diff options
author | David Green <david.green@arm.com> | 2019-08-11 08:53:18 +0000 |
---|---|---|
committer | David Green <david.green@arm.com> | 2019-08-11 08:53:18 +0000 |
commit | 11c4602fce16c22858de764bff21be4160d323af (patch) | |
tree | 9a1d2fa121501e93b09c8e75fe593f172771207b /llvm/lib/CodeGen/MachineBlockPlacement.cpp | |
parent | 44f8d635e20429365782b831ef59ee0903982d02 (diff) | |
download | llvm-11c4602fce16c22858de764bff21be4160d323af.zip llvm-11c4602fce16c22858de764bff21be4160d323af.tar.gz llvm-11c4602fce16c22858de764bff21be4160d323af.tar.bz2 |
[MVE] Don't try to unroll vectorised MVE loops
Due to the nature of the beat system in the MVE architecture, along with tail
predication and low-overhead loops, unrolling has less benefit compared to
normal loops. You can not, for example, hide the latency of a load with other
instructions as you can for scalar code. Preventing unrolling also makes the
code easier to read and reason about.
So if a loop contains vector code, don't enable the runtime unrolling. At least
for the time being.
Differential Revision: https://reviews.llvm.org/D65803
llvm-svn: 368530
Diffstat (limited to 'llvm/lib/CodeGen/MachineBlockPlacement.cpp')
0 files changed, 0 insertions, 0 deletions