit message.
2008-04-09 19:05:59 -06:00
Brian Paul
f51b76b670 Revert "Hasso Tepper <hasso@estpak.ee>"
This reverts commit 40ee989db5.
2008-04-09 19:04:36 -06:00
Brian Paul
40ee989db5 Hasso Tepper <hasso@estpak.ee> 2008-04-09 19:02:41 -06:00
Brian Paul
ef76dfc7a0 add link to 7.0.4 relnotes 2008-04-09 19:01:07 -06:00
Brian Paul
02ddc08353 initial 7.0.4 relnotes 2008-04-09 19:00:53 -06:00
Hasso Tepper
4b24d5261a patches for DragonFly OS 2008-04-09 18:59:19 -06:00
Brian
4b96a39c48 added MD5 sums for 7.0.3 2008-04-04 19:24:32 -06:00
Brian
48616e71d2 final prep/changes for 7.0.3 release 2008-04-04 19:06:30 -06:00
Brian
70d227ac62 7.0.3 release candidate 3 2008-04-01 18:04:13 -06:00
Eric Anholt
7f2c4f96f5 Bug #13492: Only call ProgramStringNotify if program parsing succeeded.
Wine intentionally tries some out-of-spec programs to test strictness, and
calling ProgramStringNotify on the results of a failed program parse resulted
in crashes in the 965 driver.
(cherry picked from commit 185320ae13)
2008-04-01 16:27:48 -07:00
Roland Scheidegger
667f0f60fc fix mistakenly set ATIFragmentShader._Enabled bit (bug 15269) 2008-04-01 23:25:04 +02:00
Brian
7592b8cc10 added some missing calls to _mesa_enable_x_y_extensions() 2008-04-01 08:41:24 -06:00
Brian
f55b831859 fix ARB f/v program comment parsing bug 2008-03-31 15:00:18 -06:00
Markus Amsler
217f7f7e5d fix parsing bug involving comments at the end of ARB v/f programs 2008-03-31 15:00:18 -06:00
Julien Cristau
2e5a63f365 Prepare changelog for upload 2008-03-31 16:47:34 +02:00
Julien Cristau
7c614b4666 Fix libgl1-mesa-swx11-i686's dependencies
Change libgl1-mesa-swx11-i686's pre-dependency on libgl1-mesa-swx11 to a
regular versioned dependency, and add ${shlibs:Depends}.
I'm not sure why this used Pre-Depends in the first place, I don't think
that's needed.
2008-03-31 16:47:17 +02:00
Julien Cristau
b23d908829 Update 02_use-ieee-fp-on-s390-and-m68k.patch. 2008-03-31 13:17:24 +02:00
Julien Cristau
5379b84e63 Update changelog and add bug closer 2008-03-31 13:11:59 +02:00
Julien Cristau
32c913308e Merge branch 'mesa_7_0_branch' of git.freedesktop.org:/git/mesa/mesa into debian-unstable 2008-03-31 13:00:20 +02:00
Xiang, Haihao
1e83d70b6d i965: Apply -ve rhw to 965. Fix the regression introduced by
commit da476ff02d. (bug#14940)
2008-03-31 09:37:55 +08:00
Brian
767dfa5b9c fix texture/renderbuffer mix-up in test_attachment_completeness() 2008-03-28 13:25:25 -06:00
Alan Hourihane
7ff5b38126 Fix build problems 2008-03-27 16:36:12 +00:00
Brian
e209f5300d updated glext.h to version 40 2008-03-25 08:11:47 -06:00
Brian
3c4b50c352 updated to version 40 2008-03-25 08:11:36 -06:00
Brian
4e7c2fcf18 add GL_READ_FRAMEBUFFER_BINDING_EXT case, regenerate get.c file 2008-03-24 19:55:46 -06:00
Brian
22534f94f5 fix wrong values for GL_READ/DRAW_FRAMEBUFFER_BINDING_EXT tokens 2008-03-24 19:55:04 -06:00
Brian
1a6928fdbe use ctx->Driver.DeleteProgram() in a few more places 2008-03-22 10:27:08 -06:00
Brian
325dbbac47 delete default programs with ctx->Driver.DeleteProgram() 2008-03-22 09:11:35 -06:00
Brian
39ac6b0481 Fix some PBO breakage.
In _mesa_Bitmap, can't early return if bitmap ptr is NULL, it may be an offset
into a PBO.  Similarly for _mesa_GetTexImage.
2008-03-21 12:32:29 -06:00
Brian
46cc4854e9 fix IEEE_ONE definition for ICC compiler (bug 15134) 2008-03-19 17:25:18 -06:00
Alan Hourihane
f93882512e set outputs_safe to 0 as it's possible for the code generation
to slip over the allocated memory for the vb.

pull in sse fixes from gallium-0.1
2008-03-19 01:03:48 +00:00
Markus Amsler
0dee2a4f6f only set InputsRead bit if input is really used 2008-03-17 08:35:34 -06:00
Brian
3cebc35669 init tmpNode to zeros 2008-03-14 14:16:00 -06:00
Brian Paul
e75a204fb9 mesa: fix emit_clamp() so that we don't use an output register as temporary
IR_CLAMP is decomposed into OPCODE_MIN+OPCODE_MAX.  Allocate a temporary
register for the intermediate value so we don't inadvertantly use an output
register (which are write-only on some GPUs).
2008-03-14 13:57:53 -06:00
Roland Scheidegger
ac06a5c16a fix state.lightprod ambient/specular w value (bug #14983) 2008-03-13 12:43:53 +01:00
Dan Nicholson
ed758fee0c Merge branch '7.0-glx' into mesa_7_0_branch 2008-03-10 16:45:49 -07:00
Jeremy Huddleston
a21c61ee8b Darwin: Fixed small error in darwin config files 2008-03-10 16:02:55 -07:00
Dan Nicholson
24697da20e glapi: Resync generated files with 7.0 branch
The previous commits pulled in the generated files from master. This
regenerates the files from the source on the 7.0 branch.
2008-03-10 15:17:59 -07:00
Jeremy Huddleston
7120c0089d Darwin: Config/source fixes to now build glxdemo apps and OSMesa
Also added darwin-fat-32bit darwin-fat-all configs and deleted old darwin-x86ppc config
2008-03-10 15:13:44 -07:00
Dan Nicholson
88a436a8f7 glapi: Generate xserver glapi sources in the mesa tree
Instead of generating the glapi sources for the xserver and commiting
them to the xserver tree, we can keep them in the mesa tree and change
the xserver build to use the files from the mesa tree.

This makes the xserver glx build more robust as it reduces the chances
for mismatches of the glX API used in the xserver vs. in mesa.
(cherry picked from commit 7688791fc5)
2008-03-10 15:03:13 -07:00
Dan Nicholson
2fdb5a245d glapi: Correct prerequisites for gl_and_glX_API.xml
The indirect_dispatch.h and indirect_table.c source files use
gl_and_glX_API.xml in their generation rather than glX_API.xml, but it
wasn't listed in their prerequisites. In turn, gl_and_glX_API.xml uses
glX_API.xml, but this is already listed in $(COMMON_GLX).
(cherry picked from commit 929b6d2ebf)
2008-03-10 15:03:01 -07:00
Dan Nicholson
8441b53538 glapi: Use make automatic variables to clean up the commands
Make use of the make automatic variables $@ (the target) and $< (first
prerequisite) to clean up the commands for the glapi generation. This
improves readability and guards against typos since words are repeated
less frequently.
(cherry picked from commit fc67d47708)
2008-03-10 15:02:51 -07:00
Dan Nicholson
d336df8b73 glapi: Use variable for indent and flags
Put the path to indent and the flags to call it with in configs/default
rather than in the Makefile. This makes it easier to change the values
globally.
(cherry picked from commit 817af9bec2)
2008-03-10 15:02:36 -07:00
Brian
8aaf805b8a fix parsing of state.texenv.color (bug 14931) 2008-03-10 07:41:31 -06:00
Markus Amsler
8161fd2785 Set normalized flag for GLubyte arrays in _mesa_VertexAttribPointerNV() 2008-03-09 17:53:00 -06:00
Markus Amsler
2f23025dfe init vertex weight attrib to (1,0,0,0) 2008-03-09 17:51:21 -06:00
Brian
1867eac230 fix __builtin_expect() definition test for IBM XLC (sf bug 1909832) 2008-03-09 10:54:50 -06:00
Roland Scheidegger
096e35d05f state.depth.range alpha value should be 1, not 0 (bug #14733) 2008-03-01 02:57:01 +01:00
Xiang, Haihao
8998f52b97 Don't Swap buffer if a DRIDrawable is entirely obscured by another window. 2008-02-28 11:43:54 +01:00
Brian
0fd38dcc83 Replace glut_fbc.c with glut_fcb.c (cb=callback) 2008-02-25 10:57:54 -07:00
Julien Cristau
d89cf0649e Add bug closer for #408679. 2008-02-24 10:23:17 +01:00
Julien Cristau
31e9d9d4f8 Prepare changelog for upload. 2008-02-23 18:18:42 +01:00
Julien Cristau
ea9c6d3d9e 03_optional-progs-and-install.patch: partly applied upstream, fixed up 2008-02-23 16:08:53 +01:00
Brian
ff63cf8068 bump libGL.so version number 2008-02-22 07:59:39 -07:00
Julien Cristau
eaac94c35a Update changelog. 2008-02-22 12:16:27 +01:00
Julien Cristau
9291506af0 Merge branch 'mesa_7_0_branch' of git.freedesktop.org:/git/mesa/mesa into debian-unstable
Conflicts:

	src/glx/mini/Makefile
2008-02-22 12:09:35 +01:00
Brian
4716670de9 prep for 7.0.3 rc-2 2008-02-21 16:59:20 -07:00
Brian
33c5b38034 latest bug fixes 2008-02-20 09:39:59 -07:00
Brian
5737d6c565 Fix point rasterization regression caused by commit 95a2eb9767
Even-sized, non-AA points need 0.5 bias in position to hit the right pixels.
Test program from bug report 11016 retested and is OK.
2008-02-20 09:38:16 -07:00
Brian
08a7f56c6a raise GL_INVALID_OPERATION if glProgramString compilation fails
cherry picked from master.
2008-02-20 09:00:49 -07:00
Brian
7916f2b4aa Fix glBegin-time test for invalid programs/shaders.
Cherry picked from master.
2008-02-20 08:56:20 -07:00
Brian
193d303ac7 Added <size name="Get" mode="get"/> lines for point parameter tokens.
Cherry picked from master
2008-02-19 08:46:39 -07:00
Jeremy Huddleston
e70609b7b8 Apple: Pulled in changes from Apple's patchset to allow mesa to build on darwin again 2008-02-19 00:54:35 -08:00
Xiang, Haihao
da476ff02d i965: new integrated graphics chipset support 2008-02-19 15:59:50 +08:00
Brian
77e3b5d28b Fix potential glDrawPixels(GL_DEPTH_COMPONENT) crashes (bug 13915)
Also, general clean-up of the Xlib-optimized glDraw/CopyPixels code.
2008-02-16 09:56:26 -07:00
Brian
bf97ca448c Fix glBindTexture crash (bug 14514) 2008-02-15 13:42:19 -07:00
Roland Scheidegger
c5f8ff8b32 fix bug with generated fragment programs if vertex shader is active
When generating a fragment program from fixed function, checking
texUnit->_ReallyEnabled is not sufficient, need texUnit->Enabled too
since the former also represents texture enables from an active vertex
shader.
2008-02-15 17:26:58 +01:00
caner
50465766d1 miniglx doc updates 2008-02-14 09:34:08 -07:00
Xiang, Haihao
7d8df58a63 _mesa_swizzle_ubyt_image: Don't use single swizzle_copy call
if components don't match. fix #13508
2008-02-13 07:07:47 -07:00
Andy Skinner
762c074012 added -altopts to allow overriding all other opts 2008-02-07 13:19:55 -07:00
Brian
51f2ee3bfb fix bug 9871 2008-02-06 07:55:26 -07:00
Brian
b5cd34aa21 Fix bug 9871: enable user-defined clip planes for R300
Patch provided by Erkin Bahceci <erkinbah@gmail.com>
2008-02-06 07:54:24 -07:00
Tony DeFeo
ae5c6dcd42 [965] Fix memory leak when deleting buffers with backing store. 2008-02-05 10:49:44 -08:00
Roland Scheidegger
97196d0c8c regenerate glsl library functions 2008-01-31 01:36:23 +01:00
Roland Scheidegger
f6de56b88a fix w component of glsl vec4 asin 2008-01-31 01:35:52 +01:00
Brian
d64ea43b76 fix bugs 13507, 14293 2008-01-30 08:12:06 -07:00
Brian
2deaf93d24 check if fb->Delete is null (bugs 13507,14293) 2008-01-30 08:12:06 -07:00
Alex Deucher
a107ec8300 Add new RV380 pci id
bug 14289
(cherry picked from commit 80efe27560)
2008-01-29 10:15:32 -05:00
Alex Deucher
120a1f9508 Merge branch 'mesa_7_0_branch' of git+ssh://agd5f@git.freedesktop.org/git/mesa/mesa into mesa_7_0_branch 2008-01-29 09:50:26 -05:00
Alex Deucher
bb84007a57 R300: RV410 SE chips have half the pipes of regular RV410
This fixes 3D rendering on x700 SE chips.  Reported
by Kano.
(cherry picked from commit 0b7e0f8159)
2008-01-29 09:46:48 -05:00
Alan Hourihane
86234e55a6 pull some more fixes for pbo access from trunk 2008-01-28 11:23:53 +00:00
Alan Hourihane
61972077cd fix some pbo path problems 2008-01-28 11:15:53 +00:00
Xiang, Haihao
5a7feb8ea2 i965: valid message length includes message header. 2008-01-25 17:23:55 +08:00
Xiang, Haihao
4b4c131cd0 i965: re-define the type of reg.loopcount.
avoid some issues such that 1 + (-2) gets a big
positive value.
2008-01-25 17:23:15 +08:00
Julien Cristau
67617c8f89 Drop -O0 workaround on hppa.
Stop building with -O0 on hppa. Bug #451047 should be fixed in recent gcc
versions.
2008-01-24 22:52:48 +01:00
Brian
0fd23f01c6 Assorted patches for miniglx/linux-solo (Gavin Li <codeview@gmail.com>) 2008-01-23 16:04:02 -07:00
Adam Jackson
557b0d9df7 glxinfo: Fix multisample visual reporting.
strstr() == 0 tests for the string _not_ being present.  Originally
Red Hat bug #351871.
2008-01-22 14:04:39 -07:00
Brian
888f4380cf fix GLX vertex array bug 14197 2008-01-22 08:41:22 -07:00
Brian
3266c5e95a fix pc vs. gc->ps usage (bug 14197) 2008-01-22 08:38:48 -07:00
Brian
834decdaae Don't build yuvrect_client by default
Added 'extra' rule to build optional yuvrect_client, shape, xdemo programs
2008-01-22 07:56:31 -07:00
Brian
18b2d83173 remove unused var 2008-01-22 07:54:37 -07:00
Alexey Sokolov
30c65c3c62 define M_PI if needed 2008-01-22 07:42:40 -07:00
Brian
04fcc4cf1e prep for 7.0.3 release 2008-01-21 16:21:39 -07:00
Julien Cristau
09504dde1f Update changelog; add Vcs-* and Homepage fields in debian/control. 2008-01-17 22:26:48 +01:00
Julien Cristau
c04885f072 Merge branch 'mesa_7_0_branch' of git://anongit.freedesktop.org/git/mesa/mesa into debian-unstable
Conflicts:

	progs/tests/.gitignore
	progs/tests/Makefile
2008-01-14 22:30:03 +01:00
Michel Dänzer
48ae5cf09d i965: Fix byte vs. pixel unit mixup for aligned texture pitch.
I sincerely hope I don't manage to mess this up yet again...

Thanks again to Todd Merrill for pointing out the problem and testing the fix
on IRC.
2008-01-14 16:30:58 +01:00
Michel Dänzer
45cdb6eb45 i965: Fix unresolved symbol intel_miptree_pitch_align.
For some reason, I thought that function was shared with i915tex... The good
news is that i965 gets to keep the lower pitch alignment requirements.

Thanks to Todd Merrill for pointing out the problem on IRC.
2008-01-11 16:32:39 +01:00
Keith Packard
b9f3f732aa i915tex: Centralize mipmap pitch computations.
mipmap pitches must account for the device alignment requirements, which
used to be fairly simple; just align to a 4-byte boundary. However, to allow
textures to be drawn to under TTM, they now need to be aligned to a 64-byte
boundary. Placing all of the alignment constraints in a single function
allows this new constraint to be applied uniformly.

There was some pitch constraining code in intel_miptree_create, but that was
modifying the pitch long after the miptree had been layed out, so it only
served to wreck the mipmap and cause rendering errors.

(cherry picked from commit a183efc132 with manual
changes for mesa_7_0_branch by Michel Dänzer. One side effect of this is that
i965 also aligns the pitch of 2D textures to multiples of 64 bytes, hopefully
that won't cause any issues)
2008-01-08 16:48:13 +01:00
Brian
17006ddd6b fix 3d proxy texture depth bug 2008-01-05 15:44:51 -07:00
Brian
b77a354df3 fix depth/1 typo in glTexImage3D proxy code 2008-01-05 15:44:51 -07:00
Brian
71d46beebf additional stub functions 2008-01-02 07:23:41 -07:00
Brian
454e296eb1 additional GL_COLOR_ATTACHMENTx_EXT cases (bug 13767) 2008-01-02 07:20:08 -07:00
Brian
f9b696be2b fix vbo display list memleak upon context destruction 2008-01-01 10:23:56 -07:00
Brian
c923edbc71 additional GL_COLOR_ATTACHMENTx_EXT cases (bug 13767) 2008-01-01 10:10:18 -07:00
Brian
e20c1d987f fix GLSL uniform/attrib bugs (13753) 2008-01-01 09:59:51 -07:00
Bruce Merry
cd354eb10f Convert to 0/1 when setting boolean uniforms
Also add some extra tests to the shader_api regression tests
2008-01-01 09:59:51 -07:00
Bruce Merry
e9ac27ee23 Make use of count in _mesa_uniform_matrix 2008-01-01 09:59:50 -07:00
Bruce Merry
3f9dc9f5b6 Add a test program to test for assorted bugs in shader_api.c 2008-01-01 09:59:50 -07:00
Bruce Merry
ba709875ae More fixes to shader_api
- return GL_INVALID_OPERATION instead of GL_INVALID_VALUE if location is bad
- correct the type-checking of uniforms from my previous commit
- accept location of -1 in _mesa_uniform_matrix
2008-01-01 09:59:50 -07:00
Bruce Merry
80f8397b35 Fix several bugs relating to uniforms and attributes in GLSL API
- fix sizes for GL_FLOAT_MAT2x3 and GL_FLOAT_MAT4x3 in sizeof_glsl_type
- fix size returns in _mesa_get_active_attrib
- fix out-of-bounds array access to vec_types in _mesa_get_active_attrib
- fix queries of matrix uniforms in _mesa_get_uniformfv
- fix _mesa_get_uniformfv to only return one base, even from an array
- allow location == -1 in _mesa_uniform
- validate types in _mesa_uniform
- allow array overruns in _mesa_uniform
2008-01-01 09:59:50 -07:00
Brian
b45fa27fa1 added 'get' info for framebuffer object tokens 2008-01-01 09:26:30 -07:00
Brian
b59480645d add 'Get' info for MAX_3D_TEXTURE_SIZE (for bug 13811) 2008-01-01 09:04:13 -07:00
Brian
148cb36979 add missing double quote (bug 13878) 2008-01-01 07:59:58 -07:00
Roland Scheidegger
c1eb78f7ff fix GL_LINE_LOOP with drivers using own render pipeline stage (#12410, #13527)
primitive needs to include the begin/end flags (broken since vbo-0.2). Should
fix missing first/last line segment on gamma, i810, i915, mga, r200, radeon,
s3v, savage, unichrome (r300 already correct). Tested on r200, fixes #13527.
2007-12-22 18:49:42 +01:00
Brian
8aa0fd6b20 glGetActiveUniform fix 2007-12-20 09:08:12 -07:00
Brian
a1b3a908f5 return correct size from glGetActiveUniform (bug 13751) 2007-12-20 09:08:12 -07:00
Xiang, Haihao
f9e70d951a i965: allocate GRF registers before building subroutines,
it ensures there are sufficient registers for all subroutines.
2007-12-19 10:44:00 +08:00
Xiang, Haihao
41ed6be1da i965: restore the flag after building the subroutine of the
GS thread. fix #13240
2007-12-19 10:36:56 +08:00
Brian
27de28fc3e Remove -I$(TOP)/src/mesa/transform 2007-12-18 17:07:11 -07:00
Brian
dd2f01c229 simplify update two-side lighting test (follow-on to previous front/back-face changes) 2007-12-18 16:28:36 -07:00
Brian
deb5c56c77 fix NEED_SECONDARY_COLOR for vert/frag progs 2007-12-18 16:28:36 -07:00
Julien Cristau
9fd22f0e3f Prepare changelog for upload. 2007-12-18 19:13:27 +01:00
Julien Cristau
942372dcdf debian/control fixups
* Bump Standards-Version to 3.7.3.
* Move libgl1-mesa-swx11-dbg, mesa-common-dev and libosmesa6-dev to section
  libdevel.
* libgl1-mesa-swx11 conflicts with libgl1-mesa-glx.  Move it and
  libgl1-mesa-swx11-dev to priority extra.
* Fix typo in mesa-common-dev's long description.
2007-12-18 13:57:06 +01:00
Julien Cristau
b7f901088b Update changelog. 2007-12-18 13:32:01 +01:00
Julien Cristau
c02990525c Merge branch 'mesa_7_0_branch' of git.freedesktop.org:/git/mesa/mesa into debian-unstable
Conflicts:

	progs/trivial/quad-clip-nearplane.c
2007-12-18 12:28:29 +01:00
Michel Dänzer
0107acded0 i915tex: Fix issues with glDrawBuffer(GL_NONE).
Don't dereference NULL renderbuffer pointer, and make sure the software
fallback sticks.

Fixes https://bugs.freedesktop.org/show_bug.cgi?id=13694 .
2007-12-17 12:32:26 +01:00
Brian
2ac5e08d1d fix polygon cull regression 2007-12-14 14:36:23 -07:00
Dan Nicholson
1837b8c214 configs: Fix linking with static libGL and --as-needed
Linking of the programs breaks when using a static libGL and the GNU ld
option --as-needed. This is because libXext is needed for the XShm
functions.
(cherry picked from commit 6aaf9bdd41)
2007-12-07 11:51:18 -08:00
Brian
03d2bc1774 fix gl_FrontFacing 2007-12-04 14:07:15 -07:00
Brian
385bddbde0 fix span->facing computation and gl_FrontFacing initialization 2007-12-04 14:07:15 -07:00
Brian
c135426267 Fix gl_FrontFacing compilation problem 2007-12-04 14:07:15 -07:00
Michel Dänzer
4be0c98120 Use -Bsymbolic for linking all shared objects.
Fixes https://bugs.freedesktop.org/show_bug.cgi?id=10132 .

Also remove comment about SONAME, as SONAME only applies to shared libraries.
(cherry picked from commit fc7ddea853)
2007-12-04 18:33:07 +01:00
Brian
e760aebd5a fix two-side lighting bug/crash 2007-12-03 14:13:08 -07:00
Brian
1f6e7d9c29 fix DD_TRI_LIGHT_TWOSIDE bug (#13368) 2007-12-03 14:12:34 -07:00
Dan Nicholson
c6d421af05 Fix the library name in glw.pc
Fix a copy and paste error s/GLU/GLw/ in glw.pc.
(cherry picked from commit 2dc85e8078)
2007-12-03 12:08:03 -08:00
Brian
952df5e493 fix two-sided stencil 2007-12-03 12:00:21 -07:00
Brian
45acb8b058 added missing quote char 2007-12-03 12:00:21 -07:00
Michel Dänzer
e8997c0183 i915tex: Fix up state changes for i8xx.
(cherry picked from commit 2af613e0b8)
2007-12-03 09:38:13 +01:00
Brian
169e62f563 fix-build: remove ctx->_Facing assignment 2007-11-30 15:52:41 -07:00
Brian
1b43babfb1 fix broken two-sided stencil 2007-11-30 13:01:42 -07:00
Brian
43e902f774 better front-plane clip test 2007-11-30 09:09:12 -07:00
Xiang, Haihao
b0a800e249 i965: use uncompressed instruction to ensure only
Pixel Mask Copy is modified as the pixel shader thread
turns off pixels based on kill instructions.
2007-11-30 12:36:07 +08:00
Brian
73d5f232ad New ctx->Driver.Map/UnmapTexture() functions for accessing textures from t_vb_program.c 2007-11-29 08:18:11 -07:00
Brian
f7209541e4 cleanups, comments
Conflicts:

	src/mesa/tnl/t_vb_program.c
2007-11-29 08:18:11 -07:00
Brian
397a32dca7 Move _mesa_load_tracked_matrices() from TNL module to prog_statevars.c 2007-11-29 08:18:11 -07:00
Roland Scheidegger
e0719d7122 make sure state token values are fully initialized 2007-11-29 03:08:48 +01:00
Michel Dänzer
3f18c0a9f2 r200: Fix texture format regression on big endian systems.
See https://bugs.freedesktop.org/show_bug.cgi?id=13324 .

Also use tx_table_be for VALID_FORMAT, in case r200SetTexImages ever gets
called for MESA_FORMAT_RGB888.
(cherry picked from commit dc88a96631)
2007-11-28 10:25:01 +01:00
Delle
b14be61938 use DEFAULT_SOFTWARE_DEPTH_BITS 2007-11-27 18:18:48 -07:00
Brian
440d620308 minor additions to avoid FAQs 2007-11-27 10:32:39 -07:00
Brian
8dd9df0f3e document GLSL float f/F suffix bug 2007-11-27 10:32:39 -07:00
Brian
6b9534eaf5 set fp->UsesKill when emitting OPCODE_KIL 2007-11-27 10:32:39 -07:00
Brian
c91d374ad7 improve 24-bit Z to 32-bit Z conversion 2007-11-27 10:32:39 -07:00
Keith Whitwell
1cab4160bc i915tex: Fix some minor batchbuffer errors. 2007-11-26 17:12:15 +01:00
Keith Whitwell
718d2dfbbd i915tex: Catch cases where not all state is emitted for a new batchbuffer.
This could lead to incorrect rendering or even lockups.
2007-11-26 17:10:26 +01:00
Michel Dänzer
0967e1270d i915tex: Some additional blit fixes and assertions. 2007-11-26 16:59:41 +01:00
Michel Dänzer
a08c02f1a8 i915tex: Actually wait for previous commands to complete for glFinish(). 2007-11-26 16:59:00 +01:00
Brian
bdfd9afead Consolidate texture fetch code and use partial derivatives when possible. 2007-11-23 12:01:39 -07:00
Brian
1c91a590d6 fix a few GLSL bugs 2007-11-23 10:28:33 -07:00
Brian
65bd7968bf Fix parsing of gl_FrontLightModelProduct.sceneColor, don't segfault on variable
array indexes.
2007-11-23 10:26:02 -07:00
Brian
6ccd23b87b need to check border width in sample_linear_2d() - fixes failed assertion in texwrap.c test 2007-11-23 09:14:27 -07:00
Brian
823409b7d0 Bump version numbers to 7.0.3 for next release 2007-11-22 09:36:35 -07:00
Brian
c85d31f4a0 Recompute ctx->Point._Size if GL_POINT_SIZE_MIN/MAX changes.
Note that all the point size clamping stuff has been redone in Mesa/master.
2007-11-22 09:36:35 -07:00
Roland Scheidegger
24af5c44da fix z buffer read/write issue with rv100-like chips and old ddx 2007-11-22 02:55:25 +01:00
Brian
75efacf8eb Initial 7.0.3 relnotes 2007-11-20 08:16:25 -07:00
Brian
825e810247 bring over Fortran fixes from master 2007-11-20 08:15:19 -07:00
Brian
e20723cfc1 Obsolete 2007-11-20 08:13:14 -07:00
Brian
9f39a67c1d Rename glut_fbc.c -> glut_fcb.c (cb=callback) 2007-11-20 08:13:03 -07:00
Brian
2044f3f791 clamp lambda to Min/MaxLod 2007-11-19 13:05:10 -07:00
Brian
30d4e8c422 fix some texture format assertions, etc 2007-11-19 10:37:41 -07:00
Brian
92d552f6a2 fix out-of-bounds array index (ix=-1) 2007-11-19 09:54:41 -07:00
Brian
5adfcbbc4f added gl_dispatch_stub_772() 2007-11-16 10:03:57 -07:00
Roland Scheidegger
69969b3819 fix position invariant vertex programs for sw-tnl
do the same math as for fixed function pipe, including
user clip planes.
(mostly resurrected from the dead t_vb_arbprogram.c code)
2007-11-15 02:06:15 +01:00
Roland Scheidegger
edfee04fcb fix bogus assumption if ddx has set up surface reg for z buffer
this is wrong since even if ddx has not set up a surface reg to cover the z
buffer we should pretend it has on those rv100 chips since they presumably do
not do z buffer tiling if not using hyperz, so we can use linear addressing
just the same. Doesn't seem to fix #13080, but it's wrong anyway and the bug
almost certainly broke newer non-tcl chips.
2007-11-15 01:16:27 +01:00
Brice Goglin
ae0bf74db9 Prepare changelog for upload 2007-11-13 21:44:00 +01:00
Julien Cristau
1bf7ce97c0 * Add build-dep on libxext-dev. Thanks, Timo Aaltonen! 2007-11-13 21:40:49 +01:00
Julien Cristau
4971a009e0 * Workaround gcc ICE on hppa: build libOSMesa with -O0 (see bug#451047). 2007-11-13 20:59:08 +01:00
Julien Cristau
e1613ab80d Don't set -fno-strict-aliasing in configs/debian-default.
Upstream uses that since 7.0.2.
2007-11-13 20:57:58 +01:00
Brian
28c9930888 add pointer to Gallium3D info 2007-11-13 09:51:14 -07:00
Brian
e3a35a123e DRI memory manager info (fixes dangling link) 2007-11-13 09:50:42 -07:00
Brian
412168f2e8 remove dependency on libGLU 2007-11-12 08:21:21 -07:00
Brian
fc0fa0d636 add glw.pc.in to tarball list, remove from DEPEND_FILES 2007-11-12 07:52:20 -07:00
Brice Goglin
caf283329b Prepare changelog for upload 2007-11-11 11:55:05 +01:00
Brice Goglin
5cedd4a43e Remove DESTDIR from INSTALL_DIR in configs/debian-default
... and update 03_optional-progs-and-install.patch accordingly.
2007-11-11 11:49:16 +01:00
Brice Goglin
0595d32d8a Refresh 03_optional-progs-and-install.patch
The default Makefile requires libGLU to be built
while linking progs/xdemos. However, we don't build
libGLU and progs/xdemos at the same time, and
these progs don't actually need libGLU.
2007-11-11 11:48:32 +01:00
Brice Goglin
e339b4e8da New upstream release 2007-11-11 00:34:46 +01:00
Brice Goglin
11c827c588 Merge tag 'mesa_7_0_2' into debian-unstable
Conflicts:

	.gitignore
	progs/demos/.gitignore
	progs/directfb/Makefile
	progs/egl/Makefile
	progs/fbdev/Makefile
	progs/fp/Makefile
	progs/miniglx/Makefile
	progs/samples/.gitignore
	progs/slang/Makefile
	progs/tests/.gitignore
	progs/tests/Makefile
	progs/tools/trace/Makefile
	progs/trivial/Makefile
	progs/vp/Makefile
	progs/xdemos/.gitignore
	src/egl/drivers/demo/Makefile
	src/egl/drivers/dri/Makefile
	src/egl/main/Makefile
	src/glu/mesa/Makefile
	src/glut/directfb/Makefile
	src/glx/mini/Makefile
2007-11-11 00:24:52 +01:00
Brian
8a40b670c3 added md5 checksums 2007-11-10 09:53:00 -07:00
Brian
d9fa5cbfc9 final changes for 7.0.2 relesae 2007-11-10 09:41:52 -07:00
Benno Schulenberg
1c21564292 code clean-ups, reformatting 2007-11-09 08:53:41 -07:00
Benno Schulenberg
6cc96bf156 remove commented-out code 2007-11-09 08:53:15 -07:00
Brian
29afd4bbc4 fix blender/unichrome crash, bug 13142 2007-11-08 07:36:30 -07:00
Benno Schulenberg
4e3db063ec fix Unichrome/Blender crash, bug 13142 2007-11-08 07:36:30 -07:00
Brian
6560744c38 added -fno-strict-aliasing and -fPIC flags 2007-11-08 07:36:30 -07:00
Brian
572ad87881 change filenames for rc2 2007-11-08 07:36:30 -07:00
Michel Dänzer
2ab75d6cfa r200: Fix SetTexOffset format for 16 bit pixmaps/textures.
Use symbolic array indices to clarify.
2007-11-06 18:39:50 +01:00
Chris Rankin
041a8eb5ec r200: Implement SetTexOffset hook.
Implementation guidance by Michel Dänzer, final testing by Timo Aaltonen.
2007-11-06 18:32:33 +01:00
Brian
c1c13bdcfa Check for NULL VB->EdgeFlag array.
There might be a bug elsewhere, but this is a simple work-around for now.
See bug 12614
2007-11-05 14:43:03 -07:00
Brian
403edd34dd fix mmx code (bug 12614) 2007-11-05 14:38:12 -07:00
Brian
f279e48416 fix mmx code (bug 12614) 2007-11-05 14:38:05 -07:00
Brian
bf854d8d27 SSE codegen fixes 2007-11-03 08:45:42 -06:00
Brian
f334121679 Fix mem leak in SSE code generation path (Michel Dänzer) and don't crash if _mesa_exec_malloc() returns NULL. 2007-11-03 08:44:48 -06:00
Dan Nicholson
b88e2be609 Add -fno-strict-aliasing workaround for all GCC targets
Use a GCC option to work around aliasing bugs. See commit 013dbcd for
more details.
(cherry picked from commit 01b18abf93)

Conflicts:

    configs/freebsd
2007-11-01 11:07:02 -07:00
Dan Nicholson
08229c8bb8 configs: Set -fexceptions for GLUT on linux-dri like other targets
Quite a while ago, the GCC option -fexceptions was added for building
libglut. See here:

  http://article.gmane.org/gmane.comp.video.mesa3d.devel/9499

This was missing in the linux-dri targets.
(cherry picked from commit dbcd20f1c2)
2007-11-01 11:02:05 -07:00
Dan Nicholson
c984017f71 configs: Always use -fPIC for shared libraries, never for static
Mesa currently disables -fPIC for DRI on x86, but most Linux distros are
re-enabling -fPIC for all DRI arches. Let's just do that here since
that's normally what's wanted for shared libraries. Some justification:

http://bugs.gentoo.org/show_bug.cgi?id=110840#c9

On the other hand, position-independent code is only necessary when
building shared libraries, so disable it for the static cases.
(cherry picked from commit c3b5adaa9a)
2007-11-01 11:01:10 -07:00
Dan Nicholson
00e7dd8a13 Add -fno-strict-aliasing workaround for Linux GCC targets
Most Linux distros work around aliasing problems in Mesa by compiling
with the GCC option -fno-strict-aliasing. Two examples:

  https://bugs.freedesktop.org/show_bug.cgi?id=6046
  http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=394311

This makes -fno-strict-aliasing the default with a comment that
developers should consider commenting it out. There is a already a note
about these bugs in docs/helpwanted.html.
(cherry picked from commit 013dbcd611)
2007-11-01 10:56:01 -07:00
Brian
d11b375b16 fix typo 2007-10-31 09:43:17 -06:00
Brian
d1afa8146f updated help/to-do list 2007-10-31 09:40:58 -06:00
Brian
31d4ba8a93 glDrawArrays(count=0) bugfix 2007-10-31 09:40:20 -06:00
Brian
1a7640958c check for count==0 in _mesa_validate_DrawArrays() 2007-10-31 09:39:20 -06:00
Brian
2e2a5a450b note G33/Q33/Q35 support 2007-10-31 08:19:14 -06:00
Brian
574fd63ff0 fix GL_ATI_separate_stencil 2007-10-30 10:46:31 -06:00
Brian
1063d47526 Alias glStencilOpSeparateATI with glStencilOpSeparate.
(Cherry-pick from master)
2007-10-30 10:45:47 -06:00
Brian
b59a892d3e Finish up ATI_separate_stencil
Add entrypoints to glapi XML file and regenerate files.
Implement glStencilOpSeparateATI().
Consolidate some code in stencil.c

Conflicts:

	src/mesa/glapi/dispatch.h
	src/mesa/glapi/glapioffsets.h
	src/mesa/glapi/glapitable.h
	src/mesa/glapi/glapitemp.h
	src/mesa/glapi/glprocs.h
	src/mesa/main/enums.c
	src/mesa/main/state.c
	src/mesa/sparc/glapi_sparc.S
	src/mesa/x86-64/glapi_x86-64.S
	src/mesa/x86/glapi_x86.S
2007-10-30 10:34:26 -06:00
Roland Scheidegger
6e5d5d77df add missing _mesa_StencilFuncSeparateATI function 2007-10-30 10:26:56 -06:00
Dan Nicholson
a6319d18cf Build xdemos programs by default on linux-dri
Since libglut is no longer hardcoded, we can build the xdemos programs
so long as a GLX enabled libGL and libGLU have been built.
(cherry picked from commit 06ca14543e)
2007-10-29 12:08:34 -07:00
Dan Nicholson
5103e883fd update APP_LIB_DEPS for static library configs (patch 3/3)
(cherry picked from commit 162914675a)
2007-10-29 12:06:50 -07:00
Dan Nicholson
98c4ea4b24 simplify APP_LIB_DEPS (patch 2/3)
(cherry picked from commit 9ef1d9f03a)
2007-10-29 12:06:23 -07:00
Dan Nicholson
29b8cac7d7 specify app lib dependencies in Makefiles (patch 1/3)
(cherry-picked from commit c05aa5ec4c)

Conflicts: progs/demos/Makefile
           progs/xdemos/Makefile
2007-10-29 12:05:54 -07:00
Dan Nicholson
cdb02d43fe linux-dri-xcb: Fix undefined refs when linking with libGL
GL_LIB_DEPS was missing -lXdamage and -lXfixes, which was causing
linker errors when trying to build the programs.
(cherry picked from commit 3b9e28d507)
2007-10-29 11:44:30 -07:00
Michel Dänzer
50cecddc4a driFenceType: Really return fence type, not flags.
This typo could lead to the i915tex driver waiting for an EXE only fence as RW,
causing it to abort with an error.
2007-10-29 13:27:59 +00:00
Alan Hourihane
177f6398e5 Only set R_MODE with NV_point_sprite 2007-10-28 20:08:15 +00:00
Dan Nicholson
887bd6a46f configs: Fix linux-static to link correctly
The linux-static target was missing necessary libraries and hardcoding
their location to /usr/X11R6/lib. This makes it comparable to the x86
and x86-64 static targets.
(cherry picked from commit 9514209593)
2007-10-26 15:49:00 -07:00
Brian
9af5153410 use grep -q 2007-10-25 14:48:08 -06:00
Dan Nicholson
0e2103689e More ignored programs merged back from master
Some more entries in progs/*/.gitignore that were in master.
2007-10-25 11:59:26 -07:00
Brian
2c496d8e46 simplify/fix the ASM_FLAGS tests (bug 12931) 2007-10-25 11:12:17 -06:00
Brian
8f0ba02e44 Cray Xt3 config 2007-10-25 11:12:07 -06:00
Brian
524c56e417 Fix merge 2007-10-24 16:36:00 -06:00
Brian
773f3a266c added catamount-osmesa-pgi config, fixes for cross-compiling 2007-10-24 16:35:47 -06:00
Brian
882ada0d22 don't build x86, x86-64 dirs if not needed 2007-10-24 16:33:24 -06:00
Brian
6229005f93 properly init dst reg's CondMask/Swizzle fields 2007-10-23 10:55:14 -06:00
Brian
6c63b35ef7 bump up MAX_INSTRUCTIONS and add an assertion to catch emitting too many instructions 2007-10-23 10:25:34 -06:00
Brian
fb85e50e09 add directfb glut sources to tarball 2007-10-23 08:22:11 -06:00
Brian
5e5f908d0c don't apply ColorMask to main gc 2007-10-22 17:54:07 -06:00
Michel Dänzer
c09e2143c7 i915tex: Don't emit 'empty' blit rectangles.
The hardware seems to interpret them differently and produce unexpected
results...
2007-10-17 18:28:46 +02:00
Brian
74ced1e67f fix fog, rescale_normals bugs (from gallium branch) 2007-10-15 18:02:14 -06:00
Andreas Micheler
dd02c1609e faster write_rgba_span_front() 2007-10-15 08:28:06 -06:00
Brian
cdf4880a79 replace 'brick' with correct program name in printfs 2007-10-10 10:57:37 -06:00
Mrc Gran
2ec8e8547d fix force_s3tc_enable option 2007-10-10 09:16:12 -06:00
Dan Nicholson
b7c36d1d33 Ignore more generated files in progs/
Many of the directories in progs/ were missing .gitignore files or
were tracking generated files. This patch is basically the process of
running `make' in each directory and then finding the untracked files
with `git-ls-files -o --exclude-per-directory=.gitignore'. The existing
files were also sorted.

Conflicts:

	progs/demos/.gitignore
	progs/tests/.gitignore
	progs/xdemos/.gitignore
2007-10-10 08:49:59 -06:00
David Nusinow
8913402b74 Revert "* Disable building i915tex."
This reverts commit 76556f8264.
2007-10-06 17:54:29 -04:00
David Nusinow
c00dfefd6f Revert "Don't release just yet."
This reverts commit c3de7b5ade.
2007-10-06 17:54:13 -04:00
David Nusinow
c3de7b5ade Don't release just yet. 2007-10-04 23:10:47 -04:00
David Nusinow
76556f8264 * Disable building i915tex.
It was previously only built on x86 and amd64.  It was never
  production-ready and relies on a kernel interface that never went
  upstream.
2007-10-04 22:49:27 -04:00
Xiang, Haihao
970d18a708 Brian's fix for bug9829 2007-10-04 00:08:18 +02:00
Roland Scheidegger
9456e7f0ff minor fog calc cleanup 2007-10-03 22:34:36 +02:00
Roland Scheidegger
a5db24adc2 fog: fix issues with negative fog coords (may fix #10529)
Rework tnl fog a bit. Make sure we always use ABS(eyez) when fog coord source
is depth, OTOH it does not seem to be necessary to use it (as was done before
in some cases) if fog coord source is fogcoord (just to save some work).
This fixes tests/fog (the first 2 cases) with i915/i915tex.
2007-10-03 22:33:26 +02:00
Roland Scheidegger
17664e2620 make sure optimized fog params get updated 2007-10-03 22:33:03 +02:00
Roland Scheidegger
43a0eb0253 unbreak 3d textures (typo when setting tex layout) 2007-10-03 22:32:33 +02:00
Michel Dänzer
bcdaed2c0a i915tex: Work around texture pitch related performance drops on i915 at least. 2007-10-03 10:48:56 +02:00
Brian
78c6637db5 updated glext.h license info (Khronos), plus other clean-ups 2007-10-01 17:56:48 -06:00
Brian
44214a98fd fix VBO-split infinite loop (bug 12164) 2007-09-29 12:01:47 -06:00
Dan Nicholson
9cc6d2310c add support for LDFLAGS env var 2007-09-28 18:47:11 -06:00
Brian
1a045954be update the DRM/DRI instructions 2007-09-28 18:43:12 -06:00
Dave Airlie
b3fc9a1585 i965: fix vblank on 965gm laptops by bringing in code from i915
jbarnes has better fixes for this in 7.1, but for 7.0.2 this seems like the best
plan, for distros to pick it up.
2007-09-28 12:02:27 +10:00
Xiang, Haihao
2a86a449ca mesa: make sure the gotten value isn't greater than the
max depth buffer value on 64bit system. fix bug #12095
2007-09-27 10:44:45 -06:00
Xiang, Haihao
44f032db61 i965: handle all unfilled mode in clip stage. fix bug #12453 2007-09-27 10:40:22 -06:00
Brian
5a520729ee prep for 7.0.2 release 2007-09-27 10:35:10 -06:00
Brian
36fbe78c60 for Miniglx, use git sources 2007-09-27 10:34:44 -06:00
Brian
029bb17770 bump versions to 7.0.2 2007-09-27 10:30:13 -06:00
Brian
7958bee6be fix blend bug 12164 2007-09-27 10:25:57 -06:00
Brian
0b9bb21ce4 Restore old _TriangleCaps code to fix Blender problem (bug 12164) 2007-09-27 10:25:33 -06:00
Roland Bär
cfcc5b794e [i965] Bug #11812: Fix fwrite return value checks in AUB file code. 2007-09-26 08:00:17 -06:00
Zou Nan hai
925ff408a6 fix fd.o bug #12217, recalcuate urb when clip plane size change 2007-09-26 07:59:41 -06:00
Xiang, Haihao
f40ca444f5 i965: The cube map texture coordinates must be devided by the
component with the largest absolute value before they are
delivered. fix bug #12421
2007-09-26 07:58:57 -06:00
Zou Nan hai
638a5e0382 fix a bug in 965 ARB_occlusion_query,
fd.o bug #12132
2007-09-26 07:58:29 -06:00
Brian
6d777bf055 fix DrawRangeElements error msg 2007-09-23 13:57:36 -06:00
Brian
ae6d4e8bc0 fix -D_BSD_SOURC 2007-09-20 18:42:59 -06:00
Brian
f2728724d3 fix for __IBMC__ 2007-09-20 15:12:42 -06:00
Brian
bc95b34cce added bluegene-xlc-osmesa, fortran fixes 2007-09-20 15:12:21 -06:00
Brian
9eb89616af fixes for bluegene-xlc-osmesa config 2007-09-20 15:10:35 -06:00
Brian
f7713a6d81 Added bluegene-xlc-osmesa config (Alexander Neundorf) 2007-09-20 15:09:54 -06:00
Brian
1cec9b727a fix bug 10604 2007-09-14 13:43:58 -06:00
Brian
18a4730f5b From trunk: mesa: bind VBO_ATTRIB_XXX to correct input array when executing a display list. fix #10604
This also fixes Bill Mitchell's Fortan/plot bug.
2007-09-14 13:43:09 -06:00
Brian
1342aefcdd updated link 2007-09-14 09:57:16 -06:00
Brian
1fd4b1e702 fix glCopyPixel/glPixelZoom bug 12417) 2007-09-14 09:57:08 -06:00
Brian
32dff44c28 fix copypixels overlap test bug (#12417) 2007-09-14 09:56:05 -06:00
Brian
e7c2ae476a fix ARB fp/vp parsing bug 12313 2007-09-13 11:42:48 -06:00
Brian
a66301f83d added program_error2() function for better error reporting 2007-09-13 11:41:44 -06:00
Brian
a5f7d56221 Fix state.texgen parsing error (bug 12313).
Replace *(*inst++) with *(*inst)++ in a few places.
2007-09-13 11:29:51 -06:00
Brian
140080c1ec added .pc.in files to tarballs 2007-09-13 09:27:06 -06:00
Brian
ef43af6d43 added DSTDIR, pkg-config items 2007-09-12 10:12:20 -06:00
Brian
949d0532f0 updated with pkg-config info 2007-09-12 10:12:06 -06:00
Dan Nicholson
96efc76d74 pkg-config support 2007-09-12 10:03:31 -06:00
Dan Nicholson
6ca0d63692 DESTDIR support. 2007-09-12 10:01:06 -06:00
Xiang, Haihao
95fc2485e2 i965: fix bug#11925 2007-09-12 16:58:45 +08:00
Xiang, Haihao
d6ce1e4834 i965: fix #11378 2007-09-12 15:29:11 +08:00
Brian
2cc2b12fc7 fix indirect rendering crash 2007-09-11 10:59:10 -06:00
Brian
2259f81cf7 In _mesa_make_current(), don't unbind FBOs from the old context.
This fixes the X server crash reported by KeithP on Aug 29.
The old context's FBOs will be unreferenced during context destruction so
there's no memleak with this change.
2007-09-11 10:59:10 -06:00
Brian
791f12880f Update depth test state when binding new framebuffer object 2007-09-11 09:25:49 -06:00
Brian
7aa7373a64 fix comment 2007-09-11 09:25:49 -06:00
Xiang, Haihao
2f186a190f i965: take the secondary color into account when drawing
bitmap. fix#10688
2007-09-11 17:03:07 +08:00
Xiang, Haihao
357893a873 i965: limit on LOD Bias, fix#11987 2007-09-11 17:02:15 +08:00
Brian
a17881da5f clean-up (see bug 12317) 2007-09-10 08:36:01 -06:00
Brian
4872ee83a9 updated VC7 project files 2007-09-05 10:05:48 -06:00
Christoph Kubisch
5ec222a832 updated VC7 project files 2007-09-05 10:01:58 -06:00
Brian
40438b3440 Use temporary matrix in __gluInvertMatrixd() to fix aliasing problem (see bugs 12269, 6748) 2007-09-03 16:21:12 -06:00
Xiang, Haihao
e0b80660c8 fix bug#11009 2007-09-01 15:31:49 +08:00
Xiang, Haihao
a2016301f5 i965: Take the upper limitation on LOD into account. 2007-09-01 15:14:58 +08:00
Brian
7ec6729fb6 sync with trunk, fixing bug 12239 2007-08-31 16:36:17 -06:00
Brian
837a2cf72e fix i965 bugs, x86-64 bugs 2007-08-31 16:35:38 -06:00
Roland Scheidegger
0adca14536 fix another occurence of movaps which might not be aligned 2007-08-31 00:06:30 +02:00
Brian
57b0f66c6d s/movaps/movups/ (see bug 12216) 2007-08-30 08:41:43 -06:00
Xiang, Haihao
e47e7a9571 i965: store read drawable info in intel_context. Some OpenGL
operations are based on read drawable. fix bug#10136.
2007-08-30 14:06:26 +08:00
Xiang, Haihao
68ca01560f i965: check NULL pointer. fix bug#12193 2007-08-30 14:06:19 +08:00
Xiang, Haihao
64f1c91fcb i965: samplers group in fours in WM_STATE. fix bug#9415 2007-08-30 14:05:56 +08:00
Xiang, Haihao
7d33f94198 i965: flush batch buffer when getting the maximum. This makes
some 3D programs such as pymol work well.
2007-08-30 14:05:39 +08:00
Julien Cristau
369cda3752 Prepare changelog for upload. 2007-08-28 12:11:47 +02:00
Julien Cristau
846078d06f Delete generated files. 2007-08-28 12:07:45 +02:00
Julien Cristau
b18500d56e * Update to latest git (from mesa_7_0_branch)
+ adds support for some new intel chipsets (i915 and i915_tex dri drivers)
    (closes: #437333)
  + broken inline asm in dri drivers fixed (closes: #423739)
2007-08-28 12:07:25 +02:00
Julien Cristau
7d92c50490 Merge branch 'mesa_7_0_branch' of git.freedesktop.org:/git/mesa/mesa into debian-unstable 2007-08-28 11:06:22 +02:00
Brian
d69a596193 Fixed bugs 6748, 12141 2007-08-27 10:48:31 -06:00
Colin McDonald
8aa4ae638e fixed problem with big glDrawArrays (see bug 12141) 2007-08-27 10:47:49 -06:00
Brian
3069f34841 new __gluInvertMatrix() function (Mesa bug 6748) 2007-08-27 10:36:25 -06:00
Adam Jackson
b3788a0e4f Unbreak Linux builds with -fvisibility=hidden. 2007-08-24 08:49:59 +01:00
Brian
582bc029ab remove CallDepth++/-- accidentally removed in prev commit 2007-08-24 08:45:27 +01:00
Brian
e9e3099433 remove unneeded CallStack array 2007-08-23 08:17:35 +01:00
Brian
4fc561cc7b refactor bounds checking code 2007-08-20 13:12:10 +01:00
Brian
ad3e3a0b99 don't map buffer in _mesa_validate_DrawElements() unless needed 2007-08-20 12:55:26 +01:00
Brian
bfb8c849cb Remove recent texobj refcount debug/logging code. 2007-08-16 09:11:11 +01:00
Michel Dänzer
09dd25b5d1 i915tex: Unreference texture buffers on context destruction.
Not doing this could cause the buffers to leak under some circumstances.
2007-08-16 08:45:16 +02:00
Roland Scheidegger
c48efb1dbe suppress warning about ncon visuals (bug #6689) 2007-08-16 02:43:45 +02:00
Brian
5408acb6b7 log deleting/binding of FBOs 2007-08-15 15:47:28 +01:00
Brian
15481160f9 more debug output (context destroy, fb destroy) 2007-08-15 12:52:00 +01:00
Brian
00ccff03a5 Rewrite quite a bit of the code for glPush/PopAttrib() for texture state.
Remove the Saved1D/2D/etc fields from gl_texture_attrib in mtypes.h
Use a new texture_state struct in attrib.c that has the extra information
for restoring texture object state and saving references to the texture
objects (so they can't accidentally get deleted while referenced by the
attribute stack).  All the texobj refcounting is mutex-protected now.
2007-08-15 11:21:06 +01:00
Brian
21594921b1 copy Target in _mesa_copy_texture_object() 2007-08-15 11:18:38 +01:00
Brian
42c91eebc9 Added _mesa_free_attrib_data() to free anything left in the attribute stack upon context destruction.
Also, a bit more refcount debug info.
2007-08-15 10:10:02 +01:00
Brian
88273e08b4 more tex obj ref count debugging (temporary) 2007-08-15 08:55:03 +01:00
Brian
08f7fb45ef added some temporary texobj ref counting debug output 2007-08-14 19:45:02 +01:00
Brian
75ebda3ce6 Fix a few more problems with freeing FBOs/textures during context destruction.
Free FBOs before textures since the later may be referenced by the former.
Need to bind the context we're destroying if there isn't a current context
so that ctx->DeleteTexture() etc can be used.
2007-08-14 11:57:16 +01:00
Brian
53ff15fcbf fix blending/banding bug 2007-08-13 21:17:04 +01:00
Dan Torop
72adb70ee2 fix spantmp2 READ_RGBA inline asm (#11931) 2007-08-13 21:14:51 +01:00
Brian
960d41fcfe added more i915/945 chipsets 2007-08-13 21:11:17 +01:00
Brian
0c4e35b070 Add PCI IDs for the G33, Q33, and Q35 chipsets.
Conflicts:

	src/mesa/drivers/dri/i915/i915_texstate.c
	src/mesa/drivers/dri/i915/intel_context.c
	src/mesa/drivers/dri/i915/intel_context.h
	src/mesa/drivers/dri/i915/intel_screen.c
	src/mesa/drivers/dri/i915/intel_tex.c
2007-08-13 21:06:33 +01:00
Wang Zhenyu
c53f1cb813 i915tex: Add support for 945GME 2007-08-13 21:00:28 +01:00
Brian
f6696cc15e free any render/framebuffers left in hash tables when freeing shared state 2007-08-13 17:39:05 +01:00
Michel Dänzer
fef9a41fe4 i915tex: Make sure pitch is aligned properly for render-to-texture.
Just always align texture pitch to multiples of 64 bytes for now, pending a
more sophisticated scheme to do it only when really necessary.
2007-08-13 13:22:43 +02:00
Brian
e279a0a076 Implement mutex/locking around texture object reference counting.
Use new _mesa_reference_texobj() function for referencing/unreferencing
textures.  Add new assertions/tests to try to detect invalid usage of
deleted textures.
2007-08-13 11:28:52 +01:00
Brian
88c8aaed96 fix link to 7.0.1 relnotes 2007-08-09 15:59:38 +01:00
Julien Cristau
34beaefe1c Prepare changelog for upload. 2007-08-09 16:00:58 +02:00
Julien Cristau
10b828f193 New upstream release. 2007-08-09 11:56:00 +02:00
Julien Cristau
43e6260ed0 Merge tag 'mesa_7_0_1' into debian-experimental 2007-08-09 11:25:42 +02:00
Brian
62cbbd7226 fix byte swap bug for GLuint stencil indexes (bug 11909) 2007-08-09 08:45:01 +01:00
Brian
f7cbe7fcee move free() after dereference (bug 11878) 2007-08-07 22:06:48 +01:00
Brian
d8e4ff9079 fix potential NULL dereference (bug 11879) 2007-08-07 21:56:39 +01:00
Brian
510ff27eda remove SHELL line, replace -e test with new logic (Daniel Stone) 2007-08-07 21:52:22 +01:00
Brian
5abf055100 fix potential NULL dereference (bug 11880) 2007-08-07 21:46:42 +01:00
Brian
4bb5721f6b fix swizzle error test (bug 11881) 2007-08-07 21:43:46 +01:00
Brian
c41ab70d3a initial 7.0.2 notes 2007-08-07 08:44:35 +01:00
Brian
a05e403bf2 fix bug 9962 (vbo splitting) as in trunk 2007-08-07 08:43:31 +01:00
Brian
2f57e0f71d Merge branch 'mesa_7_0_branch' of git+ssh://brianp@git.freedesktop.org/git/mesa/mesa into mesa_7_0_branch 2007-08-07 08:21:50 +01:00
Brian
95a2eb9767 fix even-sized point positioning (bug 11874) 2007-08-07 08:20:18 +01:00
Brian
55175daed7 added md5 sums 2007-08-03 13:54:55 -06:00
Brian
ae16a51e95 added GLU bug fixes 2007-08-03 13:39:09 -06:00
Brian
4398d67546 update 7.0.1 release info/date 2007-08-03 13:32:47 -06:00
Brian
c93738687c remove -rc2 2007-08-03 13:32:28 -06:00
Brian
799492e606 add OSMESA_LIB_NAME 2007-08-03 10:08:16 -06:00
Brian
f33ff071bb added -lm 2007-08-03 07:24:33 -06:00
Brian
99e854743c added bluegene-osmesa 2007-08-02 14:24:05 -06:00
Brian
6f79062d91 OSmesa on BlueGene (Alexander Neundorf) 2007-08-02 14:22:52 -06:00
Brian
2f682c3995 fix stencil value masking bug 11805, and fix sizeof() bug 2007-08-02 09:08:53 -06:00
Brian
067370e68f fix invalid pointer usage in bezierPatchDeleteList(), bug 11807 2007-08-02 08:40:58 -06:00
Brian
2aa439a6a4 fix double free()s (bug 11808) 2007-08-02 08:38:10 -06:00
Brian
b477182dc3 s/MAX_TEXTURE_LEVELS/SIS_MAX_TEXTURE_LEVELS/ and add assertion (bug 11806) 2007-08-02 08:35:27 -06:00
Michel Dänzer
a8964ca89e i915tex: Better attempt to release miptree when overriding texture image.
The previous approach could lead to crashes in FBO code that dereferences the
miptree struct pointer unconditionally.
2007-08-02 12:14:03 +02:00
Brian
4f0e92d07c Merge branch 'mesa_7_0_branch' of git+ssh://brianp@git.freedesktop.org/git/mesa/mesa into mesa_7_0_branch 2007-08-01 17:03:09 -06:00
Brian
577f4e8a5f Obsolete 2007-08-01 11:45:47 -06:00
Brian
0586d9fe56 fix error code test limit (bug 11795) 2007-08-01 11:27:47 -06:00
Michel Dänzer
1f0d9bf05e i915tex: Missing piece of SetTexOffset hook implementation for I830 generation. 2007-08-01 17:46:32 +02:00
Michel Dänzer
8bcae2a527 i915tex: Implement SetTexOffset hook.
Only build tested for I830 generation.
2007-08-01 17:46:31 +02:00
Michel Dänzer
35ca9aae91 r300: Implement SetTexOffset hook. 2007-08-01 17:46:30 +02:00
Michel Dänzer
6cdd2bf8d7 Add interfaces for overriding texture images with driver specific 'offsets'.
To be used by AIGLX for GLX_EXT_texture_from_pixmap without several
additional data copies.
2007-08-01 17:46:29 +02:00
Brian
ce7d175adb fix mem leak (bug 11793) 2007-08-01 08:42:13 -06:00
Brian
1904fd095f fix mem leak (bug 11791) 2007-08-01 08:42:09 -06:00
Brian
897d0ac5cc fix some FreeBSD issues 2007-07-31 14:57:54 -06:00
Brian
e262d0182f fix failure caused by undeclared variable (bug 11783) 2007-07-31 09:43:40 -06:00
Brian
aa5b748c10 glGetAttribLocation always returned 1 (bug 11774) 2007-07-31 09:29:31 -06:00
Julien Cristau
c6728df4a3 Merge tag 'mesa_7_0_1_rc2' into debian-experimental 2007-07-31 04:00:13 +02:00
Brian
3c182c5bc8 Added sunos5-v9-cc-g++ config (Roland Egger) 2007-07-30 09:00:09 -06:00
Brian
1b5677847b _mesa_pack_bitmap() fix from trunk 2007-07-30 08:13:05 -06:00
Brian
1be59a9dbe dlist polygon stipple fix from trunk 2007-07-30 08:12:29 -06:00
Brian
55c2178ad4 fix bug 11754 2007-07-29 18:28:09 -06:00
Jan Dvorak
3e3d392e1c fix glPointParameteriv bug 11754 2007-07-29 18:26:55 -06:00
Julien Cristau
593de57fa2 Prepare changelog for upload. 2007-07-27 20:18:21 +02:00
Julien Cristau
04b0e43ea6 New upstream release candidate. 2007-07-27 19:59:38 +02:00
Julien Cristau
04eee290c8 Merge branch 'mesa_7_0_branch' of git://anongit.freedesktop.org/git/mesa/mesa into debian-experimental 2007-07-27 19:58:30 +02:00
Brian
756a7a4e4b minor tweaks 2007-07-27 11:44:55 -06:00
Brian
121b4ac220 More Mingw32 fixes from Zhang (zxpmyth@yahoo.com.cn) 2007-07-27 11:12:21 -06:00
Brian
cdbd5db3af release candidate 2 2007-07-27 11:10:00 -06:00
Brian
b2240f6ad4 fix bug 11731 2007-07-26 15:33:39 -06:00
Brian
33814a55f8 Fix function call bug 11731. Also, fix up IR_CALL/IR_FUNC confusion. 2007-07-26 15:32:34 -06:00
Brian
a57c5a417b generate error upon writing to varying var in fragment program (bug 11733) 2007-07-26 11:39:37 -06:00
Brian
ff1a28de17 clamp float colors 2007-07-26 08:19:40 -06:00
Brian
772f57f99b call ctx->Driver.NewProgram() instead of _mesa_new_program() 2007-07-24 09:56:54 -06:00
Brian
dacee32b8b fix logic error, typos 2007-07-24 09:19:44 -06:00
Brian
fa1a3eb06d fix GLX_STEREO bug 2007-07-23 16:54:35 -06:00
Brian
bd4817ca54 fix GLX_STEREO handling (bug 11705) 2007-07-23 16:51:03 -06:00
Julien Cristau
d3a160e217 New upstream release candidate. 2007-07-23 16:39:48 +02:00
Julien Cristau
c5837f2791 Merge branch 'mesa_7_0_branch' of git://anongit.freedesktop.org/git/mesa/mesa into debian-experimental 2007-07-23 16:28:56 +02:00
Julien Cristau
490bd43f18 Merge branch 'debian-unstable' of git+ssh://git.debian.org/git/pkg-xorg/lib/mesa into debian-experimental
Conflicts:

	debian/changelog
2007-07-23 16:28:10 +02:00
Brian
005eea249c Fix a number of MINGW32 issues (Zhang <zxpmyth@yahoo.com.cn>) 2007-07-21 11:27:22 -06:00
Brian
9747de8ec6 drop VC6 and VC7 project files until updated 2007-07-21 10:03:30 -06:00
Brian
849a9799d0 fix bug 11588 2007-07-13 16:37:18 -06:00
Jan Dvorak
e32ae4fe33 fix shader/info string length queries (bug 11588) 2007-07-13 16:35:19 -06:00
Brian
4595389c4c fix width/depth mix-up (bug 11577) 2007-07-12 21:43:39 -06:00
Brian
e8ccc7cc49 Merge branch 'mesa_7_0_branch' of git+ssh://brianp@git.freedesktop.org/git/mesa/mesa into mesa_7_0_branch 2007-07-12 14:49:56 -06:00
Brian
673d21047a fix swizzle-related bug 11534 2007-07-12 14:49:24 -06:00
Julien Cristau
6186261589 Prepare changelog for upload. 2007-07-11 05:51:34 +02:00
Julien Cristau
45310e219c Relax libgl1-mesa-dev dependency on libgl1-mesa-glx.
Change the >= ${source:Version} dependency to >= ${source:Upstream-Version}.
This way libgl1-mesa-dev stays installable after a minor revision bump on
architectures where mesa isn't built yet.
2007-07-11 05:45:30 +02:00
Julien Cristau
250f6b2c88 Replace ${Source-Version} with ${source:Version} in one remaining spot. 2007-07-11 05:34:30 +02:00
Michel Dänzer
fdefc2bbda i915tex: Better procedure for dropping batchbuffer on virtual resolution change.
The previous procedure would often result in a GPU lockup.
2007-07-10 11:02:18 +02:00
Michel Dänzer
2079df8527 i915tex: Only wait for vblank when really necessary.
This avoids superfluous waits for vblank timing out under some circumstances.
2007-07-10 10:54:51 +02:00
Michel Dänzer
b53659452c Clear pointers to freed cliprects.
Not doing this could lead to double frees under rare circumstances.
2007-07-10 10:49:28 +02:00
Michel Dänzer
46f1d6653e Remove dubious compile-time test for pre-2.4 Linux kernels.
LINUX_VERSION_CODE shouldn't be used by userspace code, it can be defined empty
these days.

If anybody still cares about 2.2 kernels, they should reinstate this as a
proper runtime test.
2007-07-10 10:34:10 +02:00
Brian
c1938a60f7 Fix zero-size texture checking in _mesa_test_proxy_teximage(), bug 11309 2007-07-08 10:04:14 -06:00
Brian
d65110f352 Check if 'indices' is NULL, bug 11314 2007-07-08 09:21:21 -06:00
Brian
9fa3bbcb5a check depthScale value for optimized ushort->uint case (fixes bug 11474) 2007-07-08 08:53:09 -06:00
Brice Goglin
19e51ecbe6 libgl1-mesa-dev does not need to depend on libgl1-mesa-dri, libgl1-mesa-glx is enough.
libgl1-mesa-glx was included in libgl1-mesa-dri before 6.4.1-0.1,
so libgl1-mesa-dev had to depend on libgl1-mesa-dri at this point.
Since libgl1-mesa-dri only contains the DRI modules now, it is not
required anymore for libgl1-mesa-dev.
2007-07-07 22:17:35 +02:00
Brian
99d62f2922 fix i915 polygon stipple bug 2007-07-06 16:53:28 -06:00
Brian
284743cafe add GL_QUADS to i915_reduced_primitive_state() to fix redbook/polys demo 2007-07-06 16:52:34 -06:00
Xavier Bachelot
08d7307b2a call glutInit(), bug 11486 2007-07-06 12:54:40 -06:00
Brian
ef6a64abe6 fix a syntax error (missing right-paren) 2007-07-05 16:04:10 -06:00
Brian
18bfa52ef7 fix bug 11475 2007-07-05 09:42:07 -06:00
Brian
d62be2652c Copy from Mesa master. Includes fixes for bugs 11475 and 11448. 2007-07-05 09:40:37 -06:00
Julien Cristau
31f3433635 Prepare changelog for upload. 2007-07-05 00:56:56 +02:00
Samuel Thibault
639eaceee7 Fix build on hurd-i386.
Build libgl1-mesa-glx without direct rendering support on that arch, and
don't build any dri driver.
2007-07-05 00:48:30 +02:00
Eric Anholt
ea53ff80af fix LogicOp/bitmap problem, bug 11133 2007-07-04 07:37:28 -06:00
Brian
feeca1bcbc add fbdev/ to DEMO_FILES 2007-07-03 15:46:48 -06:00
Brian
12e7278c08 document GLSL fixes 2007-07-03 12:21:11 -06:00
Brian
45b5c44eb9 bring over Roland's DOT3_RGBA fix from master 2007-07-03 12:20:52 -06:00
Brian
e3456c1028 added vec2(vec4) constructor, bug 11404 2007-07-03 11:42:03 -06:00
Brian
327fb38573 add code for stpq, rgba writemasks in make_writemask(), bug 11404 2007-07-03 11:41:07 -06:00
Brian
1d25d9e15f Bump version to 7.0.1 2007-07-03 10:04:41 -06:00
Brian
cc7cee3f48 Revert the version of libOSMesa.so to 6.5.3 to avoid linking issues that were caused by bumping to 7.0 2007-07-03 10:03:01 -06:00
Brian
d7062710cd use _mesa_unpack_stencil_span() in draw_stencil_pixels(), bug 11457 2007-07-03 08:58:03 -06:00
Julien Cristau
3e21f52796 Make libgl1-mesa-swx11 depend on libosmesa6
libgl1-mesa-swx11 used to contain libOSMesa.so.6, so programs linked against
it don't necessarily depend on libosmesa6; their dependencies were broken
since mesa 6.5.2-1.
2007-07-03 05:34:07 +02:00
Brian
9bfba734d8 add const to wglSetPixelFormat() to match .h declaration 2007-07-02 10:15:39 -06:00
Julien Cristau
0511d9a9d2 Fix merge and delete symlinks which aren't in the tarball. 2007-07-01 02:55:59 +02:00
Brian
aaebf2f47e regenerated 2007-06-29 07:51:49 -06:00
Brian
23f8d77b38 Add a few missing GL 2.0 entrypoints, regenerate related files.
Specifically:
  glVertexAttrib4bv
  glVertexAttrib4iv
  glVertexAttrib4ubv
  glVertexAttrib4uiv
  glVertexAttrib4usv
2007-06-28 16:38:48 -06:00
Samuel Thibault
498b4dbba4 Fix build on hurd-i386.
Build libgl1-mesa-glx without direct rendering support on that arch, and
don't build any dri driver.
2007-06-28 22:33:05 +01:00
Julien Cristau
ecc42e79ae Merge unreleased changelog entries 2007-06-28 22:01:05 +01:00
Julien Cristau
0bf8ed523d * Add a shlibs file for libgl1-mesa-swx11-i686. 2007-06-28 21:58:42 +01:00
Brian
87d22ee0fe initial 7.0.1 relnotes 2007-06-28 08:11:36 -06:00
zhang
52e25f63d4 a variety of fixes for MingW 2007-06-28 08:10:09 -06:00
Brian
f3fb67972b remove obsolete t_save_api.c 2007-06-28 07:43:32 -06:00
Brian
6400756364 remove obsolete sources 2007-06-28 07:21:59 -06:00
David Nusinow
d71d04bf6f Bite the bullet and add myself to uploaders 2007-06-27 22:54:52 -04:00
David Nusinow
af90259e09 Merge remote branch 'origin/debian-experimental' into debian-experimental
Conflicts:

	debian/changelog
	src/mesa/drivers/dri/r300/radeon_lock.c
2007-06-27 22:53:04 -04:00
David Nusinow
7c9762f698 Changelog bump 2007-06-27 21:32:51 -04:00
David Nusinow
b2e67fdb7c Merge branch 'upstream-experimental' into debian-experimental
Conflicts:

	.gitignore
	progs/miniglx/miniglxtest.c
	src/glx/mini/Makefile
	src/glx/mini/miniglx.c
	src/mesa/drivers/dri/r300/.gitignore
2007-06-27 21:29:16 -04:00
Brian
d59f0314bd md5 sums for 7.0 2007-06-22 16:50:24 -06:00
Brian
131baefac3 fix date for 7.0 release 2007-06-22 16:41:36 -06:00
Brian
a450078b67 update version 2007-06-22 16:40:17 -06:00
Brian
f6b041bf6c bring in Eric's FreeBSD fixes for -pthread 2007-06-22 15:40:46 -06:00
Brian
4f340d181a set tentative 7.0 release date 2007-06-20 19:59:15 -06:00
Julien Cristau
74a11701f3 Shorten the one-line synopsis of various packages.
They're cut off in .changes files if they're too long.
Thanks, Marc 'HE' Brockschmidt!
2007-06-13 19:36:49 +02:00
Brian
b58e38e936 fix TEXTREL problem (bug 7459) 2007-06-12 09:52:05 -06:00
Brian
8713cb48a8 Fix TEXREL issues when using GLX_USE_TLS (see bug 7459). 2007-06-12 09:46:12 -06:00
Brian
277c5e57ed fix GLX_USE_TLS breakage 2007-06-12 08:57:39 -06:00
Brian
b3d62d5af5 disable depthBits check in check_compatible(), see bug 11161 2007-06-08 12:36:00 -06:00
Brian
9c0f0c8d81 remove old comment about DEFAULT_SOFTWARE_DEPTH_BITS 2007-06-08 12:35:37 -06:00
Brian
18a0a2a7ac document bug 11049 fix 2007-05-23 16:50:22 -06:00
Brian
8b99d9e33c restore GL_TEXTURE_LOD_BIAS in _mesa_PopAttrib(), bug 11049 2007-05-23 16:49:19 -06:00
Brian
01e7e153e3 document point attentuation fix 2007-05-23 08:59:02 -06:00
Brian
0ad4ca24d2 Fix point attentuation problem (bug 11042)
ctx->Point._Attentuation was computed in wrong place and the VB->Eye coord Z
array wasn't indexed correctly in run_point_stage().
2007-05-23 08:56:45 -06:00
Brian
724a155552 if light position is local, treat it as a homogeneous coord and divide by W (see bug 11009) 2007-05-21 10:55:15 -06:00
Julien Cristau
e1b0715dd6 Prepare changelog for upload. 2007-05-21 11:35:28 +02:00
Julien Cristau
a5c6598966 Merge remote branch 'origin/debian-unstable' into debian-experimental
Conflicts:

	debian/changelog
2007-05-21 11:33:42 +02:00
Roland Scheidegger
d38b74a316 fix miptree layout (i915) for small compressed mipmaps
This seems to work now. Also fix i945 set_level_info, need to match i915
behaviour for storing mip height, as it's assumed to be the mip width
(in texels) elsewhere and the division by 4 is done later (untested).
2007-05-19 04:43:50 +02:00
Roland Scheidegger
54cab4b47b fix copy & paste bug of previous commit, breaking dxt5 formats 2007-05-19 03:56:06 +02:00
Roland Scheidegger
c093666bc5 fix miptree comparison with compressed textures
TexelBytes is always 0 with compressed textures, thus would never match
mt->cpp. This caused constant blitting around of textures, and it fixes at
least the horrible performance of Q3 if compressed textures are enabled.
2007-05-19 01:39:54 +02:00
Roland Scheidegger
04972f6761 fix small s3tc mipmaps (#10968)
make sure that always whole blocks are uploaded.
(May still not work correctly if the top mip map is not at least a full block,
that is 4 pixels wide - not sure, but probably doesn't happen in real world)
2007-05-19 01:39:16 +02:00
Brian
67f8234622 set version to 7.0 2007-05-18 09:41:34 -06:00
Brian
201d6dbd9c fix STATE_HALF_VECTOR value (bug 10987) 2007-05-18 09:35:18 -06:00
Julien Cristau
1865ef6fd4 Prepare changelog for upload. 2007-05-18 13:36:42 +02:00
Ian Romanick
fbcac5aa83 Refactor determining whether a texture target can use compressed format 2007-05-17 14:45:25 -07:00
Brian
2b72ab8f8f document glXDestroyPbuffer fix 2007-05-17 15:30:39 -06:00
Brian
0ea97b9408 In DestroyPbuffer(), use GetReq() intead of GetReqExtra(). See bug 10983. 2007-05-17 15:26:39 -06:00
Brian
28683ac7c0 version 7.0, not 7.1 2007-05-17 09:07:25 -06:00
Brian
c72e3e210f version should be 7.0 2007-05-17 09:04:19 -06:00
Julien Cristau
a8c856d43c Cherry-pick commmit 65faf023 by Michel Dänzer to fix r300 crash.
r300: Don't crash in radeonUpdatePageFlipping when !radeon->glCtx->DrawBuffer.

This feels like a kludge, maybe there's a better solution.
2007-05-09 15:09:12 +02:00
David Nusinow
1953cfc954 Update changelog 2007-04-30 00:08:11 -04:00
David Nusinow
8e7bac950e Merge branch 'upstream-experimental' into debian-experimental 2007-04-29 23:23:00 -04:00
Julien Cristau
03cdacfb6a Prepare changelog for upload. 2007-04-25 10:37:10 +02:00
Julien Cristau
30f3c3dfad Update changelog. 2007-04-25 08:42:02 +02:00
Julien Cristau
dee1b0d5bb Merge remote branch 'upstream/master' into debian-experimental 2007-04-25 08:40:39 +02:00
Julien Cristau
6f2b0fe463 New upstream release candidate.
Refresh a couple patches.
2007-04-25 01:48:33 +02:00
Julien Cristau
0c69176498 Merge tag 'mesa_6_5_3_rc3' into debian-experimental 2007-04-25 01:41:05 +02:00
Julien Cristau
4e7a0385b5 Merge tag 'mesa_6_5_3_rc2' into debian-experimental 2007-04-25 01:40:50 +02:00
Julien Cristau
4adbd94b10 Only install stuff in mesa-swx11-source if we're building arch:all packages. 2007-04-23 11:00:09 +02:00
Julien Cristau
4a2905b645 Delete obsolete file. 2007-04-22 18:08:58 +02:00
Julien Cristau
0638c6b530 Add generated files from the tarball so they're included in mesa-swx11-source. 2007-04-22 18:08:35 +02:00
Julien Cristau
7dd35ce013 Delete .gitignore files which do more harm than good in a debian package tree. 2007-04-22 18:07:52 +02:00
Julien Cristau
c8629c3178 Resync debian/scripts/install-source.sh. 2007-04-22 00:51:10 +02:00
Julien Cristau
77ccf51cb3 New upstream release candidate. 2007-04-22 00:50:40 +02:00
Julien Cristau
01197f0c77 Merge branch 'upstream-experimental' into debian-experimental 2007-04-22 00:34:30 +02:00
Julien Cristau
69c86ff4fd Remove leftover file. 2007-04-22 00:33:38 +02:00
Julien Cristau
9a42ffad81 Add build-dependencies on libxdamage-dev and libxfixes-dev. 2007-04-21 21:23:21 +02:00
Julien Cristau
0d8633bf8c Delete autogenerated files, and files not included in the tarball. 2007-04-21 21:20:37 +02:00
Julien Cristau
d79e80a3d0 New upstream release candidate.
Drop patches included upstream.
2007-04-21 20:47:01 +02:00
Julien Cristau
d890ebdd25 Merge branch 'upstream-experimental' into debian-experimental
Conflicts:

	doxygen/Makefile
	doxygen/common.doxy
	doxygen/doxy.bat
	doxygen/glapi.doxy
	doxygen/header.html
	doxygen/main.doxy
	doxygen/math.doxy
	doxygen/shader.doxy
	doxygen/swrast.doxy
	doxygen/swrast_setup.doxy
	doxygen/tnl.doxy
	doxygen/tnl_dd.doxy
	progs/tests/Makefile
	progs/tests/fbotest1.c
	progs/tests/fbotest2.c
	progs/tests/fbotexture.c
	progs/tests/mipmap_limits.c
	progs/tests/sharedtex.c
	progs/tools/trace/gltrace_support.cc
	progs/trivial/Makefile
	src/egl/docs/EGL_MESA_screen_surface
	src/egl/main/egllog.c
	src/glx/mini/Makefile
	src/glx/mini/driver.h
	src/mesa/drivers/allegro/amesa.c
	src/mesa/drivers/windows/gldirect/dx7/gld_driver_dx7.c
	src/mesa/drivers/windows/gldirect/dx7/gld_ext_dx7.c
	src/mesa/drivers/windows/gldirect/dx7/gld_primitive_dx7.c
	src/mesa/drivers/windows/gldirect/dx8/gld_driver_dx8.c
	src/mesa/drivers/windows/gldirect/dx8/gld_ext_dx8.c
	src/mesa/drivers/windows/gldirect/dx8/gld_primitive_dx8.c
	src/mesa/drivers/windows/gldirect/dx9/gld_driver_dx9.c
	src/mesa/drivers/windows/gldirect/dx9/gld_ext_dx9.c
	src/mesa/drivers/windows/gldirect/dx9/gld_primitive_dx9.c
	src/mesa/drivers/windows/gldirect/mesasw/gld_wgl_mesasw.c
2007-04-21 20:27:55 +02:00
Brice Goglin
5ee9b3ce72 Fix conflicts for libGLw packages.
libglw1-mesa does not have to conflict with earlier libgl1-mesa-swx11
(libGLw was only shipped as development libs/headers within swx11-dev).

libglw1-mesa-dev needs to conflicts with libglw-dev.
2007-04-21 20:20:03 +02:00
Brice Goglin
031b539a09 Split out libGLw libs and headers from libgl1-mesa-swx11.
Ship both static and shared libraries for libGLw, creating libglw1-mesa and
libglw1-mesa-dev (closes: #374904).
2007-04-21 20:19:38 +02:00
Brice Goglin
47c3c706e8 Add 09_i915_always_enable_pixel_fog.patch 2007-04-21 18:00:12 +02:00
Brice Goglin
0d830a616c Add 08_r300_update_window_state_when_bound_but_stamp_changed.patch 2007-04-20 18:53:47 +02:00
Brice Goglin
e219bf8338 Add 07_call_radeonSetCliprects_from_radeonMakeCurrent.patch 2007-04-20 18:44:59 +02:00
Julien Cristau
6c0fba2480 Prepare changelog for upload. 2007-04-20 05:58:00 +02:00
Julien Cristau
f4cc159f53 Revert " * Split out libGLw libs and headers from libgl1-mesa-swx11 and ship both"
This reverts commit 856ae782c7.

Conflicts:

	debian/changelog
2007-04-20 05:55:04 +02:00
Julien Cristau
04c002abc2 Revert "Fix conflicts for libGLw packages."
This reverts commit 048144074b.
2007-04-20 05:53:59 +02:00
Julien Cristau
78fd17907e Merge branch 'debian-experimental' into debian-unstable 2007-04-20 05:52:20 +02:00
Julien Cristau
47edc83080 Merge branch 'debian-experimental' into debian-unstable
Conflicts:

	configs/debian-osmesa16
	configs/debian-osmesa16-static
	configs/debian-osmesa32
	configs/debian-osmesa32-static
	debian/changelog
	debian/compat
	debian/control
	debian/copyright
	debian/libgl1-mesa-dev.install
	debian/libgl1-mesa-swx11-dev.install
	debian/libgl1-mesa-swx11.shlibs
	debian/libglu1-mesa-dev.install
	debian/mesa-common-dev.docs
	debian/mesa-common-dev.install
	debian/rules
	src/mesa/x86-64/matypes.h
	src/mesa/x86/matypes.h
2007-04-20 05:44:24 +02:00
Julien Cristau
33e34f5bf4 Merge tag 'mesa_6_5_1' into debian-unstable
Conflicts:

	progs/util/README
	progs/util/glstate.c
	progs/util/glstate.h
	progs/util/sampleMakefile
	src/glu/sgi/libnurbs/interface/bezierEval.h
	src/glu/sgi/libnurbs/interface/bezierPatch.cc
	src/glu/sgi/libnurbs/interface/bezierPatch.h
	src/glu/sgi/libnurbs/interface/bezierPatchMesh.cc
	src/glu/sgi/libnurbs/interface/bezierPatchMesh.h
	src/glu/sgi/libnurbs/interface/glcurveval.cc
	src/glu/sgi/libnurbs/interface/glimports.h
	src/glu/sgi/libnurbs/interface/glinterface.cc
	src/glu/sgi/libnurbs/interface/glrenderer.h
	src/glu/sgi/libnurbs/interface/incurveeval.cc
	src/glu/sgi/libnurbs/interface/insurfeval.cc
	src/glu/sgi/libnurbs/interface/mystdio.h
	src/glu/sgi/libnurbs/interface/mystdlib.h
	src/glu/sgi/libnurbs/internals/arc.h
	src/glu/sgi/libnurbs/internals/arcsorter.cc
	src/glu/sgi/libnurbs/internals/arcsorter.h
	src/glu/sgi/libnurbs/internals/arctess.h
	src/glu/sgi/libnurbs/internals/backend.cc
	src/glu/sgi/libnurbs/internals/backend.h
	src/glu/sgi/libnurbs/internals/basiccrveval.h
	src/glu/sgi/libnurbs/internals/basicsurfeval.h
	src/glu/sgi/libnurbs/internals/bezierarc.h
	src/glu/sgi/libnurbs/internals/bin.cc
	src/glu/sgi/libnurbs/internals/bin.h
	src/glu/sgi/libnurbs/internals/bufpool.cc
	src/glu/sgi/libnurbs/internals/bufpool.h
	src/glu/sgi/libnurbs/internals/cachingeval.cc
	src/glu/sgi/libnurbs/internals/cachingeval.h
	src/glu/sgi/libnurbs/internals/ccw.cc
	src/glu/sgi/libnurbs/internals/coveandtiler.h
	src/glu/sgi/libnurbs/internals/curve.cc
	src/glu/sgi/libnurbs/internals/curve.h
	src/glu/sgi/libnurbs/internals/curvelist.cc
	src/glu/sgi/libnurbs/internals/curvelist.h
	src/glu/sgi/libnurbs/internals/curvesub.cc
	src/glu/sgi/libnurbs/internals/dataTransform.cc
	src/glu/sgi/libnurbs/internals/dataTransform.h
	src/glu/sgi/libnurbs/internals/defines.h
	src/glu/sgi/libnurbs/internals/displaylist.cc
	src/glu/sgi/libnurbs/internals/displaylist.h
	src/glu/sgi/libnurbs/internals/displaymode.h
	src/glu/sgi/libnurbs/internals/flist.cc
	src/glu/sgi/libnurbs/internals/flist.h
	src/glu/sgi/libnurbs/internals/flistsorter.cc
	src/glu/sgi/libnurbs/internals/flistsorter.h
	src/glu/sgi/libnurbs/internals/gridline.h
	src/glu/sgi/libnurbs/internals/gridtrimvertex.h
	src/glu/sgi/libnurbs/internals/gridvertex.h
	src/glu/sgi/libnurbs/internals/hull.cc
	src/glu/sgi/libnurbs/internals/hull.h
	src/glu/sgi/libnurbs/internals/intersect.cc
	src/glu/sgi/libnurbs/internals/jarcloc.h
	src/glu/sgi/libnurbs/internals/knotvector.h
	src/glu/sgi/libnurbs/internals/mapdesc.cc
	src/glu/sgi/libnurbs/internals/mapdesc.h
	src/glu/sgi/libnurbs/internals/mapdescv.cc
	src/glu/sgi/libnurbs/internals/maplist.cc
	src/glu/sgi/libnurbs/internals/maplist.h
	src/glu/sgi/libnurbs/internals/mesher.cc
	src/glu/sgi/libnurbs/internals/mesher.h
	src/glu/sgi/libnurbs/internals/monoTriangulationBackend.cc
	src/glu/sgi/libnurbs/internals/monotonizer.cc
	src/glu/sgi/libnurbs/internals/monotonizer.h
	src/glu/sgi/libnurbs/internals/myassert.h
	src/glu/sgi/libnurbs/internals/mycode.cc
	src/glu/sgi/libnurbs/internals/mystring.h
	src/glu/sgi/libnurbs/internals/nurbsconsts.h
	src/glu/sgi/libnurbs/internals/nurbstess.cc
	src/glu/sgi/libnurbs/internals/patch.cc
	src/glu/sgi/libnurbs/internals/patch.h
	src/glu/sgi/libnurbs/internals/patchlist.cc
	src/glu/sgi/libnurbs/internals/patchlist.h
	src/glu/sgi/libnurbs/internals/pwlarc.h
	src/glu/sgi/libnurbs/internals/quilt.cc
	src/glu/sgi/libnurbs/internals/quilt.h
	src/glu/sgi/libnurbs/internals/reader.cc
	src/glu/sgi/libnurbs/internals/reader.h
	src/glu/sgi/libnurbs/internals/renderhints.cc
	src/glu/sgi/libnurbs/internals/renderhints.h
	src/glu/sgi/libnurbs/internals/simplemath.h
	src/glu/sgi/libnurbs/internals/slicer.cc
	src/glu/sgi/libnurbs/internals/slicer.h
	src/glu/sgi/libnurbs/internals/sorter.cc
	src/glu/sgi/libnurbs/internals/sorter.h
	src/glu/sgi/libnurbs/internals/splitarcs.cc
	src/glu/sgi/libnurbs/internals/subdivider.h
	src/glu/sgi/libnurbs/internals/tobezier.cc
	src/glu/sgi/libnurbs/internals/trimline.cc
	src/glu/sgi/libnurbs/internals/trimline.h
	src/glu/sgi/libnurbs/internals/trimregion.cc
	src/glu/sgi/libnurbs/internals/trimregion.h
	src/glu/sgi/libnurbs/internals/trimvertex.h
	src/glu/sgi/libnurbs/internals/trimvertpool.cc
	src/glu/sgi/libnurbs/internals/trimvertpool.h
	src/glu/sgi/libnurbs/internals/types.h
	src/glu/sgi/libnurbs/internals/uarray.cc
	src/glu/sgi/libnurbs/internals/uarray.h
	src/glu/sgi/libnurbs/internals/varray.cc
	src/glu/sgi/libnurbs/internals/varray.h
	src/glu/sgi/libnurbs/nurbtess/definitions.h
	src/glu/sgi/libnurbs/nurbtess/directedLine.h
	src/glu/sgi/libnurbs/nurbtess/glimports.h
	src/glu/sgi/libnurbs/nurbtess/gridWrap.cc
	src/glu/sgi/libnurbs/nurbtess/gridWrap.h
	src/glu/sgi/libnurbs/nurbtess/monoChain.cc
	src/glu/sgi/libnurbs/nurbtess/monoChain.h
	src/glu/sgi/libnurbs/nurbtess/monoTriangulation.cc
	src/glu/sgi/libnurbs/nurbtess/monoTriangulation.h
	src/glu/sgi/libnurbs/nurbtess/mystdio.h
	src/glu/sgi/libnurbs/nurbtess/mystdlib.h
	src/glu/sgi/libnurbs/nurbtess/partitionX.cc
	src/glu/sgi/libnurbs/nurbtess/partitionX.h
	src/glu/sgi/libnurbs/nurbtess/partitionY.cc
	src/glu/sgi/libnurbs/nurbtess/partitionY.h
	src/glu/sgi/libnurbs/nurbtess/polyDBG.h
	src/glu/sgi/libnurbs/nurbtess/polyUtil.cc
	src/glu/sgi/libnurbs/nurbtess/polyUtil.h
	src/glu/sgi/libnurbs/nurbtess/primitiveStream.cc
	src/glu/sgi/libnurbs/nurbtess/primitiveStream.h
	src/glu/sgi/libnurbs/nurbtess/quicksort.cc
	src/glu/sgi/libnurbs/nurbtess/quicksort.h
	src/glu/sgi/libnurbs/nurbtess/rectBlock.cc
	src/glu/sgi/libnurbs/nurbtess/rectBlock.h
	src/glu/sgi/libnurbs/nurbtess/sampleComp.cc
	src/glu/sgi/libnurbs/nurbtess/sampleComp.h
	src/glu/sgi/libnurbs/nurbtess/sampleCompBot.cc
	src/glu/sgi/libnurbs/nurbtess/sampleCompBot.h
	src/glu/sgi/libnurbs/nurbtess/sampleCompRight.cc
	src/glu/sgi/libnurbs/nurbtess/sampleCompRight.h
	src/glu/sgi/libnurbs/nurbtess/sampleCompTop.cc
	src/glu/sgi/libnurbs/nurbtess/sampleCompTop.h
	src/glu/sgi/libnurbs/nurbtess/sampleMonoPoly.cc
	src/glu/sgi/libnurbs/nurbtess/sampleMonoPoly.h
	src/glu/sgi/libnurbs/nurbtess/sampledLine.cc
	src/glu/sgi/libnurbs/nurbtess/sampledLine.h
	src/glu/sgi/libnurbs/nurbtess/searchTree.cc
	src/glu/sgi/libnurbs/nurbtess/searchTree.h
	src/glu/sgi/libnurbs/nurbtess/zlassert.h
	src/glu/sgi/libtess/README
	src/glu/sgi/libtess/alg-outline
	src/glu/sgi/libtess/dict-list.h
	src/glu/sgi/libtess/dict.c
	src/glu/sgi/libtess/dict.h
	src/glu/sgi/libtess/geom.c
	src/glu/sgi/libtess/memalloc.c
	src/glu/sgi/libtess/memalloc.h
	src/glu/sgi/libtess/mesh.c
	src/glu/sgi/libtess/mesh.h
	src/glu/sgi/libtess/normal.h
	src/glu/sgi/libtess/priorityq-heap.c
	src/glu/sgi/libtess/priorityq-heap.h
	src/glu/sgi/libtess/priorityq-sort.h
	src/glu/sgi/libtess/priorityq.c
	src/glu/sgi/libtess/priorityq.h
	src/glu/sgi/libtess/render.c
	src/glu/sgi/libtess/render.h
	src/glu/sgi/libtess/sweep.h
	src/glu/sgi/libtess/tess.h
	src/glu/sgi/libtess/tessmono.c
	src/glu/sgi/libtess/tessmono.h
	src/glu/sgi/libutil/error.c
	src/glu/sgi/libutil/glue.c
	src/glu/sgi/libutil/gluint.h
	src/glu/sgi/libutil/project.c
	src/glu/sgi/libutil/registry.c
	src/glx/x11/glxext.c
	src/mesa/main/imports.h
	src/mesa/tnl/t_vb_cliptmp.h
2007-04-20 05:37:11 +02:00
Brice Goglin
3832ecadae Add 06_fix_texture_data_corruption.patch (closes: #412346) 2007-04-08 18:52:31 +02:00
Brice Goglin
285ed128ab Add a .shlibs file for libglw1-mesa. 2007-04-01 00:44:29 +02:00
Brice Goglin
048144074b Fix conflicts for libGLw packages.
libglw1-mesa does not have to conflict with earlier libgl1-mesa-swx11
(libGLw was only shipped as development libs/headers within swx11-dev).

libglw1-mesa-dev needs to conflicts with libglw-dev.
2007-04-01 00:43:20 +02:00
Brice Goglin
856ae782c7 * Split out libGLw libs and headers from libgl1-mesa-swx11 and ship both
static and shared libraries, creating libglw1-mesa and libglw1-mesa-dev
    (closes: #374904).
2007-03-31 03:28:32 +02:00
Julien Cristau
12e3f97541 Fix explanation about libglide3 in libgl1-mesa-dri's description.
Thanks to Michel Dänzer for his help.
2007-03-30 17:20:02 +02:00
Julien Cristau
30f6fe0710 Have libgl1-mesa-dri suggest libglide3.
Add explanation for this in the package description.  Closes: #387339.
2007-03-30 14:10:24 +02:00
Cyril Brulebois
0aa8f1df3c * Non-maintainer upload.
* Backport a fix from mesa 6.5.2 to mesa 6.5.1 which:
   - fixes braces around the ASSERT macro;
   - fixes a clipping problem, which leads to systematic segfaults of the
     X server (Closes: #405803).
2007-02-23 00:48:40 +01:00
Julien Cristau
d39147a8a1 Fix FTBFS on amd64.
Set LIB_DIR and EXTRA_LIB_PATH in configs/debian-default to override
settings in configs/linux-x86-64. This fixes a FTBFS on amd64, thanks to
Marc 'HE' Brockschmidt for the report (closes: #410118).
2007-02-07 23:13:59 +01:00
Julien Cristau
7991f7e936 * Add myself to Uploaders. 2007-02-04 21:17:18 +01:00
Julien Cristau
10011efcd7 Add bug closer for #409638. 2007-02-04 17:21:57 +01:00
Julien Cristau
0058815c8f Re-add src/glut/mini/ which was deleted by mistake. 2007-02-04 15:21:46 +01:00
Julien Cristau
02048984b7 Delete directories not included in the mesa tarballs. 2007-02-04 15:15:43 +01:00
Julien Cristau
544601e9f4 Fix FTBFS on non-i386 archs.
Build architecture-independant debs in binary-indep, and use the debhelper -s
option for commands in binary-arch, because architectures other than i386
don't have to build libgl1-mesa-swx11-i686.
2007-02-04 14:05:23 +01:00
Julien Cristau
2634f06c20 Merge branch 'upstream-experimental' into debian-experimental
Conflicts:

	progs/util/README
	progs/util/glstate.c
	progs/util/glstate.h
	progs/util/sampleMakefile
	src/glu/sgi/libnurbs/interface/bezierEval.h
	src/glu/sgi/libnurbs/interface/bezierPatch.cc
	src/glu/sgi/libnurbs/interface/bezierPatch.h
	src/glu/sgi/libnurbs/interface/bezierPatchMesh.cc
	src/glu/sgi/libnurbs/interface/bezierPatchMesh.h
	src/glu/sgi/libnurbs/interface/glcurveval.cc
	src/glu/sgi/libnurbs/interface/glimports.h
	src/glu/sgi/libnurbs/interface/glinterface.cc
	src/glu/sgi/libnurbs/interface/glrenderer.h
	src/glu/sgi/libnurbs/interface/incurveeval.cc
	src/glu/sgi/libnurbs/interface/insurfeval.cc
	src/glu/sgi/libnurbs/interface/mystdio.h
	src/glu/sgi/libnurbs/interface/mystdlib.h
	src/glu/sgi/libnurbs/internals/arc.h
	src/glu/sgi/libnurbs/internals/arcsorter.cc
	src/glu/sgi/libnurbs/internals/arcsorter.h
	src/glu/sgi/libnurbs/internals/arctess.h
	src/glu/sgi/libnurbs/internals/backend.cc
	src/glu/sgi/libnurbs/internals/backend.h
	src/glu/sgi/libnurbs/internals/basiccrveval.h
	src/glu/sgi/libnurbs/internals/basicsurfeval.h
	src/glu/sgi/libnurbs/internals/bezierarc.h
	src/glu/sgi/libnurbs/internals/bin.cc
	src/glu/sgi/libnurbs/internals/bin.h
	src/glu/sgi/libnurbs/internals/bufpool.cc
	src/glu/sgi/libnurbs/internals/bufpool.h
	src/glu/sgi/libnurbs/internals/cachingeval.cc
	src/glu/sgi/libnurbs/internals/cachingeval.h
	src/glu/sgi/libnurbs/internals/ccw.cc
	src/glu/sgi/libnurbs/internals/coveandtiler.h
	src/glu/sgi/libnurbs/internals/curve.cc
	src/glu/sgi/libnurbs/internals/curve.h
	src/glu/sgi/libnurbs/internals/curvelist.cc
	src/glu/sgi/libnurbs/internals/curvelist.h
	src/glu/sgi/libnurbs/internals/curvesub.cc
	src/glu/sgi/libnurbs/internals/dataTransform.cc
	src/glu/sgi/libnurbs/internals/dataTransform.h
	src/glu/sgi/libnurbs/internals/defines.h
	src/glu/sgi/libnurbs/internals/displaylist.cc
	src/glu/sgi/libnurbs/internals/displaylist.h
	src/glu/sgi/libnurbs/internals/displaymode.h
	src/glu/sgi/libnurbs/internals/flist.cc
	src/glu/sgi/libnurbs/internals/flist.h
	src/glu/sgi/libnurbs/internals/flistsorter.cc
	src/glu/sgi/libnurbs/internals/flistsorter.h
	src/glu/sgi/libnurbs/internals/gridline.h
	src/glu/sgi/libnurbs/internals/gridtrimvertex.h
	src/glu/sgi/libnurbs/internals/gridvertex.h
	src/glu/sgi/libnurbs/internals/hull.cc
	src/glu/sgi/libnurbs/internals/hull.h
	src/glu/sgi/libnurbs/internals/intersect.cc
	src/glu/sgi/libnurbs/internals/jarcloc.h
	src/glu/sgi/libnurbs/internals/knotvector.h
	src/glu/sgi/libnurbs/internals/mapdesc.cc
	src/glu/sgi/libnurbs/internals/mapdesc.h
	src/glu/sgi/libnurbs/internals/mapdescv.cc
	src/glu/sgi/libnurbs/internals/maplist.cc
	src/glu/sgi/libnurbs/internals/maplist.h
	src/glu/sgi/libnurbs/internals/mesher.cc
	src/glu/sgi/libnurbs/internals/mesher.h
	src/glu/sgi/libnurbs/internals/monoTriangulationBackend.cc
	src/glu/sgi/libnurbs/internals/monotonizer.cc
	src/glu/sgi/libnurbs/internals/monotonizer.h
	src/glu/sgi/libnurbs/internals/myassert.h
	src/glu/sgi/libnurbs/internals/mycode.cc
	src/glu/sgi/libnurbs/internals/mystring.h
	src/glu/sgi/libnurbs/internals/nurbsconsts.h
	src/glu/sgi/libnurbs/internals/nurbstess.cc
	src/glu/sgi/libnurbs/internals/patch.cc
	src/glu/sgi/libnurbs/internals/patch.h
	src/glu/sgi/libnurbs/internals/patchlist.cc
	src/glu/sgi/libnurbs/internals/patchlist.h
	src/glu/sgi/libnurbs/internals/pwlarc.h
	src/glu/sgi/libnurbs/internals/quilt.cc
	src/glu/sgi/libnurbs/internals/quilt.h
	src/glu/sgi/libnurbs/internals/reader.cc
	src/glu/sgi/libnurbs/internals/reader.h
	src/glu/sgi/libnurbs/internals/renderhints.cc
	src/glu/sgi/libnurbs/internals/renderhints.h
	src/glu/sgi/libnurbs/internals/simplemath.h
	src/glu/sgi/libnurbs/internals/slicer.cc
	src/glu/sgi/libnurbs/internals/slicer.h
	src/glu/sgi/libnurbs/internals/sorter.cc
	src/glu/sgi/libnurbs/internals/sorter.h
	src/glu/sgi/libnurbs/internals/splitarcs.cc
	src/glu/sgi/libnurbs/internals/subdivider.h
	src/glu/sgi/libnurbs/internals/tobezier.cc
	src/glu/sgi/libnurbs/internals/trimline.cc
	src/glu/sgi/libnurbs/internals/trimline.h
	src/glu/sgi/libnurbs/internals/trimregion.cc
	src/glu/sgi/libnurbs/internals/trimregion.h
	src/glu/sgi/libnurbs/internals/trimvertex.h
	src/glu/sgi/libnurbs/internals/trimvertpool.cc
	src/glu/sgi/libnurbs/internals/trimvertpool.h
	src/glu/sgi/libnurbs/internals/types.h
	src/glu/sgi/libnurbs/internals/uarray.cc
	src/glu/sgi/libnurbs/internals/uarray.h
	src/glu/sgi/libnurbs/internals/varray.cc
	src/glu/sgi/libnurbs/internals/varray.h
	src/glu/sgi/libnurbs/nurbtess/definitions.h
	src/glu/sgi/libnurbs/nurbtess/directedLine.h
	src/glu/sgi/libnurbs/nurbtess/glimports.h
	src/glu/sgi/libnurbs/nurbtess/gridWrap.cc
	src/glu/sgi/libnurbs/nurbtess/gridWrap.h
	src/glu/sgi/libnurbs/nurbtess/monoChain.cc
	src/glu/sgi/libnurbs/nurbtess/monoChain.h
	src/glu/sgi/libnurbs/nurbtess/monoTriangulation.cc
	src/glu/sgi/libnurbs/nurbtess/monoTriangulation.h
	src/glu/sgi/libnurbs/nurbtess/mystdio.h
	src/glu/sgi/libnurbs/nurbtess/mystdlib.h
	src/glu/sgi/libnurbs/nurbtess/partitionX.cc
	src/glu/sgi/libnurbs/nurbtess/partitionX.h
	src/glu/sgi/libnurbs/nurbtess/partitionY.cc
	src/glu/sgi/libnurbs/nurbtess/partitionY.h
	src/glu/sgi/libnurbs/nurbtess/polyDBG.h
	src/glu/sgi/libnurbs/nurbtess/polyUtil.cc
	src/glu/sgi/libnurbs/nurbtess/polyUtil.h
	src/glu/sgi/libnurbs/nurbtess/primitiveStream.cc
	src/glu/sgi/libnurbs/nurbtess/primitiveStream.h
	src/glu/sgi/libnurbs/nurbtess/quicksort.cc
	src/glu/sgi/libnurbs/nurbtess/quicksort.h
	src/glu/sgi/libnurbs/nurbtess/rectBlock.cc
	src/glu/sgi/libnurbs/nurbtess/rectBlock.h
	src/glu/sgi/libnurbs/nurbtess/sampleComp.cc
	src/glu/sgi/libnurbs/nurbtess/sampleComp.h
	src/glu/sgi/libnurbs/nurbtess/sampleCompBot.cc
	src/glu/sgi/libnurbs/nurbtess/sampleCompBot.h
	src/glu/sgi/libnurbs/nurbtess/sampleCompRight.cc
	src/glu/sgi/libnurbs/nurbtess/sampleCompRight.h
	src/glu/sgi/libnurbs/nurbtess/sampleCompTop.cc
	src/glu/sgi/libnurbs/nurbtess/sampleCompTop.h
	src/glu/sgi/libnurbs/nurbtess/sampleMonoPoly.cc
	src/glu/sgi/libnurbs/nurbtess/sampleMonoPoly.h
	src/glu/sgi/libnurbs/nurbtess/sampledLine.cc
	src/glu/sgi/libnurbs/nurbtess/sampledLine.h
	src/glu/sgi/libnurbs/nurbtess/searchTree.cc
	src/glu/sgi/libnurbs/nurbtess/searchTree.h
	src/glu/sgi/libnurbs/nurbtess/zlassert.h
	src/glu/sgi/libtess/README
	src/glu/sgi/libtess/alg-outline
	src/glu/sgi/libtess/dict-list.h
	src/glu/sgi/libtess/dict.c
	src/glu/sgi/libtess/dict.h
	src/glu/sgi/libtess/geom.c
	src/glu/sgi/libtess/memalloc.c
	src/glu/sgi/libtess/memalloc.h
	src/glu/sgi/libtess/mesh.c
	src/glu/sgi/libtess/mesh.h
	src/glu/sgi/libtess/normal.h
	src/glu/sgi/libtess/priorityq-heap.c
	src/glu/sgi/libtess/priorityq-heap.h
	src/glu/sgi/libtess/priorityq-sort.h
	src/glu/sgi/libtess/priorityq.c
	src/glu/sgi/libtess/priorityq.h
	src/glu/sgi/libtess/render.c
	src/glu/sgi/libtess/render.h
	src/glu/sgi/libtess/sweep.h
	src/glu/sgi/libtess/tess.h
	src/glu/sgi/libtess/tessmono.c
	src/glu/sgi/libtess/tessmono.h
	src/glu/sgi/libutil/error.c
	src/glu/sgi/libutil/glue.c
	src/glu/sgi/libutil/gluint.h
	src/glu/sgi/libutil/project.c
	src/glu/sgi/libutil/registry.c
2007-02-01 11:50:36 +01:00
Julien Cristau
7549426a16 Sync Section and Priority for all packages with the override. 2007-01-28 18:18:38 +01:00
Thierry Reding
6dcb9a3434 Give credit where credit is due. 2007-01-11 15:09:23 +01:00
Thierry Reding
e5a31c00ec libosmesa6{,-dev} replace libgl1-mesa-swx11{,-dev} and mesa-common-dev.
libosmesa6 and libosmesa6-dev are now the only packages to contain OSMesa
libraries and header files so they can be installed independent of which libGL
is installed. They statically link in the software rasterization code.
2007-01-11 15:06:42 +01:00
Thierry Reding
00295242fd mesa-common-dev needs to replace libgl1-mesa-swx11-dev and libgl1-mesa-dev.
The GLX headers were previously shipped in both packages and have been moved
to mesa-common-dev.
2007-01-11 14:56:48 +01:00
Thierry Reding
21b6ed7c29 Build the DRI modules with the default optimization flags. 2007-01-11 14:40:10 +01:00
Thierry Reding
01894cbf10 Revert the previous changes to the libgl1-mesa-{glx,swx11} shlibs. 2007-01-10 15:23:29 +01:00
Thierry Reding
28cad2b9d9 Bump libgl1-mesa-glx, libgl1-mesa-swx11 and libosmesa6 shlibs files. 2007-01-10 13:55:19 +01:00
Thierry Reding
71d7b7dbe7 Drop noop line. 2007-01-08 13:36:25 +01:00
Thierry Reding
4563723db1 Make sure all files in the mesa-swx11-source package have permissions 0644. 2007-01-08 13:36:16 +01:00
Thierry Reding
0bce6a1899 Compile with -O2 optimization instead of -O3 which might lead to problems. 2007-01-08 11:43:25 +01:00
Thierry Reding
e71d6dbd9c Make debugging symbol packages depend on the corresponding binary package and
put them into the libdevel section.
2007-01-08 11:41:02 +01:00
Thierry Reding
132ecd19f3 Do not build any linux*-static configuration with -fPIC. 2007-01-08 10:07:08 +01:00
Thierry Reding
a404143f4c Add patches 04_cleanup-osmesa-configs and 05_static-nonpic.
04_cleanup-osmesa-configs makes the OSMesa configurations behave as expected.
That is respect the OPT_FLAGS, PIC_FLAGS, ARCH_FLAGS and DEFINES variables.
The patch also makes the files more consistent by inheriting from more general
configurations.

05_static-nonpic fixes the linux-static configuration to not use -fPIC for
compiling code because it would violate Debian policy.
2007-01-08 08:35:42 +01:00
Julien Cristau
44467ca45e * Drop obsolete depends on xlibs. 2007-01-05 15:44:51 +01:00
Julien Cristau
f70ea0747d Import changelog entry from the 6.5.1-0.5 NMU. 2007-01-04 21:30:49 +01:00
Thierry Reding
c619abf18a Hijack the package (set XSF as maintainer, add myself to uploaders). \o/ 2007-01-04 16:35:14 +01:00
Julien Cristau
6da1323437 * Non-maintainer upload.
* Build with -fno-strict-aliasing to fix misbuild of i965_dri.so
  (closes: #394311).  Thanks to Michel Dänzer for the fix, and to Ryan
  Richter for the report and testing.
2007-01-03 16:59:06 +01:00
Julien Cristau
1dd29785cf Import changes from the mesa 6.5.1-0.4 NMU. 2007-01-03 13:43:46 +01:00
Julien Cristau
5681987ee4 Import changes from the mesa 6.5.1-0.3 NMU. 2007-01-03 13:41:57 +01:00
Thierry Reding
6eeb18fc68 Build the i915tex DRI module on the i386 and amd64 architectures. 2006-12-05 00:45:42 +00:00
Thierry Reding
01c90575b4 Add missing changelog entry and bug closer. 2006-12-04 07:23:14 +00:00
Thierry Reding
32e9cb0275 Update to new upstream release (6.5.2). 2006-12-02 23:46:07 +00:00
Thierry Reding
017bade93c Update to latest upstream release candidate (6.5.2 RC3). 2006-12-01 00:16:33 +00:00
Thierry Reding
c9c79e7cae Add manpages to the mesa-utils package. 2006-11-29 18:07:26 +00:00
Thierry Reding
777d41dd07 Mark the latest changelog entries as NMUs. 2006-11-26 02:40:01 +00:00
Thierry Reding
6238c30913 Revert maintainer and uploaders fields. Integrate latest NMUs into the
changelog and be a little more verbose in the debian/rules rewrite
changelog entries.
2006-11-26 02:38:31 +00:00
Thierry Reding
db164ab514 Update to latest upstream release candidate. 2006-11-22 19:50:34 +00:00
Thierry Reding
c72ef17731 Add libOSMesa16 and libOSMesa32 entries to libosmesa6.shlibs so packages
linking against those libraries will get the correct dependencies.
2006-11-22 19:47:56 +00:00
Thierry Reding
8dbee7867f Use the new upstream minstall utility to install files and directories. Using
/usr/bin/install would result in a symlink's target being copied instead of
the symlink.
2006-11-18 21:28:07 +00:00
Thierry Reding
f403108f5f Fix typo in changelog. libdrm2-dev is not a package. 2006-11-18 21:22:52 +00:00
Thierry Reding
223a97f92a Bump build-dependency on libdrm-dev (>= 2.2.0). 2006-11-18 21:21:18 +00:00
Thierry Reding
bb44a8ae1f Update to latest upstream release candidate.
Update patches.
2006-11-18 17:59:25 +00:00
Thierry Reding
b5a5062176 Target the mesa package at experimental until etch is released. 2006-11-18 17:50:44 +00:00
Thierry Reding
50faa32ce3 Move the code to choose which configurations to build to a separate script in
order to keep debian/rules cleaner.
2006-11-10 09:00:00 +00:00
Thierry Reding
b6caed527c Also add release notes of version 6.5.1 and the release notes index. 2006-11-10 08:50:35 +00:00
Thierry Reding
d5ef77353b Build Mesa utility programs (glxdemo, glxgears, glxheads, glxinfo) together
with the DRI/GLX-enabled libGL. Don't build the utilities on CPU-optimized
configurations.
2006-11-10 08:40:03 +00:00
Thierry Reding
336beba81b Update Debian changelog. 2006-11-02 21:20:33 +00:00
Thierry Reding
5d23a7905c Add code to provide a mesa-utils package since those utilities are shipped in
the MesaDemos tarball.
2006-11-02 21:05:29 +00:00
Thierry Reding
143dfcc904 Make mesa-common-dev depend on libx11-dev to provide Xlib.h and Xutil.h needed
by the GLX headers.
2006-11-01 07:54:40 +00:00
Thierry Reding
e076cd42f0 Bug #387706 should also be closed because we move libOSMesa to libosmesa6 and
statically link in the software rasterization code.
2006-10-25 00:17:42 +00:00
Thierry Reding
5bed38974e Close 392715 because we now ship the contents of the three Mesa tarballs in
the original source.
2006-10-24 14:45:18 +00:00
Thierry Reding
35bc4b82ff Since the original tarball is repackaged to include the three upstream
tarballs (MesaDemos, MesaGLUT and MesaLib), remove files that are not supposed
to be there.

Note that when importing initially, some files were not even added to the SVN
repository because SVN ignored them. These were configs/diffs~ and
docs/RELNOTES-6.1~. I removed those from the tarball as well.
2006-10-24 14:33:30 +00:00
Thierry Reding
4e5d870f72 Make libosmesa6-dev binNMU-safe again (mesa-common-dev is Architecture: all). 2006-10-24 14:29:11 +00:00
Thierry Reding
83def94db5 Make libgl1-mesa-dev Architecture: all, because all we ship with it is a
symbolic link. Make it binNMU-safe by making it depend on libgl1-mesa-glx and
libgl1-mesa-dri both >= ${Source-Version} instead of exact versions.

Add packages containing debugging symbols for libgl1-mesa-swx11,
libgl1-mesa-glx and libgl1-mesa-dri.
2006-10-24 12:16:16 +00:00
Thierry Reding
9d74c1ef2a -fno-strict-aliasing actually *does* fix #394311 according to the submitter. 2006-10-24 10:31:28 +00:00
Thierry Reding
6023d4bdb5 Build with -fno-strict-aliasing, which should fix bug #394311. 2006-10-24 10:29:21 +00:00
Thierry Reding
5e38f566f2 Provide libgl-dev as alternative to the mesa-common-dev dependency of the
libosmesa6-dev package.
2006-10-23 16:45:00 +00:00
Thierry Reding
b57341a78b Move the glx*.h headers from libgl1-mesa(-swx11)-dev into mesa-common-dev
because both packages provide identical files.
2006-10-23 16:43:41 +00:00
Thierry Reding
910c32c0e9 Install the osmesa.h header in the libosmesa6-dev package where it belongs,
now that all OSMesa libraries have been split off into libosmesa6{,-dev}.

Don't make libosmesa6-dev depend on libgl1-mesa-swx11-dev anymore because it
no longer needs a libGL, only the gl.h header provided by mesa-common-dev.
2006-10-23 16:02:22 +00:00
Thierry Reding
e49456d8d8 Don't force -fPIC everywhere, because it seems only unused on x86 which we
build with assembler optimizations by default, resulting in non-PIC libraries
anyway.
2006-10-23 14:40:10 +00:00
Thierry Reding
c46becc9b3 Move the libOSMesa shlibs entry from libgl1-mesa-swx11 to libosmesa6 because
that's where libOSMesa is now installed.
2006-10-23 08:30:53 +00:00
Thierry Reding
b6e637106f Add the static libGLw and the corresponding headers to the
libgl1-mesa-swx11-dev package again.
2006-10-23 08:23:27 +00:00
Thierry Reding
787cffe3d0 Always force -fPIC. 2006-10-23 08:22:25 +00:00
Thierry Reding
9d914e6809 Make sure the install-source.sh script is executable before trying to run it. 2006-10-20 20:56:41 +00:00
Thierry Reding
4e183c66e9 Add comments to patches and enable the previously disabled patches. 2006-10-20 19:41:00 +00:00
Thierry Reding
b1630fcbe8 Add build-dependency on lesstif2-dev needed to build the GLw libraries. 2006-10-20 19:17:34 +00:00
Thierry Reding
cd44f730cc Enable building the GLw libraries. 2006-10-20 19:07:36 +00:00
Thierry Reding
ce5a39559e Install several documents from the docs subdirectory into the mesa-common-dev
package.
2006-10-20 17:51:36 +00:00
Thierry Reding
ccffaef603 Build architecture-specific static libraries where available. 2006-10-20 12:15:29 +00:00
Thierry Reding
ae41f3a3ee Don't make libgl1-mesa-swx11-dev depend on lesstif2-dev anymore because we
don't ship anything related to LessTif in that package.
2006-10-20 12:06:20 +00:00
Thierry Reding
c7e576d6bf No longer install the default build of libOSMesa in the libgl1-mesa-swx11
package, but in the libosmesa6 package.

Modify the 01_fix-makefile patch to not build libOSMesa on stand-alone builds,
so that it is only created when building OSMesa explicitly.

Build libOSMesa with 8 bits per color channel (the default) in dynamic and
static versions separately from the software rasterization libGL with the
software rasterization code linked in statically so that libOSMesa can run in
parallel with any libGL. This should fix the problem Yann Dirson mentioned a
while ago.

Use the install-source.sh script to install the Mesa sources needed for the
mesa-swx11-source package instead of doing it in the debian/rules file itself.
2006-10-20 07:56:11 +00:00
Thierry Reding
f069b2812e Add a set of files to enable an i686 optimized build of the DRI/GLX-enabled
libGL and the DRI drivers. Disabled for now.
2006-10-20 07:47:37 +00:00
Thierry Reding
f20ff6213e Add a shell script to install the Mesa sources for the mesa-swx11-source
package.
2006-10-20 07:45:39 +00:00
Thierry Reding
24ff74bdcb Add configurations to build static versions of the default OSMesa library.
In all Debian-specific configurations, include the Debian defaults
configuration after the Mesa configurations and drop the 'override' for all
variables because it is no longer needed.
2006-10-20 07:43:50 +00:00
Thierry Reding
230e79916f Merge in the old Debian changelog entries and the old copyright file. 2006-10-17 19:35:35 +00:00
Thierry Reding
9f14b4db35 Use -march in favor of the deprecated -mcpu. 2006-10-17 19:13:27 +00:00
Thierry Reding
ec916398e7 Don't install libOSMesa development files in the libgl1-mesa-swx11 package. 2006-10-16 14:43:43 +00:00
Thierry Reding
6befad112e Actually do rename debian-static to debian-swx11+glu-static. 2006-10-16 13:54:24 +00:00
Thierry Reding
55ce23b8db Beautify IEEE floating point patch. 2006-10-16 13:34:57 +00:00
Thierry Reding
1292ec1ecf Fix typo in the filename of the IEEE FP on s390 and m68k patch. 2006-10-16 13:11:51 +00:00
Thierry Reding
65a324062f Add patches extracted from older Debian diffs but not applied yet because it's
not certain yet that they are needed.
2006-10-16 13:05:42 +00:00
Thierry Reding
c9db617e0b Make the debian-static a subclass of debian-swx11+glu-default, because that's
what it actually is. Rename it accordingly and add it to SWX11_GLU_CONFIGS.

Install the libOSMesa.so symlink and the libOSMesa static library in the
libgl1-mesa-swx11-dev package.

Add .shlibs files for the libgl1-mesa-swx11, libgl1-mesa-glx and libglu1-mesa
packages.
2006-10-16 13:04:29 +00:00
Thierry Reding
b43137d0cc Build libGLU by default for swx11+glu configurations but not in the
configurations targetting specific CPU's because it is unlikely to result in
an increase in performance.
2006-10-16 11:53:06 +00:00
Thierry Reding
eeca642778 Add versioned build-dependency on dpkg-dev (>= 1.13.19) to make sure the
binary:Version and source:Version substitution variables can be used.
2006-10-16 11:49:46 +00:00
Thierry Reding
5cef6ca36a Move the detection of architecture-specific configurations to the debian/rules
file.

Add the configs target to debian/rules for checking whether the correct
configurations will be built.
2006-10-16 10:03:32 +00:00
Thierry Reding
be2b518f7b Also build a static version of libGLU. 2006-10-16 09:53:30 +00:00
Thierry Reding
1b0d29afbc Rename the CPU optimization configurations to make it clear that they only
build swx11 and GLU.

Build architecture-optimized swx11 and GLU where possible.
2006-10-16 09:39:59 +00:00
Thierry Reding
55cf480c0b Drop the libgl1-mesa-swx11-dbg package until there's a decision as to whether
it is really needed.

Allow more than one type of optimization libraries to be built for any given
architecture. Currently optimization configurations are provided for i386
[i686], alpha [ev5], amd64, powerpc [603], ppc64 and sparc [ultrasparc].
However, only i686 and amd64 are enabled for now for safety.

Add the libgl1-mesa-swx11-i686 for i686 optimized libraries.

Drop the debian-common configuration because it is no longer used.

Don't build libGLU in all configurations, only when needed.
2006-10-16 08:58:05 +00:00
Thierry Reding
84db6912f0 Add missing build-dependencies. 2006-10-15 18:02:33 +00:00
Thierry Reding
7495cc28a0 Only build the i810, i915, i965, sis and unichrome DRI drivers on i386 and
amd64.

Only build the ffb DRI driver on sparc.
2006-10-15 16:17:53 +00:00
Thierry Reding
87744f786b Use STAMP_DIR consistently throughout the debian/rules file.
Make the BUILD_STAMPS targets depend on patch so that when they are built
separately the patches will also be applied. Thanks to Michel Dänzer for
catching this.

Make symlinking the source tree idempotent by adding the -f option to cp.
Thanks again Michel Dänzer.
2006-10-15 14:41:12 +00:00
Thierry Reding
5bcfa4c859 Build only optimized versions of the DRI drivers, only using no optimizations
when not supported.

Add configuration to build the DRI drivers on amd64.
2006-10-15 13:30:22 +00:00
Thierry Reding
8ef731bcd1 Add a patch (00_create-libdir) that fixes the upstream build system to make
sure the LIB_DIR is created or bail out.

Update the 01_fix-makefile patch to only install libGL/libOSMesa if they
exist. Don't make installation dependent on which drivers are built.

Build optimized versions of the swx11 libraries in addition to those that
should work on the least-capable supported processor for the given
architecture.

Have most configurations include debian-default instead of debian-common so
that INSTALL_DIR is correctly defined. debian-common can probably be dropped
anyway.
2006-10-15 13:15:56 +00:00
Thierry Reding
8d02f3c68a Add patch to fix upstream build system to correctly install libOSMesa and
variants. This patch also fixes the build system not to install libGL if only
the OSMesa library is built.

Update *.install files to install more files. Pretty much everything except
debugging and optimized libraries is now built and installed.

Enable building static libraries for inclusion in the -dev packages.
2006-10-15 00:49:41 +00:00
Thierry Reding
b1f8b49e8b Add build-dependency on quilt.
Remove shlibs:Depends substitution variable from -dev packages, which doesn't
make sense.
2006-10-15 00:42:25 +00:00
David Nusinow
75733e4237 Commit diff of 6.5.1-0.2 so we've got a record of it here. 2006-10-14 22:42:57 +00:00
Thierry Reding
6e318e2cd1 Base the Debian OSMesa configurations on linux-osmesa* provided upstream.
Add a configuration for building a static version of OSMesa with 32 bits per
color channel.
2006-10-14 17:04:02 +00:00
Thierry Reding
01e6b84b14 Fix some small inconsistencies involving configuration names. 2006-10-14 16:32:25 +00:00
Thierry Reding
69705d22be Rename the debian-dri-common configuration to debian-dri-default, to make it
consistent with the naming scheme of debian-default.
2006-10-14 16:29:32 +00:00
Thierry Reding
b16d1dcff5 Add a default Debian-specific configuration which overrides settings from the
default Mesa configuration.
2006-10-14 16:28:45 +00:00
Thierry Reding
10c5b5d5ba Comment the Debian-specific configurations. 2006-10-14 16:25:30 +00:00
Thierry Reding
d9c5c170f6 Rename the -default configurations to -any, which is a more appropriate name. 2006-10-14 15:44:15 +00:00
Thierry Reding
32c7138d86 Move the debian/debian-dri configurations to debian-default/debian-dri-default
respectively, and use debian/debian-dri to choose an architecture specific
configuration.
2006-10-14 15:42:59 +00:00
Thierry Reding
d29244d872 Add OSMesa configurations and made fixed several issues here and there. 2006-10-14 15:11:31 +00:00
Thierry Reding
0fca6eda1e Initial version of a new build system for Mesa. 2006-10-14 04:33:41 +00:00
Thierry Reding
1ddf606332 Import Mesa 6.5.1 (MesaLib, MesaDemos, MesaGLUT). 2006-10-14 03:46:41 +00:00
David Nusinow
bc8084da74 Fix goofed patch from before 2006-09-26 01:36:37 +00:00
David Nusinow
5eed400862 Prepare changelog for release 2006-09-26 01:21:59 +00:00
David Nusinow
a37a0959ac * Add patch from Cyril Brulebois to allow package to build on HURD, which
lacks DRI and directfb. This includes not using lib-directfb in the
  build-depends for hurd-i386. It also creates a new debian config,
  debian-indirect, which is used when building for HURD. This config is
  invoked in the debian-dri config on hurd-i386. Thanks to Cyril Brulebois
  for the patch, Michael Banck, Michel Dänzer, and Samuel Thibault for
  input on an appropriate fix. (closes: #358065)
2006-09-26 00:46:14 +00:00
David Nusinow
53845c52e5 * Re-add s390 and m68k to the USE_IEEE test in src/mesa/main/imports.h. This
package seriously needs to store patches somewhere that are easy to find
  and re-apply.
2006-09-26 00:32:06 +00:00
David Nusinow
d429f0b14f * Make sure that libGl looks for the dri drivers in the proper location. Do
this by setting the appropriate variables in the debian config
2006-09-25 02:57:24 +00:00
David Nusinow
02339187ad * Stuff not in the upstream tarballs
* Bump libdrm-dev build-dep to >= 2.0.2
* Add libdrm cflags to the debian-dri config. This allows the build system
  to find drm.h
2006-09-24 22:55:41 +00:00
David Nusinow
827f778525 * Stuff not in the upstream tarballs
+ os2 glut stuff
  + docs/gears.png
2006-09-24 21:54:25 +00:00
David Nusinow
d3fc85eea9 Remove os2 glut stuff. It's not in the tarballs 2006-09-24 21:50:01 +00:00
David Nusinow
9a62963129 Add glut 2006-09-24 21:40:02 +00:00
David Nusinow
a0b2543ba1 Add mesa demos 2006-09-24 21:29:21 +00:00
David Nusinow
387acaac69 * New upstream version 2006-09-24 21:21:15 +00:00
David Nusinow
952b775dbe Rename mesa dir 2006-09-24 21:18:17 +00:00
David Nusinow
eadb76b3f8 Bump to latest mesa in Debian 2006-09-24 21:11:46 +00:00
Steve Langasek
b3f7313ae4 Nuke the directory for an obsolete version of mesa -- no reason to merge
it anywhere, it's already superseded
2006-09-18 23:49:49 +00:00
David Nusinow
0c97d48027 * Install the glx sources with the rest in the sources package. This will
allow building of Xgl
2006-04-08 18:28:29 +00:00
David Nusinow
c4a678de9e Update source package install file, change version number to match reality, and note NMU in changelog 2006-04-07 06:36:53 +00:00
David Nusinow
2c834f4d29 * Don't build now non-existant i830 driver in debian dri configs. Fixes
FTBFS.
2006-04-07 03:11:47 +00:00
David Nusinow
b93d290c9f * Remove set_buffer function from idirectfbgl_mesa.c in directfb driver.
Also remove the line that sets SetBuffer to it. Fixes FTBFS.
2006-04-07 02:22:22 +00:00
David Nusinow
32266bb21a * Re-Add s390 and m68k to the USE_IEEE test in src/mesa/main/imports.h.
I need to submit this upstream.
2006-04-07 01:50:23 +00:00
David Nusinow
cab39103b7 * Increment libdrm-dev build-dep to 2.0.1 2006-04-07 00:38:00 +00:00
David Nusinow
5c56a80ac9 Copy over Debian packaging from 6.4.1 2006-04-07 00:35:30 +00:00
David Nusinow
d0bac31721 Copy vendor drop of Mesa 6.5 over to working branch 2006-04-07 00:30:49 +00:00
David Nusinow
9258e54cfa Move incorrect vendor drop of Mesa 6.5 2006-04-07 00:29:08 +00:00
David Nusinow
ccb68495e0 Vendor drop of Mesa 6.5 2006-04-07 00:24:21 +00:00
David Nusinow
db4236fe0b * Add versioned conflict between libgl1-mesa-dri and xlibmesa-dri so that
the xlibmesa-dri transitional upgrade package works
2006-03-07 02:48:44 +00:00
David Nusinow
a4e43427f8 * Move compatibility packages to the xorg source package. The versioning on
these packages will be too low without an epoch and they definitely fit
  with the xorg package anyway.
2006-02-24 05:43:01 +00:00
David Nusinow
829190abdd * Provide compatibility packages for the Xorg 6.9 mesa packages. This
package will be the only source of mesa packages in Debian for the
  present, and these packages will provide for smooth upgrades and
  compatibility.
2006-02-22 03:22:26 +00:00
David Nusinow
99688cd051 Add m68k to previous fix to fix its FTBFS also 2006-02-11 23:00:27 +00:00
David Nusinow
1ba0a92702 * Add s390 to the USE_IEEE test in src/mesa/main/imports.h.
(closes: #349437)
2006-02-09 01:24:52 +00:00
David Nusinow
065d5a8d17 * Remove makedepend from build-depends. Now we'll just build-dep on xutils
to get the app, which will translate over to our own xorg 7.0 plans.
2006-02-01 00:22:24 +00:00
David Nusinow
3b53c0ac19 * Re-add dh_installdirs call to binary-indep target so that we get
arch-specific dirs for the mesa-swx11-source package
2006-01-30 04:02:11 +00:00
David Nusinow
ed56fe4e82 * mesa-swrast-src.install stop looking for the swx11 dir and look for swrast 2006-01-22 19:50:19 +00:00
David Nusinow
1fbbe807ed Make libglu1-mesa default to depending on libgl1-mesa-glx. Thanks Michel Dänzer for catching this one. 2006-01-22 00:07:07 +00:00
David Nusinow
c3dccfa05b Use appropriate version number for an NMU. Thanks Julien for catching this one 2006-01-21 23:40:34 +00:00
David Nusinow
7703d9293f Incorporate other NMU for xlibs-dev transition 2006-01-21 23:32:52 +00:00
David Nusinow
bcf560e4b8 * Change libgl1-mesa to be named libgl1-mesa-glx 2006-01-19 23:45:36 +00:00
David Nusinow
9d944b501a * Change libgl1-mesa-swrast to be named libgl1-mesa-swx11 2006-01-16 22:19:03 +00:00
David Nusinow
cf013a5384 Add more closers 2006-01-11 04:10:39 +00:00
David Nusinow
2f367491c4 Add closer 2006-01-11 04:00:15 +00:00
David Nusinow
e7d93b62ec * Change libgl1-mesa-swrast Depends on libx11-6-dev to libx11-dev.
(closes: #347205)
* Add closer for packaging new upstream
2006-01-11 03:48:31 +00:00
David Nusinow
a5b309c0c9 Document that we'll be NMU'ing this for modular work 2006-01-06 03:09:12 +00:00
David Nusinow
df46b96ab0 Re-add mesa-6.4.1 which was accidentally deleted 2006-01-03 05:15:27 +00:00
David Nusinow
bda1e332ce Update libs to 7.0 release with packaging 2005-12-30 20:12:45 +00:00
David Nusinow
53b2ad5fd0 Rename mesa lib dir to appropriate version 2005-12-16 04:04:38 +00:00
David Nusinow
b44d9c657d Changelog version bump 2005-12-16 04:03:27 +00:00
David Nusinow
92ef7527d1 * New upstream version (6.4.1)
* Merge changes from Ubuntu version 6.4.1-0ubuntu1
  + Add new files required by xorg-server GL build to mesa-swrast-source.
2005-12-14 01:43:13 +00:00
David Nusinow
c37d6b4ca8 Commit Mesa 6.4 sources and packaging 2005-11-09 02:46:07 +00:00
13140 changed files with 2116192 additions and 5202572 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 +0,0 @@
((nil . ((show-trailing-whitespace . t)))
(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

10
.emacs-dirvars Normal file
View File

@@ -0,0 +1,10 @@
;; -*- emacs-lisp -*-
;;
;; This file is processed by the dirvars emacs package. Each variable
;; setting below is performed when this dirvars file is loaded.
;;
indent-tabs-mode: nil
tab-width: 8
c-basic-offset: 3
kde-emacs-after-parent-string: ""
evaluate: (c-set-offset 'inline-open '0)

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

28
.gitignore vendored
View File

@@ -1,6 +1,26 @@
.vscode*
*.a
*.dll
*.exe
*.ilk
*.o
*.obj
*.pc
*.pdb
*.pyc
*.pyo
*.out
/build
.venv/
*.so
*.sw[a-z]
*~
depend
depend.bak
lib
lib64
configure
autom4te.cache
aclocal.m4
config.log
config.status
cscope*
.scon*
config.py
build

View File

@@ -1,319 +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}/mesa-lava/${KERNEL_REPO}/${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}/mesa-lava/${KERNEL_REPO}/${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}/mesa-lava/${KERNEL_REPO}/${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}/mesa-lava/${KERNEL_REPO}/${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}/mesa-lava/${KERNEL_REPO}/${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
CI_JOB_JWT_FILE: /minio_jwt
S3_HOST: s3.freedesktop.org
# per-pipeline artifact storage on MinIO
PIPELINE_ARTIFACTS_BASE: ${S3_HOST}/artifacts/${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: "${S3_HOST}/mesa-tracie-results/$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"
default:
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" &&
. ${SCRIPTS_DIR}/setup-test-env.sh &&
echo -n "${CI_JOB_JWT}" > "${CI_JOB_JWT_FILE}" &&
unset CI_JOB_JWT # Unsetting vulnerable env variables
after_script:
# Work around https://gitlab.com/gitlab-org/gitlab/-/issues/20338
- find -name '*.log' -exec mv {} {}.txt \;
- >
set +x
test -e "${CI_JOB_JWT_FILE}" &&
export CI_JOB_JWT="$(<${CI_JOB_JWT_FILE})" &&
rm "${CI_JOB_JWT_FILE}"
# Retry when job fails. Failed jobs can be found in the Mesa CI Daily Reports:
# https://gitlab.freedesktop.org/mesa/mesa/-/issues/?sort=created_date&state=opened&label_name%5B%5D=CI%20daily
retry:
max: 1
# Ignore runner_unsupported, stale_schedule, archived_failure, or
# unmet_prerequisites
when:
- api_failure
- runner_system_failure
- script_failure
- job_execution_timeout
- scheduler_failure
- data_integrity_failure
- unknown_failure
stages:
- sanity
- container
- git-archive
- build-x86_64
- build-misc
- code-validation
- amd
- intel
- nouveau
- arm
- broadcom
- freedreno
- etnaviv
- software-renderer
- layered-backends
- deploy
include:
- project: 'freedesktop/ci-templates'
ref: 16bc29078de5e0a067ff84a1a199a3760d3b3811
file:
- '/templates/ci-fairy.yml'
- project: 'freedesktop/ci-templates'
ref: *ci-templates-commit
file:
- '/templates/alpine.yml'
- '/templates/debian.yml'
- '/templates/fedora.yml'
- local: '.gitlab-ci/image-tags.yml'
- local: '.gitlab-ci/lava/lava-gitlab-ci.yml'
- local: '.gitlab-ci/container/gitlab-ci.yml'
- local: '.gitlab-ci/build/gitlab-ci.yml'
- local: '.gitlab-ci/test/gitlab-ci.yml'
- local: '.gitlab-ci/farm-rules.yml'
- local: '.gitlab-ci/test-source-dep.yml'
- local: 'docs/gitlab-ci.yml'
- local: 'src/**/ci/gitlab-ci.yml'
# YAML anchors for rule conditions
# --------------------------------
.rules-anchors:
# Pre-merge pipeline
- &is-pre-merge '$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 ../$CI_PROJECT_NAME.tar.gz .
- ci-fairy s3cp --token-file "${CI_JOB_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
# 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

@@ -1,63 +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 }}'

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.$((48 + 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.$((48 + 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 $rootfs_dst/
date +'%F %T'
# Make JWT token available as file in the bare-metal storage to enable access
# to MinIO
cp "${CI_JOB_JWT_FILE}" "${rootfs_dst}${CI_JOB_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

@@ -1,675 +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_${CI_JOB_NAME}"
when: always
paths:
- _build/meson-logs/*.txt
- _build/meson-logs/strace
- shader-db
- artifacts
# Just Linux
.build-linux:
extends: .build-common
variables:
CCACHE_COMPILERCHECK: "content"
CCACHE_COMPRESS: "true"
CCACHE_DIR: /cache/mesa/ccache
# Use ccache transparently, and print stats before/after
before_script:
- !reference [default, before_script]
- |
export PATH="/usr/lib/ccache:$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: ${CI_JOB_NAME}
paths:
- subprojects/packagecache
.meson-build:
extends:
- .build-linux
- .use-debian/x86_64_build
stage: build-x86_64
variables:
LLVM_VERSION: 15
script:
- .gitlab-ci/meson/build.sh
debian-testing:
extends:
- .meson-build
- .ci-deqp-artifacts
variables:
UNWIND: "enabled"
DRI_LOADERS: >
-D glx=dri
-D gbm=enabled
-D egl=enabled
-D platforms=x11,wayland
GALLIUM_ST: >
-D dri3=enabled
-D gallium-nine=true
-D gallium-va=enabled
-D gallium-rusticl=true
GALLIUM_DRIVERS: "swrast,virgl,radeonsi,zink,crocus,iris,i915,r300"
VULKAN_DRIVERS: "swrast,amd,intel,intel_hasvk,virtio,nouveau"
BUILDTYPE: "debugoptimized"
EXTRA_OPTION: >
-D spirv-to-dxil=true
-D valgrind=disabled
-D perfetto=true
-D tools=drm-shim
S3_ARTIFACT_NAME: mesa-x86_64-default-${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 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
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
section_switch shader-db "shader-db"
.gitlab-ci/run-shader-db.sh
timeout: 30m
# 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 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=false
-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=true
-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 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-${BUILDTYPE}
script:
- CROSS=aarch64-linux-android GALLIUM_DRIVERS=etnaviv,freedreno,lima,panfrost,vc4,v3d VULKAN_DRIVERS=freedreno,broadcom,virtio .gitlab-ci/meson/build.sh
# x86_64 build:
# Can't do Intel because gen_decoder.c currently requires libexpat, which
# is not a dependency that AOSP wants to accept. Can't do Radeon Gallium
# drivers because they requires LLVM, which we don't have an Android build
# of.
- CROSS=x86_64-linux-android GALLIUM_DRIVERS=iris,virgl VULKAN_DRIVERS=amd,intel .gitlab-ci/meson/build.sh
- .gitlab-ci/prepare-artifacts.sh
.meson-cross:
extends:
- .meson-build
stage: build-misc
variables:
UNWIND: "disabled"
DRI_LOADERS: >
-D glx=dri
-D gbm=enabled
-D egl=enabled
-D platforms=x11,wayland
-D osmesa=false
GALLIUM_ST: >
-D dri3=enabled
-D gallium-vdpau=disabled
-D gallium-omx=disabled
-D gallium-va=disabled
-D gallium-xa=disabled
-D gallium-nine=false
.meson-arm:
extends:
- .meson-cross
- .use-debian/arm64_build
needs:
- debian/arm64_build
variables:
VULKAN_DRIVERS: freedreno,broadcom
GALLIUM_DRIVERS: "etnaviv,freedreno,kmsro,lima,nouveau,panfrost,swrast,tegra,v3d,vc4,zink"
BUILDTYPE: "debugoptimized"
tags:
- aarch64
debian-arm32:
extends:
- .meson-arm
- .ci-deqp-artifacts
variables:
CROSS: armhf
EXTRA_OPTION: >
-D llvm=disabled
-D valgrind=disabled
S3_ARTIFACT_NAME: mesa-arm32-default-${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:
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-${BUILDTYPE}
MESON_TEST_ARGS: "--no-suite mesa:compiler --no-suite mesa:util"
debian-arm64:
extends:
- .meson-arm
- .ci-deqp-artifacts
variables:
C_ARGS: >
-Wno-error=array-bounds
-Wno-error=stringop-truncation
VULKAN_DRIVERS: "freedreno,broadcom,panfrost,imagination-experimental"
EXTRA_OPTION: >
-D llvm=disabled
-D valgrind=disabled
-D imagination-srv=true
-D perfetto=true
-D freedreno-kmds=msm,virtio
-D teflon=true
S3_ARTIFACT_NAME: mesa-arm64-default-${BUILDTYPE}
script:
- .gitlab-ci/meson/build.sh
- .gitlab-ci/prepare-artifacts.sh
debian-arm64-asan:
extends:
- debian-arm64
variables:
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-${BUILDTYPE}
MESON_TEST_ARGS: "--no-suite mesa:compiler"
debian-arm64-build-test:
extends:
- .meson-arm
- .ci-deqp-artifacts
variables:
VULKAN_DRIVERS: "amd"
EXTRA_OPTION: >
-Dtools=panfrost,imagination
debian-arm64-release:
extends:
- debian-arm64
variables:
BUILDTYPE: release
S3_ARTIFACT_NAME: mesa-arm64-default-${BUILDTYPE}
C_ARGS: >
-Wno-error=array-bounds
-Wno-error=stringop-truncation
-Wno-error=stringop-overread
script:
- .gitlab-ci/meson/build.sh
- 'if [ -n "$MESA_CI_PERFORMANCE_ENABLED" ]; then .gitlab-ci/prepare-artifacts.sh; fi'
debian-clang:
extends: .meson-build
variables:
BUILDTYPE: debug
LLVM_VERSION: 15
UNWIND: "enabled"
C_ARGS: >
-Wno-error=constant-conversion
-Wno-error=enum-conversion
-Wno-error=initializer-overrides
-Wno-error=sometimes-uninitialized
-Werror=misleading-indentation
CPP_ARGS: >
-Wno-error=c99-designator
-Wno-error=overloaded-virtual
-Wno-error=tautological-constant-out-of-range-compare
-Wno-error=unused-private-field
DRI_LOADERS: >
-D glx=dri
-D gbm=enabled
-D egl=enabled
-D glvnd=true
-D platforms=x11,wayland
GALLIUM_ST: >
-D dri3=enabled
-D gallium-extra-hud=true
-D gallium-vdpau=enabled
-D gallium-omx=bellagio
-D gallium-va=enabled
-D gallium-xa=enabled
-D gallium-nine=true
-D gles1=enabled
-D gles2=enabled
-D llvm=enabled
-D microsoft-clc=disabled
-D shared-llvm=enabled
-D opencl-spirv=true
-D shared-glapi=enabled
GALLIUM_DRIVERS: "iris,nouveau,kmsro,r300,r600,freedreno,swrast,svga,v3d,vc4,virgl,etnaviv,panfrost,lima,zink,radeonsi,tegra,d3d12,crocus,i915,asahi"
VULKAN_DRIVERS: intel,amd,freedreno,broadcom,virtio,swrast,panfrost,imagination-experimental,microsoft-experimental,nouveau
EXTRA_OPTION:
-D spirv-to-dxil=true
-D osmesa=true
-D imagination-srv=true
-D tools=drm-shim,etnaviv,freedreno,glsl,intel,intel-ui,nir,nouveau,lima,panfrost,asahi,imagination
-D vulkan-layers=device-select,overlay
-D build-aco-tests=true
-D intel-clc=enabled
-D intel-rt=enabled
-D imagination-srv=true
-D teflon=true
CC: clang-${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 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
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"
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"

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

@@ -1,131 +0,0 @@
#!/bin/bash
for var in \
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 \
CI_JOB_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 \
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_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_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 \
VK_ICD_FILENAMES \
VKD3D_PROTON_RESULTS \
VKD3D_CONFIG \
VKD3D_TEST_EXCLUDE \
ZINK_DESCRIPTORS \
ZINK_DEBUG \
LVP_POISON_MEMORY \
; 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/!($MESA_LOADER_DRIVER_OVERRIDE)_dri.so
fi
ls -1 /install/lib/dri/*_dri.so
if [ "$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_ICD_FILENAMES="/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 "$HWCI_START_XORG" ]; then
echo "Please consider dropping HWCI_START_XORG and instead using Weston XWayland for testing."
WESTON_X11_SOCK="/tmp/.X11-unix/X1"
fi
export WAYLAND_DISPLAY=wayland-0
# Display server is Weston Xwayland when HWCI_START_XORG is not set or Xorg when it's
export DISPLAY=:0
mkdir -p /tmp/.X11-unix
env \
VK_ICD_FILENAMES="/install/share/vulkan/icd.d/${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 "$S3_RESULTS_UPLOAD" ]; then
tar --zstd -cf results.tar.zst results/;
ci-fairy s3cp --token-file "${CI_JOB_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

@@ -1,66 +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="${LLVM_VERSION:=16}"
EPHEMERAL=(
)
DEPS=(
bash
bison
ccache
clang16-dev
cmake
clang-dev
coreutils
curl
flex
gcc
g++
git
gettext
glslang
linux-headers
llvm16-static
llvm16-dev
meson
expat-dev
elfutils-dev
libdrm-dev
libselinux-dev
libva-dev
libpciaccess-dev
zlib-dev
python3-dev
py3-mako
py3-ply
vulkan-headers
spirv-tools-dev
util-macros
wayland-dev
wayland-protocols
)
apk --no-cache add "${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

@@ -1,275 +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=1.3.7.0
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 `$OLDPWD/.gitlab-ci/container/patches/` listed in cts_patch_files).
# Both list variables would have comments explaining the reasons behind the
# patches.
# shellcheck disable=SC2034
vk_cts_commits_to_backport=(
# Take multiview into account for task shader inv. stats
22aa3f4c59f6e1d4daebd5a8c9c05bce6cd3b63b
# Remove illegal mesh shader query tests
2a87f7b25dc27188be0f0a003b2d7aef69d9002e
# Relax fragment shader invocations result verifications
0d8bf6a2715f95907e9cf86a86876ff1f26c66fe
# Fix several issues in dynamic rendering basic tests
c5453824b498c981c6ba42017d119f5de02a3e34
# Add setVisible for VulkanWindowDirectDrm
a8466bf6ea98f6cd6733849ad8081775318a3e3e
)
# shellcheck disable=SC2034
vk_cts_patch_files=(
# Derivate subgroup fix
# https://github.com/KhronosGroup/VK-GL-CTS/pull/442
build-deqp-vk_Use-subgroups-helper-in-derivate-tests.patch
build-deqp-vk_Add-missing-subgroup-support-checks-for-linear-derivate-tests.patch
)
if [ "${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 [ "${DEQP_TARGET}" = 'android' ]; then
gl_cts_patch_files+=(
build-deqp-gl_Allow-running-on-Android-from-the-command-line.patch
build-deqp-gl_Android-prints-to-stdout-instead-of-logcat.patch
)
fi
# shellcheck disable=SC2034
# GLES builds also EGL
gles_cts_commits_to_backport=(
# Implement support for the EGL_EXT_config_select_group extension
88ba9ac270db5be600b1ecacbc6d9db0c55d5be4
)
# shellcheck disable=SC2034
gles_cts_patch_files=(
)
if [ "${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-master.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-master*
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 v1.611.0 --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="1e631479c0b477006dd7561c55e06269d2878d8d"
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="a0ccc383937903f4ca0997ce53e41ccce7f2f2ec"
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="snapshot-2024wk06"
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

@@ -1,23 +0,0 @@
#!/usr/bin/env bash
# shellcheck disable=SC2086 # we want word splitting
set -ex
export LIBWAYLAND_VERSION="1.21.0"
export WAYLAND_PROTOCOLS_VERSION="1.31"
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