aboutsummaryrefslogtreecommitdiff
path: root/lib/Transforms/Scalar
AgeCommit message (Collapse)Author
2010-04-15Fix a bunch of namespace polution.Dan Gohman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101376 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-15back out r101364, as it trips the linux nightlybot on some clang C++ testsGabor Greif
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101368 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-15rotate CallInst operands, i.e. move callee to the backGabor Greif
of the operand array the motivation for this patch are laid out in my mail to llvm-commits: more efficient access to operands and callee, faster callgraph-construction, smaller compiler binary git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101364 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-14performance: cache the dereferenced use_iteratorGabor Greif
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101253 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-14performance: cache the dereferenced use_iteratorGabor Greif
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101250 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-13Remove SCCVN from the CMake build system.Owen Anderson
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101125 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-13SCCVN, we hardly knew ye!Owen Anderson
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101117 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-13Teach IndVarSimplify how to eliminate remainder operators where theDan Gohman
numerator is an induction variable. For example, with code like this: for (i=0;i<n;++i) x[i%n] = 0; IndVarSimplify will now recognize that i is always less than n inside the loop, and eliminate the remainder. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101113 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-12Suppress LinearFunctionTestReplace when the computed backedge-takenDan Gohman
expression is a UDiv and it doesn't appear that the UDiv came from the user's source. ScalarEvolution has recently figured out how to compute a tripcount expression for the inner loop in SingleSource/Benchmarks/Shootout/sieve.c, using a udiv. Emitting a udiv instruction dramatically slows down the enclosing loop. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101068 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-12Delete this code, which is no longer needed.Dan Gohman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101033 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-12Move the EliminateIVUsers call back out to its original location. Now thatDan Gohman
a ScalarEvolution bug with overflow handling is fixed, the normal analysis code will automatically decline to operate on the icmp instructions which are responsible for the loop exit. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101032 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-12Use RecursivelyDeleteTriviallyDeadInstructions in EliminateIVComparisons,Dan Gohman
instead of deleting just the user. This makes it more consistent with other code in IndVarSimplify, and theoretically can eliminate more users earlier. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101027 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-12Re-apply r101000, with a fix: Don't eliminate an icmp which is part ofDan Gohman
the loop exit test. This usually doesn't come up for a variety of reasons, but it isn't impossible, so make IndVarSimplify handle it conservatively. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101008 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-12Revert 101000, which is breaking self-host builds.Dan Gohman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101002 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-11Teach IndVarSimplify how to eliminate comparisons involving inductionDan Gohman
variables. For example, with code like this: for (i=0;i<n;++i) if (i<n) x[i] = 0; IndVarSimplify will now recognize that i is always less than n inside the loop, and eliminate the if. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101000 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-11Rename isLoopGuardedByCond to isLoopEntryGuardedByCond, to emphasiseDan Gohman
that it's only testing for the entry condition, not full loop-invariant conditions. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100979 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-10fix PR6743, a case where we'd delete an instruction before using itChris Lattner
in some cases. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100937 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-09When determining a canonical insert position, don't climb deeperDan Gohman
into adjacent loops. Also, ensure that the insert position is dominated by the loop latch of any loop in the post-inc set which has a latch. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100906 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-09When looking for loop-invariant users, look through no-op instructions,Dan Gohman
so that an unfortunately placed bitcast doesn't pin a value in a register. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100883 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-09const-ize a predicateGabor Greif
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100856 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-09Refactor the code for computing the insertion point for an expression intoDan Gohman
a separate function. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100845 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-09fix a SCCP miscompilation that could happen when aChris Lattner
forced constant is changed to a constant, we would end up adding the instruction to the wrong worklist, preventing it from being properly revisited. This fixes rdar://7832370 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100837 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-08Avoid allocating a value of zero in a register if the initial formulaDan Gohman
inputs happen to negate each other. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100828 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-08Add variants of ult, ule, etc. which take a uint64_t RHS, for convenience.Dan Gohman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100824 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-08When expanding expressions which are using post-inc mode for multiple loops,Dan Gohman
ensure that the expansion is dominated by the increments of those loops. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100748 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-07Generalize IVUsers to track arbitrary expressions rather than expressionsDan Gohman
explicitly split into stride-and-offset pairs. Also, add the ability to track multiple post-increment loops on the same expression. This refines the concept of "normalizing" SCEV expressions used for to post-increment uses, and introduces a dedicated utility routine for normalizing and denormalizing expressions. This fixes the expansion of expressions which are post-increment users of more than one loop at a time. More broadly, this takes LSR another step closer to being able to reason about more than one loop at a time. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100699 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-06performance: get rid of repeated dereferencing of use_iterator by caching ↵Gabor Greif
its result git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100550 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-05fix a really nasty bug that Evan was tracking in SCCP. When resolvingChris Lattner
undefs in branches/switches, we have two cases: a branch on a literal undef or a branch on a symbolic value which is undef. If we have a literal undef, the code was correct: forcing it to a constant is the right thing to do. If we have a branch on a symbolic value that is undef, we should force the symbolic value to a constant, which then makes the successor block live. Forcing the condition of the branch to being a constant isn't safe if later paths become live and the value becomes overdefined. This is the case that 'forcedconstant' is designed to handle, so just use it. This fixes rdar://7765019 but there is no good testcase for this, the one I have is too insane to be useful in the future. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100478 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-05some code cleanups, use SwitchInst::findCaseValue, reduce indentationChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100468 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-05Code clean up.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100467 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-04Reapply address space patch after fixing an issue in MemCopyOptimizer.Mon P Wang
Added support for address spaces and added a isVolatile field to memcpy, memmove, and memset, e.g., llvm.memcpy.i32(i8*, i8*, i32, i32) -> llvm.memcpy.p0i8.p0i8.i32(i8*, i8*, i32, i32, i1) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100304 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-03require that the branch being controlled by the IV Chris Lattner
exits the loop. With this information we can guarantee the iteration count of the loop is bounded by the compare. I think this xforms is finally safe now. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100285 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-03add integer overflow check for the fp induction variable Chris Lattner
checker. Amusingly, we already had tests that we should have rejects because they would be miscompiled in the testsuite. The remaining issue with this is that we don't check that the branch causes us to exit the loop if it fails, so we don't actually know if we remain in bounds. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100284 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-03add a comment and fix some consistency issues, convertingChris Lattner
to a signed vs unsigned value depending on the sign of the constant fp means that we can't distinguish between a truly negative number and a positive number so large the 32nd bit is set. So, do don't this! git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100283 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-03fix PR6761, a miscompilation due to the fp->int IV conversionChris Lattner
stuff. More bugs remain though. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100282 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-03just eliminate the uitofp checks. This code isn't doingChris Lattner
the required validity checks in the first place, and supporting a condition large enough to require the 32'nd bit isn't worth it. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100280 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-03rename PH -> PN to be consistent with WeakPN and the restChris Lattner
of llvm. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100276 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-03improve comment and drop a dead check. If PH hadChris Lattner
no uses, it would have been deleted by RecursivelyDeleteTriviallyDeadInstructions git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100275 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-03strength reduce a ridiculous use of APInt.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100274 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-03rename stuff improve comment grammar.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100273 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-03simplify some code and resolve a fixme.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100272 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-03There is no guarantee that the increment and the branchChris Lattner
are in the same block. Insert the new increment in the correct location. Also, more cleanups. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100271 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-03first half of a pass through IndVarSimplify::HandleFloatingPointIV,Chris Lattner
this cleans up a bunch of code and also fixes several crashes and miscompiles. More to come unfortunately, this optimization is quite broken. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100270 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-03Code refactoring.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100262 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-02Revert r100191 since it breaks objc in clang Mon P Wang
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100199 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-02Reapply address space patch after fixing an issue in MemCopyOptimizer.Mon P Wang
Added support for address spaces and added a isVolatile field to memcpy, memmove, and memset, e.g., llvm.memcpy.i32(i8*, i8*, i32, i32) -> llvm.memcpy.p0i8.p0i8.i32(i8*, i8*, i32, i32, i1) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100191 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-02Manually notify ScalarEvolution before making an operand replacement, sinceDan Gohman
it can't currently observe such changes automatically. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100186 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-01Introduce ImmutableCallSite, useful for contexts where no mutationGabor Greif
is necessary. Inherits from new templated baseclass CallSiteBase<> which is highly customizable. Base CallSite on it too, in a configuration that allows full mutation. Adapt some call sites in analyses to employ ImmutableCallSite. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100100 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-31Fix a nasty dangling-pointer heisenbug that couldDale Johannesen
generate wrong code pretty much anywhere AFAICT. A case that hits the bug reproducibly is impossible, but the situation was like this: Addr = ... Store -> Addr Addr2 = GEP , 0, 0 Store -> Addr2 Handling the first store, the code changed replaced Addr with a sunkaddr and deleted Addr, but not its table entry. Code in OptimizedBlock replaced Addr2 with a bitcast; if that happened to reuse the memory of Addr, the old table entry was erroneously found when handling the second store. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100044 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-30Revert Mon Ping's change 99928, since it broke all the llvm-gcc buildbots.Bob Wilson
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@99948 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-18-g5258 (git 2.32.0) at 2025-08-04 16:19:06 +0000