Sanjay Patel 0e15de2d0c [InstCombine] fold reassociative FP add into start value of fadd reduction ...

This pattern is visible in unrolled and vectorized loops.
Although the backend seems to be able to reassociate to
ideal form in the examples I looked at, we might as well
do that in IR for efficiency.

2021-07-18 06:26:20 -04:00

..

addsub-inseltpoison.ll

…

addsub.ll

…

ctlz-loop.ll

Reapply [InstCombine] Fold multiuse shr eq zero

2021-05-22 14:46:50 +02:00

earlycse-after-simplifycfg-two-entry-phi-node-folding.ll

[SimplifyCFG] Rerun PHI deduplication after common code sinkinkg (PR51092)

2021-07-15 16:34:34 +03:00

horiz-math-inseltpoison.ll

…

horiz-math.ll

…

lit.local.cfg

…

loop-idiom-vs-indvars.ll

…

masked-memory-ops.ll

…

nancvt.ll

…

peel-before-lv-to-enable-vectorization.ll

[PhaseOrdering] Update tests after 23c2f2e6b2.

2021-06-07 10:59:30 +01:00

pixel-splat.ll

[PhaseOrdering] Update tests after 23c2f2e6b2.

2021-06-07 10:59:30 +01:00

pr48844-br-to-switch-vectorization.ll

…

scalarization-inseltpoison.ll

…

scalarization.ll

Update InstCombine to use undef matcher instead

2021-04-18 11:05:36 +09:00

shuffle-inseltpoison.ll

…

shuffle.ll

…

simplifycfg-late.ll

[SimplifyCFG] Tail-merging all blocks with ret terminator

2021-06-24 13:15:39 +03:00

spurious-peeling.ll

[PassManager] Run additional LICM before LoopRotate

2021-04-02 11:11:42 +03:00

SROA-after-loop-unrolling.ll

…

vdiv-nounroll.ll

[PhaseOrdering] Update tests after 23c2f2e6b2.

2021-06-07 10:59:30 +01:00

vdiv.ll

[PhaseOrdering] Update tests after 23c2f2e6b2.

2021-06-07 10:59:30 +01:00

vector-reductions-expanded.ll

[InstCombine] fold reassociative FP add into start value of fadd reduction

2021-07-18 06:26:20 -04:00

vector-reductions-logical.ll

[SLP] avoid leaking poison in reduction of safe boolean logic ops

2021-07-15 17:33:06 -04:00

vector-reductions.ll

[SimplifyCFG] Tail-merging all blocks with ret terminator

2021-06-24 13:15:39 +03:00