Changes in directory llvm/tools/llvm-upgrade:
UpgradeParser.y updated: 1.35 - 1.36
---
Log message:
Upgrade the llvm.isunordered intrinsics to fcmp uno instruction.
---
Diffs of the changes: (+24 -14)
UpgradeParser.y | 38 --
1 files changed, 24
Changes in directory llvm/tools/llvm-upgrade:
UpgradeParser.cpp updated: 1.37 - 1.38
UpgradeParser.cpp.cvs updated: 1.35 - 1.36
UpgradeParser.y.cvs updated: 1.34 - 1.35
---
Log message:
Regenerate.
---
Diffs of the changes: (+104 -74)
UpgradeParser.cpp | 70
Changes in directory llvm/include/llvm:
Module.h updated: 1.78 - 1.79
---
Log message:
Change the interface to Module::getOrInsertFunction to be easier to use,
to resolve PR1088: http://llvm.org/PR1088 , and to help PR411:
http://llvm.org/PR411 .
---
Diffs of the changes: (+10 -4)
Changes in directory llvm/lib/VMCore:
Module.cpp updated: 1.71 - 1.72
---
Log message:
Change the interface to Module::getOrInsertFunction to be easier to use,
to resolve PR1088: http://llvm.org/PR1088 , and to help PR411:
http://llvm.org/PR411 .
---
Diffs of the changes: (+24 -12)
Changes in directory llvm/lib/CodeGen:
IntrinsicLowering.cpp updated: 1.55 - 1.56
---
Log message:
Change the interface to Module::getOrInsertFunction to be easier to use,to
resolve PR1088: http://llvm.org/PR1088 , and to help PR411:
http://llvm.org/PR411 .
This simplifies many clients also
Changes in directory llvm/lib/Transforms/Utils:
LowerAllocations.cpp updated: 1.69 - 1.70
LowerInvoke.cpp updated: 1.49 - 1.50
---
Log message:
Change the interface to Module::getOrInsertFunction to be easier to use,to
resolve PR1088: http://llvm.org/PR1088 , and to help PR411:
Changes in directory llvm/tools/bugpoint:
Miscompilation.cpp updated: 1.83 - 1.84
---
Log message:
relax type
---
Diffs of the changes: (+1 -1)
Miscompilation.cpp |2 +-
1 files changed, 1 insertion(+), 1 deletion(-)
Index: llvm/tools/bugpoint/Miscompilation.cpp
diff -u
Changes in directory llvm/examples/Fibonacci:
fibonacci.cpp updated: 1.14 - 1.15
---
Log message:
add some casts to support a change in the getOrInsertFunction interface
---
Diffs of the changes: (+3 -2)
fibonacci.cpp |5 +++--
1 files changed, 3 insertions(+), 2 deletions(-)
Index:
Changes in directory llvm/lib/Target/X86:
X86TargetAsmInfo.cpp updated: 1.17 - 1.18
---
Log message:
relax type
---
Diffs of the changes: (+1 -1)
X86TargetAsmInfo.cpp |2 +-
1 files changed, 1 insertion(+), 1 deletion(-)
Index: llvm/lib/Target/X86/X86TargetAsmInfo.cpp
diff -u
Changes in directory llvm/examples/ParallelJIT:
ParallelJIT.cpp updated: 1.8 - 1.9
---
Log message:
add some casts to support a change in the getOrInsertFunction interface
---
Diffs of the changes: (+8 -8)
ParallelJIT.cpp | 16
1 files changed, 8 insertions(+), 8
Changes in directory llvm/lib/Transforms/Utils:
InlineFunction.cpp updated: 1.48 - 1.49
---
Log message:
prepare for adjustment to getOrInsertFunction method
---
Diffs of the changes: (+8 -6)
InlineFunction.cpp | 14 --
1 files changed, 8 insertions(+), 6 deletions(-)
Changes in directory llvm/lib/Transforms/Scalar:
LowerGC.cpp updated: 1.17 - 1.18
---
Log message:
relax some types
---
Diffs of the changes: (+1 -1)
LowerGC.cpp |2 +-
1 files changed, 1 insertion(+), 1 deletion(-)
Index: llvm/lib/Transforms/Scalar/LowerGC.cpp
diff -u
Changes in directory llvm/lib/Transforms/Instrumentation:
ProfilingUtils.cpp updated: 1.13 - 1.14
RSProfiling.cpp updated: 1.14 - 1.15
TraceBasicBlocks.cpp updated: 1.20 - 1.21
TraceValues.cpp updated: 1.79 - 1.80
---
Log message:
relax some types
---
Diffs of the changes: (+20 -21)
Changes in directory llvm/lib/Transforms/IPO:
LowerSetJmp.cpp updated: 1.35 - 1.36
---
Log message:
relax types
---
Diffs of the changes: (+7 -7)
LowerSetJmp.cpp | 14 +++---
1 files changed, 7 insertions(+), 7 deletions(-)
Index: llvm/lib/Transforms/IPO/LowerSetJmp.cpp
diff
Changes in directory llvm/tools/lli:
lli.cpp updated: 1.63 - 1.64
---
Log message:
this is an overcomplex way to call exit :)
---
Diffs of the changes: (+1 -12)
lli.cpp | 13 +
1 files changed, 1 insertion(+), 12 deletions(-)
Index: llvm/tools/lli/lli.cpp
diff -u
Changes in directory llvm/lib/Transforms/Scalar:
InstructionCombining.cpp updated: 1.585 - 1.586
---
Log message:
relax some types
---
Diffs of the changes: (+2 -3)
InstructionCombining.cpp |5 ++---
1 files changed, 2 insertions(+), 3 deletions(-)
Index:
Changes in directory llvm/lib/Analysis:
ConstantFolding.cpp updated: 1.8 - 1.9
---
Log message:
remove llvm.isunordered
---
Diffs of the changes: (+0 -5)
ConstantFolding.cpp |5 -
1 files changed, 5 deletions(-)
Index: llvm/lib/Analysis/ConstantFolding.cpp
diff -u
Changes in directory llvm/test/Regression/CodeGen/CBackend:
2007-01-06-Signless.c added (r1.1)
---
Log message:
Add a test case for PR1086: http://llvm.org/PR1086
---
Diffs of the changes: (+6 -0)
2007-01-06-Signless.c |6 ++
1 files changed, 6 insertions(+)
Index:
Update code to work with new Module::getOrInsertFunction
prototype. Fix unordered comparison emission from being this:
%tmp = fcmp uno float %tmp, %tmp1 ; bool
[#uses=1]
%tmp2 = call bool %llvm.isunordered.f32( float %tmp, float %
tmp1 )
%tmp3 = or bool
Changes in directory llvm/lib/CodeGen/SelectionDAG:
SelectionDAGISel.cpp updated: 1.340 - 1.341
---
Log message:
remove support for llvm.isunordered
---
Diffs of the changes: (+0 -22)
SelectionDAGISel.cpp | 22 --
1 files changed, 22 deletions(-)
Index:
Changes in directory llvm/projects/Stacker/lib/compiler:
StackerCompiler.cpp updated: 1.29 - 1.30
---
Log message:
relax type
---
Diffs of the changes: (+1 -1)
StackerCompiler.cpp |2 +-
1 files changed, 1 insertion(+), 1 deletion(-)
Index:
Changes in directory llvm/lib/CodeGen:
IntrinsicLowering.cpp updated: 1.56 - 1.57
---
Log message:
llvm 2.0 doesn't support llvm.isunordered.*
---
Diffs of the changes: (+0 -17)
IntrinsicLowering.cpp | 17 -
1 files changed, 17 deletions(-)
Index:
Changes in directory llvm/include/llvm:
Intrinsics.td updated: 1.37 - 1.38
---
Log message:
llvm 2.0 doesn't support llvm.isunordered.*
---
Diffs of the changes: (+0 -4)
Intrinsics.td |4
1 files changed, 4 deletions(-)
Index: llvm/include/llvm/Intrinsics.td
diff -u
Changes in directory llvm/docs:
LangRef.html updated: 1.188 - 1.189
---
Log message:
llvm.isunordered is no longer alive.
---
Diffs of the changes: (+1 -37)
LangRef.html | 38 +-
1 files changed, 1 insertion(+), 37 deletions(-)
Index:
Changes in directory llvm-test/SingleSource/UnitTests/Vector/Altivec:
2007-01-07-lvsl-lvsr-Regression.c added (r1.1)
---
Log message:
Regression test for llvm-rs6000.cpp function cache overlap.
---
Diffs of the changes: (+28 -0)
2007-01-07-lvsl-lvsr-Regression.c | 28
When I converted the megamacro over to C, I should have listened to
the alarms going off in my head when I saw static Function *Cache =
NULL;. There should have been a per call site cache. The result of
this mistake was that the same function prototype was being used for
several altivec
Ignore previous patch. This one doesn't trounce Chris' patch.
FunctionCache.patch
Description: Binary data
On 7-Jan-07, at 09:49 AM, Jim Laskey wrote:
When I converted the megamacro over to C, I should have listened to
the alarms going off in my head when I saw static Function *Cache
Changes in directory llvm/include/llvm/Transforms:
Instrumentation.h updated: 1.10 - 1.11
---
Log message:
remove an old instrumentation pass that is not used anymore.
---
Diffs of the changes: (+0 -12)
Instrumentation.h | 12
1 files changed, 12 deletions(-)
Index:
Changes in directory llvm/lib/Transforms/Instrumentation:
TraceValues.cpp (r1.80) removed
---
Log message:
this pass is unused
---
Diffs of the changes: (+0 -0)
0 files changed
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
Changes in directory llvm/runtime/libtrace:
Makefile (r1.16) removed
README.txt (r1.3) removed
tracelib.c (r1.14) removed
tracelib.h (r1.5) removed
---
Log message:
the old trace values pass has been removed, remove its runtime library.
---
Diffs of the changes: (+0 -0)
0 files changed
Changes in directory llvm/runtime:
Makefile updated: 1.30 - 1.31
---
Log message:
the old trace values pass has been removed, remove its runtime library.
---
Diffs of the changes: (+1 -1)
Makefile |2 +-
1 files changed, 1 insertion(+), 1 deletion(-)
Index: llvm/runtime/Makefile
Changes in directory llvm/include/llvm:
LinkAllPasses.h updated: 1.6 - 1.7
---
Log message:
remove an old instrumentation pass that is not used anymore.
---
Diffs of the changes: (+0 -2)
LinkAllPasses.h |2 --
1 files changed, 2 deletions(-)
Index: llvm/include/llvm/LinkAllPasses.h
Changes in directory llvm/test/Feature:
calltest.ll updated: 1.6 - 1.7
instructions.ll updated: 1.4 - 1.5
---
Log message:
Redefinition of functions is no longer permitted.
---
Diffs of the changes: (+3 -4)
calltest.ll |3 +--
instructions.ll |4 ++--
2 files changed, 3
The SVN mirror is still not updating. It is stuck at r240 which is now
several days old. There have been many important changes and fixes since
then. Could someone please create a cumulative patch from r240 so the
mirror users only have one patch to apply?
Thanks,
Reid.
Changes in directory llvm/lib/Transforms/IPO:
SimplifyLibCalls.cpp updated: 1.82 - 1.83
---
Log message:
Types should be const.
---
Diffs of the changes: (+2 -2)
SimplifyLibCalls.cpp |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
Index:
Changes in directory llvm/test/Regression/Transforms/DeadStoreElimination:
2006-06-27-AST-Remove.ll updated: 1.3 - 1.4
---
Log message:
llvm.isunordered.* are gone.
---
Diffs of the changes: (+0 -2)
2006-06-27-AST-Remove.ll |2 --
1 files changed, 2 deletions(-)
Index:
All,
Attached is a cumulative patch with all changes in the Apple repository
since the mirror's r240 revision. If you haven't applied the various
patches sent in the last few days, this one will be much easier.
The attached patch also contains one thing that the Apple repository
doesn't have:
Changes in directory llvm/lib/Analysis:
ConstantRange.cpp updated: 1.23 - 1.24
ScalarEvolution.cpp updated: 1.80 - 1.81
---
Log message:
Convert uses of getPrimitiveSize that should be getPrimitiveSizeInBits.
---
Diffs of the changes: (+14 -15)
ConstantRange.cpp | 17 -
Changes in directory llvm:
CREDITS.TXT updated: 1.66 - 1.67
---
Log message:
Testing commit, and adding an entry to CREDITS.TXT
---
Diffs of the changes: (+4 -0)
CREDITS.TXT |4
1 files changed, 4 insertions(+)
Index: llvm/CREDITS.TXT
diff -u llvm/CREDITS.TXT:1.66
Chris,
This patch breaks:
SingleSource/UnitTests/2003-05-14-AtExit
for sure and possibly:
External/SPEC/CINT2000/252.eon/252.eon
Please revert or fix.
Reid.
On Sun, 2007-01-07 at 00:43 -0600, Chris Lattner wrote:
Changes in directory llvm/tools/lli:
lli.cpp updated: 1.63 - 1.64
Changes in directory llvm/lib/Analysis:
ConstantRange.cpp updated: 1.24 - 1.25
---
Log message:
Fix a bug in an assert that would never trigger.
---
Diffs of the changes: (+1 -1)
ConstantRange.cpp |2 +-
1 files changed, 1 insertion(+), 1 deletion(-)
Index:
Changes in directory llvm/tools:
Makefile updated: 1.60 - 1.61
---
Log message:
Build libLLVMlto on non-Darwin architectures. Resolves PR1055:
http://llvm.org/PR1055 : http://llvm.org/PR1055: http://llvm.org/PR1055
---
Diffs of the changes: (+2 -2)
Makefile |4 ++--
1 files changed,
Changes in directory llvm/tools/lto:
Makefile updated: 1.6 - 1.7
lto.cpp updated: 1.28 - 1.29
---
Log message:
Build libLLVMlto on non-Darwin architectures. Resolves PR1055:
http://llvm.org/PR1055 : http://llvm.org/PR1055: http://llvm.org/PR1055
---
Diffs of the changes: (+10 -3)
Changes in directory llvm/lib/Target/CBackend:
Writer.cpp updated: 1.304 - 1.305
---
Log message:
Fix PR1090: http://llvm.org/PR1090 :
Implemented some llvm_fcmp_{pred} functions at the start of the function bodies
and use them for fcmp instructions and constant expressions. These help
+ // Emit some helper functions for dealing with FCMP instruction's
+ // predicates
+ Out static inline int llvm_fcmp_ord(double X, double Y) { ;
+ Out return X == X Y == Y; }\n;
+ Out static inline int llvm_fcmp_uno(double X, double Y) { ;
+ Out return X != X || Y != Y; }\n;
+
Changes in directory llvm/tools/lli:
lli.cpp updated: 1.64 - 1.65
---
Log message:
fix atexit. This is an overcomplex way of calling exit, but it is required,
as the jit intercepts exit calls to implement atexit handlers. This
fixes SingleSource/UnitTests/2003-05-14-AtExit
---
Diffs of the
This corrects llvm-gcc to always emit the llvm.va* intrinsics with
i8* as the valist pointer type. This fixes llvm-gcc to comply with
PR1093.
Index: llvm-convert.cpp
===
--- llvm-convert.cpp(revision 122093)
+++
Changes in directory llvm/docs:
LangRef.html updated: 1.189 - 1.190
---
Log message:
Fully specify the type of the llvm.va* intrinsics. This helps resolve
Pr1093: http://llvm.org/PR1093
---
Diffs of the changes: (+16 -9)
LangRef.html | 25 -
1 files changed,
48 matches
Mail list logo
