aboutsummaryrefslogtreecommitdiff
path: root/lib/Transforms
AgeCommit message (Collapse)Author
2013-04-20SimplifyCFG: If convert single conditional storesArnold Schwaighofer
This transformation will transform a conditional store with a preceeding uncondtional store to the same location: a[i] = may-alias with a[i] load if (cond) a[i] = Y into an unconditional store. a[i] = X may-alias with a[i] load tmp = cond ? Y : X; a[i] = tmp We assume that on average the cost of a mispredicted branch is going to be higher than the cost of a second store to the same location, and that the secondary benefits of creating a bigger basic block for other optimizations to work on outway the potential case were the branch would be correctly predicted and the cost of the executing the second store would be noticably reflected in performance. hmmer's execution time improves by 30% on an imac12,2 on ref data sets. With this change we are on par with gcc's performance (gcc also performs this transformation). There was a 1.2 % performance improvement on a ARM swift chip. Other tests in the test-suite+external seem to be mostly uninfluenced in my experiments: This optimization was triggered on 41 tests such that the executable was different before/after the patch. Only 1 out of the 40 tests (dealII) was reproducable below 100% (by about .4%). Given that hmmer benefits so much I believe this to be a fair trade off. I am going to watch performance numbers across the builtbots and will revert this if anything unexpected comes up. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179957 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-20VecUtils: Clean up uses of dyn_cast.Benjamin Kramer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179936 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-20SLPVectorizer: Strength reduce SmallVectors to ArrayRefs.Benjamin Kramer
Avoids a couple of copies and allows more flexibility in the clients. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179935 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-20SLPVectorizer: Reduce the compile time by eliminating the search for some of ↵Nadav Rotem
the more expensive patterns. After this change will only check basic arithmetic trees that start at cmpinstr. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179933 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-20refactor tryToVectorizePair to a new method that supports vectorization of ↵Nadav Rotem
lists. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179932 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-20Fix an unused variable warning.Nadav Rotem
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179931 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-20SLPVectorizer: Improve the cost model for loop invariant broadcast values.Nadav Rotem
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179930 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-20Report the number of stores that were found in the debug message.Nadav Rotem
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179929 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-20Fix the header comment.Nadav Rotem
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179928 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-20Use 64bit arithmetic for calculating distance between pointers.Nadav Rotem
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179927 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-19MergeFunc: Make pointer and integer types generate the same hash.Benjamin Kramer
The logic that actually compares the types considers pointers and integers the same if they are of the same size. This created a strange mismatch between hash and reality and made the test case for this fail on some platforms (yay, test cases). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179905 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-19LoopVectorizer: Use matcher from PatternMatch.h for the min/max patternsArnold Schwaighofer
Also make some static function class functions to avoid having to mention the class namespace for enums all the time. No functionality change intended. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179886 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-19Keep coding stanard. Don't use "else if" after "return".Jakub Staszak
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179826 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-18Implement a better fix for PR15185.Bill Wendling
If the return type is a pointer and the call returns an integer, then do the inttoptr convertions. And vice versa. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179817 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-18Fix a -Wdocumentation warningDmitri Gribenko
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179789 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-18In the function InstCombiner::visitExtractElementInst() removed the ↵Anat Shemer
limitation that extract is promoted over a cast only if the cast has only one use. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179786 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-18Added a function scalarizePHI() that sclarizes a vector phi instruction if ↵Anat Shemer
it has only 2 uses: one to promote the vector phi in a loop and the other use is an extract operation of one element at a constant location. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179783 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-18Fix a comment, PR15777.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179775 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-18LoopVectorizer: Recognize min/max reductionsArnold Schwaighofer
A min/max operation is represented by a select(cmp(lt/le/gt/ge, X, Y), X, Y) sequence in LLVM. If we see such a sequence we can treat it just as any other commutative binary instruction and reduce it. This appears to help bzip2 by about 1.5% on an imac12,2. radar://12960601 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179773 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-18LoopVectorize: Use a set to avoid longer cycles in the reduction chain too.Benjamin Kramer
Fixes PR15748. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179757 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-18Revert "Combine bit test + conditional or into simple math"David Majnemer
It is causing stage2 builds to fail, let's get them running again. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179750 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-18Combine bit test + conditional or into simple mathDavid Majnemer
Simplify: (select (icmp eq (and X, C1), 0), Y, (or Y, C2)) Into: (or (shl (and X, C1), C3), y) Where: C3 = Log(C2) - Log(C1) If: C1 and C2 are both powers of two git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179748 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-18[objc-arc] Do not mismatch up retains inside a for loop with releases ↵Michael Gottesman
outside said for loop in the presense of differing provenance caused by escaping blocks. This occurs due to an alloca representing a separate ownership from the original pointer. Thus consider the following pseudo-IR: objc_retain(%a) for (...) { objc_retain(%a) %block <- %a F(%block) objc_release(%block) } objc_release(%a) From the perspective of the optimizer, the %block is a separate provenance from the original %a. Thus the optimizer pairs up the inner retain for %a and the outer release from %a, resulting in segfaults. This is fixed by noting that the signature of a mismatch of retain/releases inside the for loop is a Use/CanRelease top down with an None bottom up (since bottom up the Retain-CanRelease-Use-Release sequence is completed by the inner objc_retain, but top down due to the differing provenance from the objc_release said sequence is not completed). In said case in CheckForCFGHazards, we now clear the state of %a implying that no pairing will occur. Additionally a test case is included. rdar://12969722 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179747 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-18Removed trailing whitespace.Michael Gottesman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179746 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-17[objc-arc] Added annotation option to only emit annotations for a specific ↵Michael Gottesman
ssa identifier. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179729 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-17Fixed typo.Michael Gottesman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179721 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-17[objc-arc] Added descriptions for EnableARCAnnotations, ↵Michael Gottesman
EnableCheckForCFGHazards, EnableARCOptimizations. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179718 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-17[objc-arc] Added an option to arc-annotations for turning off CheckForCFGHazard.Michael Gottesman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179717 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-17Do not optimise fprintf() calls if its return value is used.Peter Collingbourne
Differential Revision: http://llvm-reviews.chandlerc.com/D620 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179661 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-16simplifycfg: Fix integer overflow converting switch into icmp.Hans Wennborg
If a switch instruction has a case for every possible value of its type, with the same successor, SimplifyCFG would replace it with an icmp ult, but the computation of the bound overflows in that case, which inverts the test. Patch by Jed Davis! git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179587 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-15We are not able to bitcast a pointer to an integral value.Bill Wendling
Two return types are not equivalent if one is a pointer and the other is an integral. This is because we cannot bitcast a pointer to an integral value. PR15185 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179569 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-15SLPVectorizer: Make it a function pass and add code for hoisting the ↵Nadav Rotem
vector-gather sequence out of loops. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179562 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-15Fix a typo in comment.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179542 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-15Add an option -vectorize-slp-aggressive for running the BB vectorizer. Make ↵Nadav Rotem
-fslp-vectorize run the slp-vectorizer. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179508 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-15Rename the slp-vectorizer clang/llvm flags. No functionality change.Nadav Rotem
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179505 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-15SLPVectorizer: Add support for vectorizing trees that start at compare ↵Nadav Rotem
instructions. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179504 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-14Reorders two transforms that collide with each otherDavid Majnemer
One performs: (X == 13 | X == 14) -> X-13 <u 2 The other: (A == C1 || A == C2) -> (A & ~(C1 ^ C2)) == C1 The problem is that there are certain values of C1 and C2 that trigger both transforms but the first one blocks out the second, this generates suboptimal code. Reordering the transforms should be better in every case and allows us to do interesting stuff like turn: %shr = lshr i32 %X, 4 %and = and i32 %shr, 15 %add = add i32 %and, -14 %tobool = icmp ne i32 %add, 0 into: %and = and i32 %X, 240 %tobool = icmp ne i32 %and, 224 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179493 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-14Miscellaneous cleanups for VecUtils.hBenjamin Kramer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179483 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-14SLP: Document the scalarization cost method.Nadav Rotem
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179479 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-14SLPVectorizer: Add support for trees that don't start at binary operators, ↵Nadav Rotem
and add the cost of extracting values from the roots of the tree. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179475 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-14SLPVectorizer: add initial support for reduction variable vectorization.Nadav Rotem
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179470 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-13GlobalDCE: Fix an oversight in my last commit that could lead to crashes.Benjamin Kramer
There is a Constant with non-constant operands: blockaddress. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179460 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-13Fix a scalability issue with complex ConstantExprs.Benjamin Kramer
This is basically the same fix in three different places. We use a set to avoid walking the whole tree of a big ConstantExprs multiple times. For example: (select cmp, (add big_expr 1), (add big_expr 2)) We don't want to visit big_expr twice here, it may consist of thousands of nodes. The testcase exercises this by creating an insanely large ConstantExprs out of a loop. It's questionable if the optimizer should ever create those, but this can be triggered with real C code. Fixes PR15714. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179458 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-12InstCombine: Check the operand types before merging fcmp ord & fcmp ord.Benjamin Kramer
Fixes PR15737. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179417 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-12SLPVectorizer: add support for vectorization of diamond shaped trees. We now ↵Nadav Rotem
perform a preliminary traversal of the graph to collect values with multiple users and check where the users came from. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179414 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-12Add debug prints.Nadav Rotem
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179412 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-12Simplify (A & ~B) in icmp if A is a power of 2David Majnemer
The transform will execute like so: (A & ~B) == 0 --> (A & B) != 0 (A & ~B) != 0 --> (A & B) == 0 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179386 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-12LoopVectorizer: integer division is not a reduction operationArnold Schwaighofer
Don't classify idiv/udiv as a reduction operation. Integer division is lossy. For example : (1 / 2) * 4 != 4/2. Example: int a[] = { 2, 5, 2, 2} int x = 80; for() x /= a[i]; Scalar: x /= 2 // = 40 x /= 5 // = 8 x /= 2 // = 4 x /= 2 // = 2 Vectorized: <80, 1> / <2,5> //= <40,0> <40, 0> / <2,2> //= <20,0> 20*0 = 0 radar://13640654 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179381 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-11Optimize icmp involving addition betterDavid Majnemer
Allows LLVM to optimize sequences like the following: %add = add nsw i32 %x, 1 %cmp = icmp sgt i32 %add, %y into: %cmp = icmp sge i32 %x, %y as well as: %add1 = add nsw i32 %x, 20 %add2 = add nsw i32 %y, 57 %cmp = icmp sge i32 %add1, %add2 into: %add = add nsw i32 %y, 37 %cmp = icmp sle i32 %cmp, %x git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179316 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-11Fix for wrong instcombine on vector insert/extractBenjamin Kramer
When trying to collapse sequences of insertelement/extractelement instructions into single shuffle instructions, there is one specific case where the Instruction Combiner wrongly updates the resulting Mask of shuffle indexes. The problem is in function CollectShuffleElments. If we have a sequence of insert/extract element instructions like the one below: %tmp1 = extractelement <4 x float> %LHS, i32 0 %tmp2 = insertelement <4 x float> %RHS, float %tmp1, i32 1 %tmp3 = extractelement <4 x float> %RHS, i32 2 %tmp4 = insertelement <4 x float> %tmp2, float %tmp3, i32 3 Where: . %RHS will have a mask of [4,5,6,7] . %LHS will have a mask of [0,1,2,3] The Mask of shuffle indexes is wrongly computed to [4,1,6,7] instead of [4,0,6,7]. When analyzing %tmp2 in order to compute the Mask for the resulting shuffle instruction, the algorithm forgets to update the mask index at position 1 with the index associated to the element extracted from %LHS by instruction %tmp1. Patch by Andrea DiBiagio! git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179291 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-18-g5258 (git 2.32.0) at 2025-07-16 08:17:33 +0000