projects / gcc.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: fe74f9b) | patch
re PR tree-optimization/84037 (Speed regression of polyhedron benchmark since r256644)
authorRichard Biener <rguenther@suse.de>
Fri, 16 Feb 2018 13:47:25 +0000 (13:47 +0000)
committerRichard Biener <rguenth@gcc.gnu.org>
Fri, 16 Feb 2018 13:47:25 +0000 (13:47 +0000)
commitbe77ba2a461eefdf4a2676b19025f36ec092c598
tree7a02106de91bc43d8e50c5adb8b842c3ba9b3fc1tree
parentfe74f9b4ed572a3bb2f8a110ea178578df8e0563commit | diff
re PR tree-optimization/84037 (Speed regression of polyhedron benchmark since r256644)

2018-02-16  Richard Biener  <rguenther@suse.de>

PR tree-optimization/84037
PR tree-optimization/84016
PR target/82862
* config/i386/i386.c (ix86_builtin_vectorization_cost):
Adjust vec_construct for the fact we need additional higher latency
128bit inserts for AVX256 and AVX512 vector builds.
(ix86_add_stmt_cost): Scale vector construction cost for
elementwise loads.

From-SVN: r257734
gcc/ChangeLog diff | blob | history
gcc/config/i386/i386.c diff | blob | history