Changes in directory llvm/test/Regression/CodeGen/X86:
vec_shuffle-2.ll added (r1.1)
---
Log message:
Added more shuffle tests
---
Diffs of the changes: (+46 -0)
vec_shuffle-2.ll | 46 ++
1 files changed, 46 insertions(+)
Index: llvm/test/Regr
Changes in directory llvm/lib/Bytecode/Reader:
Reader.cpp updated: 1.190 -> 1.191
---
Log message:
We have an assertion that checks that we do not encode null values into the
.bc file if they are supposed to be implicit. This is cool, except that it
checked *after* constant expr folding: impro
Changes in directory llvm/lib/VMCore:
ConstantFolding.cpp updated: 1.85 -> 1.86
---
Log message:
Constant fold extractelement(zero, x) -> zero
---
Diffs of the changes: (+3 -0)
ConstantFolding.cpp |3 +++
1 files changed, 3 insertions(+)
Index: llvm/lib/VMCore/ConstantFolding.cpp
dif
Changes in directory llvm/lib/Transforms/Utils:
ValueMapper.cpp updated: 1.23 -> 1.24
---
Log message:
Fix inlining of insert/extract element constantexprs
---
Diffs of the changes: (+9 -0)
ValueMapper.cpp |9 +
1 files changed, 9 insertions(+)
Index: llvm/lib/Transforms/Uti
Changes in directory llvm/lib/Linker:
LinkModules.cpp updated: 1.111 -> 1.112
---
Log message:
Add code to RemapOperand() to handle Instruction::ExtractElement and
Instruction::InsertElement.
---
Diffs of the changes: (+12 -0)
LinkModules.cpp | 12
1 files changed, 12 insert
Changes in directory llvm/lib/Target/X86:
X86InstrSSE.td updated: 1.65 -> 1.66
---
Log message:
A MOVPS2SSmr, i.e. _mm_store_ss, encoding bug.
Also MOVPDI2DIrr.
---
Diffs of the changes: (+2 -2)
X86InstrSSE.td |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
Index: llvm/lib/
Changes in directory llvm/include/llvm:
IntrinsicsX86.td updated: 1.10 -> 1.11
---
Log message:
Remove int_x86_sse_loadh_ps and int_x86_sse_loadl_ps. These are now lowered
by the frontend to shuffles.
---
Diffs of the changes: (+0 -4)
IntrinsicsX86.td |4
1 files changed, 4 deleti
Changes in directory llvm/lib/Target/X86:
X86ISelLowering.cpp updated: 1.158 -> 1.159
X86ISelLowering.h updated: 1.52 -> 1.53
X86InstrSSE.td updated: 1.64 -> 1.65
---
Log message:
- movlp{s|d} and movhp{s|d} support.
- Normalize shuffle nodes so result vector lower half elements come from the
Changes in directory llvm/lib/Target/X86:
README.txt updated: 1.79 -> 1.80
---
Log message:
New entries.
---
Diffs of the changes: (+56 -0)
README.txt | 56
1 files changed, 56 insertions(+)
Index: llvm/lib/Target/X86/README.txt
Changes in directory llvm/lib/CodeGen/SelectionDAG:
DAGCombiner.cpp updated: 1.144 -> 1.145
---
Log message:
1. If both vector operands of a vector_shuffle are undef, turn it into an undef.
2. A shuffle mask element can also be an undef.
---
Diffs of the changes: (+6 -3)
DAGCombiner.cpp |
Changes in directory llvm/lib/Target/Alpha:
AlphaISelLowering.cpp updated: 1.45 -> 1.46
---
Log message:
This may be overconservative, but it lets the new cfe compile
---
Diffs of the changes: (+2 -1)
AlphaISelLowering.cpp |3 ++-
1 files changed, 2 insertions(+), 1 deletion(-)
Index:
Changes in directory llvm/lib/Target/PowerPC:
README_ALTIVEC.txt updated: 1.18 -> 1.19
---
Log message:
Add an item
---
Diffs of the changes: (+6 -1)
README_ALTIVEC.txt |7 ++-
1 files changed, 6 insertions(+), 1 deletion(-)
Index: llvm/lib/Target/PowerPC/README_ALTIVEC.txt
diff
Changes in directory llvm/lib/Target/PowerPC:
PPCISelLowering.cpp updated: 1.137 -> 1.138
---
Log message:
Make sure to return the result in the right type.
---
Diffs of the changes: (+6 -4)
PPCISelLowering.cpp | 10 ++
1 files changed, 6 insertions(+), 4 deletions(-)
Index: ll
Changes in directory llvm/lib/VMCore:
ConstantRange.cpp (r1.14) removed
---
Log message:
Move this to lib/Analysis.
This reduces Core size, yay.
---
Diffs of the changes: (+0 -0)
0 files changed
___
llvm-commits mailing list
llvm-commits@cs.u
Changes in directory llvm/lib/Target/PowerPC:
PPCISelLowering.cpp updated: 1.136 -> 1.137
PPCISelLowering.h updated: 1.43 -> 1.44
PPCInstrAltivec.td updated: 1.45 -> 1.46
---
Log message:
Match vpku[hw]um(x,x).
Convert vsldoi(x,x) to work the same way other (x,x) cases work.
---
Diffs of the
Changes in directory llvm/test/Regression/CodeGen/PowerPC:
vec_shuffle.ll updated: 1.4 -> 1.5
---
Log message:
Add testcases for vpku[hw]um(x,x)
---
Diffs of the changes: (+71 -1)
vec_shuffle.ll | 72 -
1 files changed, 71 insertion
Changes in directory llvm/test:
Makefile updated: 1.84 -> 1.85
---
Log message:
Add the variable llvmgccmajvers to the site.exp file. This will contain
the major version number of llvm-gcc, as configured.
---
Diffs of the changes: (+1 -0)
Makefile |1 +
1 files changed, 1 insertion(+)
Changes in directory llvm/runtime:
Makefile updated: 1.25 -> 1.26
---
Log message:
Avoid building the runtime libraries if llvm-gcc version 4 is detected.
---
Diffs of the changes: (+8 -0)
Makefile |8
1 files changed, 8 insertions(+)
Index: llvm/runtime/Makefile
diff -u llv
Changes in directory llvm:
Makefile updated: 1.51 -> 1.52
---
Log message:
Avoid building the runtime libraries if llvm-gcc version 4 is detected.
---
Diffs of the changes: (+11 -6)
Makefile | 17 +++--
1 files changed, 11 insertions(+), 6 deletions(-)
Index: llvm/Makefile
Changes in directory llvm/test/Regression/CodeGen/PowerPC:
vec_shuffle.ll updated: 1.3 -> 1.4
---
Log message:
Add vmrg(x,x) tests
---
Diffs of the changes: (+156 -2)
vec_shuffle.ll | 158 -
1 files changed, 156 insertions(+), 2 del
Changes in directory llvm/lib/Target/PowerPC:
PPCISelLowering.cpp updated: 1.135 -> 1.136
PPCISelLowering.h updated: 1.42 -> 1.43
PPCInstrAltivec.td updated: 1.44 -> 1.45
---
Log message:
Add support for matching vmrg(x,x) patterns
---
Diffs of the changes: (+81 -41)
PPCISelLowering.cpp |
Changes in directory llvm:
configure updated: 1.215 -> 1.216
Makefile.config.in updated: 1.53 -> 1.54
---
Log message:
For support of new GCC v4, obtain the full and major versions of the
llvm-gcc command. This will help distinguish which version is being used
so the makefiles can accommodate.
Changes in directory llvm/autoconf:
configure.ac updated: 1.213 -> 1.214
---
Log message:
For support of new GCC v4, obtain the full and major versions of the
llvm-gcc command. This will help distinguish which version is being used
so the makefiles can accommodate.
---
Diffs of the changes:
Changes in directory llvm/lib/Target/Alpha:
AlphaISelLowering.cpp updated: 1.44 -> 1.45
---
Log message:
fix some linking problems with the new gcc
---
Diffs of the changes: (+3 -1)
AlphaISelLowering.cpp |4 +++-
1 files changed, 3 insertions(+), 1 deletion(-)
Index: llvm/lib/Target/A
Changes in directory llvm/test/Regression/CodeGen/PowerPC:
vec_shuffle.ll updated: 1.2 -> 1.3
---
Log message:
Add tests for vmrg[hl]*
---
Diffs of the changes: (+164 -2)
vec_shuffle.ll | 166 -
1 files changed, 164 insertions(+), 2
Changes in directory llvm/include/llvm:
IntrinsicsPowerPC.td updated: 1.22 -> 1.23
---
Log message:
These intrinsics are now lowered by the FE.
---
Diffs of the changes: (+0 -20)
IntrinsicsPowerPC.td | 20
1 files changed, 20 deletions(-)
Index: llvm/include/llvm/I
Changes in directory llvm/lib/Target/PowerPC:
PPCISelLowering.cpp updated: 1.134 -> 1.135
PPCISelLowering.h updated: 1.41 -> 1.42
PPCInstrAltivec.td updated: 1.43 -> 1.44
README_ALTIVEC.txt updated: 1.17 -> 1.18
---
Log message:
Pattern match vmrg* instructions, which are now lowered by the CFE
Changes in directory llvm/utils/TableGen:
DAGISelEmitter.cpp updated: 1.193 -> 1.194
---
Log message:
Infer element types for shuffle masks
---
Diffs of the changes: (+20 -0)
DAGISelEmitter.cpp | 20
1 files changed, 20 insertions(+)
Index: llvm/utils/TableGen/DAGI
Changes in directory llvm/utils/TableGen:
DAGISelEmitter.cpp updated: 1.192 -> 1.193
---
Log message:
rename a method, to avoid confusion with llvm intrinsics.
---
Diffs of the changes: (+3 -4)
DAGISelEmitter.cpp |7 +++
1 files changed, 3 insertions(+), 4 deletions(-)
Index: llv
Changes in directory llvm/test/Regression/CodeGen/PowerPC:
vec_shuffle.ll updated: 1.1 -> 1.2
---
Log message:
test vperm promotion
---
Diffs of the changes: (+16 -0)
vec_shuffle.ll | 16
1 files changed, 16 insertions(+)
Index: llvm/test/Regression/CodeGen/PowerPC/vec
Changes in directory llvm/lib/Target/PowerPC:
README_ALTIVEC.txt updated: 1.16 -> 1.17
---
Log message:
remove two done items
---
Diffs of the changes: (+0 -10)
README_ALTIVEC.txt | 10 --
1 files changed, 10 deletions(-)
Index: llvm/lib/Target/PowerPC/README_ALTIVEC.txt
diff -
Changes in directory llvm/lib/Transforms/Scalar:
InstructionCombining.cpp updated: 1.461 -> 1.462
---
Log message:
Lower vperm(x,y, mask) -> shuffle(x,y,mask) if mask is constant. This allows
us to compile oh-so-realistic stuff like this:
vec_vperm(A, B, (vector unsigned char){14});
to:
Changes in directory llvm/include/llvm:
IntrinsicsPowerPC.td updated: 1.21 -> 1.22
---
Log message:
this is no longer an intrinsic.
---
Diffs of the changes: (+0 -5)
IntrinsicsPowerPC.td |5 -
1 files changed, 5 deletions(-)
Index: llvm/include/llvm/IntrinsicsPowerPC.td
diff -u l
Changes in directory llvm/test/Regression/CodeGen/PowerPC:
vec_shuffle.ll added (r1.1)
---
Log message:
new testcase
---
Diffs of the changes: (+86 -0)
vec_shuffle.ll | 86 +
1 files changed, 86 insertions(+)
Index: llvm/test/Regr
Changes in directory llvm/lib/Target/PowerPC:
PPCISelLowering.cpp updated: 1.133 -> 1.134
PPCISelLowering.h updated: 1.40 -> 1.41
PPCInstrAltivec.td updated: 1.42 -> 1.43
README_ALTIVEC.txt updated: 1.15 -> 1.16
---
Log message:
Support pattern matching vsldoi(x,y) and vsldoi(x,x), which allows
Changes in directory llvm/lib/Target/PowerPC:
PPCISelLowering.cpp updated: 1.132 -> 1.133
PPCISelLowering.h updated: 1.39 -> 1.40
PPCInstrAltivec.td updated: 1.41 -> 1.42
---
Log message:
Compile the vpkuhum/vpkuwum intrinsics into vpkuhum/vpkuwum instead of into
vperm with a perm mask lvx'd fr
36 matches
Mail list logo
