diff options
author | Mark Mendell <mark.p.mendell@intel.com> | 2015-05-07 11:25:03 -0400 |
---|---|---|
committer | Mark Mendell <mark.p.mendell@intel.com> | 2015-05-07 12:33:00 -0400 |
commit | 92e83bf8c0b2df8c977ffbc527989631d94b1819 (patch) | |
tree | fb23a1d027549a68a9ea88bf0eb63f85bc116aee /compiler/optimizing/intrinsics_x86_64.cc | |
parent | dceab011395333520959cf743d680bcf5dc78797 (diff) | |
download | art-92e83bf8c0b2df8c977ffbc527989631d94b1819.zip art-92e83bf8c0b2df8c977ffbc527989631d94b1819.tar.gz art-92e83bf8c0b2df8c977ffbc527989631d94b1819.tar.bz2 |
[optimizing] Tune some x86_64 moves
Generate Moves of constant FP values by loading from the constant table.
Use 'movl' to load a 64 bit register for positive 32-bit values, saving
a byte in the generated code by taking advantage of the implicit
zero extension.
Change a couple of xorq(reg, reg) to xorl to (potentially) save a byte
of code per xor.
Change-Id: I5b2a807f0d3b29294fd4e7b8ef6d654491fa0b01
Signed-off-by: Mark Mendell <mark.p.mendell@intel.com>
Diffstat (limited to 'compiler/optimizing/intrinsics_x86_64.cc')
-rw-r--r-- | compiler/optimizing/intrinsics_x86_64.cc | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/compiler/optimizing/intrinsics_x86_64.cc b/compiler/optimizing/intrinsics_x86_64.cc index 1fc5432..3dbcd4c 100644 --- a/compiler/optimizing/intrinsics_x86_64.cc +++ b/compiler/optimizing/intrinsics_x86_64.cc @@ -783,7 +783,7 @@ void IntrinsicCodeGeneratorX86_64::VisitMathRoundDouble(HInvoke* invoke) { __ Bind(&nan); // output = 0 - __ xorq(out, out); + __ xorl(out, out); __ Bind(&done); } |