Changes in bswap.ll

e2926501 - [AMDGPU] Aggressively fold immediates in SIShrinkInstructions

Mon, 16 May 2022 14:53:03 +0000

[AMDGPU] Aggressively fold immediates in SIShrinkInstructionsFold immediates regardless of how many uses they have. This is expectedto increase overall code size, but decrease register usage.Differential Revision: https://reviews.llvm.org/D114644 List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll

da067ed5 - [AMDGPU] Set most sched model resource's BufferSize to one

Wed, 10 Nov 2021 17:59:31 +0000

[AMDGPU] Set most sched model resource's BufferSize to oneUsing a BufferSize of one for memory ProcResources will result in betterILP since it more accurately models the dependencies between memory opsand their consumers on an in-order processor. After this change, thescheduler will treat the data edges from loads as blocking so thatstalls are guaranteed when waiting for data to be retreaved from memory.Since we don't actually track waitcnt here, this should do a better jobat modeling their behavior.Practically, this means that the scheduler will trigger the 'STALL'heuristic more often.This type of change needs to be evaluated experimentally. Preliminaryresults are positive.Fixes: SWDEV-282962Reviewed By: rampitecDifferential Revision: https://reviews.llvm.org/D114777 List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll

1ebe86ad - [NFC] Removed unused prefixes in test/CodeGen/AMDGPU

Tue, 05 Jan 2021 22:15:20 +0000

[NFC] Removed unused prefixes in test/CodeGen/AMDGPUMore patches to follow.Differential Revision: https://reviews.llvm.org/D94121 List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll

fdc90277 - [DAG][AMDGPU][X86] Add SimplifyMultipleUseDemandedBits handling for SIGN/ZERO_EXTEND + SIGN/ZERO_EXTEND_VECTOR_INREG

Wed, 29 Jul 2020 16:33:56 +0000

[DAG][AMDGPU][X86] Add SimplifyMultipleUseDemandedBits handling for SIGN/ZERO_EXTEND + SIGN/ZERO_EXTEND_VECTOR_INREGPeek through multiple use ops like we already do for ANY_EXTEND/ANY_EXTEND_VECTOR_INREGDifferential Revision: https://reviews.llvm.org/D84863 List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll

62fd7f76 - [MachineScheduler] Fix the TopDepth/BotHeightReduce latency heuristics

Tue, 07 Jan 2020 15:43:46 +0000

[MachineScheduler] Fix the TopDepth/BotHeightReduce latency heuristicstryLatency compares two sched candidates. For the top zone it prefersthe one with lesser depth, but only if that depth is greater than thetotal latency of the instructions we've already scheduled -- otherwiseits latency would be hidden and there would be no stall.Unfortunately it only tests the depth of one of the candidates. This canlead to situations where the TopDepthReduce heuristic does not kick in,but a lower priority heuristic chooses the other candidate, whose depth*is* greater than the already scheduled latency, which causes a stall.The fix is to apply the heuristic if the depth of *either* candidate isgreater than the already scheduled latency.All this also applies to the BotHeightReduce heuristic in the bottomzone.Differential Revision: https://reviews.llvm.org/D72392 List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll

8c2c0b36 - AMDGPU: Improve i16/v2i16 bswap

Fri, 14 Feb 2020 15:31:38 +0000

AMDGPU: Improve i16/v2i16 bswap List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll

e0fd2d6d - AMDGPU: Add baseline tests for 16-bit bswap

Fri, 14 Feb 2020 15:15:20 +0000

AMDGPU: Add baseline tests for 16-bit bswap List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll

bfe37794 - AMDGPU: Use v_perm_b32 to implement bswap

Thu, 13 Feb 2020 01:28:38 +0000

AMDGPU: Use v_perm_b32 to implement bswapAlso greatly improve i64 lowering. LegalizeIntegerTypes does thecorrect narrowing if i64 isn't legal. Just workaround this forSelectionDAG by making i64 legal and splitting in the patterns. List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll

2252cac6 - [ANDGPU] getMemOperandsWithOffset: support BUF non-stack-access instructions with resource but no vaddr

Fri, 20 Dec 2019 14:59:15 +0000

[ANDGPU] getMemOperandsWithOffset: support BUF non-stack-access instructions with resource but no vaddrSummary:This enables clustering for many more BUF instructions.Reviewers: rampitec, arsenm, nhaehnleSubscribers: jvesely, wdng, hiraditya, kerbowa, llvm-commitsTags: #llvmDifferential Revision: https://reviews.llvm.org/D73868 List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll

01a4b831 - [codegen,amdgpu] Enhance MIR DIE and re-arrange it for AMDGPU.

Wed, 08 Jan 2020 15:50:23 +0000

[codegen,amdgpu] Enhance MIR DIE and re-arrange it for AMDGPU.Summary:- `dead-mi-elimination` assumes MIR in the SSA form and cannot be arranged after phi elimination or DeSSA. It's enhanced to handle the dead register definition by skipping use check on it. Once a register def is `dead`, all its uses, if any, should be `undef`.- Re-arrange the DIE in RA phase for AMDGPU by placing it directly after `detect-dead-lanes`.- Many relevant tests are refined due to different register assignment.Reviewers: rampitec, qcolombet, sunfishSubscribers: arsenm, kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, hiraditya, llvm-commitsTags: #llvmDifferential Revision: https://reviews.llvm.org/D72709 List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll

2afe8641 - [DAG] Add SimplifyDemandedBits support for BSWAP

Sun, 15 Dec 2019 13:52:34 +0000

[DAG] Add SimplifyDemandedBits support for BSWAPThis exposes a shortcoming for AArch64, and that is tracked by PR40881:https://bugs.llvm.org/show_bug.cgi?id=40881Patch by: @RKSimon (Simon Pilgrim)Differential Revision: https://reviews.llvm.org/D58017 List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll

d88db6d7 - AMDGPU: Always use s33 for global scratch wave offset

Thu, 20 Jun 2019 21:58:24 +0000

AMDGPU: Always use s33 for global scratch wave offsetEvery called function could possibly need this to calculate theabsolute address of stack objectst, and this avoids inserting a copyaround every call site in the kernel. It's also somewhat cleaner tokeep this in a callee saved SGPR.llvm-svn: 363990 List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll

913ba8ee - Revert LIS handling in MachineDCE

Tue, 09 Apr 2019 16:13:53 +0000

Revert LIS handling in MachineDCEOne of out of tree targets has regressed with this patch. Revertingit for now and let liveness to be fully reconstructed in case passwas used after the LIS is created to resolve the regression.Differential Revision: https://reviews.llvm.org/D60466llvm-svn: 358015 List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll

2ccc120d - [AMDGPU] Regenerate bswap/bitreverse tests.

Tue, 26 Feb 2019 11:01:08 +0000

[AMDGPU] Regenerate bswap/bitreverse tests.Make codegen changes more obvious in D58017llvm-svn: 354863 List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll

df0f2507 - DAG: Fix not truncating when promoting bswap/bitreverse

Wed, 31 Jan 2018 23:54:16 +0000

DAG: Fix not truncating when promoting bswap/bitreverseThese need to convert back to the original type, like anyother promotion.llvm-svn: 323932 List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll

982aee6a - [AMDGPU] Switch scalarize global loads ON by default

Tue, 04 Jul 2017 17:32:00 +0000

[AMDGPU] Switch scalarize global loads ON by defaultDifferential revision: https://reviews.llvm.org/D34407llvm-svn: 307097 List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll

e4a74137 - Revert r307026, "[AMDGPU] Switch scalarize global loads ON by default"

Tue, 04 Jul 2017 02:14:18 +0000

Revert r307026, "[AMDGPU] Switch scalarize global loads ON by default"It broke a testcase. Failing Tests (1): LLVM :: CodeGen/AMDGPU/alignbit-pat.llllvm-svn: 307054 List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll

ea7f08be - [AMDGPU] Switch scalarize global loads ON by default

Mon, 03 Jul 2017 14:54:11 +0000

[AMDGPU] Switch scalarize global loads ON by defaultDifferential revision: https://reviews.llvm.org/D34407llvm-svn: 307026 List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll

3dbeefa9 - AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernel

Tue, 21 Mar 2017 21:39:51 +0000

AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernelCurrently the default C calling convention functions are treatedthe same as compute kernels. Make this explicit so the defaultcalling convention can be changed to a non-kernel.Converted with perl -pi -e 's/define void/define amdgpu_kernel void/'on the relevant test directories (and undoing in one place that actuallywanted a non-kernel).llvm-svn: 298444 List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll

7aad8fd8 - Enable FeatureFlatForGlobal on Volcanic Islands

Tue, 24 Jan 2017 22:02:15 +0000

Enable FeatureFlatForGlobal on Volcanic IslandsThis switches to the workaround that HSA defaults tofor the mesa path.This should be applied to the 4.0 branch.Patch by Vedran Miletić llvm-svn: 292982 List of files: /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/bswap.ll