| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
| |
It selects v_med3_f32, which has the same rate & size.
Reviewed-by: Dave Airlie <[email protected]>
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
|
| |
Reviewed-by: Dave Airlie <[email protected]>
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
|
|
|
| |
v2: do it correctly
Reviewed-by: Dave Airlie <[email protected]>
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
|
| |
Reviewed-by: Dave Airlie <[email protected]>
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
| |
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
| |
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
|
|
| |
All call sites set bind = 0. The next commit will use this.
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
|
|
|
|
| |
This lowers lgkm wait cycles by 30% on VI and normal conditions.
The might be a measurable improvement when CE is disabled (radeon)
or under L2 thrashing.
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
| |
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
| |
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
|
|
|
| |
Non-VBO descriptors won't be smaller than the cache line, so simply use
the cache line size.
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
|
|
| |
This results in a very tiny decrease in lgkm wait cycles.
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
|
|
| |
The other codepaths don't need this.
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
|
|
| |
for consistency; no change in behavior
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
|
|
| |
and improve some comments
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
|
|
|
| |
This might reduce the on-demand compilation if the initial VS/LS/ES
determination is wrong.
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
|
|
| |
there is no point, we have to wait anyway.
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
|
|
|
|
|
| |
So that we can disable u_vbuf for GL core profiles.
This is a v2 of the previous VI-only patch.
It requires SH_MEM_CONFIG.ALIGNMENT_MODE = UNALIGNED on CIK-VI.
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
|
|
| |
not needed with the shader fallback
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
| |
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
|
|
| |
so that we can add 3-component fallbacks.
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
|
|
| |
there's no error checking, because the previous code didn't do it either.
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
|
| |
Reviewed-by: Nicolai Hähnle <[email protected]>
Reviewed-by: Edmondo Tommasina <[email protected]>
|
|
|
|
| |
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
| |
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
| |
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
| |
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
| |
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
| |
trivial
|
|
|
|
| |
The code for this landed a few days ago.
|
|
|
|
|
|
| |
sffbh needs to be suffixed with ".i32"
Reviewed-by: Dave Airlie <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
If an unsized declared array is not the last in an SSBO
and an implicit size can not be defined on linking time,
the linker should raise an error instead of reaching
an assertion on GL.
This reverts part of commit 3da08e166415a745139c1127040a24e8a45dc553
getting back to the behavior of commit 5b2675093e863a52b610f112884ae12d42513770
The original patch was correct for GLES that should produce
a compile-time error but the linker error is still necessary
in desktop GL.
Fixes the following piglit tests:
tests/spec/arb_shader_storage_buffer_object/non_integral_size_array_member.shader_test
tests/spec/arb_shader_storage_buffer_object/unsized_array_member.shader_test
Reviewed-by: Dave Airlie <[email protected]>
Reviewed-by: Iago Toral Quiroga <[email protected]>
Signed-off-by: Jose Maria Casanova Crespo <[email protected]>
|
|
|
|
|
|
|
|
| |
Found while running shader-db under valgrind.
Signed-off-by: Lionel Landwerlin <[email protected]>
Reviewed-by: Jordan Justen <[email protected]>
Cc: "13.0 17.0" <[email protected]>
|
|
|
|
| |
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
| |
Reviewed-by: Samuel Iglesias Gonsálvez <[email protected]>
|
|
|
|
|
|
|
|
|
| |
This one only keeps allocated memory in the list, and list nodes
in the descriptor sets. Thsi doesn't need messing around with
max_sets, and we get automatic merging of free regions.
Signed-off-by: Bas Nieuwenhuizen <[email protected]>
Reviewed-by: Dave Airlie <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
| |
We only use the freed ones after all free space has been used. If
the app only allocates small descriptor sets, we might go over
max_sets before the memory is full.
Signed-off-by: Bas Nieuwenhuizen <[email protected]>
Reviewed-by: Dave Airlie <[email protected]>
CC: <[email protected]>
Fixes: f4e499ec79147f4172f3669ae9dafd941aaeeb65
|
|
|
|
|
|
|
|
|
|
|
| |
The 32-bit to 64-bit conversions need to have the 32-bit
data source elements aligned to 64-bit but only with doubles as
destination type.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=99660
Signed-off-by: Samuel Iglesias Gonsálvez <[email protected]>
Tested-by: Mark Janes <[email protected]>
Reviewed-by: Matt Turner <[email protected]>
|
|
|
|
|
| |
For now its disabled for scons so wrap glsl cache calls in a
define conditional.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
The optimization in unpack_64 is clearly subsumed with the opt_algebraic
optimizations in the previous commit. The pack optimization may not be
quite handled by opt_algebraic but opt_algebraic should get the really
bad cases. Also, it's been broken since it was merged and we've never
noticed so it must not be doing anything.
Reviewed-by: Kenneth Graunke <[email protected]>
|
|
|
|
|
|
| |
This reduces the instruction count in some fp64 and int64 piglit tests
Reviewed-by: Kenneth Graunke <[email protected]>
|
|
|
|
|
|
|
| |
There's nothing "double" about it other than, perhaps, the fact that it
packs two 32-bit values.
Reviewed-by: Kenneth Graunke <[email protected]>
|
|
|
|
|
|
|
| |
NIR is a typeless IR and the two opcodes, when considered bitwise, do
exactly the same thing. There's no reason to have two versions.
Reviewed-by: Kenneth Graunke <[email protected]>
|
|
|
|
|
|
|
| |
We can only do the optimization if the source *is* SSA.
Reviewed-by: Kenneth Graunke <[email protected]>
Cc: "13.0 17.0" <[email protected]>
|
|
|
|
|
|
| |
the guards have been added to the header files that needed them.
Reviewed-by: Ilia Mirkin <[email protected]>
|
|
|
|
|
|
| |
Added extern "C" __cplusplus guards on headers that did not have them.
Reviewed-by: Ilia Mirkin <[email protected]>
|
|
|
|
| |
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
| |
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In order to avoid costly fallback recompiles when cache items are
created with an old version of Mesa or for a different gpu on the
same system we want to create directories that look like this:
./{TIMESTAMP}_{LLVM_TIMESTAMP}/{GPU_ID}
Note: The disk cache util will take a single timestamp string, it is
up to the backend to concatenate the llvm string with the mesa string
if applicable.
Reviewed-by: Nicolai Hähnle <[email protected]>
|