AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
119,593 Commits 1 Branch 0 Tags
72e9a2344346f4e2eff240e475463eeb046fe9be
Commit Graph

4567 Commits

Author SHA1 Message Date
Rhys Perry 72e9a23443 radv/aco: use ACO for GS copy shaders 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>
 2020-01-24 13:35:07 +00:00
Rhys Perry f8f7712666 aco: implement GS copy shaders 
v5: rebase on float_controls changes
v7: rebase after shader args MR and load/store vectorizer MR

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>
 2020-01-24 13:35:07 +00:00
Rhys Perry de4ce66f5c aco: remove needs_instance_id 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>
 2020-01-24 13:35:07 +00:00
Rhys Perry e192e268de aco: explicitly mark end blocks for exports 
For GS copy shaders, whether we want to do exports is conditional. By
explicitly marking the end blocks, we can mark an IF's then branch as an
export block and ensure that's where the assembler inserts null exports.

v6: only fixup exports in the end block, like before
v8: simplify some code

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>
 2020-01-24 13:35:07 +00:00
Rhys Perry d46a54ecff radv/aco: allow ACO for GS 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>
 2020-01-24 13:35:07 +00:00
Rhys Perry 8bad100f83 aco: implement GS on GFX7-8 
GS is the same on GFX6, but GFX6 isn't fully supported yet.

v4: fix regclass
v7: rebase after shader args MR

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>
 2020-01-24 13:35:07 +00:00
Rhys Perry 40bb81c9dd radv/aco,aco: implement GS on GFX9+ 
v2: implement GFX10
v3: rebase
v7: rebase after shader args MR
v8: fix gs_vtx_offset usage on GFX9/GFX10
v8: use unreachable() instead of printing intrinsic
v8: rename output_state to ge_output_state
v8: fix formatting around nir_foreach_variable()
v8: rename some helpers in the scheduler
v8: rename p_memory_barrier_all to p_memory_barrier_common
v8: fix assertion comparing ctx.stage against vertex_geometry_gs

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>
 2020-01-24 13:35:07 +00:00
Rhys Perry 70f63c1988 aco: improve support for s_sendmsg 
In particular, the messages needed for GS.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>
 2020-01-24 13:35:07 +00:00
Rhys Perry 0da7b3b18b radv: move gs copy shader creation before other variants 
ACO lowers output derefs which breaks the shader_info pass used by gs copy
shader creation.

v3: rebase

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>
 2020-01-24 13:35:07 +00:00
Timur Kristóf 23edcf6490 aco: Make a better guess at which instructions need the VCC hint. 
Previously, bool_to_vector_condition would always set the VCC hint
on its result. This commit improves it by having the optimizer set
the VCC hint only when the result really needs to be in the VCC.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3451>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3451>
 2020-01-24 13:14:23 +00:00
Bas Nieuwenhuizen 0890482969 radv: Allow DCC & TC-compat HTILE with VK_IMAGE_CREATE_EXTENDED_USAGE_BIT. 
I misunderstood the flag when initially disabling. But this flag
only does something with mutable formats. If we have DCC and
mutable formats, the formats are close enough that the allowed
usage flags are not meaningfully different nor used during
allocation.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3424>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3424>
 2020-01-24 11:16:39 +00:00
Bas Nieuwenhuizen 1b447bd2e6 radv: Expose VK_KHR_swapchain_mutable_format. 
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2354
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3425>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3425>
 2020-01-24 10:47:07 +00:00
Samuel Pitoiset a31bcf2be6 ac/llvm: fix missing casts in ac_build_readlane() 
Because ac_build_optimization_barrier() overwrites the original
src_type, we have to keep track of it before emitting that barrier.
Otherwise, wrong conversions are expected for pointers or small
bitsizes.

By doing this, we no longer need to do the cast dance in
ac_build_readlane_no_opt_barrier(), it was just necessary for
ac_build_optimization_barrier().

This fixes a bunch of crashes with subgroups related tests when
RADV_DEBUG=checkir is enabled, and it also fixes a compiler crash
with The Surge 2.

Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2395
Fixes: 0f45d4dc2b ("ac: add ac_build_readlane without optimization barrier")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3535>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3535>
 2020-01-24 07:40:07 +01:00
Samuel Pitoiset 8d5203dad2 aco: implement nir_op_f2i64/nir_op_f2u64 on GFX6 
V_TRUNC_F64 and V_FLOOR_F64 needs to be lowered on GFX6.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3477>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3477>
 2020-01-23 14:40:48 +01:00
Samuel Pitoiset 4d92601715 aco: implement 64-bit nir_op_ffloor on GFX6 
GFX6 doesn't have V_FLOOR_F64, it needs to be lowered. Loosely based
on the AMDGPU LLVM backend.

Introduce a new function because it will be useful for some other
64-bit operations.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3477>
 2020-01-23 14:40:45 +01:00
Samuel Pitoiset fbd169e421 aco: implement 64-bit nir_op_fround_even on GFX6 
GFX6 doesn't have V_RNDNE_F64, it needs to be lowered. Loosely based
on the AMDGPU LLVM backend.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3477>
 2020-01-23 14:40:42 +01:00
Samuel Pitoiset 87588801d3 aco: implement 64-bit nir_op_fceil on GFX6 
GFX6 doesn't have V_CEIL_F64, it needs to be lowered. Loosely based
on the AMDGPU LLVM backend.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3477>
 2020-01-23 14:40:38 +01:00
Samuel Pitoiset aad5176c58 aco: implement 64-bit nir_op_ftrunc on GFX6 
GFX6 doesn't have V_TRUNC_F64, it needs to be lowered. Loosely based
on the AMDGPU LLVM backend.

Introduce a new function because it will be useful for some other
64-bit operations.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3477>
 2020-01-23 14:40:34 +01:00
Samuel Pitoiset 36e7a5f5b9 aco: implement nir_intrinsic_global_atomic_* on GFX6 
GFX6 doesn't have FLAT instructions, use MUBUF instructions instead.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3477>
 2020-01-23 14:40:30 +01:00
Samuel Pitoiset 22d8822683 aco: implement nir_intrinsic_load_global on GFX6 
GFX6 doesn't have FLAT instructions, use MUBUF instructions instead.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3477>
 2020-01-23 14:40:27 +01:00
Samuel Pitoiset d6af7571c2 aco: implement nir_intrinsic_store_global on GFX6 
GFX6 doesn't have FLAT instructions, use MUBUF instructions instead.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3477>
 2020-01-23 14:40:24 +01:00
Samuel Pitoiset 01f0bef71e aco: fix wrong IR in nir_intrinsic_load_barycentric_at_sample 
Only GFX6 was affected, my mistake. The total number of SGPR operands
should be 4 when we want to create a vec4.

Fixes: dbdf3b3ef9 ("aco: implement nir_intrinsic_load_barycentric_at_sample on GFX6")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3477>
 2020-01-23 14:40:21 +01:00
Samuel Pitoiset 54e54ec3e8 aco: fix printing assembly with CLRXdisasm on GFX6 
We thought that CLRXdisasm allowed gfx600 as well as gfx700 but
it actually doesn't. Use the family for GFX6 chips instead.

Fixes: 0099f85232 ("aco: print assembly with CLRXdisasm for GFX6-GFX7 if found on the system")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3531>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3531>
 2020-01-23 11:34:37 +00:00
Samuel Pitoiset 12fe19ba3b radv: advertise VK_AMD_shader_fragment_mask 
Only for GFX8+ because it's untested on older generations.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3304>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3304>
 2020-01-23 10:48:02 +00:00
Samuel Pitoiset e030aef32c aco: add support for nir_texop_fragment_{mask}_fetch 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3304>
 2020-01-23 10:48:02 +00:00
Samuel Pitoiset 9e477d79b7 ac/nir: add support for nir_texop_fragment_{mask}_fetch 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3304>
 2020-01-23 10:48:02 +00:00
Samuel Pitoiset e60de08547 radv: handle missing implicit subpass dependencies 
When a subpass doesn't declare an explicit dependency from/to
VK_SUBPASS_EXTERNAL, Vulkan says there is an implicit dependency.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3330>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3330>
 2020-01-23 11:25:41 +01:00
Samuel Pitoiset 0d2da2a8c0 radv: add explicit external subpass dependencies to meta operations 
No functional changes because a subpass dependency with dstStageMask
set to VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT is a no-op.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3330>
 2020-01-23 11:25:38 +01:00
Rhys Perry 15a1cc00d3 aco: fix off-by-one error when initializing sgpr_live_in 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2394
Fixes: 93c8ebfa78 ('aco: Initial commit of independent AMD compiler')
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3511>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3511>
 2020-01-22 17:23:30 +00:00
Samuel Pitoiset bd51538d28 radv: fix double free corruption in radv_alloc_memory() 
If the driver fails to allocate memory for some reasons, it shouldn't
free the 'mem' object twice.

Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2302
Fixes: 825ddfee59 ("radv: Handle device memory alloc failure with normal free.")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3508>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3508>
 2020-01-22 17:01:16 +00:00
Rhys Perry 3f96a1ed86 aco: fix operand kill flags when a temporary is used more than once 
Helps create v_mac_f32 from v_mad_f32(b, a, b)

Totals from affected shaders:
SGPRS: 35824 -> 35824 (0.00 %)
VGPRS: 33460 -> 33456 (-0.01 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Private memory VGPRs: 0 -> 0 (0.00 %)
Scratch size: 0 -> 0 (0.00 %) dwords per thread
Code Size: 2187264 -> 2180976 (-0.29 %) bytes
LDS: 127 -> 127 (0.00 %) blocks
Max Waves: 3802 -> 3802 (0.00 %)

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3486>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3486>
 2020-01-22 15:55:00 +00:00
Timur Kristóf 1c9ecb2123 aco: Fix signedness compare warning. 
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3483>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3483>
 2020-01-22 11:09:17 +01:00
Timur Kristóf 533a20dbd5 aco: Fix maybe-uninitialized warnings. 
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3483>
 2020-01-22 11:09:14 +01:00
Timur Kristóf 6fb3df2786 aco: Fix -Wstringop-overflow warnings in aco_span. 
GCC does not understand how aco_span works.
This patch fixes it by casting the aco_span's this pointer
to uintptr_t rather than to a char pointer, effectively
telling GCC not to try to figure it out.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3483>
 2020-01-22 11:09:10 +01:00
Bas Nieuwenhuizen bd4380c63c radv: Remove syncobj_handle variable in header. 
I strongly suspect it was supposed to be a typedef. However, used
nowhere, we should remove it.

Fixes: eaa56eab6d "radv: initial support for shared semaphores (v2)"
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2385
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3479>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3479>
 2020-01-21 12:28:00 +00:00
Marek Olšák 4e4b2d13f0 ac: add helper ac_build_triangle_strip_indices_to_triangle 
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
 2020-01-20 16:16:11 -05:00
Marek Olšák 0f45d4dc2b ac: add ac_build_readlane without optimization barrier 
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
 2020-01-20 16:16:11 -05:00
Marek Olšák 77393cf39b ac: add prefix bitcount functions 
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
 2020-01-20 16:16:11 -05:00
Samuel Pitoiset dbdf3b3ef9 aco: implement nir_intrinsic_load_barycentric_at_sample on GFX6 
GFX6 doesn't have FLAT instructions which means we have to emit
a 64-bit MUBUF load.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3432>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3432>
 2020-01-20 16:24:55 +00:00
Samuel Pitoiset 9e2fde84fc aco: add new addr64 bit to MUBUF instructions on GFX6-GFX7 
According to the different ISA docs (and to LLVM), this bit seems
to only exists on GFX6-GFX7.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3432>
 2020-01-20 16:24:55 +00:00
Samuel Pitoiset fe9157a700 aco: do not use the vec3 variant for loads on GFX6 
GFX6 only supports vec3 with load/store format.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3432>
 2020-01-20 16:24:55 +00:00
Samuel Pitoiset 1b5bb204d9 aco: do not use the vec3 variant for stores on GFX6 
GFX6 only supports vec3 with load/store format.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3432>
 2020-01-20 16:24:55 +00:00
Samuel Pitoiset b8abfafe86 aco: fix constant folding of SMRD instructions on GFX6 
SMRD instructions have an 8-bit dword offset on SI.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3432>
 2020-01-20 16:24:55 +00:00
Rhys Perry 29bfe18abd aco: fix fall-through test in try_remove_simple_block() with back-edges 
3bca0af2 enhanced empty block determination which exposed this bug and
created an infinite loop in a Guild Wars 2 shader.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Fixes: 3bca0af25d
     ('aco: ignore parallelcopies to the same register on jump threading')

Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2364
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3452>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3452>
 2020-01-20 11:51:45 +00:00
Rhys Perry e151398de6 aco: fix stack buffer overflow in apply_sgprs() 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Fixes: cef7879719 ('aco: rewrite apply_sgprs()')
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2361
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3442>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3442>
 2020-01-20 11:13:11 +00:00
Samuel Pitoiset 0099f85232 aco: print assembly with CLRXdisasm for GFX6-GFX7 if found on the system 
LLVM only supports GFX8+. Using CLRXdisasm works most of the time,
so it's useful to add support for it.

Original patch by Daniel Schürmann.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3439>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3439>
 2020-01-17 17:41:32 +00:00
Samuel Pitoiset b9b393f0ce aco: fix emitting slc for MUBUF instructions on GFX6-GFX7 
Same as GFX10, only GFX8/GFX9 moved that bit near the opcode.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3437>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3437>
 2020-01-17 16:56:04 +01:00
Daniel Schürmann 3bca0af25d aco: ignore parallelcopies to the same register on jump threading 
The more conservative lowering to CSSA inserts unnecessary parallelcopies
which might get coalesced and can be ignored on jump threading.

v2: outline is_empty_block() check.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3385>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3385>
 2020-01-16 16:01:59 +01:00
Daniel Schürmann 427e5eeb02 aco: handle phi affinities transitively through parallelcopies 
This can coalesce most unnecessarily inserted parallelcopies
from lowering to CSSA.

v2: refactor loop a bit to make it more efficient and readable.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3385>
 2020-01-16 16:01:59 +01:00
Daniel Schürmann d098024c40 aco: rework lower_to_cssa() 
This patch changes lower_to_cssa to be much more conservative
about assumptions which phi operands might interfere.
Previously, this pass wasn't exhaustive and could miss some corner cases.

v2: remove optimizations to find better insertion points as it's hard
to guarantee that they are always correct and have overall no benefit.

Fixes: 0b8216b2cd ('aco: Lower to CSSA')

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3385>
 2020-01-16 16:01:59 +01:00