OSchip
/
llvm-project
mirror of https://github.com/THU-DSP-LAB/llvm-project.git
13
0
Fork 6
Code Issues Packages Projects Releases Wiki Activity
234,982 Commits 32 Branches 0 Tags 2.2 GiB
Commit Graph

4 Commits

Author SHA1 Message Date
Michael Kuperstein 3a3c64d23e [LV] For some IVs, use vector phis instead of widening in the loop body 
Previously, whenever we needed a vector IV, we would create it on the fly,
by splatting the scalar IV and adding a step vector. Instead, we can create a
real vector IV. This tends to save a couple of instructions per iteration.

This only changes the behavior for the most basic case - integer primary
IVs with a constant step.

Differential Revision: http://reviews.llvm.org/D20315

llvm-svn: 271410
 2016-06-01 17:16:46 +00:00
Elena Demikhovsky 2b06b0fe2a LoopVectorizer - skip 'bitcast' between GEP and load. 
Skipping 'bitcast' in this case allows to vectorize load:

  %arrayidx = getelementptr inbounds double*, double** %in, i64 %indvars.iv
  %tmp53 = bitcast double** %arrayidx to i64*
  %tmp54 = load i64, i64* %tmp53, align 8

Differential Revision http://reviews.llvm.org/D14112

llvm-svn: 251907
 2015-11-03 10:29:34 +00:00
NAKAMURA Takumi 7ef7293b40 Revert r251291, "Loop Vectorizer - skipping "bitcast" before GEP" 
It causes miscompilation of llvm/lib/ExecutionEngine/Interpreter/Execution.cpp.
See also PR25324.

llvm-svn: 251436
 2015-10-27 19:02:36 +00:00
Elena Demikhovsky 7a77149391 Loop Vectorizer - skipping "bitcast" before GEP 
Vectorization of memory instruction (Load/Store) is possible when the pointer is coming from GEP. The GEP analysis allows to estimate the profit.
In some cases we have a "bitcast" between GEP and memory instruction.
I added code that skips the "bitcast".

http://reviews.llvm.org/D13886

llvm-svn: 251291
 2015-10-26 13:42:41 +00:00