diff options
| author | Zi Xuan Wu <wuzish@cn.ibm.com> | 2019-04-12 05:21:31 +0000 |
|---|---|---|
| committer | Zi Xuan Wu <wuzish@cn.ibm.com> | 2019-04-12 05:21:31 +0000 |
| commit | ac79ef8f0ec23aaaf485ec2da4ef7cadd36e3aa6 (patch) | |
| tree | d9af13064f5c4a678e4acd4b85bca94118202ef6 /llvm/lib/CodeGen/MachineFunction.cpp | |
| parent | aa1cad1591bf4fd55b0d7f28e616329c4943e27d (diff) | |
| download | bcm5719-llvm-ac79ef8f0ec23aaaf485ec2da4ef7cadd36e3aa6.tar.gz bcm5719-llvm-ac79ef8f0ec23aaaf485ec2da4ef7cadd36e3aa6.zip | |
[PowerPC] More precise exploitation of P9 maddld instruction when operands are constant
There are 3 operands of maddld, (add (mul %1, %2), %3) and sometimes
they are constant. If there is constant operand, it takes extra li to
materialize the operand, and one more extra register too. So it's not
profitable to use maddld to optimize mul-add pattern.
Differential Revision: https://reviews.llvm.org/D60181
llvm-svn: 358253
Diffstat (limited to 'llvm/lib/CodeGen/MachineFunction.cpp')
0 files changed, 0 insertions, 0 deletions
