summaryrefslogtreecommitdiffstats
path: root/llvm/test/CodeGen/AMDGPU/fcopysign.f16.ll
Commit message (Collapse)AuthorAgeFilesLines
* AMDGPU: Add pass to lower kernel arguments to loadsMatt Arsenault2018-06-261-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This replaces most argument uses with loads, but for now not all. The code in SelectionDAG for calling convention lowering is actively harmful for amdgpu_kernel. It attempts to split the argument types into register legal types, which results in low quality code for arbitary types. Since all kernel arguments are passed in memory, we just want the raw types. I've tried a couple of methods of mitigating this in SelectionDAG, but it's easier to just bypass this problem alltogether. It's possible to hack around the problem in the initial lowering, but the real problem is the DAG then expects to be able to use CopyToReg/CopyFromReg for uses of the arguments outside the block. Exposing the argument loads in the IR also has the advantage that the LoadStoreVectorizer can merge them. I'm not sure the best approach to dealing with the IR argument list is. The patch as-is just leaves the IR arguments in place, so all the existing code will still compute the same kernarg size and pointlessly lowers the arguments. Arguably the frontend should emit kernels with an empty argument list in the first place. Alternatively a dummy array could be inserted as a single argument just to reserve space. This does have some disadvantages. Local pointer kernel arguments can no longer have AssertZext placed on them as the equivalent !range metadata is not valid on pointer typed loads. This is mostly bad for SI which needs to know about the known bits in order to use the DS instruction offset, so in this case this is not done. More importantly, this skips noalias arguments since this pass does not yet convert this to the equivalent !alias.scope and !noalias metadata. Producing this metadata correctly seems to be tricky, although this logically is the same as inlining into a function which doesn't exist. Additionally, exposing these loads to the vectorizer may result in degraded aliasing information if a pointer load is merged with another argument load. I'm also not entirely sure this is preserving the current clover ABI, although I would greatly prefer if it would stop widening arguments and match the HSA ABI. As-is I think it is extending < 4-byte arguments to 4-bytes but doesn't align them to 4-bytes. llvm-svn: 335650
* AMDGPU/GCN: Bring processors in sync with AMDGPUUsageKonstantin Zhuravlyov2017-12-081-1/+1
| | | | | | | | | | | | - Add gfx704 - Change bonaire to gfx704 - Remove gfx804 - Remove gfx901 - Remove gfx903 Differential Revision: https://reviews.llvm.org/D40046 llvm-svn: 320194
* [AMDGPU] Prevent post-RA scheduler from breaking memory clausesStanislav Mekhanoshin2017-09-191-2/+2
| | | | | | | | | The pre-RA scheduler does load/store clustering, but post-RA scheduler undoes it. Add mutation to prevent it. Differential Revision: https://reviews.llvm.org/D38014 llvm-svn: 313670
* AMDGPU: Cleanup subtarget featuresMatt Arsenault2017-08-071-3/+3
| | | | | | | | | | | | Try to avoid mutually exclusive features. Don't use a real default GPU, and use a fake "generic". The goal is to make it easier to see which set of features are incompatible between feature strings. Most of the test changes are due to random scheduling changes from not having a default fullspeed model. llvm-svn: 310258
* AMDGPU: Start selecting global instructionsMatt Arsenault2017-07-291-14/+14
| | | | llvm-svn: 309470
* [AMDGPU] Switch scalarize global loads ON by defaultAlexander Timofeev2017-07-041-34/+57
| | | | | | Differential revision: https://reviews.llvm.org/D34407 llvm-svn: 307097
* Revert r307026, "[AMDGPU] Switch scalarize global loads ON by default"NAKAMURA Takumi2017-07-041-57/+34
| | | | | | | | | It broke a testcase. Failing Tests (1): LLVM :: CodeGen/AMDGPU/alignbit-pat.ll llvm-svn: 307054
* [AMDGPU] Switch scalarize global loads ON by defaultAlexander Timofeev2017-07-031-34/+57
| | | | | | Differential revision: https://reviews.llvm.org/D34407 llvm-svn: 307026
* AMDGPU: Replace fp16SrcZerosHighBits with a whitelistMatt Arsenault2017-04-061-20/+21
| | | | | | | FCOPYSIGN is lowered to bit operations which don't clear the high bits. llvm-svn: 299708
* AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernelMatt Arsenault2017-03-211-12/+12
| | | | | | | | | | | | Currently the default C calling convention functions are treated the same as compute kernels. Make this explicit so the default calling 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 actually wanted a non-kernel). llvm-svn: 298444
* [AMDGPU] Remove getBidirectionalReasonRankStanislav Mekhanoshin2017-03-111-3/+3
| | | | | | | | | | | | | | This method inverts the Reason field of a scheduling candidate. It does right comparison between RegCritical and RegExcess, but everything else is broken. In fact it can prefer less strong reason such as Weak over RegCritical because Weak > -RegCritical. The CandReason enum is properly sorted, so just remove artificial ranking. Differential Revision: https://reviews.llvm.org/D30557 llvm-svn: 297536
* AMDGPU: Support v2i16/v2f16 packed operationsMatt Arsenault2017-02-271-14/+15
| | | | llvm-svn: 296396
* Enable FeatureFlatForGlobal on Volcanic IslandsMatt Arsenault2017-01-241-1/+1
| | | | | | | | | | | This switches to the workaround that HSA defaults to for the mesa path. This should be applied to the 4.0 branch. Patch by Vedran Miletić <vedran@miletic.net> llvm-svn: 292982
* [AMDGPU] Implement f16 fcopysign and fcopysign(f32, f64)Konstantin Zhuravlyov2017-01-131-0/+262
Differential Revision: https://reviews.llvm.org/D28496 llvm-svn: 291954
OpenPOWER on IntegriCloud