summaryrefslogtreecommitdiffstats
path: root/llvm/lib/CodeGen/MachineCombiner.cpp
diff options
context:
space:
mode:
authorSanjay Patel <spatel@rotateright.com>2015-11-10 19:24:31 +0000
committerSanjay Patel <spatel@rotateright.com>2015-11-10 19:24:31 +0000
commitaf1b48bfdcc4cf1602dca45fe2511a426f48879e (patch)
tree52be4df7ebb4815e028d6745789aeb7220867e5e /llvm/lib/CodeGen/MachineCombiner.cpp
parent8cf44d9684a7c1e24246eee7184c7bc0d4c4bd54 (diff)
downloadbcm5719-llvm-af1b48bfdcc4cf1602dca45fe2511a426f48879e.tar.gz
bcm5719-llvm-af1b48bfdcc4cf1602dca45fe2511a426f48879e.zip
[ARM] add overrides for isCheapToSpeculateCttz() and isCheapToSpeculateCtlz()
ARM V6T2 has instructions for efficient count-leading/trailing-zeros, so this should be considered a cheap operation (and therefore fair game for speculation) for any ARM V6T2 implementation. The net result of allowing this speculation for the regression tests in this patch is that we get this code: ctlz: clz r0, r0 bx lr cttz: rbit r0, r0 clz r0, r0 bx lr Instead of: ctlz: cmp r0, #0 moveq r0, #32 clzne r0, r0 bx lr cttz: cmp r0, #0 moveq r0, #32 rbitne r0, r0 clzne r0, r0 bx lr This will help solve a general speculation/despeculation problem noted in PR24818: https://llvm.org/bugs/show_bug.cgi?id=24818 Differential Revision: http://reviews.llvm.org/D14469 llvm-svn: 252639
Diffstat (limited to 'llvm/lib/CodeGen/MachineCombiner.cpp')
0 files changed, 0 insertions, 0 deletions
OpenPOWER on IntegriCloud