diff options
| author | Jonas Paulsson <paulsson@linux.vnet.ibm.com> | 2017-04-12 12:41:37 +0000 |
|---|---|---|
| committer | Jonas Paulsson <paulsson@linux.vnet.ibm.com> | 2017-04-12 12:41:37 +0000 |
| commit | da74ed42dadbbf62644fb8fe0ff1e7885c7b2a04 (patch) | |
| tree | 5ead03687c900acdb6bf62c7c34fa34f62b03480 /llvm/lib/Target | |
| parent | 12194e9bec6f5c4dc887659983764e6cfb708205 (diff) | |
| download | bcm5719-llvm-da74ed42dadbbf62644fb8fe0ff1e7885c7b2a04.tar.gz bcm5719-llvm-da74ed42dadbbf62644fb8fe0ff1e7885c7b2a04.zip | |
[LoopVectorizer, TTI] New method supportsEfficientVectorElementLoadStore()
Since SystemZ supports vector element load/store instructions, there is no
need for extracts/inserts if a vector load/store gets scalarized.
This patch lets Target specify that it supports such instructions by means of
a new TTI hook that defaults to false.
The use for this is in the LoopVectorizer getScalarizationOverhead() method,
which will with this patch produce a smaller sum for a vector load/store on
SystemZ.
New test: test/Transforms/LoopVectorize/SystemZ/load-store-scalarization-cost.ll
Review: Adam Nemet
https://reviews.llvm.org/D30680
llvm-svn: 300056
Diffstat (limited to 'llvm/lib/Target')
| -rw-r--r-- | llvm/lib/Target/SystemZ/SystemZTargetTransformInfo.h | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/llvm/lib/Target/SystemZ/SystemZTargetTransformInfo.h b/llvm/lib/Target/SystemZ/SystemZTargetTransformInfo.h index d2639cb271d..3766ed45b8c 100644 --- a/llvm/lib/Target/SystemZ/SystemZTargetTransformInfo.h +++ b/llvm/lib/Target/SystemZ/SystemZTargetTransformInfo.h @@ -55,6 +55,7 @@ public: unsigned getNumberOfRegisters(bool Vector); unsigned getRegisterBitWidth(bool Vector); + bool supportsEfficientVectorElementLoadStore() { return true; } bool enableInterleavedAccessVectorization() { return true; } int getArithmeticInstrCost( |
