- 07 Dec, 2021 40 commits
-
-
Sometimes as a developer when a new version of libdrm is required, it's nice to not have to install it somewhere and set up pkg-confing, and just let Meson do it for us.
-
Signed-off-by:
Samuel Pitoiset <samuel.pitoiset@gmail.com> Reviewed-by:
Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl> Part-of: <mesa/mesa!13857>
-
Signed-off-by:
-
Signed-off-by:
-
Inferring from blob's cmdstream the size of shader instruction cache for: - a630 is 64 - a650 is 128 - a660 is 128 On a650 and a660 gpu could hang if we exceed the limit. Though it is not reproducible with computerator or a single amber test. Also while blob limits the size to 128 - Turnip still hangs with it but does not hang with the limit of 127. On a630 there seem to be no hang when limit is exceeded. Fixes the hang of compute shader in Alien Isolation on a650/a660. Signed-off-by:
Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <mesa/mesa!14044> -
This moves the memory and semaphore objects to direct calls Reviewed-by:
Marek Olšák <marek.olsak@amd.com> Part-of: <!14073>
-
Reviewed-by:
-
Reviewed-by:
-
Reviewed-by:
-
These were all interrelated, avoid the indirect calls here, and call directly between main and state tracker Reviewed-by:
-
Reviewed-by:
-
If I start direct linking the state tracker to mesa, the tests fail to build because they don't have gallium linked. Reviewed-by:
Dylan Baker <dylan.c.baker@intel.com> Part-of: <!14073>
-
Only the nir options are used now, since i965 was dropped, the glsl options come from the state tracker Reviewed-by:
Caio Oliveira <caio.oliveira@intel.com> Part-of: <mesa/mesa!14102>
-
Cuts 28% of the remaining relocations in libvulkan_intel.so, shrinks binary size by 290kb. Reviewed-by:
Adam Jackson <ajax@redhat.com> Part-of: <mesa/mesa!13987>
-
Moves it to .rodata instead of .data. Reviewed-by:
-
This gets our union's size down to 22 bytes (now smaller than any of the union's types were before we made the union!). Cuts another 48kb off of the drivers. Reviewed-by:
-
This helps concentrate the dirty pages from the relocations, reduces how many relocations there are, and reduces the size of each variable assuming variables mostly don't have conditions or the conditions are mostly reused). Reduces libvulkan_intel.so size by 49kb. Reviewed-by:
-
This helps concentrate the dirty pages from the relocations, reduces how many relocations there are, and reduces the size of each expression (assuming expressions mostly don't have conditions or the conditions are mostly reused). Reduces libvulkan_intel.so size by 8.7kb. Reviewed-by:
-
You can't have an array of them after removing many-comm-expr, there's no space in the struct. Reviewed-by:
-
Even with packing all 3 types into a 40-byte union (nir_search_constant being 24 bytes and nir_search_expression having formerly been 32), and having a single array of them, this cuts 1.7MB from each of libvulkan_intel.so and libgallium_dri.so. Reviewed-by:
-
The cache is always set. Reviewed-by:
-
I'm going to be adding some more tables to reduce relocations in the generated code, so move the current tables to a struct for arg-passing sanity. Reviewed-by:
-
Fixes the dEQP regressions in dEQP-GLES2.functional.shaders.indexing.*. TGSI used the address reg for these offsets too. Reviewed-by:
Karol Herbst <kherbst@redhat.com>
Part-of: <mesa/mesa!14083> -
Now that the classic drivers that were mixing the use of these asm and glsl shader fields are gone we can finally use a union here. This basically reverts commit 9d99dc4b but also moves a read of IsPositionInvariant inside an arb asm only code block for safety. Reviewed-by:
-
Double buffered pbuffer need to update the front buffer, otherwise we always get wrong value when glReadPixels(). Reviewed-by:
Qiang Yu <yuq825@gmail.com> Part-of: <mesa/mesa!13750>
-
We are sure to have a back buffer in swap buffer now. Reviewed-by:
-
Sometimes this is the real front buffer depends on the place called. Since it's the same LOADER_DRI3_FRONT_ID slot, just name it dri3_front_buffer. Reviewed-by:
-
Reviewed-by:
-
If we already know the drawable type, setup event in a simpler way. Reviewed-by:
-
For simplicity and latter commits. Reviewed-by:
-
The query result is not used anywhere. Reviewed-by:
-
Drawable type include more information which can be used to distinguish pixmap and pbuffer which both treated as pixmap before. Reviewed-by:
-
For distinguish different behavior of pixmap and pbuffer in latter commits. Reviewed-by:
-
If glxDrawable is not a X window ID, we can only support GLXPbuffer now. Reviewed-by:
-
XServer already created a pixmap with same id as pbuffer, so that other client can use the pbuffer id to do glXMakeCurrent(). But with a hidden pixmap, we can't do this. Reviewed-by:
-
The game makes use of builtin functions that were moved to compatibility shaders in GLSL 4.20 in its GLSL 4.20 shaders without declaring the compatibility token. Closes: mesa/mesa#5706 Reviewed-by:
Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> Part-of: <mesa/mesa!14040>
-
This allows us to force all shaders to offer shader features only provided to compatibility shaders. Reviewed-by:
-
Kenneth Graunke authoredOn XeHP, the XY_BLOCK_COPY_BLT command has a number of fields that describe the layout of the surface, much like SURFACE_STATE does. Several of them are encoded in such a similar manner that we really would like to reuse the isl helpers for emitting those. This commit moves them into a new isl_genX_helpers.h file which I can include from the BLORP code. (The alternative would be to add XY_BLOCK_COPY_BLT filling commands to isl, but that...seems more like a BLORP feature.) Reviewed-by:
Jason Ekstrand <jason@jlekstrand.net>
Part-of: <!14094> -
Kenneth Graunke authored
When targeting the blitter or compute engines, the destination is not really a render target. But it's still useful to know whether we're talking about the source or destination. Reviewed-by:
Jason Ekstrand <jason@jlekstrand.net>
Part-of: <mesa/mesa!14094> -
Cc: mesa-stable Reviewed-by:
-
