Commit Graph

21 Commits

Author SHA1 Message Date
Nadav Rotem
d60cb11afd When unsafe math is used, we can use commutative FMAX and FMIN. In some cases
this allows for better code generation.

Added a new DAGCombine transformation to convert FMAX and FMIN to FMANC and
FMINC, which are commutative.

For example:

  movaps  %xmm0, %xmm1
  movsd LC(%rip), %xmm0
  minsd %xmm1, %xmm0

becomes:

  minsd LC(%rip), %xmm0




git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162187 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-19 13:06:16 +00:00
Manman Ren
127eea87d6 X86 Peephole: fold loads to the source register operand if possible.
Add more comments and use early returns to reduce nesting in isLoadFoldable.
Also disable folding for V_SET0 to avoid introducing a const pool entry and
a const pool load.

rdar://10554090 and rdar://11873276


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@161207 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-02 19:37:32 +00:00
Manman Ren
d7d003c2b7 X86 Peephole: fold loads to the source register operand if possible.
Machine CSE and other optimizations can remove instructions so folding
is possible at peephole while not possible at ISel.

This patch is a rework of r160919 and was tested on clang self-host on my local
machine.

rdar://10554090 and rdar://11873276


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@161152 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-02 00:56:42 +00:00
Manman Ren
e8b4a4a9d1 Revert r160920 and r160919 due to dragonegg and clang selfhost failure
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@160927 91177308-0d34-0410-b5e6-96231b3b80d8
2012-07-29 02:44:09 +00:00
Manman Ren
14148c41d9 X86 Peephole: fold loads to the source register operand if possible.
Trying to fix the bot by specifying a triple in the failing testing cases.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@160920 91177308-0d34-0410-b5e6-96231b3b80d8
2012-07-28 17:51:24 +00:00
Manman Ren
0eb3edea9c X86 Peephole: fold loads to the source register operand if possible.
Machine CSE and other optimizations can remove instructions so folding
is possible at peephole while not possible at ISel.

rdar://10554090 and rdar://11873276


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@160919 91177308-0d34-0410-b5e6-96231b3b80d8
2012-07-28 16:48:01 +00:00
Nadav Rotem
fcb2c3cf5e Remove the "-promote-elements" flag. This flag is now enabled by default.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@157925 91177308-0d34-0410-b5e6-96231b3b80d8
2012-06-04 11:27:21 +00:00
Jakob Stoklund Olesen
ed18a3e6b2 Remove -join-physregs from the test suite.
This option has been disabled for a while, and it is going away so I can
clean up the coalescer code.

The tests that required physreg joining to be enabled were almost all of
the form "tiny function with interference between arguments and return
value". Such functions are usually inlined in the real world.

The problem exposed by phys_subreg_coalesce-3.ll is real, but fairly
rare.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@157027 91177308-0d34-0410-b5e6-96231b3b80d8
2012-05-17 23:44:19 +00:00
Craig Topper
76c5897eae Add mcpu to tests to prevent them from using AVX instructions on Sandy Bridge after r155618.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@155696 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-27 07:11:58 +00:00
Jakob Stoklund Olesen
0edd83bfff Make X86::FsFLD0SS / FsFLD0SD real pseudo-instructions.
Like V_SET0, these instructions are expanded by ExpandPostRA to xorps /
vxorps so they can participate in execution domain swizzling.

This also makes the AVX variants redundant.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@145440 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-29 22:27:25 +00:00
Duncan Sands
6bcd2196e5 Synthesize x86 max/min instructions also for vectors (i.e. produce
maxps and maxpd).  This broke the sse41-blend.ll testcase by causing
maxpd to be produced rather than a cmp+blend pair, which is the reason
I tweaked it.  Gives a small speedup on doduc with dragonegg when the
GCC vectorizer is used.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139986 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-17 16:49:39 +00:00
Jakob Stoklund Olesen
49b2803ec2 Prepare remaining tests for -join-physreg going away.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@130893 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-04 23:54:59 +00:00
Evan Cheng
60108e96bb Split -enable-finite-only-fp-math to two options:
-enable-no-nans-fp-math and -enable-no-infs-fp-math. All of the current codegen fp math optimizations only care whether the fp arithmetics arguments and results can never be NaN.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@108465 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-15 22:07:12 +00:00
Evan Cheng
ea4cdb7ead -enable-unsafe-fp-math should not imply -enable-finite-only-fp-math.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@108254 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-13 18:46:14 +00:00
Jakob Stoklund Olesen
9c50e8b890 Fix a few tests
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@108011 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-09 20:43:09 +00:00
Dan Gohman
e832693acb When forming SSE min and max nodes for UGE and ULE comparisons, it's
necessary to swap the operands to handle NaN and negative zero properly.

Also, reintroduce logic for checking for NaN conditions when forming
SSE min and max instructions, fixed to take into consideration NaNs and
negative zeros. This allows forming min and max instructions in more
cases.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97025 91177308-0d34-0410-b5e6-96231b3b80d8
2010-02-24 06:52:40 +00:00
Dan Gohman
fc90d4c0f4 Actually enable the -enable-unsafe-fp-math tests.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@96796 91177308-0d34-0410-b5e6-96231b3b80d8
2010-02-22 18:53:26 +00:00
Dan Gohman
8ce05daf54 Remove the logic for reasoning about NaNs from the code that forms
SSE min and max instructions. The real thing this code needs to be
concerned about is negative zero.

Update the sse-minmax.ll test accordingly, and add tests for
-enable-unsafe-fp-math mode as well.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@96775 91177308-0d34-0410-b5e6-96231b3b80d8
2010-02-22 04:03:39 +00:00
Dan Gohman
670e53977b Recognize SSE min and max opportunities in even more cases.
And fix a bug with the behavior of min/max instructions formed from
fcmp uge comparisons.

Also, use FiniteOnlyFPMath() for this code instead of UnsafeFPMath,
as it is more specific.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@82466 91177308-0d34-0410-b5e6-96231b3b80d8
2009-09-21 18:03:22 +00:00
Dan Gohman
36a0947820 Eliminate more uses of llvm-as and llvm-dis.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@81290 91177308-0d34-0410-b5e6-96231b3b80d8
2009-09-08 23:54:48 +00:00
Dan Gohman
8d44b28bc6 Recognize more opportunities to use SSE min and max instructions,
swapping the operands if necessary.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@80940 91177308-0d34-0410-b5e6-96231b3b80d8
2009-09-03 20:34:31 +00:00