1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2025-02-01 13:11:39 +01:00

24 Commits

Author SHA1 Message Date
Sanjay Patel
b3d997fc00 [PatternMatch] define m_FNeg using m_FSub
Using cstfp_pred_ty in the definition allows us to match vectors with undef elements.

This replicates the change for m_Not from D44076 / rL326823 and continues
towards making all pattern matchers allow undef elements in vectors.

llvm-svn: 329303
2018-04-05 15:36:55 +00:00
Sanjay Patel
42a0576ba2 [InstCombine, InstSimplify] add tests with undef elements in constant FP vectors; NFC
llvm-svn: 326148
2018-02-26 23:23:02 +00:00
Sanjay Patel
473137160c [InstCombine] allow fdiv folds with less than fully 'fast' ops
Note: gcc appears to allow this fold with -freciprocal-math alone, 
but clang/llvm require more than that with this patch. The wording
in the definitions seems fuzzy enough that it could go either way,
but we'll err on the conservative side of FMF interpretation.

This patch also changes the newly created fmul to have FMF propagated
by the last fdiv rather than intersecting the FMF of the fdivs. This
matches the behavior of other folds near here. The new fmul is only 
used to produce an intermediate op for the final fdiv result, so it
shouldn't be any stricter than that result. The previous behavior
could result in dropping FMF via other folds in instcombine or CSE.

Differential Revision: https://reviews.llvm.org/D43398

llvm-svn: 326098
2018-02-26 16:02:45 +00:00
Sanjay Patel
ec1f3bb249 [InstCombine] C / -X --> -C / X
We already do this in DAGCombiner, but it should
also be good to eliminate the fsub use in IR.

This is similar to rL325648.

llvm-svn: 325649
2018-02-21 00:01:45 +00:00
Sanjay Patel
7b47a584af [InstCombine] -X / C --> X / -C for FP
We already do this in DAGCombiner, but it should 
also be good to eliminate the fsub use in IR.

llvm-svn: 325648
2018-02-20 23:51:16 +00:00
Sanjay Patel
75f1f50880 [InstCombine] add tests for fdiv with negated op and constant op; NFC
llvm-svn: 325644
2018-02-20 23:34:43 +00:00
Sanjay Patel
a19199e5bb [PatternMatch] allow vector matches with m_FNeg
llvm-svn: 325642
2018-02-20 23:29:05 +00:00
Sanjay Patel
7831c64abe [InstCombine] add test for vector -X/-Y; NFC
m_FNeg doesn't match vector types.

llvm-svn: 325637
2018-02-20 22:46:38 +00:00
Sanjay Patel
fa138ca656 [InstCombine] fold fdiv with non-splat divisor to fmul: X/C --> X * (1/C)
llvm-svn: 325590
2018-02-20 16:08:15 +00:00
Sanjay Patel
54877d006d [InstCombine] allow fdiv with constant dividend folds with less than full -ffast-math
It's possible that we could allow this either 'arcp' or 'reassoc' alone, but this
should be conservatively better than what we have right now. GCC allows this with
only -freciprocal-math.

The last test is changed to show a case that is expected to fold, but we need D43398.

llvm-svn: 325533
2018-02-19 21:46:52 +00:00
Sanjay Patel
4fe25aad51 [InstCombine] move fdiv tests; NFC
Also, use vector constants just to prove that already works.

llvm-svn: 325530
2018-02-19 21:13:39 +00:00
Sanjay Patel
2925a2a3c3 [InstCombine] add FMF to better show current fdiv fold behavior; NFC
llvm-svn: 325365
2018-02-16 17:46:50 +00:00
Sanjay Patel
806f07ed48 [InstCombine] test fdiv folds better; NFC
We had redundant tests, but no tests for extra uses or vectors.
'fast' is an overly conservative requirement for these folds.

llvm-svn: 325262
2018-02-15 16:28:15 +00:00
Sanjay Patel
76d5ee7595 [InstCombine] allow X / C -> X * (1.0/C) for vector splat FP constants
llvm-svn: 325237
2018-02-15 13:55:52 +00:00
Sanjay Patel
7662be63cf [InstCombine] add tests and comments for fdiv X, C; NFC
llvm-svn: 325161
2018-02-14 19:54:51 +00:00
Sanjay Patel
65fb2ea1a9 [InstCombine] X / (X * Y) --> 1.0 / Y
This is similar to the instsimplify fold added with D42385 
( rL323716 )
...but this can't be in instsimplify because we're creating/morphing
a different instruction.

llvm-svn: 324927
2018-02-12 19:39:21 +00:00
Sanjay Patel
860ba4b209 [InstCombine] add tests for missing fdiv fold; NFC
llvm-svn: 324926
2018-02-12 19:23:39 +00:00
Sanjay Patel
493cd4b305 [InstCombine] regenerate checks; NFC
llvm-svn: 324924
2018-02-12 19:14:01 +00:00
Matt Arsenault
0be1430dec InstCombine: fdiv -x, -y -> fdiv x, y
llvm-svn: 291611
2017-01-10 23:08:54 +00:00
Owen Anderson
e0205fdcd8 Fix all the remaining lost-fast-math-flags bugs I've been able to find. The most important of these are cases in the generic logic for combining BinaryOperators.
This logic hadn't been updated to handle FastMathFlags, and it took me a while to detect it because it doesn't show up in a simple search for CreateFAdd.

llvm-svn: 199629
2014-01-20 07:44:53 +00:00
Owen Anderson
dbdd830886 Fix an instance where we would drop fast math flags when performing an fdiv to reciprocal multiply transformation.
llvm-svn: 199425
2014-01-16 21:07:52 +00:00
Stephen Lin
cf082ae903 Update Transforms tests to use CHECK-LABEL for easier debugging. No functionality change.
This update was done with the following bash script:

  find test/Transforms -name "*.ll" | \
  while read NAME; do
    echo "$NAME"
    if ! grep -q "^; *RUN: *llc" $NAME; then
      TEMP=`mktemp -t temp`
      cp $NAME $TEMP
      sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \
      while read FUNC; do
        sed -i '' "s/;\(.*\)\([A-Za-z0-9_]*\):\( *\)@$FUNC\([( ]*\)\$/;\1\2-LABEL:\3@$FUNC(/g" $TEMP
      done
      mv $TEMP $NAME
    fi
  done

llvm-svn: 186268
2013-07-14 01:42:54 +00:00
Benjamin Kramer
e6e5b11a65 Avoid turning a floating point division with a constant power of two into a denormal multiplication.
Some platforms may treat denormals as zero, on other platforms multiplication
with a subnormal is slower than dividing by a normal.

llvm-svn: 128555
2011-03-30 17:02:54 +00:00
Benjamin Kramer
310f9bb68e InstCombine: If the divisor of an fdiv has an exact inverse, turn it into an fmul.
Fixes PR9587.

llvm-svn: 128546
2011-03-30 15:42:35 +00:00