Commit Graph

18 Commits

Author SHA1 Message Date
Stanislav Mekhanoshin a0ab45799b [AMDGPU] Move atomic expand past infer address spaces
There are cases where infer address spaces pass cannot yet
infer an address space in the opt pipeline and then in the
llc pipeline it runs too late for atomic expand pass to
benefit from a specific address space.

Move atomic expand pass past the infer address spaces.

Fixes: SWDEV-293410

Differential Revision: https://reviews.llvm.org/D105511
2021-07-06 15:53:32 -07:00
Tony Tye 7f19aa73c2 [AMDGPU] Update gfx90a memory model support
Update AMDGPU gfx90a memory model to make coarse grain memory allocations
consistent when fine grained system scope atomic acquire and release is
performed.

Reviewed By: rampitec

Differential Revision: https://reviews.llvm.org/D105137
2021-06-30 04:05:22 +00:00
Stanislav Mekhanoshin 8f98356bb5 [AMDGPU] Only allow global fp atomics with unsafe option
Previously we were allowing to use FP atomics without
-amdgpu-unsafe-fp-atomics option if a scope is less then
system. This is not safe just as well if we have UC memory.

This change only allows global and flat FP atomics with
the unsafe option. Consequentially that makes a check for
denorm mode redundant since we skip it with the unsafe
option and do not have a way to produce these instructions
without it anyway.

Differential Revision: https://reviews.llvm.org/D102347
2021-05-13 08:52:20 -07:00
Stanislav Mekhanoshin 189310a140 [AMDGPU] Allow -amdgpu-unsafe-fp-atomics to ignore denorm mode
Fixes: SWDEV-274276

Differential Revision: https://reviews.llvm.org/D100072
2021-04-08 12:46:36 -07:00
Tony Tye 4658cd4c18 [AMDGPU] Update gfx90a memory model support
Reviewed By: rampitec

Differential Revision: https://reviews.llvm.org/D100070
2021-04-07 22:17:58 +00:00
Stanislav Mekhanoshin 30b3aab329 Copy syncscope when expanding atomicrmw into cmpxchg loop
Fixes: SWDEV-280070

Differential Revision: https://reviews.llvm.org/D99902
2021-04-05 17:29:38 -07:00
Petar Avramovic b082e6f88a [AMDGPU] Extend gfx10 test coverage. NFC.
Differential Revision: https://reviews.llvm.org/D99267
2021-03-29 11:13:55 +02:00
Stanislav Mekhanoshin 574a9dabc6 [AMDGPU] Always expand system scope fp atomics on gfx90a
FP atomics in system scope cannot be used and shall always
be expanded in a CAS loop.

Differential Revision: https://reviews.llvm.org/D98085
2021-03-10 12:35:23 -08:00
Matt Arsenault 589223e044 AMDGPU: Remove special case in shouldCoalesce
Unaligned registers are now constrained with classes, rather than
specially reserving a subset of the whole class.
2021-02-24 14:49:44 -05:00
Stanislav Mekhanoshin a8d9d50762 [AMDGPU] gfx90a support
Differential Revision: https://reviews.llvm.org/D96906
2021-02-17 16:01:32 -08:00
Konstantin Zhuravlyov 6054a456da AMDGPU: Add support for amdgpu-unsafe-fp-atomics attribute
If amdgpu-unsafe-fp-atomics is specified, allow {flat|global}_atomic_add_f32 even if atomic modes don't match.

Differential Revision: https://reviews.llvm.org/D95391
2021-02-04 08:09:34 -05:00
Matt Arsenault d2e52eec51 AMDGPU: Select global saddr mode from SGPR pointer
Use the 64-bit SGPR base with a 0 offset, since it's 1 fewer
instruction to materialize the 0 vs. the 64-bit copy.
2020-11-16 11:51:06 -05:00
Scott Linder d5f2c3e7c0 [NFC][AMDGPU] Clean up some lit test prefixes
Replace some instances of "ALL" with "GCN" where it applies. Committed
as obvious.
2020-11-11 17:12:37 +00:00
Stanislav Mekhanoshin 43804364e2 [AMDGPU] Fixes typo in the test. NFC.
denormal-fp-math-fp32 -> denormal-fp-math-f32
2020-09-24 16:07:15 -07:00
Matt Arsenault af0207f2ba AMDGPU: Check global FP atomics match default FP mode
We would always select global FP atomics from atomicrmw fadd, although
they have a hardcoded FP mode.
2020-09-23 09:07:50 -04:00
Stanislav Mekhanoshin 277de43d88 [AMDGPU] Unify intrinsic ret/nortn interface
We have a single noret intrinsic an a lot of special handling
around it. Declare it just as any other but do not define rtn
instructions itself instead.

Differential Revision: https://reviews.llvm.org/D87719
2020-09-15 15:26:42 -07:00
Matt Arsenault 1657f0ebc2 AMDGPU: Fix overriding global FP atomic feature predicates
Global TableGen let override blocks are pretty dangerous and override
any local special cases. In this case, the broader HasFlatGlobalInsts
was overriding the more specific predicate for
FeatureAtomicFaddInsts. Make sure HasFlatGlobalInsts is implied by
FeatureAtomicFaddInsts, and make sure the right predicate is used.

One issue with independently setting the subtarget features on
incompatible targets is all of the encoding families do not define all
opcodes. This will hit an assert on gfx10 for example, since we set
the encoding independently based on the generation and not based on a
feature.
2020-06-04 17:50:38 -04:00
Matt Arsenault e16a71382d AMDGPU: Select global atomicrmw fadd
This only works if there is no use of the return value.
2019-11-06 16:06:38 -08:00