shylie/llvm-project
1
0
Fork 0
Code Issues Pull Requests Actions 6 Packages Projects Releases Wiki Activity
llvm-project/llvm/test/Transforms/LoopUnroll
History
Kunqiu Chen 85e07bad93
[InstructionSimplify] Extend simplifyICmpWithZero to handle equivalent zero RHS (#179055) 
Add a new helper function `matchEquivZeroRHS()` that recognizes
comparisons with constants that are equivalent to comparisons with zero,
and transforms the predicate accordingly.

This handles the following transformations:
- icmp sgt X, -1 --> icmp sge X, 0
- icmp sle X, -1 --> icmp slt X, 0
- icmp [us]ge X, 1 --> icmp [us]gt X, 0
- icmp [us]lt X, 1 --> icmp [us]le X, 0

This enables more optimization opportunities in `simplifyICmpWithZero`,
such as folding icmp sgt X, -1 when X is known to be non-negative.

---

- IR Impact: https://github.com/dtcxzyw/llvm-opt-benchmark/pull/3414
2026-02-13 00:06:32 +08:00
..
AArch64
[IR] Reland Optimize PHINode::removeIncomingValue() and PHINode::removeIncomingValueIf() to use the swapping strategy. (#174274)
2026-01-04 09:24:53 +08:00
AMDGPU
[LoopUnroll] Skip remainder loop guard if skip unrolled loop (#156549)
2025-10-07 10:45:49 -04:00
ARM
[IR] Replace alignment argument with attribute on masked intrinsics (#163802)
2025-10-20 08:50:09 +00:00
branch-weights-freq
[LoopUnroll][NFC] Simplify recent block frequency tests (#177025)
2026-01-21 10:40:10 -05:00
Hexagon
[LoopUnroll] Skip remainder loop guard if skip unrolled loop (#156549)
2025-10-07 10:45:49 -04:00
PowerPC
[LoopUnroll] Skip remainder loop guard if skip unrolled loop (#156549)
2025-10-07 10:45:49 -04:00
RISCV
[RISCV] Don't unroll vectorized loops with vector operands (#171089)
2025-12-09 12:42:41 +08:00
WebAssembly
[InstCombine] fold icmp ne (and X, 1), 0 --> trunc X to i1 (#178977)
2026-02-03 19:14:27 +01:00
X86
[RemoveDIs] Remove "try-debuginfo-iterators..." test flags (#130298)
2025-03-14 15:50:49 +00:00
2004-05-13-DontUnrollTooMuch.ll
…
2005-03-06-BadLoopInfoUpdate.ll
…
2006-08-24-MultiBlockLoop.ll
…
2007-04-16-PhiUpdate.ll
…
2007-05-05-UnrollMiscomp.ll
…
2007-05-09-UnknownTripCount.ll
…
2007-11-05-Crash.ll
…
2011-08-08-PhiUpdate.ll
…
2011-08-09-IVSimplify.ll
…
2011-08-09-PhiUpdate.ll
…
2011-10-01-NoopTrunc.ll
…
2012-04-09-unroll-indirectbr.ll
…
add-folding-overflow.ll
…
assume-timeout.ll
…
basic.ll
…
callbr.ll
…
complete_unroll_profitability_with_assume.ll
…
convergent.controlled.ll
[LoopUnroll] Do not copy !llvm.loop from latch to non-latch (#165635)
2026-01-14 10:40:31 -05:00
convergent.ll
…
dce.ll
…
debug-info.ll
[RemoveDIs] Remove "try-debuginfo-iterators..." test flags (#130298)
2025-03-14 15:50:49 +00:00
debug.ll
[LoopUnrollPass] Indent LLVM_DEBUG() messages based on our depth in the tryToUnrollLoop() call graph (#178945)
2026-02-11 17:59:42 +00:00
disable_nonforced_count.ll
…
disable_nonforced_enable.ll
…
disable_nonforced_full.ll
…
disable_nonforced.ll
…
disable-full-unroll-by-opt.ll
…
disable-loop-unrolling_forced.ll
…
ephemeral.ll
…
epilog_const_phi.ll
…
followup.ll
[LoopUnroll] Do not copy !llvm.loop from latch to non-latch (#165635)
2026-01-14 10:40:31 -05:00
full-unroll-avoid-partial.ll
[LoopUnroll] Prevent LoopFullUnrollPass from performing partial unrolling when trip counts are unknown (#165013)
2025-11-04 09:20:01 -08:00
full-unroll-bad-cost.ll
…
full-unroll-crashers.ll
…
full-unroll-heuristics-2.ll
…
full-unroll-heuristics-cmp.ll
…
full-unroll-heuristics-dce.ll
…
full-unroll-heuristics-geps.ll
…
full-unroll-heuristics-load-const.ll
…
full-unroll-heuristics-phi-prop.ll
…
full-unroll-heuristics.ll
…
full-unroll-invariant.ll
…
full-unroll-keep-first-exit.ll
…
full-unroll-one-unpredictable-exit.ll
[IR] Reland Optimize PHINode::removeIncomingValue() and PHINode::removeIncomingValueIf() to use the swapping strategy. (#174274)
2026-01-04 09:24:53 +08:00
FullUnroll.ll
…
gh-issue77118-broken-lcssa-form.ll
…
guard-cost-for-unrolling.ll
…
high-cost-trip-count-computation.ll
…
ignore-annotation-intrinsic-cost.ll
…
invalidate_right_loop.ll
…
loop-block-disposition-after-full-unroll.ll
…
loop-block-disposition-after-loop-peeling.ll
…
loop-branch-folding.ll
…
loop-probability-one.ll
[LoopUnroll] Fix division by zero (#166258)
2025-11-04 12:49:33 -05:00
loop-remarks-with-hotness.ll
…
loop-remarks.ll
…
multiple-exits.ll
…
noalias.ll
…
nonlatchcondbr.ll
…
not-rotated.ll
…
opt-levels.ll
…
optsize-loop-size.ll
…
partial-unroll-const-bounds.ll
…
partial-unroll-dead-instructions.ll
…
partial-unroll-non-latch-exit.ll
[IR] Reland Optimize PHINode::removeIncomingValue() and PHINode::removeIncomingValueIf() to use the swapping strategy. (#174274)
2026-01-04 09:24:53 +08:00
partial-unroll-reductions.ll
[LV] Vectorize conditional scalar assignments (#158088)
2026-01-14 14:59:18 +00:00
partially-unroll-unconditional-latch.ll
…
peel-branch-weights.ll
[LoopPeel] Fix branch weights' effect on block frequencies (#128785)
2025-10-02 16:07:55 +00:00
peel-last-iteration-debug.ll
[LoopPeel] Add peeling tests with debug value and pointer inductions
2025-05-28 10:07:02 +01:00
peel-last-iteration-expansion-cost.ll
[LP] Assign weights when peeling last iteration. (#166858)
2025-11-15 10:01:04 -08:00
peel-last-iteration-minmax.ll
[LoopPeel] Support last iteration peeling of min/max intrinsics (#143598)
2025-06-17 11:22:23 -07:00
peel-last-iteration-multi-exit.ll
[LoopPeel] Add tests for peeling last iteration with multiple exits.
2025-05-23 15:46:34 +01:00
peel-last-iteration-pointer-inductions.ll
[LoopPeel] Add peeling tests with debug value and pointer inductions
2025-05-28 10:07:02 +01:00
peel-last-iteration-with-constant-trip-count.ll
Reapply "[LoopPeel] Remove known trip count restriction when peeling last. (#140792)"
2025-05-28 13:02:03 +01:00
peel-last-iteration-with-guards.ll
[SCEV] Add tests that benefit from rewriting SCEVAddExpr with guards.
2025-09-20 19:24:19 +01:00
peel-last-iteration-with-variable-trip-count.ll
[LoopPeel] Add test coverage for edge case for peel last
2025-06-10 11:46:06 -07:00
peel-loop2.ll
…
peel-loop-and-unroll.ll
…
peel-loop-conditions-pgo-1.ll
…
peel-loop-conditions-pgo-2.ll
…
peel-loop-conditions.ll
…
peel-loop-inner.ll
…
peel-loop-irreducible.ll
…
peel-loop-min-max-intrinsics.ll
…
peel-loop-negative.ll
…
peel-loop-nests.ll
…
peel-loop-noalias-scope-decl.ll
…
peel-loop-not-forced.ll
…
peel-loop-pgo-deopt-idom-2.ll
…
peel-loop-pgo-deopt-idom.ll
…
peel-loop-pgo-deopt.ll
[LoopPeel] Fix branch weights' effect on block frequencies (#128785)
2025-10-02 16:07:55 +00:00
peel-loop-pgo.ll
[LoopPeel] Fix branch weights' effect on block frequencies (#128785)
2025-10-02 16:07:55 +00:00
peel-loop-phi-analysis-iv-not-triggered.ll
[LoopPeel] Add new option to peeling loops to convert PHI into IV (#121104)
2025-08-20 13:44:56 +00:00
peel-loop-phi-analysis-iv.ll
[LoopUnroll] Remove preceding whitespace in loop peeling optimization remark (#178951)
2026-02-02 09:32:57 -08:00
peel-loop-phi-analysis.ll
…
peel-loop-scev-invalidate-with-overflow-inst.ll
…
peel-loop-scev-invalidate.ll
…
peel-loop.ll
[IR] Reland Optimize PHINode::removeIncomingValue() and PHINode::removeIncomingValueIf() to use the swapping strategy. (#174274)
2026-01-04 09:24:53 +08:00
peel-multiple-unreachable-exits.ll
[IndVars] Strengthen inference of samesign flags (#170363)
2025-12-03 16:16:22 +00:00
peel-to-turn-invariant-accesses-dereferenceable.ll
[LoopPeel] Check for onlyAccessesInaccessibleMemory instead of llvm.assume in peelToTurnInvariantLoadsDereferenceable. (#171910)
2025-12-12 10:45:41 -08:00
pr10813.ll
…
pr11361.ll
…
pr14167.ll
…
pr18861.ll
…
pr27157.ll
…
pr28132.ll
…
pr31718.ll
…
pr33437.ll
…
pr45939-peel-count-and-complete-unroll.ll
…
pr77842.ll
…
pr109333.ll
…
pr114879.ll
…
pr131465.ll
[LoopUnroll] Do not copy !llvm.loop from latch to non-latch (#165635)
2026-01-14 10:40:31 -05:00
preserve-branch-debuglocs.ll
…
rebuild_lcssa.ll
…
revert-D110922.ll
…
revisit.ll
…
runtime-epilog-debuginfo.ll
[test][Transforms] Remove unsafe-fp-math uses part 2 (NFC) (#164786)
2025-10-23 20:31:31 +08:00
runtime-exit-phi-scev-invalidation.ll
[LoopUnroll] Fix block frequencies for epilogue (#159163)
2025-10-31 11:01:42 -04:00
runtime-i128.ll
[LoopUnroll] Skip remainder loop guard if skip unrolled loop (#156549)
2025-10-07 10:45:49 -04:00
runtime-li.ll
…
runtime-loop1.ll
[LoopUnroll] Skip remainder loop guard if skip unrolled loop (#156549)
2025-10-07 10:45:49 -04:00
runtime-loop2.ll
[LoopUnroll] Skip remainder loop guard if skip unrolled loop (#156549)
2025-10-07 10:45:49 -04:00
runtime-loop3.ll
…
runtime-loop4.ll
…
runtime-loop5.ll
[LoopUnroll] Skip remainder loop guard if skip unrolled loop (#156549)
2025-10-07 10:45:49 -04:00
runtime-loop-at-most-two-exits.ll
[LoopUnroll] Skip remainder loop guard if skip unrolled loop (#156549)
2025-10-07 10:45:49 -04:00
runtime-loop-branchweight.ll
[LoopUnroll] Fix block frequencies for epilogue (#159163)
2025-10-31 11:01:42 -04:00
runtime-loop-known-exit.ll
…
runtime-loop-multiexit-dom-verify.ll
[IndVars] Strengthen inference of samesign flags (#170363)
2025-12-03 16:16:22 +00:00
runtime-loop-multiple-exits.ll
[InstructionSimplify] Extend simplifyICmpWithZero to handle equivalent zero RHS (#179055)
2026-02-13 00:06:32 +08:00
runtime-loop-non-exiting-latch.ll
…
runtime-loop.ll
[LoopUnroll] Fix block frequencies for epilogue (#159163)
2025-10-31 11:01:42 -04:00
runtime-multiexit-heuristic.ll
[LoopUnroll] Skip remainder loop guard if skip unrolled loop (#156549)
2025-10-07 10:45:49 -04:00
runtime-small-upperbound.ll
…
runtime-unroll-assume-no-remainder.ll
[LoopUnroll] Skip remainder loop guard if skip unrolled loop (#156549)
2025-10-07 10:45:49 -04:00
runtime-unroll-reductions.ll
[LoopUnroll] Introduce parallel accumulators when unrolling FP reductions. (#166630)
2025-11-27 15:03:36 +00:00
runtime-unroll-remainder.ll
[LoopUnroll] Skip remainder loop guard if skip unrolled loop (#156549)
2025-10-07 10:45:49 -04:00
scev-dispositions-type-not-scevable.ll
…
scev-forget-exit-phis.ll
…
scev-invalidation-lcssa.ll
[LoopUnroll] Skip remainder loop guard if skip unrolled loop (#156549)
2025-10-07 10:45:49 -04:00
scevunroll.ll
[IR] Reland Optimize PHINode::removeIncomingValue() and PHINode::removeIncomingValueIf() to use the swapping strategy. (#174274)
2026-01-04 09:24:53 +08:00
shifted-tripcount.ll
…
tripcount-overflow.ll
[LoopUnroll] Skip remainder loop guard if skip unrolled loop (#156549)
2025-10-07 10:45:49 -04:00
unloop.ll
…
unroll-after-peel.ll
…
unroll-and-peel-last-iteration.ll
[LoopPeel] Handle non-local instructions/arguments when updating exiting values (#142993)
2025-06-06 12:56:28 +08:00
unroll-cleanup.ll
Revert "[PGO] Add llvm.loop.estimated_trip_count metadata" (#151585)
2025-07-31 15:56:31 -04:00
unroll-cleanuppad.ll
…
unroll-cost-symbolic-execute.ll
…
unroll-count.ll
…
unroll-dont-copy-latch-loop-id.ll
[LoopUnroll] Do not copy !llvm.loop from latch to non-latch (#165635)
2026-01-14 10:40:31 -05:00
unroll-header-exiting-with-phis-multiple-exiting-blocks.ll
[IR] Reland Optimize PHINode::removeIncomingValue() and PHINode::removeIncomingValueIf() to use the swapping strategy. (#174274)
2026-01-04 09:24:53 +08:00
unroll-header-exiting-with-phis.ll
[IndVars] Strengthen inference of samesign flags (#170363)
2025-12-03 16:16:22 +00:00
unroll-heuristics-pgo.ll
[LoopUnroll] Fix block frequencies for epilogue (#159163)
2025-10-31 11:01:42 -04:00
unroll-loads-cse.ll
[LoopUnroll] Do not copy !llvm.loop from latch to non-latch (#165635)
2026-01-14 10:40:31 -05:00
unroll-loop-invalidation.ll
…
unroll-maxcount.ll
…
unroll-multi-exit-loop-heuristics.ll
[NFC][LoopUnroll] Add -unroll-runtime-other-exit-predictable=false to unroll-multi-exit-loop-heuristics.ll (#179198)
2026-02-02 11:11:41 -05:00
unroll-opt-attribute.ll
…
unroll-pragmas-disabled.ll
…
unroll-pragmas.ll
…
unroll-preserve-scev-lcssa.ll
…
unroll-remove-redundant-dbg.ll
…
unroll-unconditional-latch.ll
…
update-loop-info-in-subloops.ll
…
verify-dominators.ll
…
wrong_assert_in_peeling.ll
…
zeroed-branch-weights.ll
[LoopUnroll] Fix assert fail on zeroed branch weights (#165938)
2025-11-03 10:19:12 -05:00