This is the opposite direction of D62158 (we have to choose 1 form or the other).
Now that we have FMF on the select, this becomes more palatable. And the benefits
of having a single IR instruction for this operation (less chances of missing folds
based on extra uses, etc) overcome my previous comments about the potential advantage
of larger pattern matching/analysis.
Differential Revision: https://reviews.llvm.org/D62414
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@364721 91177308-0d34-0410-b5e6-96231b3b80d8
We have a known-never-nan value via 'nnan', so an unordered predicate
is the same as its ordered sibling.
Similar to:
rL362937
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@362954 91177308-0d34-0410-b5e6-96231b3b80d8
Similar to rL362909:
This isn't the ideal fix (use FMF on the select), but it's still an
improvement until we have better FMF propagation to selects and other
FP math operators.
I don't think there's much risk of regression from this change by
not including the FMF on the fcmp any more. The nsz/nnan FMF
should be the same on the fcmp and the fsub because they have the
same operand.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@362943 91177308-0d34-0410-b5e6-96231b3b80d8
This isn't the ideal fix (use FMF on the select), but it's still an
improvement until we have better FMF propagation to selects and other
FP math operators.
I don't think there's much risk of regression from this change by
not including the FMF on the fcmp any more. The nsz/nnan FMF
should be the same on the fcmp and the fneg (fsub) because they
have the same operand.
This works around the most glaring FMF logical inconsistency cited
in PR38086:
https://bugs.llvm.org/show_bug.cgi?id=38086
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@362909 91177308-0d34-0410-b5e6-96231b3b80d8
We don't always get this:
Cond ? -X : -Y --> -(Cond ? X : Y)
...even with the legacy IR form of fneg in the case with extra uses,
and we miss matching with the newer 'fneg' instruction because we
are expecting binops through the rest of the path.
Differential Revision: https://reviews.llvm.org/D61604
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360075 91177308-0d34-0410-b5e6-96231b3b80d8
Start using the uadd.sat and usub.sat intrinsics for the existing
canonicalizations. These intrinsics should optimize better than
expanded IR, have better handling in the X86 backend and should
be no worse than expanded IR in other backends, as far as we know.
rL357012 already introduced use of uadd.sat for the add+umin pattern.
Differential Revision: https://reviews.llvm.org/D58872
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@357103 91177308-0d34-0410-b5e6-96231b3b80d8
This is no-functional-change-intended, but that was also
true when it was part of rL354276, and I managed to lose
2 predicates for the fold with constant...causing much bot
distress. So this time I'm adding a couple of negative tests
to avoid that.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@354384 91177308-0d34-0410-b5e6-96231b3b80d8
This reverts commit 079b610c29b4a428b3ae7b64dbac0378facf6632.
Bots are failing after this change on a stage 2 compile of clang.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@354277 91177308-0d34-0410-b5e6-96231b3b80d8
We want to use the sum in the icmp to allow matching with
m_UAddWithOverflow and eliminate the 'not'. This is discussed
in D51929 and is another step towards solving PR14613:
https://bugs.llvm.org/show_bug.cgi?id=14613
(The matching here is incomplete. Trying to take minimal steps
to make sure we don't induce infinite looping from existing
canonicalizations of the 'select'.)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@354221 91177308-0d34-0410-b5e6-96231b3b80d8
I'm circling back around to a loose end from D51929.
The backend (either CGP or DAG) doesn't recognize this pattern, so we end up with different asm for these IR variants.
Regardless of any future changes to canonicalize to saturation/overflow intrinsics, we want to get raw IR variations
into the minimal number of raw IR forms. If/when we can canonicalize to intrinsics, that will make that step easier.
Pre: C2 == ~C1
%a = add i32 %x, C1
%c = icmp ugt i32 %x, C2
%r = select i1 %c, i32 -1, i32 %a
=>
%a = add i32 %x, C1
%c2 = icmp ult i32 %x, C2
%r = select i1 %c2, i32 %a, i32 -1
https://rise4fun.com/Alive/pkH
Differential Revision: https://reviews.llvm.org/D57352
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@352536 91177308-0d34-0410-b5e6-96231b3b80d8
to reflect the new license.
We understand that people may be surprised that we're moving the header
entirely to discuss the new license. We checked this carefully with the
Foundation's lawyer and we believe this is the correct approach.
Essentially, all code in the project is now made available by the LLVM
project under our new license, so you will see that the license headers
include that license only. Some of our contributors have contributed
code under our old license, and accordingly, we have retained a copy of
our old license notice in the top-level files in each project and
repository.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@351636 91177308-0d34-0410-b5e6-96231b3b80d8
This is matching the equivalent of the DAG expansion,
so it should never end up with worse perf than the
original code even if the target doesn't have a rotate
instruction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@350672 91177308-0d34-0410-b5e6-96231b3b80d8
The cttz/ctlz intrinsics have a parameter specifying whether the
result is undefined for zero. cttz(x, false) can be relaxed to
cttz(x, true) if x is known non-zero, and in fact such an optimization
is already performed. However, this currently doesn't work if x is
non-zero as a result of a select rather than an explicit branch.
This patch adds handling for this case, thus allowing
x != 0 ? cttz(x, false) : y to simplify to x != 0 ? cttz(x, true) : y.
Differential Revision: https://reviews.llvm.org/D55786
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@350463 91177308-0d34-0410-b5e6-96231b3b80d8
The final piece of IR-level analysis to allow this was committed with:
rL350188
Using the intrinsics should improve transforms based on cost models
like vectorization and inlining.
The backend should be prepared too, so we can now canonicalize more
sequences of shift/logic to the intrinsics and know that the end
result should be equal or better to the original code even if the
target does not have an actual rotate instruction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@350199 91177308-0d34-0410-b5e6-96231b3b80d8
This is an almost direct move of the functionality from InstCombine to
InstSimplify. There's no reason not to do this in InstSimplify because
we never create a new value with this transform.
(There's a question of whether any dominance-based transform belongs in
either of these passes, but that's a separate issue.)
I've changed 1 of the conditions for the fold (1 of the blocks for the
branch must be the block we started with) into an assert because I'm not
sure how that could ever be false.
We need 1 extra check to make sure that the instruction itself is in a
basic block because passes other than InstCombine may be using InstSimplify
as an analysis on values that are not wired up yet.
The 3-way compare changes show that InstCombine has some kind of
phase-ordering hole. Otherwise, we would have already gotten the intended
final result that we now show here.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@347896 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
These asserts are based on the assumption that the order of true/false operands in a select and those in the compare would always be the same.
This fixes PR39595.
Reviewers: craig.topper, spatel, dmgreen
Reviewed By: craig.topper
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D54359
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346874 91177308-0d34-0410-b5e6-96231b3b80d8
The cmp+branch variant of this pattern is shown in:
https://bugs.llvm.org/show_bug.cgi?id=34924
...and as discussed there, we probably can't transform
that without a rotate intrinsic. We do have that now
via funnel shift, but we're not quite ready to
canonicalize IR to that form yet. The case with 'select'
should already be transformed though, so that's this patch.
The sequence with negation followed by masking is what we
use in the backend and partly in clang (though that part
should be updated).
https://rise4fun.com/Alive/TplC
%cmp = icmp eq i32 %shamt, 0
%sub = sub i32 32, %shamt
%shr = lshr i32 %x, %shamt
%shl = shl i32 %x, %sub
%or = or i32 %shr, %shl
%r = select i1 %cmp, i32 %x, i32 %or
=>
%neg = sub i32 0, %shamt
%masked = and i32 %shamt, 31
%maskedneg = and i32 %neg, 31
%shl2 = lshr i32 %x, %masked
%shr2 = shl i32 %x, %maskedneg
%r = or i32 %shl2, %shr2
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346807 91177308-0d34-0410-b5e6-96231b3b80d8
This is NFCI for InstCombine because it calls InstSimplify,
so I left the tests for this transform there. As noted in
the code comment, we can allow this fold more often by using
FMF and/or value tracking.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346169 91177308-0d34-0410-b5e6-96231b3b80d8
The IRBuilder CreateIntrinsic method wouldn't allow you to specify the
types that you wanted the intrinsic to be mangled with. To fix this
I've:
- Added an ArrayRef<Type *> member to both CreateIntrinsic overloads.
- Used that array to pass into the Intrinsic::getDeclaration call.
- Added a CreateUnaryIntrinsic to replace the most common use of
CreateIntrinsic where the type was auto-deduced from operand 0.
- Added a bunch more unit tests to test Create*Intrinsic calls that
weren't being tested (including the FMF flag that wasn't checked).
This was suggested as part of the AMDGPU specific atomic optimizer
review (https://reviews.llvm.org/D51969).
Differential Revision: https://reviews.llvm.org/D52087
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@343962 91177308-0d34-0410-b5e6-96231b3b80d8
Summary: This restores the combine that was reverted in r341883. The infinite loop from the failing test no longer occurs due to changes from r342163.
Reviewers: spatel, dmgreen
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D52070
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@342797 91177308-0d34-0410-b5e6-96231b3b80d8
I accidentally committed this diff with rL342147 because
I had applied D51964. We probably do need those checks,
but D51964 has tests and more discussion/motivation,
so they should be re-added with that patch.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@342149 91177308-0d34-0410-b5e6-96231b3b80d8
I don't have a test case for this, but it's motivated by
the discussion in D51964, and I've added TODO comments for
the better fix - move simplifications into instsimplify
because that's more efficient and reduces risk of infinite
loops in instcombine caused by transforms trying to do the
opposite folds.
In this case, we know that the transform that tries to move
'not' through min/max can be fooled by the multiple uses
of a value in another min/max, so try to squash the
foldSPFofSPF() patterns first.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@342147 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
Revert min/max changes in rL341674 dues to high compile times causing timeouts (PR38897).
Checking in to unblock failing builds. Patch available for post-commit review and re-revert once resolved.
Working on a smaller reproducer for PR38897.
Reviewers: craig.topper, spatel
Subscribers: sanjoy, jlebar, llvm-commits
Differential Revision: https://reviews.llvm.org/D51897
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@341883 91177308-0d34-0410-b5e6-96231b3b80d8
If the ~X wasn't able to simplify above the max/min, we might be able to simplify it by moving it below the max/min.
I had to modify the ~(min/max ~X, Y) transform to prevent getting stuck in a loop when we saw the new ~(max/min X, ~Y) before the ~Y had been folded away to remove the new not.
Differential Revision: https://reviews.llvm.org/D51398
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@341674 91177308-0d34-0410-b5e6-96231b3b80d8
If OtherOpT or OtherOpF have scalar types and the condition is a vector,
we would create an invalid select.
Reviewers: spatel, john.brawn, mssimpso, craig.topper
Reviewed By: spatel
Differential Revision: https://reviews.llvm.org/D51781
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@341666 91177308-0d34-0410-b5e6-96231b3b80d8
We were calling getNumUses to check for 1 or 2 uses. But getNumUses is linear in the number of uses. We can instead use !hasNUsesOrMore(3) which will stop the linear scan as soon as it determines there are at least 3 uses even if there are more.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@340939 91177308-0d34-0410-b5e6-96231b3b80d8
This is a retry of rL339439 with a fix for the problem that
caused the original commit to be reverted at rL339446.
That problem was that the compare can be integer while
the binop is FP or vice-versa, so we need to use the binop
type when we ask for the identity constant.
A test to guard against the problem was added at rL339453.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@339469 91177308-0d34-0410-b5e6-96231b3b80d8
