After some investigation, it seems like that disabling the UNK02C4
command avoid a read fault with texelFetch() from a compute shader.
I have no clue on what this method actually does, but this avoid the
GPU to hang with basic-texelFetch.shader_test without introducing any
compute-related regressions.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
Normally, we split uniforms at the end but in Vulkan, we bail because we
don't want pull constants. However, we still need them split because
pack_uniforms relies on it.
I really don't like this patch not because it doesn't work (it does) but
because now that we're using MOV_INDIRECT, uniform numbers and sizes don't
really matter anymore. In the FS backend, uniform splitting and packing is
handled all at once (actual re-assignment of locations happens later) and
we really should do it that way in vec4 eventually as well.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94998
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=95001
All of the code that did something special based on vec4 vs. scalar is
bogus. In the backend, everything is now in units of bytes and the vec4
backend can handle full std140 packing so we don't need to do anything
special anymore.
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94998
Code was using an incorrect address for the base pointer.
v2: use swr_resource_data() utility function.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94979
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tested-by: Markus Wick <markus@selfnet.de>
Add support for OpenCL global memory buffers, note this has only
been tested with regular load and stores and likely needs more work
for e.g. atomic ops.
Tested with piglet on a gf119 and a gk107:
./piglit run -o shader -t '.*arb_shader_storage_buffer_object.*' results/shader
[9/9] pass: 9 /
./piglit run -o shader -t '.*arb_compute_shader.*' results/shader
[20/20] skip: 4, pass: 16 |
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Some of the lowering steps we currently do for FILE_MEMORY_GLOBAL only
apply to buffers, making it impossible to use FILE_MEMORY_GLOBAL for
OpenCL global buffers.
This commits changes the buffer code to use FILE_MEMORY_BUFFER at the
ir_from_tgsi and lowering steps, freeing use of FILE_MEMORY_GLOBAL
for use with OpenCL global buffers.
Note that after lowering buffer accesses use the FILE_MEMORY_GLOBAL
register file.
Tested with piglet on a gf119 and a gk107:
./piglit run -o shader -t '.*arb_shader_storage_buffer_object.*' results/shader
[9/9] pass: 9 /
./piglit run -o shader -t '.*arb_compute_shader.*' results/shader
[20/20] skip: 4, pass: 16 |
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
v2: - set interop_version
- simplify the offset_after macro
v2.1: - use version numbers, remove offset_after
- set "out_driver_data_written"
v2.2: - set buf_offset & buf_size for GL_ARRAY_BUFFER too
- add whandle.offset to buf_offset
- disable the minmax cache for GL_TEXTURE_BUFFER
v2: - use "enum" to define stuff
v3: - more comments, define MESA_GLINTEROP_UNSUPPORTED
v4: - add mesa_glinterop_device_info::interop_version
- more comments
- remove #define MESA_GLINTEROP_VERSION
- use const for "in"
v4.1: - use version numbers for structures
- add "out_driver_data_written"
v4.2: - buf_offset & buf_size affect GL_ARRAY_BUFFER too, this is required
for sharing suballocations within a larger buffer
When creating egl images we do a bytes to pixel conversion by deviding
by 4 regardless of the pixel format. This does not work for RGB565. In
this patch, we avoid useless conversion and use proper API when the
conversion cannot be avoided.
Signed-off-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
This code is already duplicated twice and will be useful again. This
will also help when adding formats.
Signed-off-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
This *seems* like a hw bug, and maybe only applies to certain a4xx
variants/revisions. But setting the SRGB bit in sampler view state
(texconst0) causes invalid alpha for ASTC textures. Work around this
by doing the srgb->linear conversion in the shader instead.
This fixes 392 dEQP tests: dEQP-GLES3.functional.texture.*astc*srgb*
(The remaining fails seem to be a bug w/ ASTC + linear filtering, also
possibly a420.0 specific.)
Signed-off-by: Rob Clark <robclark@freedesktop.org>
No need for it not to be const, and lets caller declare it const if
desired.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Provide an improved lowering for LRP, which can be implemented in two
MAD instructions with a bit of rearranging of the equation, rather
than the literal implementation of two multiplies, an add and a
subtract.
Signed-off-by: Russell King <rmk@arm.linux.org.uk>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Improve XPD lowering to consume less instructions by using the
MAD instruction to perform the multiply and subtraction together.
Signed-off-by: Russell King <rmk@arm.linux.org.uk>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Add support for lowering TRUNC using the following sequence:
FRC tmpA, |src|
SUB tmpA, |src|, tmpA
CMP dst, -tmpA, tmpA
Note that this is incompatible with FRC lowering.
Signed-off-by: Russell King <rmk@arm.linux.org.uk>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Add support for lowering FLR and CEIL to FRC/SUB and FRC/ADD
instructions for GPUs that support FRC but not FLR or CEIL. Since
these uses FRC, it is invalid to ask for FLR or CEIL to be lowered
along with FRC, so add an assert to catch this invalid configuration.
We also need to deal with FLR instructions emitted by the lowering
code. Fix these up with the FRC+SUB equivalent when FLR lowering is
enabled.
Signed-off-by: Russell King <rmk@arm.linux.org.uk>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Signed-off-by: Rob Clark <robclark@freedesktop.org>
v2: Use chip_class instead of family.
v3: Check kernel version for SI.
v4: Preemptively allow amdgpu winsys for SI.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
si_shader_create corrects the SGPR count with si_fix_num_sgprs. We then
recompute the rsrc1 register to use the new SGPR count.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
v2: Also depend on atomic counters.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
v2: Add more CS_PARTIAL_FLUSH events.
Essentially every place with waits on finishing for pixel shaders
also has a write after read hazard with compute shaders.
Invalidating L2 waits implicitly on pixel and compute shaders,
so, we don't need a CS_PARTIAL_FLUSH for switching FBO.
v3: Add CS_PARTIAL_FLUSH events even if we already have INV_GLOBAL_L2.
According to Marek the INV_GLOBAL_L2 events don't wait for compute
shaders to finish, so wait for them explicitly.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
v2: - Use radeon_set_sh_reg_seq.
- Set predicate bit for conditional rendering.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
v2: - Do check if anything changed earlier
- Use emitted_program instead of emitted_bo to prevent
shaders with shader->bo = NULL confusing the check
- Use radeon_set_sh_reg*
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Instead of having a scratch buffer per program, have one per
context.
Also removed the per kernel wave count calculations, but
that only helped if the total number of waves in the dispatch
was smaller than sctx->scratch_waves.
v2: Fix style issue.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Also removes PKT3_CONTEXT_CONTROL as that is already being done
by si_begin_new_cs, when emitting init_config.
v2: - Use radeon_set_sh_reg_seq.
- Also set COMPUTE_STATIC_THREAD_MGMT_SE2 / SE3 for CIK+
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
As far as I can see we use relocations for clover too.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Also uses a dynamically allocated buffer using u_upload_alloc.
The old buffer per program approach required serializing all
dispatches of the same program.
v2: - Clarified commit message.
- Use radeon_set_sh_reg_seq.
- Also upload input buffer for clover kernels, even when
input_size is 0, as it contains grid parameters.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
v2: Moved scratch_enabled initialization after compile.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
v2: - Use single region
- Use get_memory_ptr
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
v2: - Use single region
- Combine address calculation
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Declares the shared memory as a global variable so that
LLVM is aware of it and it does not conflict with passes
like AMDGPUPromoteAlloca.
v2: - Use ctx->i8.
- Dropped null-check for declare_memory_region.
- Changed memory region array to single region.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
v2: Load previous list for new CS instead of re-emitting
all descriptors.
v3: Do radeon_add_to_buffer_list in si_ce_upload.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>