Marek Olšák [Thu, 28 Apr 2011 17:34:56 +0000 (19:34 +0200)]
mesa: remove unused variable bytesPerRow
Marek Olšák [Wed, 27 Apr 2011 14:30:28 +0000 (16:30 +0200)]
r600g: implement EXT_packed_float
Marek Olšák [Tue, 26 Apr 2011 00:28:33 +0000 (02:28 +0200)]
st/mesa: support EXT_packed_float
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Tue, 26 Apr 2011 00:27:25 +0000 (02:27 +0200)]
mesa: implement EXT_packed_float
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Marek Olšák [Wed, 27 Apr 2011 11:37:27 +0000 (13:37 +0200)]
util: implement R11G11B10_FLOAT pack/unpack functions
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Wed, 27 Apr 2011 10:52:10 +0000 (12:52 +0200)]
mesa, util: move RGB9E5 conversion functions to gallium/util
Also use MAX3 and incorporate Ian's suggestion in texformat.c.
I don't think wrapping u_format_rgb9e5.h in another header and thus making it
more complicated is worth it.
Marek Olšák [Tue, 26 Apr 2011 00:24:47 +0000 (02:24 +0200)]
r600g: trivially implement EXT_texture_shared_exponent
Nothing else needed.
Marek Olšák [Tue, 26 Apr 2011 00:29:00 +0000 (02:29 +0200)]
util: implement R9G9B9E5 pack and unpack functions
softpipe and llvmpipe support done (sampler only).
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Tue, 26 Apr 2011 00:23:33 +0000 (02:23 +0200)]
st/mesa: support EXT_texture_shared_exponent
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Tue, 26 Apr 2011 00:18:24 +0000 (02:18 +0200)]
mesa: implement EXT_texture_shared_exponent
swrast support done.
There is no renderbuffer support in swrast, because it's not required
by the extension.
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Mon, 25 Apr 2011 23:53:20 +0000 (01:53 +0200)]
mesa: set reasonable defaults in update_wrapper
I was wondering why I had been getting GL_RGBA for GL_RGB9_E5.
Instead of setting GL_RGBA and CHAN_TYPE for most types,
use the helper functions to obtain the info.
Reviewed-by: Brian Paul <brianp@vmware.com>
Christopher Egert [Fri, 29 Apr 2011 00:13:35 +0000 (02:13 +0200)]
i915g: Enable S3TC texture support
Reviewed-by: Jakob Bornecrantz <wallbraker@gmail.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Christopher Egert <cme3000@gmail.com>
Signed-off-by: Jakob Bornecrantz <wallbraker@gmail.com>
Kenneth Graunke [Tue, 26 Apr 2011 23:14:06 +0000 (16:14 -0700)]
prog_print: Add support for printing the TXD opcode.
Reviewed-by: Brian Paul <brianp@vmware.com>
Brian Paul [Thu, 28 Apr 2011 18:41:45 +0000 (12:41 -0600)]
llvmpipe: move active_query assignment
Fixes piglit regression.
Brian Paul [Thu, 28 Apr 2011 13:38:53 +0000 (07:38 -0600)]
llvmpipe: move active_query assignment in lp_setup_begin_query()
If we run out of bin memory and do an early return from
lp_setup_begin_query() we'd omit setting the setup->active_query
pointer. Then, when lp_setup_end_query() was later called, the
assertion for setup->active_query == pq would fail. Moving the
assigment in lp_setup_begin_query() avoids that.
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
José Fonseca [Wed, 27 Apr 2011 18:05:29 +0000 (19:05 +0100)]
sgi: Fix MSVC build.
Including windows.h was ineffective on MSVC because we define the NOGDI macro,
which skips the wingdi.h include.
Unsetting NOGDI is also a bad idea because it causes all sort of symbol
clashes with SGI code.
The real problem is that WINGDAPI was not being defined, also due to NOGDI,
so simply define it to blank if not done already. This seems to make
everybody happy.
Brian Paul [Wed, 27 Apr 2011 15:31:30 +0000 (09:31 -0600)]
st/mesa: raise shader MaxParameters if driver supports more
The default value is 64 but drivers usually advertise more, like 4096.
Allows ARB vp/fp programs to use more parameters.
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Brian Paul [Wed, 27 Apr 2011 15:23:16 +0000 (09:23 -0600)]
mesa: emit more info in program parser error message
Brian Paul [Mon, 25 Apr 2011 16:26:19 +0000 (10:26 -0600)]
svga: emit user-defined clip plane state
User-defined clip planes were a swtnl fallback before.
Eric Anholt [Tue, 26 Apr 2011 22:14:01 +0000 (15:14 -0700)]
docs: Note that ES2_compatibility is already done.
Eric Anholt [Mon, 25 Apr 2011 17:36:29 +0000 (10:36 -0700)]
Revert "intel: use throttle ioctl for throttling"
This reverts commit
50ade6ea697953bb17e3ca7210515fbd0411cd1e.
Fixes jerky rendering again on apps that don't block on the GPU per
frame and are GPU bound (e.g. 3DMMES on Ironlake). The whole point of
this complicated throttle scheme is to wait on frame n-1 to have
started rendering before starting frame n's rendering. Otherwise, the
GPU-bound app will race ahead and call the GL to draw many
nearly-identical frames, then >0ms later get stuck waiting for them
(all dispatched at about the same time) to retire, then render a new
batch of nearly-identical frames.
Brian Paul [Wed, 27 Apr 2011 15:06:07 +0000 (09:06 -0600)]
st/mesa: choose 3-component float formats before 4-component formats
If GL_RGB16F or GL_RGB32F is specified let's try the 3-component float
texture formats before trying the 4-component ones. Before this,
GL_RGB16/32F were treated the same as GL_RGBA16/32F.
Reviewed-by: Marek Olšák <maraeo@gmail.com>
Brian Paul [Wed, 27 Apr 2011 15:05:30 +0000 (09:05 -0600)]
st/mesa: whitespace fixes
Marek Olšák [Wed, 27 Apr 2011 10:54:24 +0000 (12:54 +0200)]
util: make macros MIN3, MAX3, MIN4, MAX4 little more efficient
Marek Olšák [Wed, 27 Apr 2011 10:53:58 +0000 (12:53 +0200)]
mesa: add macros MIN3 and MAX3
Marek Olšák [Wed, 27 Apr 2011 11:08:25 +0000 (13:08 +0200)]
gallium/nouveau: fix printf warnings
Marek Olšák [Wed, 27 Apr 2011 11:05:53 +0000 (13:05 +0200)]
gallivm: fix warning: ‘value’ may be used uninitialized in this function
The path where it's uninitialized is guarded by an assert.
Marek Olšák [Wed, 27 Apr 2011 11:04:21 +0000 (13:04 +0200)]
st/mesa: fix warning: ‘user_memory’ may be used uninitialized in this function
It's initialized later in a conditional the condition of which is always true
the first time it's evaluated.
Marek Olšák [Wed, 27 Apr 2011 11:02:32 +0000 (13:02 +0200)]
svga: fix warning: ‘uc.ui’ may be used uninitialized in this function
This is safe because it's initialized if buffers & PIPE_CLEAR_COLOR and
probably doesn't have any effect otherwise.
nobled [Sun, 18 Jul 2010 16:08:19 +0000 (16:08 +0000)]
os: simplify ifdef
The actual code that needs this include is just using
"if defined (PIPE_OS_UNIX)", and the two conditions should match.
This should also make the file compile under Hurd.
Brian Paul [Tue, 26 Apr 2011 20:58:33 +0000 (14:58 -0600)]
docs: fix typo, wrap lines in 7.11 release notes
Brian Paul [Tue, 26 Apr 2011 20:57:57 +0000 (14:57 -0600)]
docs: add GL_ARB_robustness to 7.11 release notes
Brian Paul [Tue, 26 Apr 2011 20:54:41 +0000 (14:54 -0600)]
Squashed commit of the following:
commit
864fe253b04105b7469e5f7b064dc37637b944f8
Author: Brian Paul <brianp@vmware.com>
Date: Thu Apr 21 20:13:07 2011 -0600
mesa: s/exec/disp/ in _mesa_init_histogram_dispatch()
This function isn't normally compiled (FEATURE_histogram).
commit
f4bf45e2b94b582cacd19cdca873c5be627e4250
Author: nobled <nobled@dreamwidth.org>
Date: Thu Apr 21 07:53:58 2011 -0600
mesa: hook up GL_ARB_robustness dispatch functions
...and advertise the extension.
Signed-off-by: Brian Paul <brianp@vmware.com>
commit
2b89e38e5f572dc40cebc06381ae7c5d04386998
Author: nobled <nobled@dreamwidth.org>
Date: Thu Apr 21 07:53:58 2011 -0600
mesa: regenerated API files for GL_ARB_robustness
Signed-off-by: Brian Paul <brianp@vmware.com>
commit
5d5ebfb7135cec9d833adef86cbf4d0f3d9beca8
Author: nobled <nobled@dreamwidth.org>
Date: Thu Apr 21 07:53:57 2011 -0600
glapi: add ARB_robustness xml
Signed-off-by: Brian Paul <brianp@vmware.com>
commit
0159d1d6d99f4bbc18381dc2081c20d3aff17ac9
Author: nobled <nobled@dreamwidth.org>
Date: Thu Apr 21 07:53:57 2011 -0600
mesa: implement GL_ARB_robustness functions
Signed-off-by: Brian Paul <brianp@vmware.com>
commit
938fd71f4c4742f274922d53492a7290ab8d9c9b
Author: nobled <nobled@dreamwidth.org>
Date: Thu Apr 21 07:53:57 2011 -0600
mesa: add context fields for GL_ARB_robustness
Signed-off-by: Brian Paul <brianp@vmware.com>
commit
72075137bc79e65be03dac7e97b6dba93c3a86a4
Author: nobled <nobled@dreamwidth.org>
Date: Thu Apr 21 07:53:57 2011 -0600
mesa: standardize more bounds-checking error messages
Signed-off-by: Brian Paul <brianp@vmware.com>
commit
32a3fc23746db49da903fbc08afa0135af3007d2
Author: nobled <nobled@dreamwidth.org>
Date: Thu Apr 21 07:53:57 2011 -0600
mesa: standardize some bounds-checking error messages
Signed-off-by: Brian Paul <brianp@vmware.com>
commit
cecbf1f4d164207de373dec0cadee2e84e1f9656
Author: nobled <nobled@dreamwidth.org>
Date: Thu Apr 21 07:53:57 2011 -0600
mesa: add more bounds-checking support for client memory buffers
Signed-off-by: Brian Paul <brianp@vmware.com>
commit
edc895b52383d5bd274422db56adead1d81daf5f
Author: nobled <nobled@dreamwidth.org>
Date: Thu Apr 21 07:53:57 2011 -0600
mesa: add bounds-checking support for client memory buffers
Signed-off-by: Brian Paul <brianp@vmware.com>
commit
3a96ef28a538f158a219b406cd090dee70470c85
Author: nobled <nobled@dreamwidth.org>
Date: Thu Apr 21 07:53:57 2011 -0600
mesa: use is_bufferobj() helper function
Signed-off-by: Brian Paul <brianp@vmware.com>
Eric Anholt [Mon, 28 Mar 2011 23:54:39 +0000 (16:54 -0700)]
i965/fs: Add support for compute-to-mrf in 16-wide mode.
This is more painful than instruction scheduling, as we have to
compare two MRF writes to see if they coincide, and have to handle
partial GRF writes before that (for example, the result of a math
instruction written to color).
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Eric Anholt [Mon, 28 Mar 2011 23:55:14 +0000 (16:55 -0700)]
i965/fs: Typo fix a comment.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Eric Anholt [Fri, 25 Mar 2011 18:19:53 +0000 (11:19 -0700)]
i965/fs: Enable constant propagation in 16-wide.
All that needed fixing was skipping the newly-possible
uncompressed/sechalf partial GRF constant writes.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Eric Anholt [Wed, 23 Mar 2011 20:53:26 +0000 (13:53 -0700)]
i965/fs: Fix and enable the instruction scheduler for 16-wide.
Most of the work of the scheduler is agnostic to wide dispatch. It
operates on our virtual GRF file, which means instructions are
generally referring to 8 or 16 wide naturally. For the MRF file
management we're trying to track the actual hardware MRF file, so we
need to watch if an instruction writes multiple MRFs.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Eric Anholt [Wed, 23 Mar 2011 19:50:53 +0000 (12:50 -0700)]
i965/fs: Add support for 16-wide dispatch with uniforms in use.
This is glued in in a bit of an ugly way -- we rely on the uniforms
having been set up by 8-wide dispatch, and we just reuse them without
the ability to add new uniforms for any reason, since the 8-wide
compile is already completed. Today, this all works out because our
optimization passes are effectively the same for both and even if they
weren't, we don't reduce the set of uniforms pushed after
optimization.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Eric Anholt [Wed, 23 Mar 2011 19:05:56 +0000 (12:05 -0700)]
hash_table: Add an iterator for doing things like cleanup of the HT.
Without this, consumers often have to keep linked lists of the
entries, at additional malloc cost.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Eric Anholt [Wed, 23 Mar 2011 19:36:52 +0000 (12:36 -0700)]
i965/fs: Add a little whitespace between shader dumping debug.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Eric Anholt [Mon, 28 Mar 2011 18:29:55 +0000 (11:29 -0700)]
i965/fs: Add support for compr4 MRF writes.
These reduce an emitted (not decoded) instruction per shader on
g4x/gen5, but may allow for additional register coalescing as well.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Eric Anholt [Mon, 14 Mar 2011 17:29:12 +0000 (10:29 -0700)]
i965/fs: Add support for 16-wide dispatch on gen5.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Eric Anholt [Sat, 12 Mar 2011 03:19:01 +0000 (19:19 -0800)]
i965/fs: Add initial support for 16-wide dispatch on gen6.
At this point it doesn't do uniforms, which have to be laid out the
same between 8 and 16. Other than that, it supports everything but
flow control, which was the thing that forced us to choose 8-wide for
general GLSL support.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Eric Anholt [Thu, 24 Mar 2011 16:15:55 +0000 (09:15 -0700)]
i965/fs: Add support for discard instructions in 16-wide mode.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Eric Anholt [Tue, 29 Mar 2011 01:03:37 +0000 (18:03 -0700)]
i965/fs: Add support for math instructions in 16-wide mode.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Eric Anholt [Thu, 24 Mar 2011 16:08:58 +0000 (09:08 -0700)]
i965/fs: Fix interference calculation of pixel_[xy] in 16-wide.
Fixes glsl-fs-ceil in that mode, which produced the code in the comment.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Eric Anholt [Wed, 23 Mar 2011 21:00:01 +0000 (14:00 -0700)]
i965/fs: Disable some optimization passes under 16-wide for now.
These are fixable for 16, but that can wait until after it's basically
working.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Eric Anholt [Sun, 13 Mar 2011 20:27:59 +0000 (13:27 -0700)]
i965/fs: Add support for 16-wide texturing on gen5+.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Eric Anholt [Mon, 21 Mar 2011 22:54:18 +0000 (15:54 -0700)]
i965/fs: Add support for computing pixel_[xy] in 16-wide.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Eric Anholt [Sun, 13 Mar 2011 20:26:09 +0000 (13:26 -0700)]
i965/fs: Add support for 16-wide dispatch to the register allocator.
Note that the virtual grfs are in increments of the dispatch_width,
not hardware registers -- this makes the 16-wide emit and 8-wide emit
mostly the same.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Eric Anholt [Mon, 14 Mar 2011 17:51:19 +0000 (10:51 -0700)]
i965: Move the destination reg setup for 8/16 wide to the emit code.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Kristian Høgsberg [Mon, 25 Apr 2011 13:53:22 +0000 (09:53 -0400)]
intel: Use tiling for dri2AllocateBuffer implementation
Kristian Høgsberg [Mon, 25 Apr 2011 13:49:09 +0000 (09:49 -0400)]
intel: Set gen in intelInitScreen, just copy value in intelInitContext
Kristian Høgsberg [Sat, 23 Apr 2011 01:05:40 +0000 (21:05 -0400)]
intel: Use X tiling for DRM EGL Images
Ian Romanick [Fri, 22 Apr 2011 00:36:17 +0000 (17:36 -0700)]
mesa: Remove SWcontext::_FogMode, use gl_context::gl_fog_attrib::Mode everywhere
Reviewed-by: Brian Paul <brianp@vmware.com>
Kristian Høgsberg [Mon, 25 Apr 2011 19:34:04 +0000 (15:34 -0400)]
egl: Use the right extension name in #ifdef
Kristian Høgsberg [Mon, 25 Apr 2011 19:06:41 +0000 (15:06 -0400)]
wayland: Fix prototypes for EGL_WL_bind_wayland_display extension
Nicolas Kaiser [Mon, 25 Apr 2011 16:37:08 +0000 (18:37 +0200)]
main: remove duplicated includes
Remove duplicated includes of guarded headers.
Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
Signed-off-by: Brian Paul <brianp@vmware.com>
Benjamin Franzke [Mon, 25 Apr 2011 16:16:35 +0000 (18:16 +0200)]
winsys/wayland: Fix typo in Makefile
Reported by dir1212 on irc.
Benjamin Franzke [Sat, 23 Apr 2011 12:21:05 +0000 (14:21 +0200)]
st/egl: Add wayland shm softpipe support
Benjamin Franzke [Sat, 23 Apr 2011 12:20:24 +0000 (14:20 +0200)]
st/egl: Generalize wayland backend a bit
Benjamin Franzke [Sat, 23 Apr 2011 12:14:29 +0000 (14:14 +0200)]
winsys: Add wayland shm sw winsys
Henri Verbeet [Mon, 25 Apr 2011 11:28:55 +0000 (13:28 +0200)]
r600g: Unify comment style somewhat.
Signed-off-by: Henri Verbeet <hverbeet@gmail.com>
Henri Verbeet [Mon, 25 Apr 2011 11:28:55 +0000 (13:28 +0200)]
r600g: Cleanup the big endian support a bit.
In particular, make sure the code is at least compiled on little endian
systems.
Signed-off-by: Henri Verbeet <hverbeet@gmail.com>
Henri Verbeet [Mon, 25 Apr 2011 11:28:55 +0000 (13:28 +0200)]
r600g: Use EG constants in EG r600_colorformat_endian_swap().
This would actually fail to compile when PIPE_ARCH_BIG_ENDIAN is defined.
Signed-off-by: Henri Verbeet <hverbeet@gmail.com>
Marek Olšák [Thu, 21 Apr 2011 11:33:29 +0000 (13:33 +0200)]
configure.ac: require LLVM for r300g on x86 and x86_64
Marek Olšák [Thu, 21 Apr 2011 11:27:55 +0000 (13:27 +0200)]
configure.ac: enable LLVM by default on x86 and x86_64
Marek Olšák [Fri, 22 Apr 2011 17:29:55 +0000 (19:29 +0200)]
mesa: add stricter checks for float formats in the texstore memcpy path
E.g. when the internal format was RGBA16F and the source was RG, it would use
memcpy.
Marek Olšák [Fri, 22 Apr 2011 16:22:51 +0000 (18:22 +0200)]
mesa: implement generate-mipmap fallback for RGB10_A2
I hit this when testing RV350, which lacks RGB10_A2 render target
support. It had been missed when implementing the format and probably
unused by anything else too.
Not applicable to 7.10.
Reviewed-by: Eric Anholt <eric@anholt.net>
Vinson Lee [Mon, 25 Apr 2011 06:58:39 +0000 (23:58 -0700)]
r600g: Remove r600_helper.c from SConscript.
This is a follow-up to commit
d737857ed2ff4313fd6046dcd80018c6308a53c5.
Marek Olšák [Sun, 24 Apr 2011 23:52:33 +0000 (01:52 +0200)]
docs: update GL3 status
Marek Olšák [Thu, 21 Apr 2011 15:24:50 +0000 (17:24 +0200)]
r600g: remove some pointless and unused functions
Marek Olšák [Thu, 21 Apr 2011 15:20:27 +0000 (17:20 +0200)]
r600g: do not reset device to 0 when doing unrelated operations
Seems to be a copy-paste bug.
Marek Olšák [Sun, 24 Apr 2011 12:00:55 +0000 (14:00 +0200)]
r600g: trivially implement LATC/3DC
Passes fbo-generatemipmap-formats.
Dave Airlie [Sun, 24 Apr 2011 23:05:08 +0000 (09:05 +1000)]
r600g: drop r600_helper.c no point in it
move the one function into state common
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Sun, 24 Apr 2011 20:55:09 +0000 (06:55 +1000)]
r600g: enable EXT_draw_buffers2
Doesn't cause any piglit regression and passes the fbo-draw-buffers-blend
test.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Sun, 24 Apr 2011 07:35:45 +0000 (17:35 +1000)]
st/mesa: fix regression since
a22aba4eae9b29db731487bce90e8292f7e82c72
"st/mesa: check image size before copy_image_data_to_texture()" caused
a regression in piglit fbo-generatemipmap-formats test on all gallium drivers.
Level 0 for NPOT textures will not match minified values, so don't do this
check for level 0.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Sun, 24 Apr 2011 10:20:55 +0000 (20:20 +1000)]
r600g: fix glean clipflat test.
the provoking vertex doesn't apply to quad/strip/polygon.
This fixes clipFlat on r600g.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Marek Olšák [Sun, 24 Apr 2011 09:59:16 +0000 (11:59 +0200)]
r300g: reorder caps
Marek Olšák [Sun, 24 Apr 2011 09:48:46 +0000 (11:48 +0200)]
r300g: fix exposing caps on r300-r400
Broken with
72239d16cd08113e994ea3508f91193c682b0930.
Bryan Cain [Sun, 24 Apr 2011 00:29:15 +0000 (19:29 -0500)]
glsl: Fix typos in comments.
Dave Airlie [Sun, 24 Apr 2011 01:04:46 +0000 (11:04 +1000)]
r600g: fix glsl-fs-abs-neg
the hw does neg after abs, so don't neg the source in the ABS instruction case.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Sat, 23 Apr 2011 22:33:19 +0000 (08:33 +1000)]
r600g: fix bank swizzle calcs for scalar only operations.
In the initial code if we had nothing in the vector slots r would
never get reset to 0, so we'd fail to compile shaders, after the previous
commit this would happen for the LIT tests. When I fixed that we did a lot
of unnecessary loops through all the vector states when we had no vector
slots filled. So this patch optimises thing for the scalar only state.
This fixes the 3 LIT piglit tests on r600g.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Sat, 23 Apr 2011 22:30:08 +0000 (08:30 +1000)]
r600g: PV/PS have cycle restrictions in scalar operations
In the R600 ISA document:
Section 4.7.5 Cycle restrictions for the ALU.trans states that
PV/PS have cycle restrictions wrt constants.
This is part of a fix for the LIT tests
Signed-off-by: Dave Airlie <airlied@redhat.com>
Eric Anholt [Thu, 21 Apr 2011 00:55:54 +0000 (17:55 -0700)]
mesa: Add some comments about FRAG_RESULT_COLOR vs FRAG_RESULT_DATAn.
This came from reading what swrast does, and 965 now behaves the same
and gallium appears to as well.
Eric Anholt [Wed, 20 Apr 2011 23:49:19 +0000 (16:49 -0700)]
mesa: Fix fragment.color (no index) writes with OPTION ARB_draw_buffers.
Fixes a bug in Trine where fragment.color would write
FRAG_RESULT_COLOR (which is interpreted by drivers as being the "write
this to all color buffers" option) instead of FRAG_RESULT_DATA0 (just
the first target).
Fixes piglit ATI_draw_buffers/arbfp-no-index.
Eric Anholt [Wed, 20 Apr 2011 23:10:14 +0000 (16:10 -0700)]
i965: Don't double-emit fragment.color writes for MRT with ARB_fp.
Eric Anholt [Wed, 20 Apr 2011 21:03:21 +0000 (14:03 -0700)]
i965: Fill in the remaining fields of gen5+ sampler default color.
Still doesn't fix texwrap.
Eric Anholt [Wed, 20 Apr 2011 20:59:13 +0000 (13:59 -0700)]
i965: Fix batch decode for the gen5+ sampler default color.
Eric Anholt [Wed, 20 Apr 2011 18:48:31 +0000 (11:48 -0700)]
intel: Add support for ARB_sampler_objects.
This extension support consists of replacing
"gl_texture_obj->Sampler." with "_mesa_get_samplerobj(ctx, unit)->".
One instance of referencing the texture's base sampler remains in the
initial miptree allocation, where I'm not sure we have a clear
association with any texture unit.
Tested with piglit ARB_sampler_objects/sampler-objects.
Reviewed-by: Brian Paul <brianp@vmware.com>
Eric Anholt [Wed, 20 Apr 2011 04:40:48 +0000 (21:40 -0700)]
i965: Add support for NV_conditional_render.
Since we lack hardware support for it, this is a simple matter of
checking _mesa_check_conditional_render at the entrypoints, and
suppressing it for the metaops where it doesn't apply.
Reviewed-by: Brian Paul <brianp@vmware.com>
Eric Anholt [Thu, 21 Apr 2011 23:52:21 +0000 (16:52 -0700)]
swrast: Disable glAccum drawing during conditional rendering.
Eric Anholt [Wed, 20 Apr 2011 07:21:01 +0000 (00:21 -0700)]
meta: Don't do conditional rendering on GenerateMipmaps and BlitFramebuffer.
The NV_conditional_render spec calls out specific operations that
conditional rendering applies to, which doesn't include these.
Fixes NV_conditional_render/generatemipmap on swrast.
Reviewed-by: Brian Paul <brianp@vmware.com>
Eric Anholt [Mon, 18 Apr 2011 21:28:14 +0000 (14:28 -0700)]
i965: Add support for ARB_texture_compression_rgtc.
Tested with rgtc-teximage-0[12].
EXT_texture_compression_rgtc/fbo-generatemipmap-formats fails in NPOT
just like S3TC does.
Reviewed-by: Brian Paul <brianp@vmware.com>
Eric Anholt [Wed, 20 Apr 2011 16:33:59 +0000 (09:33 -0700)]
mesa: Don't assert on the compressed convertformat for GenerateMipmaps.
This assertion doesn't make any sense to me -- the convertFormat is
already something valid (tested above), and the BaseFormat dictated by
convertFormat doesn't matter to the function about to be called (it's
the datatype/comps that were pulled out of convertFormat).
Fixes assertion failure in
GL_EXT_texture_compression_rgtc/fbo-generatemipmap-formats
(still has a rendering failure in NPOT like S3TC does).
Reviewed-by: Brian Paul <brianp@vmware.com>
Eric Anholt [Wed, 20 Apr 2011 16:31:34 +0000 (09:31 -0700)]
mesa: Choose RGTC formats for GL_COMPRESSED_RED, GL_COMPRESSED_RG.
We were falling through to the default R8 and RG88 formats instead of
compressing when possible. Noticed by swrast fbo-blending-formats
actually doing rendering.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
José Fonseca [Sat, 23 Apr 2011 19:25:11 +0000 (20:25 +0100)]
mesa/gdi: Address compiler warnings.
In particular the function prototypes.
José Fonseca [Sat, 23 Apr 2011 17:44:19 +0000 (18:44 +0100)]
mesa/gdi: Remove InitCritSections.cpp.
We are now able to declare global critical sections through other
mechanisms so this is useless code.
José Fonseca [Sat, 23 Apr 2011 12:21:41 +0000 (13:21 +0100)]
Drop fx windows driver.
Irrelevant now that glide driver was removed.
Jakob Bornecrantz [Sat, 23 Apr 2011 15:01:19 +0000 (17:01 +0200)]
glu: Fix linux build