projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 22d275f) | patch
Refine fix to bug 15041.
authorBill Schmidt <wschmidt@linux.vnet.ibm.com>
Fri, 8 Feb 2013 18:19:17 +0000 (18:19 +0000)
committerBill Schmidt <wschmidt@linux.vnet.ibm.com>
Fri, 8 Feb 2013 18:19:17 +0000 (18:19 +0000)
commit62fe7a5b173b29bfeaafdb78977c18000960c2f3
tree0876c8c36e0999d55d32460bb1bac088d96da6b6tree | snapshot
parent22d275f7b80234429957013dbef5e0a10b2b28dacommit | diff
Refine fix to bug 15041.

Thanks to help from Nadav and Hal, I have a more reasonable (and even
correct!) approach.  This specifically penalizes the insertelement
and extractelement operations for the performance hit that will occur
on PowerPC processors.

llvm-svn: 174725
llvm/lib/Target/PowerPC/PPCTargetTransformInfo.cpp diff | blob | history
llvm/test/Analysis/CostModel/PowerPC/insert_extract.ll [new file with mode: 0644] blob
Domain: System / Toolchain;