maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
262,225 Commits 23 Branches 0 Tags 2.2 GiB
Commit Graph

3 Commits

Author SHA1 Message Date
Simon Pilgrim ef46c2762a [x86, SSE] AVX1 PR28129 (256-bit all-ones rematerialization) 
Further perf tests on Jaguar indicate that:

vxorps  %ymm0, %ymm0, %ymm0
vcmpps  $15, %ymm0, %ymm0, %ymm0

is consistently faster (by about 9%) than:

vpcmpeqd  %xmm0, %xmm0, %xmm0
vinsertf128  $1, %xmm0, %ymm0, %ymm0

Testing equivalent code on a SandyBridge (E5-2640) puts it slightly (~3%) faster as well.

Committed on behalf of @dtemirbulatov

Differential Revision: https://reviews.llvm.org/D32416

llvm-svn: 302989
 2017-05-13 13:42:35 +00:00
Simon Pilgrim 380ce75687 [X86][SSE] Replace insert_vector_elt(vec, -1, idx) with shuffle 
Similar to what we already do for zero elt insertion, we can quickly rematerialize 'allbits' vectors so to avoid a unnecessary gpr value and insertion into a vector

llvm-svn: 294162
 2017-02-05 22:50:29 +00:00
Simon Pilgrim 2bec3c2f4b [X86][AVX] Add 8i32 -> 8f32 sitofp tests with constant insertion 
Some elements are constant inserted into the source integer vector before conversion.

llvm-svn: 294147
 2017-02-05 21:40:25 +00:00