[PATCH] D57056: [MC][X86] Correctly model additional operand latency caused by transfer delays from the integer to the floating point unit.
Matt Davis via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Tue Jan 22 10:11:58 PST 2019
mattd added a comment.
LGTM. I'd wait a bit to see if others have anything else to add.
================
Comment at: include/llvm/MC/MCSchedule.h:375
+ /// writes of scheduling class WriteResourceIdx.
+ static unsigned getForwardingDelayCycles(ArrayRef<MCReadAdvanceEntry> Entries,
+ unsigned WriteResourceIdx = 0);
----------------
You can probably constify Entries.
================
Comment at: lib/Target/X86/X86Schedule.td:19
def ReadAfterVecYLd : SchedRead;
+def ReadInt2Fpu : SchedRead;
----------------
I think it would be helpful to have a comment here, describing ReadInt2Fpu. Something similar to what you described above in the Details.
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D57056/new/
https://reviews.llvm.org/D57056
More information about the llvm-commits
mailing list