Adel Gadllah [Mon, 24 Oct 2011 17:41:03 +0000 (19:41 +0200)]
nv50: fix max texture levels
MAX_TEXTURE_2D_LEVELS and MAX_TEXTURE_CUBE_LEVELS are supposed to be
14 not 13, while MAX_TEXTURE_3D_LEVELS should be 12 not 10.
Marek Olšák [Mon, 24 Oct 2011 17:44:49 +0000 (19:44 +0200)]
winsys/radeon: don't use the new GEM_WAIT ioctl for now
Tom Fogal [Wed, 19 Oct 2011 20:41:35 +0000 (14:41 -0600)]
Only use gcc visibility support with gcc4+.
I had a colleague hitting issues compiling with an old gcc3.2
system. These patches got them through.
NOTE: This is a candidate for the 7.11 branch.
Reviewed-by: Brian Paul <brianp@vmware.com>
Mathias Fröhlich [Sat, 8 Oct 2011 19:33:23 +0000 (21:33 +0200)]
winsys/radeon: restore the old r600g winsys memory characteristics.
Use VRAM for static and immutable buffers. This restores the
recently removed r600g winsys behaviour for memory locations.
This also improoves rendering times on the gpu for some
OpenSceneGraph based test cases by about 15%.
Signed-off-by: Marek Olšák <maraeo@gmail.com>
Brian Paul [Sun, 23 Oct 2011 16:14:27 +0000 (10:14 -0600)]
util: remove gotos in u_upload_mgr.c
We can trivially remove the gotos in two places in this code and make
it a bit more readable.
Reviewed-by: José Fonseca <jfonseca@vmware.com>
Mathias Fröhlich [Sat, 22 Oct 2011 10:57:01 +0000 (12:57 +0200)]
mesa: Avoid ABA problem on buffer object bind.
Make sure we do not run into the classic ABA problem on buffer object bind,
reusing this name and may be never rebind since we get an new name
that was just deleted and never rebound in between.
The explicit rebinding to the debault object in the current context
prevents the above in the current context, but another context
sharing the same objects might suffer from this problem.
Minor var renaming and comments edited by Brian.
Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Signed-off-by: Brian Paul <brianp@vmware.com>
Brian Paul [Mon, 24 Oct 2011 02:21:18 +0000 (20:21 -0600)]
swrast: fix comment typo (s/texure/texture/)
Brian Paul [Mon, 24 Oct 2011 02:17:53 +0000 (20:17 -0600)]
mesa: remove dead code from teximage.c
Mathias Fröhlich [Wed, 19 Oct 2011 05:54:20 +0000 (07:54 +0200)]
mesa: Fix multithreaded buffer object refcounting.
Buffer objects may be shared across contexts.
Rework the array attrib push/pop implementation
to be thread safe. Make use of more library functions
for this purpose.
Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Sun, 23 Oct 2011 18:29:06 +0000 (20:29 +0200)]
r300g: don't call u_trim_pipe_prim in r300_swtcl_draw_vbo
This was dead code anyway.
Marek Olšák [Fri, 21 Oct 2011 01:32:29 +0000 (03:32 +0200)]
u_blitter: accept overriden width0 and height0
We'll use this soon.
Marek Olšák [Fri, 21 Oct 2011 00:09:40 +0000 (02:09 +0200)]
u_blitter: add a copy_texture function which uses views and not resources
The views (sampler views and surfaces) are great tools for changing resource
properties without having to change pipe_resource.
Brian Paul [Sun, 23 Oct 2011 17:19:46 +0000 (11:19 -0600)]
intel: remove duplicated #include of texstore.h
Brian Paul [Sun, 23 Oct 2011 17:18:22 +0000 (11:18 -0600)]
radeon: remove unnecessary #includes of texstore.h
Brian Paul [Sun, 23 Oct 2011 16:39:38 +0000 (10:39 -0600)]
configs: add vmgfx-related dirs to configs/linux-dri
Brian Paul [Sun, 23 Oct 2011 16:44:47 +0000 (10:44 -0600)]
swrast: update renderbuffer format assertion
Failed when exercising i965 swrast fallback rendering.
Brian Paul [Sun, 23 Oct 2011 16:44:47 +0000 (10:44 -0600)]
mesa: add swrast_texture_image::Buffer
In the past, swrast_texture_image::Data has been overloaded. It could
either point to malloc'd memory storing texture data, or it could point
to a current mapping of GPU memory.
Now, Buffer always points to malloc'd memory (if we're not using GPU
memory) and Data always points to mapped memory. The next step would
be to rename Data -> Map.
This change also involves adding swrast functions for mapping textures
and renderbuffers prior to rendering to setup the Data pointer. Plus,
corresponding functions to unmap texures and renderbuffers. This is
very much like similar code in the dri drivers.
Brian Paul [Sun, 23 Oct 2011 16:44:47 +0000 (10:44 -0600)]
mesa: remove _mesa_alloc_texmemory(), _mesa_free_texmemory()
Core Mesa no longer does any texture memory allocation.
Brian Paul [Sun, 23 Oct 2011 16:44:47 +0000 (10:44 -0600)]
mesa: move gl_texture_image::Data, RowStride, ImageOffsets to swrast
Only swrast and the drivers that fall back to swrast need these fields now.
This removes the last of the fields related to software rendering from
gl_texture_image.
Brian Paul [Sun, 23 Oct 2011 16:08:06 +0000 (10:08 -0600)]
llvmpipe: point out that there's two stencil writemasks
In lp_build_stencil_op() the incoming 'stencil' var is a 2-element array.
There's a front-face writemask and a back-face writemask but we're ignoring
the later. This patch doesn't fix anything but at least points out the
problem.
Brian Paul [Sun, 23 Oct 2011 16:05:14 +0000 (10:05 -0600)]
llvmpipe: compare front_facing to NULL to improve readability
Compare 'front_facing' to NULL to make it more obvious that front_facing
is a pointer and not a simple boolean value.
Brian Paul [Sun, 23 Oct 2011 16:02:53 +0000 (10:02 -0600)]
gallivm: added lp_build_print_ivec4() function
Brian Paul [Sun, 23 Oct 2011 16:00:00 +0000 (10:00 -0600)]
mesa: improve the warning message in _mesa_choose_tex_format()
Bug 42128 hits this _mesa_warning() call.
Eric Anholt [Tue, 27 Sep 2011 21:54:10 +0000 (14:54 -0700)]
glsl: Add support for constant expression evaluation on round(), roundEven().
v2: Avoid the C99 rounding functions, because I don't trust
get/setting the C99 rounding mode from inside our library not having
other side effects. Instead, open-code roundEven() behavior around
Mesa's IROUND, which we're already testing for C99 rounding mode
safety.
Fixes glsl-1.30/compiler/built-in-functions/round*
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Mathias Fröhlich [Fri, 14 Oct 2011 14:25:21 +0000 (16:25 +0200)]
dri-r600: Hook up a drm_descriptor configuration function
Returns a configuration that makes the dri state-tracker-manager
throttle.
Mathias Fröhlich [Sun, 23 Oct 2011 06:57:43 +0000 (08:57 +0200)]
r600g: make r[67]00 not bail out on PRED_SETNE_INT.
Dave Airlie [Sat, 22 Oct 2011 19:13:20 +0000 (20:13 +0100)]
r600g: make if's use PRED_SETNE_INT no matter what.
This is more correct for TGSI if, and with native ints enabled
it fixes 25 piglit fails.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Sat, 22 Oct 2011 19:11:26 +0000 (20:11 +0100)]
r600g: print inst in hex in dumps
since we have them as hex in the headers.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Sat, 22 Oct 2011 19:08:44 +0000 (20:08 +0100)]
r600g: drop specific i2f it should be a trans only op2.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Mathias Froehlich [Wed, 3 Aug 2011 06:10:52 +0000 (08:10 +0200)]
vbo: Clean up unused variables in the vbo module.
Remove some unused or unused but set variables
from the vbo module.
Alan Coopersmith [Fri, 21 Oct 2011 22:43:28 +0000 (15:43 -0700)]
Convert additional GNUC_MINOR checks to multiplied version
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Alan Coopersmith [Fri, 21 Oct 2011 00:14:05 +0000 (17:14 -0700)]
Fix gcc version checks for _mesa_bitcount
- Fix _GNUC__ typo in both checks
- Fix logic error in check for gcc < 3.4 that breaks for gcc 2.x & older
Without this fix, builds with gcc 3.4.x end up depending on undefined
_mesa_bitcount instead of gcc's __builtin_popcount.
NOTE: This is a candidate for the stable branches.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Maarten Lankhorst [Tue, 18 Oct 2011 12:24:30 +0000 (14:24 +0200)]
gallium/targets: Add vdpau target for nouveau
Should fall back to shader based decoding (g3dvl) for now.
This is probably broken on systems that support xvmc, because
nouveau_video_buffer_create has no way to know for what api
the buffer is created, so I think this call might need a
separate argument as workaround.
Signed-off-by: Maarten Lankhorst <m.b.lankhorst@gmail.com>
Christoph Bumiller [Thu, 20 Oct 2011 21:36:15 +0000 (23:36 +0200)]
d3d1x: set primitive restart in the correct draw calls
Christoph Bumiller [Tue, 18 Oct 2011 16:36:57 +0000 (18:36 +0200)]
nv50/ir: import SM4 converter
Christoph Bumiller [Thu, 20 Oct 2011 20:42:59 +0000 (22:42 +0200)]
nouveau,nvc0: fix/improve handling of multiple constant buffers
Christoph Bumiller [Tue, 18 Oct 2011 10:08:19 +0000 (12:08 +0200)]
nvc0: add support for linear and buffer textures and RTs
Christoph Bumiller [Tue, 18 Oct 2011 10:07:01 +0000 (12:07 +0200)]
nvc0: add support for clip distance shader outputs
Christoph Bumiller [Fri, 21 Oct 2011 20:58:18 +0000 (22:58 +0200)]
nvc0: handle more query types
Christoph Bumiller [Sat, 15 Oct 2011 09:49:55 +0000 (11:49 +0200)]
nvc0: fix location of the PrimitiveID output
Christoph Bumiller [Mon, 17 Oct 2011 21:26:57 +0000 (23:26 +0200)]
nvc0: prevent VERTEXID/INSTANCEID from consuming input slots
Christoph Bumiller [Mon, 17 Oct 2011 21:30:02 +0000 (23:30 +0200)]
nvc0: fixes for program tessellation parameters
Christoph Bumiller [Sat, 15 Oct 2011 10:03:03 +0000 (12:03 +0200)]
nv50,nvc0: reset per-instance state for inactive vertex elements
Christoph Bumiller [Fri, 14 Oct 2011 20:22:04 +0000 (22:22 +0200)]
nv50,nvc0: reset base element in draw_arrays
It affects VERTEX_BUFFER_FIRST,COUNT submission, too.
Christoph Bumiller [Tue, 11 Oct 2011 09:41:47 +0000 (11:41 +0200)]
nvc0: apply first_layer offset to all resources with array_size
Makes CUBE arrays work with d3d1x.
Christoph Bumiller [Fri, 14 Oct 2011 20:19:01 +0000 (22:19 +0200)]
nvc0: emit state to allow GP to select the RT layer
Christoph Bumiller [Fri, 14 Oct 2011 20:16:48 +0000 (22:16 +0200)]
nvc0: validate GP samplers, textures
Christoph Bumiller [Fri, 14 Oct 2011 20:12:42 +0000 (22:12 +0200)]
nvc0: fix clear_render_target/depth_stencil region
In all 3 dimensions (now clearing all layers too).
Christoph Bumiller [Sun, 25 Sep 2011 16:00:07 +0000 (18:00 +0200)]
nvc0: fix assertion that immediate array buffer fits
Christoph Bumiller [Mon, 17 Oct 2011 21:04:11 +0000 (23:04 +0200)]
nv50/ir: use RDSV to fetch FrontFacing before lowering
Christoph Bumiller [Tue, 11 Oct 2011 15:58:14 +0000 (17:58 +0200)]
nv50/ir: fix textureGrad with offsets and in non-FPs
Christoph Bumiller [Mon, 17 Oct 2011 21:02:16 +0000 (23:02 +0200)]
nv50/ir: add wrap mode for shift operations
D3D1x specifies that only the low 5 bit of the shift are used.
Christoph Bumiller [Thu, 6 Oct 2011 18:45:08 +0000 (20:45 +0200)]
nv50/ir: initialize RelocInfo to 0
Christoph Bumiller [Thu, 6 Oct 2011 12:32:58 +0000 (14:32 +0200)]
nvc0/ir: fix emission of cvt when register and type size differ
Christoph Bumiller [Fri, 14 Oct 2011 17:58:04 +0000 (19:58 +0200)]
nv50/ir: fix argument count for CUBE_ARRAY texture target
Christoph Bumiller [Mon, 17 Oct 2011 21:00:59 +0000 (23:00 +0200)]
nvc0/ir: GP emit address must end up in $r0
Christoph Bumiller [Fri, 14 Oct 2011 17:56:33 +0000 (19:56 +0200)]
nvc0/ir: TXQ requires different lowering from normal TEX
Christoph Bumiller [Fri, 14 Oct 2011 17:54:34 +0000 (19:54 +0200)]
nv50/ir: initialize default prog_info values for GP,TP
Christoph Bumiller [Fri, 14 Oct 2011 17:49:22 +0000 (19:49 +0200)]
nv50/ir: fix memory value equality check
Christoph Bumiller [Fri, 14 Oct 2011 17:47:45 +0000 (19:47 +0200)]
nv50/ir: fix leak in removal of graph root
Christoph Bumiller [Thu, 20 Oct 2011 18:43:11 +0000 (20:43 +0200)]
d3d1x: fix/improve OMSetRenderTargets
Don't count trailing NULL RTVs.
Don't skip update if only DSV has changed.
Christoph Bumiller [Thu, 13 Oct 2011 20:36:27 +0000 (22:36 +0200)]
d3d1x: avoid translating invalid blend, depth or stencil state
Christoph Bumiller [Thu, 13 Oct 2011 20:29:12 +0000 (22:29 +0200)]
d3d1x: add support for buffer views
Christoph Bumiller [Thu, 13 Oct 2011 20:27:40 +0000 (22:27 +0200)]
d3d1x: improve CreateInputLayout
Christoph Bumiller [Wed, 5 Oct 2011 14:56:24 +0000 (16:56 +0200)]
d3d1x/context: fix IASetVertexBuffers stride comparison
Christoph Bumiller [Thu, 13 Oct 2011 19:27:09 +0000 (21:27 +0200)]
d3d1x: use ZS formats for TYPELESS resources with DS binding
Christoph Bumiller [Mon, 17 Oct 2011 18:50:23 +0000 (20:50 +0200)]
d3d1x: use resource format if view format is UNKNOWN
Christoph Bumiller [Sun, 25 Sep 2011 17:03:44 +0000 (19:03 +0200)]
d3d1x: fix shadow comparison to gallium enum off by one
Christoph Bumiller [Mon, 17 Oct 2011 18:49:56 +0000 (20:49 +0200)]
d3d1x: improve CheckFormatSupport
Christoph Bumiller [Thu, 13 Oct 2011 17:31:51 +0000 (19:31 +0200)]
d3d1x: update format mapping table
Use the proper integer formats.
Fix mapping of some PIPE depth formats to DXGI.
Christoph Bumiller [Thu, 20 Oct 2011 18:12:51 +0000 (20:12 +0200)]
d3d1x: fix translation of subresource to layer
Christoph Bumiller [Thu, 13 Oct 2011 12:25:44 +0000 (14:25 +0200)]
d3d1x: fix initial data upload for 3D textures
Christoph Bumiller [Thu, 13 Oct 2011 12:22:37 +0000 (14:22 +0200)]
d3d1x: MipLevels -1 in resource view desc means all
Christoph Bumiller [Tue, 11 Oct 2011 12:22:32 +0000 (14:22 +0200)]
d3d1x/dxgi: fix initialization of pipe_box for Present copy
Christoph Bumiller [Thu, 13 Oct 2011 12:40:44 +0000 (14:40 +0200)]
d3d1x: implement array textures
Christoph Bumiller [Thu, 13 Oct 2011 11:53:39 +0000 (13:53 +0200)]
nvc0/ir: handle levelZero modifier in TEX emission
Christoph Bumiller [Fri, 23 Sep 2011 16:39:48 +0000 (18:39 +0200)]
nvc0/ir: fix lowering of DIV F32
Christoph Bumiller [Wed, 12 Oct 2011 19:08:40 +0000 (21:08 +0200)]
d3d1x: fix refcounting of GalliumD3D11DeviceChild objects
An external Release would have lowered the device reference count,
but an internal reference in the context does not raise it (by
design).
Christoph Bumiller [Wed, 12 Oct 2011 18:44:53 +0000 (20:44 +0200)]
nvc0/ir: fix xy confusion typo in readTessCoord
Christoph Bumiller [Wed, 12 Oct 2011 17:37:02 +0000 (19:37 +0200)]
d3d1x/sm4: save CUSTOMDATA
This is the data for the immediate constant buffer.
Christoph Bumiller [Wed, 12 Oct 2011 16:04:50 +0000 (18:04 +0200)]
d3d1x: add FORK/JOIN phase opcodes to declarations
The phase instance counts are not necessarily redeclared so with
the separation of declarations and instructions we wouldn't know
which instance count applies to which phase.
Christoph Bumiller [Thu, 20 Oct 2011 16:37:07 +0000 (18:37 +0200)]
d3d1x: add shader signature to sm4_program
Correct linkage requires examining the signature itself, it cannot
be reconstructed from declarations only since unused registers may
have been omitted from them.
Christoph Bumiller [Wed, 12 Oct 2011 15:35:30 +0000 (17:35 +0200)]
d3d1x: attempt to detect D3D10 feature level
Virtually all applications refuse to use the D3D10 backend if this
level is not supported, which makes testing annoying.
Christoph Bumiller [Thu, 20 Oct 2011 16:32:23 +0000 (18:32 +0200)]
d3d1x: switch to TGSI SAMPLE opcodes
We don't want to clutter the code or handicap new hardware for
the sake of ancient GPUs on which d3d1x won't ever be used,
much less be fully compliant, anyway.
Brian Paul [Fri, 21 Oct 2011 16:09:48 +0000 (10:09 -0600)]
st/mesa: fix a bug in and re-org setup_interleaved_attribs()
We were mis-computing the size of the user-space vertex buffer in
some circumstances. This led to a failed assertion at u_inlines.h:222
when using the VMware svga driver.
For example, if we had arrays such as:
array[0]: element_offset = 12, stride = 24
array[1]: element_offset = 0, stride = 24
We'd mistakenly compute 'bytes' to be 12 bytes too small.
I've reorganized the function too. By time it's called, we know that
we've got interleaved arrays either all in one VBO or all in user memory
and the stride is equal for all arrays.
Move the code that lived inside the attr==0 test after the loop.
In the loop we compute the true vertex size. That size factors into the
pipe->redefine_user_buffer() call later. Using the vertex size instead
of array[0]'s element_offset fixes the failed assertion.
Reviewed-by: José Fonseca <jfonseca@vmware.com>
Kenneth Graunke [Tue, 18 Oct 2011 22:04:37 +0000 (15:04 -0700)]
glsl: Short-circuit lower_if_to_cond_assign when MaxIfDepth is UINT_MAX.
Setting MaxIfDepth to UINT_MAX effectively means "don't lower anything."
Explicitly checking for this common case allows us to avoid walking the
IR, computing nesting levels, and so on.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Bryan Cain <bryancain3@gmail.com>
Kenneth Graunke [Tue, 20 Sep 2011 06:47:56 +0000 (23:47 -0700)]
i965: Set MaxIfDepth to UINT_MAX on Gen6+ and 16 on prior generations.
Commit
488fe51cf823ccd137c667f1e92dd86f8323b723 converted the EmitNoIfs
flag to MaxIfDepth, an unsigned integer saying "flatten if-statements
nested beyond this depth."
Unfortunately, i965 left this initialized to 0, which made ir_to_mesa
attempt to flatten all if-statements. We didn't notice right away
because we usually throw away ir_to_mesa's code in favor of the native
VS and FS backends...but this still creates a lot of unnecessary work.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Jeremy Huddleston [Fri, 21 Oct 2011 07:22:40 +0000 (00:22 -0700)]
apple: Use the correct (OpenGL.framework) glViewport and glScissor during init
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
Jeremy Huddleston [Fri, 21 Oct 2011 05:54:08 +0000 (22:54 -0700)]
apple: Silence some debug spew
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
Marek Olšák [Fri, 21 Oct 2011 00:00:40 +0000 (02:00 +0200)]
r600g: cleanup r600_reset_blittable_to_compressed
Vinson Lee [Thu, 20 Oct 2011 23:52:53 +0000 (16:52 -0700)]
st/mesa: Initialize variable.
ptr is uninitialized if ib is NULL.
Fixes Coverity uninitialized pointer read defect.
Reviewed-by: Jakob Bornecrantz <jakob@vmware.com>
Alan Coopersmith [Tue, 18 Oct 2011 21:04:15 +0000 (14:04 -0700)]
Add solaris detection for PIPE_ARCH_LITTLE_ENDIAN/PIPE_ARCH_BIG_ENDIAN
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Kenneth Graunke [Wed, 19 Oct 2011 00:05:30 +0000 (17:05 -0700)]
i965: Remove copy and pasted gen7_wm_constants state atom.
Now that this is identical to gen6_wm_constants, just use that instead.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Kenneth Graunke [Wed, 19 Oct 2011 00:05:29 +0000 (17:05 -0700)]
i965: Use AUB_TRACE_WM_CONSTANTS in gen7_prepare_wm_push_constants.
This makes it match gen6_prepare_wm_push_constants. For some reason, it
had been using AUB_TRACE_NO_TYPE.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Kenneth Graunke [Wed, 19 Oct 2011 00:05:28 +0000 (17:05 -0700)]
i965: Fix incorrect dirty bit in gen6_prepare_wm_push_constants.
We definitely want CACHE_NEW_WM_PROG, not CACHE_NEW_VS_PROG.
NOTE: This is a candidate for the 7.11 branch.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Marek Olšák [Thu, 20 Oct 2011 21:14:36 +0000 (23:14 +0200)]
r300g: don't return NULL in resource_from_handle if the resource is too small
The DDX may allocate a buffer with a too small size.
Instead of failing, let's pretend everything's alright.
Such bugs should be fixed in the DDX, of course.
NOTE: This is a candidate for the stable branches.
Brian Paul [Thu, 20 Oct 2011 21:13:17 +0000 (15:13 -0600)]
docs: document known issues with Viewperf 11
Eric Anholt [Mon, 3 Oct 2011 22:31:52 +0000 (15:31 -0700)]
i965/vs: Fix comparisons with uint negation.
The condmod instruction ends up generating garbage condition codes,
because apparently the comparison happens on the accumulator value (33
bits for UD), not the truncated value that would be written.
Fixes vs-op-neg-*
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Eric Anholt [Mon, 3 Oct 2011 22:12:10 +0000 (15:12 -0700)]
i965/fs: Fix comparisions with uint negation.
The condmod instruction ends up generating garbage condition codes,
because apparently the comparison happens on the accumulator value (33
bits for UD), not the truncated value that would be written.
Fixes fs-op-neg-*
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Christoph Bumiller [Sat, 15 Oct 2011 14:44:04 +0000 (16:44 +0200)]
d3d1x: make use of new query types