Author: evancheng
Date: Fri Jan 18 02:39:55 2008
New Revision: 46160
URL: http://llvm.org/viewvc/llvm-project?rev=46160&view=rev
Log:
Eliminate a compilation warning: mixing declarations and code is badness.
Forgot it's not C++.
Modified:
llvm-gcc-4.2/trunk/gcc/config/i386/i386.c
Modified:
Hi Evan,
> Do not pass zero sized array, struct, or class using byval attribute. This is
> true on both x86-32 and x86-64.
is this because LLVM can't handle it, or is it an optimization?
If LLVM can't handle passing zero-sized objects using byval, that
sounds like a bug...
Thanks,
Duncan.
On Jan 18, 2008, at 10:13 AM, Tanya Lattner wrote:
> Should the test/CodeGen/X86/aligned-comm.ll be un-XFAILED? Its now
> XPASSing.
>
> -Tanya
I guess so; given the current state of llvm it is supposed to pass.
Longterm what should happen is that tentative definitions ("common")
get represe
Should the test/CodeGen/X86/aligned-comm.ll be un-XFAILED? Its now
XPASSing.
-Tanya
On Jan 17, 2008, at 3:04 PM, Dale Johannesen wrote:
> Author: johannes
> Date: Thu Jan 17 17:04:07 2008
> New Revision: 46144
>
> URL: http://llvm.org/viewvc/llvm-project?rev=46144&view=rev
> Log:
> Revert the
On Jan 17, 2008, at 10:03 PM, Duncan Sands wrote:
> Hi Chris,
>> + LegalizeAction getTruncStoreAction(MVT::ValueType ValVT,
>> + MVT::ValueType MemVT) const {
>> +assert(ValVT < MVT::LAST_VALUETYPE && MemVT < 32 &&
>
> what is 32? Did you mean <= LAST_INTEG
On Jan 17, 2008, at 1:31 AM, Duncan Sands wrote:
> Hi Dale,
>
>> Do not mark EH tables no-dead-strip unless the
>> associated function is so marked.
>
> can you please explain what this means?
Dead stripping is a linktime optimization, where code not reachable
from main is removed. Functions
Author: dpatel
Date: Fri Jan 18 16:41:06 2008
New Revision: 46181
URL: http://llvm.org/viewvc/llvm-project?rev=46181&view=rev
Log:
Do not raise error if system llvm (e.g. --enable-llvm=/usr/local) is used.
Modified:
llvm-gcc-4.2/trunk/gcc/configure
llvm-gcc-4.2/trunk/gcc/configure.ac
Mod
Author: evancheng
Date: Fri Jan 18 15:01:00 2008
New Revision: 46180
URL: http://llvm.org/viewvc/llvm-project?rev=46180&view=rev
Log:
Fix makefiles to enable Apply style debug build.
Modified:
llvm/trunk/utils/buildit/GNUmakefile
llvm/trunk/utils/buildit/build_llvm
Modified: llvm/trunk/u
Author: johannes
Date: Fri Jan 18 13:55:37 2008
New Revision: 46174
URL: http://llvm.org/viewvc/llvm-project?rev=46174&view=rev
Log:
Implement flt_rounds for PowerPC.
Added:
llvm/trunk/test/CodeGen/PowerPC/frounds.ll
Modified:
llvm/trunk/lib/Target/PowerPC/PPCISelLowering.cpp
Modified:
Author: lattner
Date: Fri Jan 18 13:53:43 2008
New Revision: 46173
URL: http://llvm.org/viewvc/llvm-project?rev=46173&view=rev
Log:
remove extraneous &&'s from tests, as Scott is apparently not going to.
Modified:
llvm/trunk/test/CodeGen/CellSPU/call.ll
llvm/trunk/test/CodeGen/CellSPU/cal
Author: lattner
Date: Fri Jan 18 10:54:56 2008
New Revision: 46161
URL: http://llvm.org/viewvc/llvm-project?rev=46161&view=rev
Log:
Fix a latent bug exposed by my truncstore patch. We compiled stfiwx-2.ll to:
_test:
fctiwz f0, f1
stfiwx f0, 0, r4
blr
instead of:
_test:
Author: lattner
Date: Fri Jan 18 13:36:20 2008
New Revision: 46171
URL: http://llvm.org/viewvc/llvm-project?rev=46171&view=rev
Log:
remove extraneous &'s.
Modified:
llvm/trunk/lib/CodeGen/SelectionDAG/TargetLowering.cpp
Modified: llvm/trunk/lib/CodeGen/SelectionDAG/TargetLowering.cpp
URL:
h
Author: tbrethou
Date: Fri Jan 18 13:13:34 2008
New Revision: 46167
URL: http://llvm.org/viewvc/llvm-project?rev=46167&view=rev
Log:
Merge from mainline.
Revert the part of 45848 that treated weak globals
as weak globals rather than commons. While not wrong,
this change tickled a latent bug in Da
Author: laurov
Date: Fri Jan 18 12:14:23 2008
New Revision: 46163
URL: http://llvm.org/viewvc/llvm-project?rev=46163&view=rev
Log:
Eliminate a compilation warning.
Modified:
llvm-gcc-4.2/trunk/gcc/config/arm/arm.c
Modified: llvm-gcc-4.2/trunk/gcc/config/arm/arm.c
URL:
http://llvm.org/viewv
Author: dpatel
Date: Fri Jan 18 14:02:18 2008
New Revision: 46175
URL: http://llvm.org/viewvc/llvm-project?rev=46175&view=rev
Log:
Place for Apple specific branches.
Added:
llvm/branches/Apple/
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
On Jan 18, 2008, at 11:40 AM, Eric Christopher wrote:
>
> On Jan 18, 2008, at 11:35 AM, Devang Patel wrote:
>
>> llvm-gcc-4.2/trunk/gcc/autom4te.cache/output.0
>> llvm-gcc-4.2/trunk/gcc/autom4te.cache/requests
>> llvm-gcc-4.2/trunk/gcc/autom4te.cache/traces.0
>
> Any reason why the auto-cach
Author: johannes
Date: Fri Jan 18 13:53:31 2008
New Revision: 46172
URL: http://llvm.org/viewvc/llvm-project?rev=46172&view=rev
Log:
Test is correct again for the moment.
Modified:
llvm/trunk/test/CodeGen/X86/aligned-comm.ll
Modified: llvm/trunk/test/CodeGen/X86/aligned-comm.ll
URL:
http:/
On Jan 18, 2008, at 11:35 AM, Devang Patel wrote:
> llvm-gcc-4.2/trunk/gcc/autom4te.cache/output.0
>llvm-gcc-4.2/trunk/gcc/autom4te.cache/requests
>llvm-gcc-4.2/trunk/gcc/autom4te.cache/traces.0
Any reason why the auto-cache is checked into svn?
-eric
_
Author: tbrethou
Date: Fri Jan 18 13:18:51 2008
New Revision: 46169
URL: http://llvm.org/viewvc/llvm-project?rev=46169&view=rev
Log:
UnXFAIL this test.
Modified:
llvm/branches/release_22/test/CodeGen/X86/aligned-comm.ll
Modified: llvm/branches/release_22/test/CodeGen/X86/aligned-comm.ll
URL:
Author: tbrethou
Date: Fri Jan 18 13:15:06 2008
New Revision: 46168
URL: http://llvm.org/viewvc/llvm-project?rev=46168&view=rev
Log:
Merge from mainline.
Revert the part of 45849 that treated weak globals
as weak globals rather than commons. While not wrong,
this change tickled a latent bug in Da
On Jan 18, 2008, at 10:19 AM, Dale Johannesen wrote:
> On Jan 18, 2008, at 10:13 AM, Tanya Lattner wrote:
>> Should the test/CodeGen/X86/aligned-comm.ll be un-XFAILED? Its now
>> XPASSing.
>>
>> -Tanya
>
> I guess so; given the current state of llvm it is supposed to pass.
> Longterm what should ha
Hi Evan,
> It's for ABI compatibility reason. Also, it makes sense.
presumably this results in the parameter being dropped?
If so, the LLVM codegenerators could do that too. And
either they should, or the verifier should check for this
case.
Ciao,
Duncan.
It's for ABI compatibility reason. Also, it makes sense.
Evan
On Jan 18, 2008, at 1:39 AM, Duncan Sands wrote:
> Hi Evan,
>
>> Do not pass zero sized array, struct, or class using byval
>> attribute. This is true on both x86-32 and x86-64.
>
> is this because LLVM can't handle it, or is it an
Author: lattner
Date: Fri Jan 18 11:13:03 2008
New Revision: 46162
URL: http://llvm.org/viewvc/llvm-project?rev=46162&view=rev
Log:
remove magic numbers.
Modified:
llvm/trunk/include/llvm/Target/TargetLowering.h
Modified: llvm/trunk/include/llvm/Target/TargetLowering.h
URL:
http://llvm.org/
Author: lattner
Date: Fri Jan 18 12:51:16 2008
New Revision: 46165
URL: http://llvm.org/viewvc/llvm-project?rev=46165&view=rev
Log:
get symbolic information for ppc ldbl nodes.
Modified:
llvm/trunk/lib/Target/PowerPC/PPCISelLowering.cpp
Modified: llvm/trunk/lib/Target/PowerPC/PPCISelLowering
Author: evancheng
Date: Fri Jan 18 12:35:18 2008
New Revision: 46164
URL: http://llvm.org/viewvc/llvm-project?rev=46164&view=rev
Log:
i32 / i64 all integer structs are not passed byval.
Modified:
llvm-gcc-4.2/trunk/gcc/config/i386/llvm-i386-target.h
llvm-gcc-4.2/trunk/gcc/config/i386/llvm
Author: lattner
Date: Fri Jan 18 12:54:31 2008
New Revision: 46166
URL: http://llvm.org/viewvc/llvm-project?rev=46166&view=rev
Log:
don't form an std::string with a null pointer, it aborts.
Modified:
llvm/trunk/include/llvm/ADT/StringExtras.h
Modified: llvm/trunk/include/llvm/ADT/StringExtra
Author: kremenek
Date: Fri Jan 18 22:22:50 2008
New Revision: 46187
URL: http://llvm.org/viewvc/llvm-project?rev=46187&view=rev
Log:
Made 'FoldingSetNodeID' a proper class instead of a nested class in
'FoldingSetNodeImpl' (previously 'FoldingSetNodeID' was a typedef of
'FoldingSetNodeImpl::NodeID'
Author: kremenek
Date: Fri Jan 18 22:23:33 2008
New Revision: 46188
URL: http://llvm.org/viewvc/llvm-project?rev=46188&view=rev
Log:
Added FoldingSet style 'profiling' support for APInt.
Modified:
llvm/trunk/include/llvm/ADT/APInt.h
llvm/trunk/lib/Support/APInt.cpp
Modified: llvm/trunk/i
Author: kremenek
Date: Fri Jan 18 21:58:00 2008
New Revision: 46186
URL: http://llvm.org/viewvc/llvm-project?rev=46186&view=rev
Log:
Made 'profiling' of objects in a FoldingSet trait-based using FoldingSetTrait
instead of always assuming that the stored objects had a method called
'Profile'. The d
Hi Chris,
> 32 is sizeof(uint64_t)*8 / 2, because each entry takes two bits in a
> uint64_t.
>
> I converted this to symbolic math to make it more obvious, and
> switched to using array_lengthof instead of comparing against
> MVT::LAST_VALUETYPE directly.
thanks!
> > This may be wrong for
Author: dpatel
Date: Fri Jan 18 14:03:19 2008
New Revision: 46176
URL: http://llvm.org/viewvc/llvm-project?rev=46176&view=rev
Log:
Create llvmCore-2010 branch based on llvmCore-2009 tag.
Added:
llvm/branches/Apple/llvmCore-2010/
- copied from r46175, llvm/tags/Apple/llvmCore-2009/
Author: dpatel
Date: Fri Jan 18 14:10:27 2008
New Revision: 46178
URL: http://llvm.org/viewvc/llvm-project?rev=46178&view=rev
Log:
Tag llvmCore-2010
Added:
llvm/tags/Apple/llvmCore-2010/
- copied from r46177, llvm/branches/Apple/llvmCore-2010/
__
Author: kremenek
Date: Fri Jan 18 22:51:55 2008
New Revision: 46190
URL: http://llvm.org/viewvc/llvm-project?rev=46190&view=rev
Log:
Modified ImmutableSet/ImmutableMap to use FoldingSet profiling using
FoldingSetTrait instead of directly calling a 'Profile' method.
Modified:
llvm/trunk/includ
Author: kremenek
Date: Fri Jan 18 22:31:12 2008
New Revision: 46189
URL: http://llvm.org/viewvc/llvm-project?rev=46189&view=rev
Log:
Added FoldingSet style 'profiling' support for APSInt.
Added:
llvm/trunk/lib/Support/APSInt.cpp
Modified:
llvm/trunk/include/llvm/ADT/APSInt.h
Modified: ll
Author: evancheng
Date: Sat Jan 19 00:03:38 2008
New Revision: 46191
URL: http://llvm.org/viewvc/llvm-project?rev=46191&view=rev
Log:
Much improved x86-64 ABI compliance.
Modified:
llvm-gcc-4.2/trunk/gcc/config/i386/i386-protos.h
llvm-gcc-4.2/trunk/gcc/config/i386/i386.c
llvm-gcc-4.2/
36 matches
Mail list logo
