projects
/
oota-llvm.git
/ history
commit
grep
author
committer
pickaxe
?
search:
re
summary
|
shortlog
|
log
|
commit
|
commitdiff
|
tree
first ⋅ prev ⋅
next
X86: fcmov doesn't handle all possible EFLAGS, fall back to a branch for the others.
[oota-llvm.git]
/
lib
/
Target
/
X86
/
X86ISelLowering.cpp
2012-10-04
Preston Gurd
This patch corrects commit 165126 by using an integer...
blob
|
commitdiff
|
raw
2012-10-04
Michael Liao
Add register encoding support in X86 backend
blob
|
commitdiff
|
raw
|
diff to current
2012-10-04
Bill Wendling
Use new accessor methods to query for attributes.
blob
|
commitdiff
|
raw
|
diff to current
2012-10-03
Michael Liao
Clean up tailing whitespaces
blob
|
commitdiff
|
raw
|
diff to current
2012-09-30
Craig Topper
Change getX86SubSuperRegister to take an MVT::SimpleVal...
blob
|
commitdiff
|
raw
|
diff to current
2012-09-26
Bill Wendling
Remove the `hasFnAttr' method from Function.
blob
|
commitdiff
|
raw
|
diff to current
2012-09-25
Michael Liao
Add missing i64 max/min/umax/umin on 32-bit target
blob
|
commitdiff
|
raw
|
diff to current
2012-09-25
Evan Cheng
Fix an illegal tailcall opt where the callee returns...
blob
|
commitdiff
|
raw
|
diff to current
2012-09-21
Michael Liao
Add missing i8 max/min/umax/umin support
blob
|
commitdiff
|
raw
|
diff to current
2012-09-21
Michael Liao
Revise td of X86 atomic instructions
blob
|
commitdiff
|
raw
|
diff to current
2012-09-20
Michael Liao
Re-work X86 code generation of atomic ops with spin...
blob
|
commitdiff
|
raw
|
diff to current
2012-09-15
Benjamin Kramer
X86: Emitting x87 fsin/fcos for sinf/cosf is not safe...
blob
|
commitdiff
|
raw
|
diff to current
2012-09-13
Michael Liao
Fix comment
blob
|
commitdiff
|
raw
|
diff to current
2012-09-13
Michael Liao
Add wider vector/integer support for PR12312
blob
|
commitdiff
|
raw
|
diff to current
2012-09-12
Michael Liao
Fix PR11985
blob
|
commitdiff
|
raw
|
diff to current
2012-09-12
Craig Topper
Indentation fixes. No functional change.
blob
|
commitdiff
|
raw
|
diff to current
2012-09-11
Craig Topper
Make a bunch of lowering helper functions static instea...
blob
|
commitdiff
|
raw
|
diff to current
2012-09-10
Dmitri Gribenko
Remove redundant semicolons which are null statements.
blob
|
commitdiff
|
raw
|
diff to current
2012-09-10
Michael Liao
Enhance PR11334 fix to support extload from v2f32/v4f32
blob
|
commitdiff
|
raw
|
diff to current
2012-09-10
Michael Liao
Add boolean simplification support from CMOV
blob
|
commitdiff
|
raw
|
diff to current
2012-09-10
Elena Demikhovsky
The VPSHUFB 256-bit instruction may be generated when...
blob
|
commitdiff
|
raw
|
diff to current
2012-09-08
Craig Topper
Add instruction selection for ffloor of vectors when...
blob
|
commitdiff
|
raw
|
diff to current
2012-09-08
Craig Topper
Use 256-bit alignment for constant pool value for 256...
blob
|
commitdiff
|
raw
|
diff to current
2012-09-08
Craig Topper
Add support for lowering FABS of vector types.
blob
|
commitdiff
|
raw
|
diff to current
2012-09-08
Craig Topper
Set operation action for FFLOOR to Expand for all vecto...
blob
|
commitdiff
|
raw
|
diff to current
2012-09-06
Elena Demikhovsky
AVX2 optimization.
blob
|
commitdiff
|
raw
|
diff to current
2012-09-06
Michael Liao
Remove duplicated helper function
blob
|
commitdiff
|
raw
|
diff to current
2012-09-06
Craig Topper
Use iPTR instead of i32 for extract_subvector/insert_su...
blob
|
commitdiff
|
raw
|
diff to current
2012-09-05
Roman Divacky
Stop casting away const qualifier needlessly.
blob
|
commitdiff
|
raw
|
diff to current
2012-09-04
Preston Gurd
Generic Bypass Slow Div
blob
|
commitdiff
|
raw
|
diff to current
2012-09-04
Elena Demikhovsky
This patch optimizes shuffle instruction - generates...
blob
|
commitdiff
|
raw
|
diff to current
2012-09-01
Craig Topper
Typos
blob
|
commitdiff
|
raw
|
diff to current
2012-08-31
Manman Ren
SelectionDAG: when constructing VZEXT_LOAD from other...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-31
Michael Liao
Fix PR12359
blob
|
commitdiff
|
raw
|
diff to current
2012-08-31
Craig Topper
Add support for converting llvm.fma to fma4 instructions.
blob
|
commitdiff
|
raw
|
diff to current
2012-08-30
Craig Topper
Only perform DAG combine on FMAs of legal types.
blob
|
commitdiff
|
raw
|
diff to current
2012-08-29
Craig Topper
Convert FMA4 patterns to use target specific nodes...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-28
Michael Liao
Add comments on the literal value used.
blob
|
commitdiff
|
raw
|
diff to current
2012-08-28
Michael Liao
Explicitly update the number of nodes to be traversed
blob
|
commitdiff
|
raw
|
diff to current
2012-08-28
Michael Liao
Fix PR12312
blob
|
commitdiff
|
raw
|
diff to current
2012-08-27
Craig Topper
Remove MMX shift intrinsic handling code that also...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-24
Craig Topper
Custom lower FMA intrinsics to target specific nodes...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-20
Michael Liao
fix a case where all operands of BUILD_VECTOR are undefined
blob
|
commitdiff
|
raw
|
diff to current
2012-08-19
Nadav Rotem
When unsafe math is used, we can use commutative FMAX...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-18
Nadav Rotem
Reapply r162160 with a fix: Optimize Arith->Trunc-...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-18
Craig Topper
Refactor code a bit to reduce number of calls in the...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-18
Nadav Rotem
Revert r162160 because it made a few buildbots fail.
blob
|
commitdiff
|
raw
|
diff to current
2012-08-18
Nadav Rotem
The X86 backend has a number of optimizations for SETCC...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-17
Craig Topper
Use nested switch to select arguments to reduce calls...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-17
Craig Topper
Make ReplaceATOMIC_BINARY_64 a static function. Use...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-14
Michael Liao
minor fix of X86ISD::VSEXT_MOVL dump
blob
|
commitdiff
|
raw
|
diff to current
2012-08-14
Michael Liao
fix PR11334
blob
|
commitdiff
|
raw
|
diff to current
2012-08-14
Craig Topper
Factor duplicate calls to getUNDEF in several functions.
blob
|
commitdiff
|
raw
|
diff to current
2012-08-14
Craig Topper
Re-factor intrinsic lowering to combine common parts...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-13
Craig Topper
Tidy up VSETCC lowering code a bit more by adding an...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-13
Craig Topper
Refactor code a bit to share commonalities. No function...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-13
Craig Topper
Fix an unused variable warning from r161742.
blob
|
commitdiff
|
raw
|
diff to current
2012-08-13
Craig Topper
Remove the LowerMMXCONCAT_VECTORS function. It could...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-12
Craig Topper
Remove call to setOperationAction for SETCC of v4f32...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-12
Craig Topper
Remove unnecessary call to setOperationAction for SETCC...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-12
Craig Topper
Make replace many calls to getSizeInBits() with is128Bi...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-12
Craig Topper
Use MVT.isXBitVector instead of EVT.isXBitVector when...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-11
Michael Liao
fix PR13577, an issue introduced by r161687
blob
|
commitdiff
|
raw
|
diff to current
2012-08-11
Craig Topper
Move setOperationAction for CONCAT_VECTORS for 256...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-10
Michael Liao
add X86-specific DAG optimization to simplify boolean...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-10
Michael Liao
remove tailing whitespaces and test commit
blob
|
commitdiff
|
raw
|
diff to current
2012-08-10
Joerg Sonnenberger
Add some missing includes for the build against stdcxx.
blob
|
commitdiff
|
raw
|
diff to current
2012-08-08
Manman Ren
X86: enable CSE between CMP and SUB
blob
|
commitdiff
|
raw
|
diff to current
2012-08-07
Evan Cheng
X86 cmp lowering is looking past truncate on the condit...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-06
Craig Topper
Implement proper handling for pcmpistri/pcmpestri intri...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-05
Craig Topper
Remove custom inserter for MWAIT. It doesn't do anythin...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-05
Craig Topper
Use a COPY node instead of an explicit MOVA opcode...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-03
Bob Wilson
Fall back to selection DAG isel for calls to builtin...
blob
|
commitdiff
|
raw
|
diff to current
2012-08-01
Chad Rosier
Whitespace.
blob
|
commitdiff
|
raw
|
diff to current
2012-08-01
Elena Demikhovsky
Added FMA functionality to X86 target.
blob
|
commitdiff
|
raw
|
diff to current
2012-07-25
Rafael Espindola
When a return struct pointer is passed in registers...
blob
|
commitdiff
|
raw
|
diff to current
2012-07-23
Sylvestre Ledru
Fix a typo (the the => the)
blob
|
commitdiff
|
raw
|
diff to current
2012-07-17
Evan Cheng
Back out r160101 and instead implement a dag combine...
blob
|
commitdiff
|
raw
|
diff to current
2012-07-17
Evan Cheng
Implement r160312 as target indepedenet dag combine.
blob
|
commitdiff
|
raw
|
diff to current
2012-07-17
Evan Cheng
This is another case where instcombine demanded bits...
blob
|
commitdiff
|
raw
|
diff to current
2012-07-16
Evan Cheng
For something like
blob
|
commitdiff
|
raw
|
diff to current
2012-07-15
Nadav Rotem
Teach getTargetVShiftNode about TargetConstant nodes.
blob
|
commitdiff
|
raw
|
diff to current
2012-07-14
Nadav Rotem
AVX: Fix a bug in getTargetVShiftNode. The shift amount...
blob
|
commitdiff
|
raw
|
diff to current
2012-07-12
Benjamin Kramer
Give the rdrand instructions a SideEffect flag and...
blob
|
commitdiff
|
raw
|
diff to current
2012-07-12
Benjamin Kramer
Add intrinsics for Ivy Bridge's rdrand instruction.
blob
|
commitdiff
|
raw
|
diff to current
2012-07-11
Nadav Rotem
When ext-loading and trunc-storing vectors to memory...
blob
|
commitdiff
|
raw
|
diff to current
2012-07-10
Nadav Rotem
Improve the loading of load-anyext vectors by allowing...
blob
|
commitdiff
|
raw
|
diff to current
2012-07-04
Jakob Stoklund Olesen
Make X86 call and return instructions non-variadic.
blob
|
commitdiff
|
raw
|
diff to current
2012-07-04
Jakob Stoklund Olesen
Ensure CopyToReg nodes are always glued to the call...
blob
|
commitdiff
|
raw
|
diff to current
2012-07-01
Elena Demikhovsky
Optimization of shuffle node that can fit to the regist...
blob
|
commitdiff
|
raw
|
diff to current
2012-06-29
Rafael Espindola
In the initial exec mode we always do a load to find...
blob
|
commitdiff
|
raw
|
diff to current
2012-06-26
Elena Demikhovsky
Removed unused variable
blob
|
commitdiff
|
raw
|
diff to current
2012-06-26
Bill Wendling
Rename to match other X86_64* names.
blob
|
commitdiff
|
raw
|
diff to current
2012-06-26
Elena Demikhovsky
Shuffle optimization for AVX/AVX2.
blob
|
commitdiff
|
raw
|
diff to current
2012-06-25
Eli Friedman
Make some ugly hacks for inline asm operands which...
blob
|
commitdiff
|
raw
|
diff to current
2012-06-24
Jakob Stoklund Olesen
%RCX is not a function live-out in eh.return functions.
blob
|
commitdiff
|
raw
|
diff to current
2012-06-24
Pete Cooper
Remove code i'd been testing with but didn't mean to...
blob
|
commitdiff
|
raw
|
diff to current
2012-06-24
Pete Cooper
DAG legalisation can now handle illegal fma vector...
blob
|
commitdiff
|
raw
|
diff to current
2012-06-23
Rafael Espindola
Handle aliases to tls variables in all architectures...
blob
|
commitdiff
|
raw
|
diff to current
2012-06-20
Craig Topper
Don't insert 128-bit UNDEF into 256-bit vectors. Just...
blob
|
commitdiff
|
raw
|
diff to current
next