This is an automated email from the git hooks/post-receive script.
unknown user pushed a change to branch master in repository llvm.
from 22e14e2efdc [DAG] Extend visitSCALAR_TO_VECTOR optimization to truncate [...] new 20e83eb1936 Revert "[SLP] General improvements of SLP vectorization process." new 2e48864110d AMDGPU: Cleanup subtarget features new aa4edb14222 Fix typo in comment
The 3 revisions listed above as "new" are entirely new to this repository and will be described in separate emails. The revisions listed as "adds" were already present in the repository and have only been added to this reference.
Summary of changes: include/llvm/Analysis/TargetTransformInfoImpl.h | 4 +- include/llvm/Transforms/Vectorize/SLPVectorizer.h | 13 -- lib/Target/AMDGPU/AMDGPU.td | 12 +- lib/Target/AMDGPU/AMDGPUSubtarget.cpp | 15 +- lib/Target/AMDGPU/AMDGPUSubtarget.h | 1 - lib/Target/AMDGPU/AMDGPUTargetMachine.cpp | 3 +- lib/Target/AMDGPU/FLATInstructions.td | 17 +- lib/Target/AMDGPU/Processors.td | 6 + lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp | 15 +- lib/Transforms/Vectorize/SLPVectorizer.cpp | 215 ++++++++++----------- test/CodeGen/AMDGPU/addrspacecast.ll | 2 +- test/CodeGen/AMDGPU/array-ptr-calc-i32.ll | 4 +- test/CodeGen/AMDGPU/bitcast-vector-extract.ll | 2 +- test/CodeGen/AMDGPU/br_cc.f16.ll | 2 +- test/CodeGen/AMDGPU/branch-relaxation.ll | 2 +- test/CodeGen/AMDGPU/callee-special-input-sgprs.ll | 2 +- test/CodeGen/AMDGPU/callee-special-input-vgprs.ll | 2 +- test/CodeGen/AMDGPU/captured-frame-index.ll | 2 +- test/CodeGen/AMDGPU/collapse-endcf.ll | 2 +- test/CodeGen/AMDGPU/ctpop.ll | 2 +- test/CodeGen/AMDGPU/ctpop64.ll | 2 +- test/CodeGen/AMDGPU/cvt_f32_ubyte.ll | 32 +-- test/CodeGen/AMDGPU/fabs.f16.ll | 2 +- test/CodeGen/AMDGPU/fadd.f16.ll | 2 +- test/CodeGen/AMDGPU/fcopysign.f16.ll | 6 +- test/CodeGen/AMDGPU/fence-amdgiz.ll | 3 +- test/CodeGen/AMDGPU/fmax3.f64.ll | 2 +- test/CodeGen/AMDGPU/fmul.f16.ll | 4 +- test/CodeGen/AMDGPU/fpext.f16.ll | 6 +- test/CodeGen/AMDGPU/fptosi.f16.ll | 4 +- test/CodeGen/AMDGPU/fptoui.f16.ll | 4 +- test/CodeGen/AMDGPU/frame-index-amdgiz.ll | 5 +- test/CodeGen/AMDGPU/frame-index-elimination.ll | 2 +- test/CodeGen/AMDGPU/fsub.f16.ll | 6 +- test/CodeGen/AMDGPU/half.ll | 4 +- .../CodeGen/AMDGPU/indirect-addressing-si-noopt.ll | 2 +- test/CodeGen/AMDGPU/indirect-addressing-si.ll | 8 +- test/CodeGen/AMDGPU/insert_vector_elt.ll | 2 +- test/CodeGen/AMDGPU/llvm.ceil.f16.ll | 4 +- test/CodeGen/AMDGPU/llvm.cos.f16.ll | 4 +- test/CodeGen/AMDGPU/llvm.exp2.f16.ll | 4 +- test/CodeGen/AMDGPU/llvm.floor.f16.ll | 4 +- test/CodeGen/AMDGPU/llvm.fma.f16.ll | 4 +- test/CodeGen/AMDGPU/llvm.fmuladd.f16.ll | 8 +- test/CodeGen/AMDGPU/llvm.log2.f16.ll | 4 +- test/CodeGen/AMDGPU/llvm.maxnum.f16.ll | 4 +- test/CodeGen/AMDGPU/llvm.memcpy.ll | 2 +- test/CodeGen/AMDGPU/llvm.minnum.f16.ll | 4 +- test/CodeGen/AMDGPU/llvm.rint.f16.ll | 6 +- test/CodeGen/AMDGPU/llvm.round.ll | 2 +- test/CodeGen/AMDGPU/llvm.sin.f16.ll | 4 +- test/CodeGen/AMDGPU/llvm.sqrt.f16.ll | 4 +- test/CodeGen/AMDGPU/llvm.trunc.f16.ll | 4 +- .../AMDGPU/macro-fusion-cluster-vcc-uses.mir | 2 +- test/CodeGen/AMDGPU/merge-stores.ll | 4 +- .../AMDGPU/promote-alloca-mem-intrinsics.ll | 2 +- test/CodeGen/AMDGPU/promote-alloca-no-opts.ll | 4 +- .../AMDGPU/promote-alloca-padding-size-estimate.ll | 2 +- test/CodeGen/AMDGPU/s_addk_i32.ll | 2 +- test/CodeGen/AMDGPU/s_mulk_i32.ll | 2 +- test/CodeGen/AMDGPU/sad.ll | 2 +- test/CodeGen/AMDGPU/schedule-kernel-arg-loads.ll | 2 +- test/CodeGen/AMDGPU/schedule-regpressure-limit2.ll | 4 +- test/CodeGen/AMDGPU/select-vectors.ll | 6 +- test/CodeGen/AMDGPU/select.f16.ll | 4 +- test/CodeGen/AMDGPU/setcc-fneg-constant.ll | 2 +- test/CodeGen/AMDGPU/shift-and-i128-ubfe.ll | 2 +- test/CodeGen/AMDGPU/shift-and-i64-ubfe.ll | 3 +- test/CodeGen/AMDGPU/shl_add_constant.ll | 4 +- test/CodeGen/AMDGPU/sitofp.f16.ll | 4 +- test/CodeGen/AMDGPU/trunc.ll | 6 +- test/CodeGen/AMDGPU/udivrem.ll | 2 +- test/CodeGen/AMDGPU/uitofp.f16.ll | 4 +- test/CodeGen/AMDGPU/v_mac_f16.ll | 2 +- test/CodeGen/AMDGPU/vector-extract-insert.ll | 4 +- .../SLPVectorizer/AArch64/gather-root.ll | 100 +++++----- test/Transforms/SLPVectorizer/X86/horizontal.ll | 64 +++--- .../X86/insert-element-build-vector.ll | 40 ++-- 78 files changed, 371 insertions(+), 383 deletions(-)