diff options
| author | Eli Friedman <efriedma@codeaurora.org> | 2016-12-16 18:44:08 +0000 |
|---|---|---|
| committer | Eli Friedman <efriedma@codeaurora.org> | 2016-12-16 18:44:08 +0000 |
| commit | f624ec27b75701f1227eb03a44f3da84920c798d (patch) | |
| tree | d10b422365e2b461b34e1a53f1f29a72cb5b5b00 /llvm/lib/CodeGen/DetectDeadLanes.cpp | |
| parent | 79b4f0ad9cc5b019cf71bac388f1da1de4dd4e34 (diff) | |
| download | bcm5719-llvm-f624ec27b75701f1227eb03a44f3da84920c798d.tar.gz bcm5719-llvm-f624ec27b75701f1227eb03a44f3da84920c798d.zip | |
[ARM] Add ARMISD::VLD1DUP to match vld1_dup more consistently.
Currently, there are substantial problems forming vld1_dup even if the
VDUP survives legalization. The lack of an actual node
leads to terrible results: not only can we not form post-increment vld1_dup
instructions, but we form scalar pre-increment and post-increment
loads which force the loaded value into a GPR. This patch fixes that
by combining the vdup+load into an ARMISD node before DAGCombine
messes it up.
Also includes a crash fix for vld2_dup (see testcase @vld2dupi8_postinc_variable).
Recommiting with fix to avoid forming vld1dup if the type of the load
doesn't match the type of the vdup (see
https://llvm.org/bugs/show_bug.cgi?id=31404).
Differential Revision: https://reviews.llvm.org/D27694
llvm-svn: 289972
Diffstat (limited to 'llvm/lib/CodeGen/DetectDeadLanes.cpp')
0 files changed, 0 insertions, 0 deletions
