diff options
| author | Evan Cheng <evan.cheng@apple.com> | 2012-04-11 00:13:00 +0000 |
|---|---|---|
| committer | Evan Cheng <evan.cheng@apple.com> | 2012-04-11 00:13:00 +0000 |
| commit | 82509e5c62a99912c636b22e227b810eaf6eda78 (patch) | |
| tree | 4c7f36ce001590abdda1adea175a40a34ae0eb7c /lib/Target/ARM/ARM.td | |
| parent | 71fbed45d9f4e2e886afc7f22c058087e7872dc6 (diff) | |
| download | llvm-82509e5c62a99912c636b22e227b810eaf6eda78.tar.gz llvm-82509e5c62a99912c636b22e227b810eaf6eda78.tar.bz2 llvm-82509e5c62a99912c636b22e227b810eaf6eda78.tar.xz | |
Fix a number of problems with ARM fused multiply add/subtract instructions.
1. The new instruction itinerary entries are not properly described.
2. The asm parser can't handle vfms and vfnms.
3. There were no assembler, disassembler test cases.
4. HasNEON2 has the wrong assembler predicate.
rdar://10139676
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154456 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'lib/Target/ARM/ARM.td')
| -rw-r--r-- | lib/Target/ARM/ARM.td | 2 |
1 files changed, 0 insertions, 2 deletions
diff --git a/lib/Target/ARM/ARM.td b/lib/Target/ARM/ARM.td index b05fe629b7..85c41fc75d 100644 --- a/lib/Target/ARM/ARM.td +++ b/lib/Target/ARM/ARM.td @@ -76,8 +76,6 @@ def FeatureVMLxForwarding : SubtargetFeature<"vmlx-forwarding", def FeatureNEONForFP : SubtargetFeature<"neonfp", "UseNEONForSinglePrecisionFP", "true", "Use NEON for single precision FP">; -// Allow more precision in FP computation -def FPContractions : Predicate<"!TM.Options.NoExcessFPPrecision">; // Disable 32-bit to 16-bit narrowing for experimentation. def FeaturePref32BitThumb : SubtargetFeature<"32bit", "Pref32BitThumb", "true", |