| Age | Commit message (Collapse) | Author |
|---|
|
Most of the failure from using uninlined function calls ends up being
just bad rendering, but nested function calls in the VS currently hang
the GPU, so reject them and explain why.
|
|
We were doubling up the offsets for the mipmap levels for CPU access.
Instead of reimplementing i945_miptree_layout_2d with 6 cube images
separated by qpitch, share that function and provide the level offsets
later.
Fixes piglit cubemap and fbo-cubemap.
|
|
This should be functionally equivalent, with the possible exception of
NaN handling.
|
|
We could use this to reduce constant register pressure, but for now it
makes the resulting program assembly much more readable.
|
|
This struct is used to generate a hash, ignoring the entry boundaries.
Signed-off-by: Eric Anholt <eric@anholt.net>
|
|
Fixes assertion failure in fbo-generatemipmap-npot.
|
|
We would run into trouble due to the hardware using inclusive numbers
and the subtraction to handle that producing negative (meaning large
positive) coordinates.
Bug #27643.
|
|
Rename old IGDNG to Ironlake, and set 'gen' number for
Ironlake as 5, so tracking the features with generation num
instead of special is_ironlake flag.
Reviewed-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
|
|
|
|
|
|
Fixes piglit glsl-orangebook-ch06-bump, regressed with
4fc57322258a750c0a9cabc77372b5ccde1fa877
|
|
|
|
Rejecting all doesn't seem to be helping get the pipeline lit up.
|
|
|
|
|
|
It appears that the thing that was killing VS threads was the
gratuitous NOP that replaced the gratuitous jump from OPCODE_END to
the nearby OPCODE_END implementation. With that gone, we can move on
to the rest of the pipeline.
|
|
Just emit the URB write at END time. Subroutine code that sits after
OPCODE_END won't be executed since we've ended the thread at the point
that the URB write is done.
|
|
Otherwise, we may not get the FFTID set up which would break freeing
of resources.
|
|
The hope is to later take advantage of the reduced constant usage to
free up regs. This only covers the GLSL path at the moment, because
the brw_wm_emit path doesn't get the information as to whether a float
value is a constant or a uniform.
|
|
I keep finding the desire to force this path to debug it instead of
cooking up goofy-looking testcases to do so.
|
|
Tested with piglit glsl-fs-sqrt-branch, fp-cmp.vpfp.
|
|
The pitch is not really an inherent part of the miptree, since it's
not part of any of the layout calculations, and it's dictated by the
libdrm-allocated region pitch now.
|
|
MOV, MOV."
This reverts commit 8ef3b1834a896927bdd4f2aea552cdb732849da9. Fixes
piglit glsl-vs-if.
|
|
MOV, MOV."
This reverts commit 46450c1f3f93bf4dc96696fc7e0f0eb808d9c08a. I was
wrong about null reg behavior -- it reads undefined, not 0. And
they're not kidding.
|
|
|
|
|
|
|
|
|
|
This is recommended by the B-Spec. I wasn't able to measure any
difference in ETQW.
|
|
This was obvious when looking at the compiled output of ETQW's
shaders.
|
|
Saves an instruction over doing conditional moves.
|
|
Saves an instruction in PINTERP, LINTERP, and PIXEL_W from
brw_wm_glsl.c For non-GLSL it isn't used yet because the deltas have
to be laid out differently.
|
|
Fixes hangs with texturing in the non-GLSL path since
f6d210c284751ac50a8d6358de7e75a1ff1e4ac7
|
|
This would be triggered by use of sqrt() along with control flow.
Fixes piglit-fs-sqrt-branch and a bug in Yo Frankie!.
|
|
|
|
The bad response length would hang the GPU with a masked sample in a
shader using control flow. For 8-wide, the response length is always
4, and masked slots are just not written to. brw_wm_glsl.c already
allocates registers in the right locations.
Fixes piglit glsl-fs-bug25902 (fd.o bug #25902).
|
|
|
|
Corresponds to d225a25e21a24508aea3b877c78beb35502e942d and fixes
piglit glsl-fs-loop-nested. Bug #25173.
|
|
We were doing it ad-hoc before, as instructions with potential
aliasing problems were identified. But thanks to swizzling basically
anything can have aliasing, so just do it generally at source reg
setup time. This is somewhat inefficient, because sometimes an
operation doesn't need unaliasing protection if the swizzling is safe,
but the unaliasing before didn't cover those cases either.
Fixes piglit glsl-fs-loop.
|
|
|
|
We were patching up all the break and continues between the start of
our loop and the end of our loop, even if they were breaks/continues
for an inner loop. Avoiding patching already patched breaks/continues
fixes piglit glsl-vs-loop-nested.
|
|
|
|
The code was walking over the regs of pairs of attributes and checking
whether the attribute with a given reg index had point sprite enabled.
So the point sprite setup code was rarely even getting executed.
Instead, we need to determine which channels of a reg need point
sprite coordinate replacement. In addition, it was multiplying the
attribute by 1/w, when it's supposed to cover (0, 1) in each direction
regardless of w, and it wasn't filling in the Z and W components of
the texcoord as specified.
Fixes piglit point-sprite and the spriteblast demo. Bug #24431, #22245.
|
|
|
|
|
|
The whole structure is 836 bytes, but if only the first one or two
samplers are used (as is common), the part that matters is only 56 or
108 bytes. By using just that subset as the key (since the key size
is also part of the key), we improve firefox-talos-gfx performance by
1%.
|
|
|
|
|
|
Since it's a BO pointer, it already lives outside the key in
brw->wm.sdc_bo[] which is used for sampler state lookup and setup.
|
|
|
