[InstCombine] reduce disguised mul+add factorization

~(A * C1) + A --> (A * (1 - C1)) - 1

This is a non-obvious mix of bitwise logic and math:
https://alive2.llvm.org/ce/z/U7ACVT

The pattern may b

[InstCombine] reduce disguised mul+add factorization

~(A * C1) + A --> (A * (1 - C1)) - 1

This is a non-obvious mix of bitwise logic and math:
https://alive2.llvm.org/ce/z/U7ACVT

The pattern may be produced by Negator from the more typical
code seen in issue #57255.

show more ...

[InstCombine] Canonicalize ((X & -X) - 1) --> ((X - 1) & ~X) (PR51784)

Enables the ctpop((x & -x ) - 1) -> cttz(x, false) fold

Alive2: https://alive2.llvm.org/ce/z/EDk4h7 (((X & -X) - 1) --> (~X &

[InstCombine] Canonicalize ((X & -X) - 1) --> ((X - 1) & ~X) (PR51784)

Enables the ctpop((x & -x ) - 1) -> cttz(x, false) fold

Alive2: https://alive2.llvm.org/ce/z/EDk4h7 (((X & -X) - 1) --> (~X & (X - 1)) )

Alive2: https://alive2.llvm.org/ce/z/8Yr3XG (CTPOP -> CTTZ)

Fixes #51126

Differential Revision: https://reviews.llvm.org/D110488

show more ...

Revert rGc360955c4804e9b25017372cb4c6be7adcb216ce "[InstCombine] Canonicalize ((X & -X) - 1) --> (~X & (X - 1)) (PR51784)"

The test changes are failing on some buildbots (but not others.....).

[InstCombine] Canonicalize ((X & -X) - 1) --> (~X & (X - 1)) (PR51784)

Enables the ctpop((x & -x ) - 1) -> cttz(x, false) fold

Alive2: https://alive2.llvm.org/ce/z/EDk4h7 (((X & -X) - 1) --> (~X &

[InstCombine] Canonicalize ((X & -X) - 1) --> (~X & (X - 1)) (PR51784)

Enables the ctpop((x & -x ) - 1) -> cttz(x, false) fold

Alive2: https://alive2.llvm.org/ce/z/EDk4h7 (((X & -X) - 1) --> (~X & (X - 1)) )

Alive2: https://alive2.llvm.org/ce/z/8Yr3XG (CTPOP -> CTTZ)

Fixes #51126

Differential Revision: https://reviews.llvm.org/D110488

show more ...

[InstCombine] Change order of canonicalization of ADD and AND

Canonicalize ((x + C1) & C2) --> ((x & C2) + C1) for suitable constants
C1 and C2, instead of the other way round. This should allow mor

[InstCombine] Change order of canonicalization of ADD and AND

Canonicalize ((x + C1) & C2) --> ((x & C2) + C1) for suitable constants
C1 and C2, instead of the other way round. This should allow more
constant ADDs to be matched as part of addressing modes for loads and
stores.

Differential Revision: https://reviews.llvm.org/D130080

show more ...

[InstCombine] fold reassociative fadd with negated operand

We manage to iteratively achieve this result with no extra
uses, and the reassociate pass can also do this, but this
pattern falls through

[InstCombine] fold reassociative fadd with negated operand

We manage to iteratively achieve this result with no extra
uses, and the reassociate pass can also do this, but this
pattern falls through the cracks in the example from
issue #57053.

show more ...

[InstCombine] Avoid ConstantExpr::getFNeg() calls (NFCI)

Instead call the constant folding API, which can fail. For now,
this should be NFC, as we still allow the creation of fneg
constant expressio

[InstCombine] Avoid ConstantExpr::getFNeg() calls (NFCI)

Instead call the constant folding API, which can fail. For now,
this should be NFC, as we still allow the creation of fneg
constant expressions.

show more ...

Revert "[InstCombine] enhance fold for subtract-from-constant -> xor"

This reverts commit 79bb915fb60b2cd220d89e3bb54f67abb8cdb7bd.
This caused regressions because SCEV works better with sub.

[InstCombine] Avoid creating float binop ConstantExprs

Replace ConstantExpr:getFAdd etc with call to
ConstantFoldBinaryOpOperands(). I'm using the constant folding API
rather than IRBuilder here to

[InstCombine] Avoid creating float binop ConstantExprs

Replace ConstantExpr:getFAdd etc with call to
ConstantFoldBinaryOpOperands(). I'm using the constant folding API
rather than IRBuilder here to ensure that this does actually
constant fold. These transforms don't use m_ImmConstant(), so this
would not otherwise be guaranteed (and apparently, they can't use
m_ImmConstant because they want to handle scalable vector splats).

There is an opportunity here to further migrate these to the
ConstantFoldFPInstOperands() API, which would respect the denormal
mode. I've held off on doing so here, because some of this code
explicitly checks for denormal results, and I don't want to touch
it in a mostly NFC change.

show more ...

[InstCombine] enhance fold for subtract-from-constant -> xor

A low-bit mask is not required:
https://alive2.llvm.org/ce/z/yPShss

This matches the SDAG implementation that was updated at:
8b75671314

[InstCombine] enhance fold for subtract-from-constant -> xor

A low-bit mask is not required:
https://alive2.llvm.org/ce/z/yPShss

This matches the SDAG implementation that was updated at:
8b756713140f

show more ...

[InstCombine] fold sub of min/max of sub with common operand

x - max(x - y, 0) --> min(x, y)
x - min(x - y, 0) --> max(x, y)

https://alive2.llvm.org/ce/z/2YkqFe

issue #55470

[InstCombine] add helper function for sub-of-min/max folds; NFC

The test diffs are cosmetic -- but improvements -- because we
let instcombine handle replacement. Instead of dropping the
old value na

[InstCombine] add helper function for sub-of-min/max folds; NFC

The test diffs are cosmetic -- but improvements -- because we
let instcombine handle replacement. Instead of dropping the
old value name, it propagates to the new instruction.

show more ...

[NFC] format InstructionSimplify & lowerCaseFunctionNames

Clang-format InstructionSimplify and convert all "FunctionName"s to
"functionName". This patch does touch a lot of files but gets done with

[NFC] format InstructionSimplify & lowerCaseFunctionNames

Clang-format InstructionSimplify and convert all "FunctionName"s to
"functionName". This patch does touch a lot of files but gets done with
the cleanup of InstructionSimplify in one commit.

This is the alternative to the less invasive clang-format only patch: D126783

Reviewed By: spatel, rengolin

Differential Revision: https://reviews.llvm.org/D126889

show more ...

[InstCombine] Strip bitcasts in GEP diff fold

Bitcasts were stripped in one case, but not the other. Of course,
this no longer really matters with opaque pointers, but as I went
through the trouble

[InstCombine] Strip bitcasts in GEP diff fold

Bitcasts were stripped in one case, but not the other. Of course,
this no longer really matters with opaque pointers, but as I went
through the trouble of tracking this down, we may as well remove
one typed vs opaque pointer optimization discrepancy.

show more ...

[IR] create and use helper functions that test the signbit; NFCI

[InstCombine] fix sub(add(X,Y),umin(Y,Z)) --> add(X,usub.sat(Y,Z))

This patch fix bug left in D124503. We should do
sub(add(X,Z),umin(Y,Z)) --> add(X,usub.sat(Z,Y)) instead of
sub(add(X,Z),umin(Y,Z)

[InstCombine] fix sub(add(X,Y),umin(Y,Z)) --> add(X,usub.sat(Y,Z))

This patch fix bug left in D124503. We should do
sub(add(X,Z),umin(Y,Z)) --> add(X,usub.sat(Z,Y)) instead of
sub(add(X,Z),umin(Y,Z)) --> add(X,usub.sat(Y,Z)).

Reviewed By: spatel

Differential Revision: https://reviews.llvm.org/D125352

show more ...

[InstCombine] sub(add(X,Y),umin(Y,Z)) --> add(X,usub.sat(Y,Z))

Alive2: https://alive2.llvm.org/ce/z/2UNVbp

Reviewed By: RKSimon, spatel

Differential Revision: https://reviews.llvm.org/D124503

[InstCombine] Fold (A & 2^C1) + A => A & (2^C1 - 1) iff bit C1 in A is a sign bit (PR21929)

Alive2: https://alive2.llvm.org/ce/z/Ygq26C

This is the final missing fold to handle the modulo2 simplifi

[InstCombine] Fold (A & 2^C1) + A => A & (2^C1 - 1) iff bit C1 in A is a sign bit (PR21929)

Alive2: https://alive2.llvm.org/ce/z/Ygq26C

This is the final missing fold to handle the modulo2 simplification: https://github.com/llvm/llvm-project/issues/22303

Fixes #22303

Differential Revision: https://reviews.llvm.org/D123374

show more ...

[InstCombine] Fold sub(add(x,y),min/max(x,y)) -> max/min(x,y) (PR38280)

As discussed on Issue #37628, we can flip a min/max node if we're subtracting from the sum of the node's operands

Alive2: htt

[InstCombine] Fold sub(add(x,y),min/max(x,y)) -> max/min(x,y) (PR38280)

As discussed on Issue #37628, we can flip a min/max node if we're subtracting from the sum of the node's operands

Alive2: https://alive2.llvm.org/ce/z/W_KXfy

Differential Revision: https://reviews.llvm.org/D123399

show more ...

[NFCI] Fix set-but-unused warning in InstCombineAddSub.cpp

[InstCombine] try harder to propagate 'nsz' through fneg-of-select

This can be viewed as swapping the select arms:
https://alive2.llvm.org/ce/z/jUvFMJ
...so we don't have the 'nsz' problem with the

[InstCombine] try harder to propagate 'nsz' through fneg-of-select

This can be viewed as swapping the select arms:
https://alive2.llvm.org/ce/z/jUvFMJ
...so we don't have the 'nsz' problem with the more general fold.

This unlocks other folds for the motivating fabs example.
This was discussed in issue #38828.

show more ...

[InstCombine] Remove sub of SPF min/max fold (NFCI)

This isn't necessary anymore, now that we canonicalize SPF min/max
to intrinsics. Might not be strictly NFC due to worklist order
changes.

[InstCombine] try harder to preserve 'nsz' in fneg-of-select transform

The corner case where 'nsz' needs to be removed is very narrow
as discussed here:
https://reviews.llvm.org/rG3cdd05e519dd

If t

[InstCombine] try harder to preserve 'nsz' in fneg-of-select transform

The corner case where 'nsz' needs to be removed is very narrow
as discussed here:
https://reviews.llvm.org/rG3cdd05e519dd

If the select condition is not undef, there's no problem with
propagating 'nsz':
https://alive2.llvm.org/ce/z/4GWJdq

show more ...

[InstCombine] Fold sub of umin to usub.sat

We were handling sub of umax, but not the conjugated umin case.

https://alive2.llvm.org/ce/z/4fdZfy
https://alive2.llvm.org/ce/z/BhUQBM

Revert "Rename llvm::array_lengthof into llvm::size to match std::size from C++17"

This reverts commit ef8206320769ad31422a803a0d6de6077fd231d2.

- It conflicts with the existing llvm::size in STLEx

Revert "Rename llvm::array_lengthof into llvm::size to match std::size from C++17"

This reverts commit ef8206320769ad31422a803a0d6de6077fd231d2.

- It conflicts with the existing llvm::size in STLExtras, which will now
never be called.
- Calling it without llvm:: breaks C++17 compat

show more ...