summaryrefslogtreecommitdiff
path: root/lib/Transforms/InstCombine
Commit message (Collapse)AuthorAge
* Fix a bug about generating undef operand when optimising shuffle vector and ↵Hao Liu2014-01-08
| | | | | | insert element in instruction combine. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198730 91177308-0d34-0410-b5e6-96231b3b80d8
* Stay classy (and legal) LLVM. Remove links to 3rd party SMT solver whose ↵Kay Tiong Khoo2013-12-19
| | | | | | links may not be permanent. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197713 91177308-0d34-0410-b5e6-96231b3b80d8
* Improved fix for PR17827 (instcombine of shift/and/compare).Kay Tiong Khoo2013-12-19
| | | | | | | | | This change fixes the case of arithmetic shift right - do not attempt to fold that case. This change also relaxes the conditions when attempting to fold the logical shift right and shift left cases. No additional IR-level test cases included at this time. See http://llvm.org/bugs/show_bug.cgi?id=17827 for proofs that these are correct transformations. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197705 91177308-0d34-0410-b5e6-96231b3b80d8
* Fix assert with copy from global through addrspacecastMatt Arsenault2013-12-07
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@196638 91177308-0d34-0410-b5e6-96231b3b80d8
* Don't use isNullValue to evaluate ConstantExprDuncan P. N. Exon Smith2013-12-06
| | | | | | | | ConstantExpr can evaluate to false even when isNullValue gives false. Fixes PR18143. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@196611 91177308-0d34-0410-b5e6-96231b3b80d8
* Use local variable for repeated use rather than 'get' method. No functional ↵Kay Tiong Khoo2013-12-02
| | | | | | change intended. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@196164 91177308-0d34-0410-b5e6-96231b3b80d8
* Move variables to where they are used and give them better names. No ↵Kay Tiong Khoo2013-12-02
| | | | | | functional change intended. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@196163 91177308-0d34-0410-b5e6-96231b3b80d8
* Rename variables to be consistent (CST -> Cst). No functional change intended.Kay Tiong Khoo2013-12-02
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@196161 91177308-0d34-0410-b5e6-96231b3b80d8
* Conservative fix for PR17827 - don't optimize a shift + and + compare ↵Kay Tiong Khoo2013-12-02
| | | | | | sequence where the shift is logical unless the comparison is unsigned git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@196129 91177308-0d34-0410-b5e6-96231b3b80d8
* Rein in overzealous InstCombine of fptrunc(OP(fpextend, fpextend)).Stephen Canon2013-11-28
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@195934 91177308-0d34-0410-b5e6-96231b3b80d8
* Apply the InstCombine fptrunc sqrt optimization to llvm.sqrtHal Finkel2013-11-16
| | | | | | | | | | | | | | | | | | | InstCombine, in visitFPTrunc, applies the following optimization to sqrt calls: (fptrunc (sqrt (fpext x))) -> (sqrtf x) but does not apply the same optimization to llvm.sqrt. This is a problem because, to enable vectorization, Clang generates llvm.sqrt instead of sqrt in fast-math mode, and because this optimization is being applied to sqrt and not applied to llvm.sqrt, sometimes the fast-math code is slower. This change makes InstCombine apply this optimization to llvm.sqrt as well. This fixes the specific problem in PR17758, although the same underlying issue (optimizations applied to libcalls are not applied to intrinsics) exists for other optimizations in SimplifyLibCalls. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194935 91177308-0d34-0410-b5e6-96231b3b80d8
* InstCombine: fold (A >> C) == (B >> C) --> (A^B) < (1 << C) for constant Cs.Benjamin Kramer2013-11-16
| | | | | | This is common in bitfield code. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194925 91177308-0d34-0410-b5e6-96231b3b80d8
* Add instcombine visitor for addrspacecastMatt Arsenault2013-11-15
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194786 91177308-0d34-0410-b5e6-96231b3b80d8
* Update the docs to match the function name.Nadav Rotem2013-11-13
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194537 91177308-0d34-0410-b5e6-96231b3b80d8
* Fold (iszero(A&K1) | iszero(A&K2)) -> (A&(K1|K2)) != (K1|K2) if we know ↵Nadav Rotem2013-11-12
| | | | | | that K1 and K2 are 'one-hot' (only one bit is on). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194525 91177308-0d34-0410-b5e6-96231b3b80d8
* Scalarize select vector arguments when extracted.Matt Arsenault2013-11-04
| | | | | | | | When the elements are extracted from a select on vectors or a vector select, do the select on the extracted scalars from the input if there is only one use. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194013 91177308-0d34-0410-b5e6-96231b3b80d8
* Remove x86_sse42_crc32_64_8 intrinsic. It has no functional difference from ↵Craig Topper2013-10-15
| | | | | | x86_sse42_crc32_32_8 and was not mapped to a clang builtin. I'm not even sure why this form of the instruction is even called out explicitly in the docs. Also add AutoUpgrade support to convert it into the other intrinsic with appropriate trunc and zext. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@192672 91177308-0d34-0410-b5e6-96231b3b80d8
* Pull fptrunc's upwards through selects when one of the select's selectands ↵Owen Anderson2013-10-03
| | | | | | was a constant. This has a number of benefits, including producing small immediates (easier to materialize, smaller constant pools) as well as being more likely to allow the fptrunc to fuse with a preceding instruction (truncating selects are unusual). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@191929 91177308-0d34-0410-b5e6-96231b3b80d8
* Make gep i8* X, -(ptrtoint Y) transform work with address spacesMatt Arsenault2013-10-03
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@191920 91177308-0d34-0410-b5e6-96231b3b80d8
* Use right address space size in InstCombineComparesMatt Arsenault2013-09-30
| | | | | | | The test's output doesn't change, but this ensures this is actually hit with a different address space. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@191701 91177308-0d34-0410-b5e6-96231b3b80d8
* Constant fold ptrtoint + compare with address spacesMatt Arsenault2013-09-30
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@191699 91177308-0d34-0410-b5e6-96231b3b80d8
* InstCombine: Replace manual fast math flag copying with the new IRBuilder ↵Benjamin Kramer2013-09-30
| | | | | | | | | RAII helper. Defines away the issue where cast<Instruction> would fail because constant folding happened. Also slightly cleaner. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@191674 91177308-0d34-0410-b5e6-96231b3b80d8
* Fix a bug in InstCombine where it attempted to cast a Value* to an Instruction*Joey Gouly2013-09-30
| | | | | | | | | | when it was actually a Constant*. There are quite a few other casts to Instruction that might have the same problem, but this is the only one I have a test case for. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@191668 91177308-0d34-0410-b5e6-96231b3b80d8
* Use type helper functionsMatt Arsenault2013-09-27
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@191574 91177308-0d34-0410-b5e6-96231b3b80d8
* InstCombine: Only foldSelectICmpAndOr for integer typesJustin Bogner2013-09-27
| | | | | | | | | | Currently foldSelectICmpAndOr asserts if the "or" involves a vector containing several of the same power of two. We can easily avoid this by only performing the fold on integer types, like foldSelectICmpAnd does. Fixes <rdar://problem/15012516> git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@191552 91177308-0d34-0410-b5e6-96231b3b80d8
* Push analysis passes to InstSimplify when they're around anyways.Benjamin Kramer2013-09-24
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@191309 91177308-0d34-0410-b5e6-96231b3b80d8
* InstCombine: Remove unused argument. No functionality change.Benjamin Kramer2013-09-20
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@191112 91177308-0d34-0410-b5e6-96231b3b80d8
* InstCombine: Canonicalize (gep i8* X, -(ptrtoint Y)) to (sub (ptrtoint X), ↵Benjamin Kramer2013-09-20
| | | | | | | | | | (ptrtoint Y)) The GEP pattern is what SCEV expander emits for "ugly geps". The latter is what you get for pointer subtraction in C code. The rest of instcombine already knows how to deal with that so just canonicalize on that. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@191090 91177308-0d34-0410-b5e6-96231b3b80d8
* [Fast-math] Disable "(C1/X)*C2 => (C1*C2)/X" if C1/X has multiple uses.Shuxin Yang2013-09-19
| | | | | | | | | | | | | | | | | | | If "C1/X" were having multiple uses, the only benefit of this transformation is to potentially shorten critical path. But it is at the cost of instroducing additional div. The additional div may or may not incur cost depending on how div is implemented. If it is implemented using Newton–Raphson iteration, it dosen't seem to incur any cost (FIXME). However, if the div blocks the entire pipeline, that sounds to be pretty expensive. Let CodeGen to take care this transformation. This patch sees 6% on a benchmark. rdar://15032743 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@191037 91177308-0d34-0410-b5e6-96231b3b80d8
* InstCombine: Don't allow turning vector-of-pointer loads into vector-of-integer.Benjamin Kramer2013-09-19
| | | | | | The code below can't handle any pointers. PR17293. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@191036 91177308-0d34-0410-b5e6-96231b3b80d8
* Revert the load slicing done in r190870.Quentin Colombet2013-09-17
| | | | | | | | To avoid regressions with bitfield optimizations, this slicing should take place later, like ISel time. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@190891 91177308-0d34-0410-b5e6-96231b3b80d8
* Cleanup handling of constant function casts.Matt Arsenault2013-09-17
| | | | | | | | | | Some of this code is no longer necessary since int<->ptr casts are no longer occur as of r187444. This also fixes handling vectors of pointers, and adds a bunch of new testcases for vectors and address spaces. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@190885 91177308-0d34-0410-b5e6-96231b3b80d8
* [InstCombiner] Slice a big load in two loads when the elements are next to eachQuentin Colombet2013-09-17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | other in memory. The motivation was to get rid of truncate and shift right instructions that get in the way of paired load or floating point load. E.g., Consider the following example: struct Complex { float real; float imm; }; When accessing a complex, llvm was generating a 64-bits load and the imm field was obtained by a trunc(lshr) sequence, resulting in poor code generation, at least for x86. The idea is to declare that two load instructions is the canonical form for loading two arithmetic type, which are next to each other in memory. Two scalar loads at a constant offset from each other are pretty easy to detect for the sorts of passes that like to mess with loads. <rdar://problem/14477220> git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@190870 91177308-0d34-0410-b5e6-96231b3b80d8
* Get rid of unused isPodLike definitions.Eli Friedman2013-09-11
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@190461 91177308-0d34-0410-b5e6-96231b3b80d8
* [InstCombiner] Expose opportunities to merge subtract and comparison.Quentin Colombet2013-09-09
| | | | | | | | | | | | | | | | | | | | | | Several architectures use the same instruction to perform both a comparison and a subtract. The instruction selection framework does not allow to consider different basic blocks to expose such fusion opportunities. Therefore, these instructions are “merged” by CSE at MI IR level. To increase the likelihood of CSE to apply in such situation, we reorder the operands of the comparison, when they have the same complexity, so that they matches the order of the most frequent subtract. E.g., icmp A, B ... sub B, A <rdar://problem/14514580> git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@190352 91177308-0d34-0410-b5e6-96231b3b80d8
* Use type helper functions.Matt Arsenault2013-09-06
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@190113 91177308-0d34-0410-b5e6-96231b3b80d8
* Consistently use dbgs() in debug printingMatt Arsenault2013-09-05
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@190093 91177308-0d34-0410-b5e6-96231b3b80d8
* InstCombine: allow unmasked icmps to be combined with logical opsTim Northover2013-09-04
| | | | | | | | | | "(icmp op i8 A, B)" is equivalent to "(icmp op i8 (A & 0xff), B)" as a degenerate case. Allowing this as a "masked" comparison when analysing "(icmp) &/| (icmp)" allows us to combine them in more cases. rdar://problem/7625728 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@189931 91177308-0d34-0410-b5e6-96231b3b80d8
* InstCombine: look for masked compares with subset relationTim Northover2013-09-04
| | | | | | | | | | | Even in cases which aren't universally optimisable like "(A & B) != 0 && (A & C) != 0", the masks can make one of the comparisons completely redundant. In this case, since we've gone to the effort of spotting masked comparisons we should combine them. rdar://problem/7625728 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@189930 91177308-0d34-0410-b5e6-96231b3b80d8
* Teach InstCombineLoadCast about address spaces.Matt Arsenault2013-09-03
| | | | | | | | This is another one that doesn't matter much, but uses the right GEP index types in the first place. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@189854 91177308-0d34-0410-b5e6-96231b3b80d8
* Use type form of getIntPtrType in alloca visitor.Matt Arsenault2013-09-03
| | | | | | | This doesn't actually matter, since alloca is always 0 address space, but this is more consistent. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@189853 91177308-0d34-0410-b5e6-96231b3b80d8
* InstCombine: Check for zero shift amounts before subtracting one causing ↵Benjamin Kramer2013-08-30
| | | | | | | | | integer overflow. PR17026. Also avoid undefined shifts and shift amounts larger than 64 bits (those are always undef because we can't represent integer types that large). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@189672 91177308-0d34-0410-b5e6-96231b3b80d8
* Fix typo.Matt Arsenault2013-08-28
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@189524 91177308-0d34-0410-b5e6-96231b3b80d8
* Teach InstCombine about address spacesMatt Arsenault2013-08-21
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@188926 91177308-0d34-0410-b5e6-96231b3b80d8
* Use pop_back_val() instead of both back() and pop_back().Jakub Staszak2013-08-19
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@188723 91177308-0d34-0410-b5e6-96231b3b80d8
* Teach InstCombine visitGetElementPtr about address spacesMatt Arsenault2013-08-19
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@188721 91177308-0d34-0410-b5e6-96231b3b80d8
* Cleanup visitGetElementPtr to make address space change easierMatt Arsenault2013-08-19
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@188720 91177308-0d34-0410-b5e6-96231b3b80d8
* commonPointerCast cleanups to make address space change easierMatt Arsenault2013-08-19
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@188719 91177308-0d34-0410-b5e6-96231b3b80d8
* Revert non-test parts of r188507Matt Arsenault2013-08-19
| | | | | | Re-add the inboundsless tests I didn't add originally git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@188710 91177308-0d34-0410-b5e6-96231b3b80d8
* InstCombine: Use isAllOnesValue() instead of explicit -1.Jim Grosbach2013-08-16
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@188563 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3 (git 2.25.1) at 2024-05-16 03:10:42 +0000