Hi Dale,
While that's nonstandard, it's hard to imagine a memcpy
implementation that would screw it up.
But they can partially overlap:
how does mainline handle this? I know it used to get it
wrong (the testcase I sent was filched from gcc bugzilla),
but the bug was marked as resolved,
Author: void
Date: Thu Nov 8 02:58:47 2007
New Revision: 43886
URL: http://llvm.org/viewvc/llvm-project?rev=43886view=rev
Log:
Removing to sync with Apple GCC 4.2
Removed:
llvm-gcc-4.2/trunk/gnattools/
___
llvm-commits mailing list
Author: evancheng
Date: Thu Nov 8 03:25:29 2007
New Revision: 43888
URL: http://llvm.org/viewvc/llvm-project?rev=43888view=rev
Log:
If both parts of smul_lohi, etc. are used, don't simplify. If only one part is
used, try simplify it.
Added:
llvm/trunk/test/CodeGen/X86/2007-11-07-MulBy4.ll
On Nov 8, 2007, at 1:18 AM, Duncan Sands wrote:
Removing to sync with Apple GCC 4.2
Removed:
llvm-gcc-4.2/trunk/gnattools/
Argh! The Ada front-end won't build without this - is it
going to come back at some point?
Readded. Sorry about this.
-bw
Author: void
Date: Thu Nov 8 03:11:48 2007
New Revision: 43887
URL: http://llvm.org/viewvc/llvm-project?rev=43887view=rev
Log:
Removing libraries to sync with Apple GCC 4.2.
Removed:
llvm-gcc-4.2/trunk/boehm-gc/
llvm-gcc-4.2/trunk/libffi/
llvm-gcc-4.2/trunk/libjava/
Removing to sync with Apple GCC 4.2
Removed:
llvm-gcc-4.2/trunk/gnattools/
Argh! The Ada front-end won't build without this - is it
going to come back at some point?
D.
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
Also, Bill, after Duncan has tested and commits his patch to 4.2, can
you please backport it to 4.0?
Duncan mentioned that he was going to attempt this. :-)
I will if no-one else does, but I'd rather not. Pretty please? :)
D.
___
llvm-commits
On Nov 8, 2007, at 1:15 AM, Duncan Sands wrote:
Also, Bill, after Duncan has tested and commits his patch to 4.2,
can
you please backport it to 4.0?
Duncan mentioned that he was going to attempt this. :-)
I will if no-one else does, but I'd rather not. Pretty please? :)
I'm willing to
Shoot! I thought it wasn't used...I'll resurrect it.
-bw
On Nov 8, 2007, at 1:18 AM, Duncan Sands wrote:
Removing to sync with Apple GCC 4.2
Removed:
llvm-gcc-4.2/trunk/gnattools/
Argh! The Ada front-end won't build without this - is it
going to come back at some point?
D.
Bill,
llvm-gcc-4.2/trunk/libobjc/
I expect, that nobody will use objc fe on non-darwin, but anyway - why
removing libobjc (and not just excluding from build)?
--
With best regards, Anton Korobeynikov.
Faculty of Mathematics Mechanics, Saint Petersburg State University.
Author: laurov
Date: Thu Nov 8 11:20:05 2007
New Revision: 43892
URL: http://llvm.org/viewvc/llvm-project?rev=43892view=rev
Log:
[ARM] Implement __builtin_thread_pointer.
Added:
llvm/trunk/include/llvm/IntrinsicsARM.td
llvm/trunk/test/CodeGen/ARM/thread_pointer.ll
Modified:
On Nov 8, 2007, at 1:42 AM, Bill Wendling wrote:
On Nov 8, 2007, at 1:15 AM, Duncan Sands wrote:
Also, Bill, after Duncan has tested and commits his patch to 4.2,
can
you please backport it to 4.0?
Duncan mentioned that he was going to attempt this. :-)
I will if no-one else does, but
On Nov 8, 2007, at 9:20 AM, Lauro Ramos Venancio wrote:
URL: http://llvm.org/viewvc/llvm-project?rev=43892view=rev
Log:
[ARM] Implement __builtin_thread_pointer.
Hey Lauro,
Is __builtin_thread_pointer ARM specific?
-Chris
___
llvm-commits
2007/11/8, Chris Lattner [EMAIL PROTECTED]:
On Nov 8, 2007, at 9:20 AM, Lauro Ramos Venancio wrote:
URL: http://llvm.org/viewvc/llvm-project?rev=43892view=rev
Log:
[ARM] Implement __builtin_thread_pointer.
Hey Lauro,
Is __builtin_thread_pointer ARM specific?
-Chris
Yes. It is
On Nov 8, 2007, at 9:39 AM, Andrew Lenharth wrote:
Author: alenhar2
Date: Thu Nov 8 11:39:28 2007
New Revision: 43893
URL: http://llvm.org/viewvc/llvm-project?rev=43893view=rev
Log:
Fix PR1780
Nice, thanks Andrew!
It would be slightly more robust to check for Ty-isSized() instead
of
On Nov 8, 2007, at 9:25 AM, Lauro Ramos Venancio wrote:
2007/11/8, Chris Lattner [EMAIL PROTECTED]:
On Nov 8, 2007, at 9:20 AM, Lauro Ramos Venancio wrote:
URL: http://llvm.org/viewvc/llvm-project?rev=43892view=rev
Log:
[ARM] Implement __builtin_thread_pointer.
Hey Lauro,
Is
Author: alenhar2
Date: Thu Nov 8 11:39:28 2007
New Revision: 43893
URL: http://llvm.org/viewvc/llvm-project?rev=43893view=rev
Log:
Fix PR1780
Added:
llvm/trunk/test/Transforms/InstCombine/2007-11-07-OpaqueAlignCrash.ll
Modified:
llvm/trunk/lib/Transforms/Scalar/InstructionCombining.cpp
Author: alenhar2
Date: Thu Nov 8 12:45:15 2007
New Revision: 43897
URL: http://llvm.org/viewvc/llvm-project?rev=43897view=rev
Log:
Better check
Modified:
llvm/trunk/lib/Transforms/Scalar/InstructionCombining.cpp
llvm/trunk/test/Transforms/InstCombine/2007-11-07-OpaqueAlignCrash.ll
Hi Anton,
llvm-gcc-4.2/trunk/libobjc/
I expect, that nobody will use objc fe on non-darwin, but anyway - why
removing libobjc (and not just excluding from build)?
Good question. Right now, I'm just trying to sync our tree with what's
in Apple GCC 4.2's TOT. My guess is that if we don't
On Nov 8, 2007, at 10:54 AM, Bill Wendling wrote:
Hi Anton,
llvm-gcc-4.2/trunk/libobjc/
I expect, that nobody will use objc fe on non-darwin, but anyway -
why
removing libobjc (and not just excluding from build)?
Good question. Right now, I'm just trying to sync our tree with what's
2007/11/8, Chris Lattner [EMAIL PROTECTED]:
On Nov 8, 2007, at 2:10 PM, Lauro Ramos Venancio wrote:
Author: laurov
Date: Thu Nov 8 16:10:12 2007
New Revision: 43909
URL: http://llvm.org/viewvc/llvm-project?rev=43909view=rev
Log:
Implement the '%@' inline asm ARM extension.
Hey
If BreakCriticalMachineEdges is going to require analysis info (e.g.
dom info) then it may be a good idea to use BCME as a pass, if
possible. Otherwise, BCME user won't know that about required analysis
and it may not let pass manager know about those requirements.
We have seen such
On Nov 8, 2007, at 3:23 PM, Tanya Lattner wrote:
Ok, I can do this but I may need a day or so to get to it.
geez.. two changes to the dejagnu testing framework in the same week.
I usually try to avoid writing tcl code as much as possible ;)
Thank you!
-
Devang
Author: johannes
Date: Thu Nov 8 17:32:31 2007
New Revision: 43914
URL: http://llvm.org/viewvc/llvm-project?rev=43914view=rev
Log:
Really disable test.
Modified:
llvm-gcc-4.2/trunk/gcc/testsuite/g++.dg/tree-ssa/pr23948.C
Modified: llvm-gcc-4.2/trunk/gcc/testsuite/g++.dg/tree-ssa/pr23948.C
On Nov 8, 2007, at 3:26 PM, Dale Johannesen wrote:
Well, OK. It looks like this. To my thinking the duplicated
registers and unnatural alphabetical
ordering makes this harder to read than the other way, and I have
to admit that having a human
(even me) type in these tables is less
On Nov 7, 2007, at 11:43 PM, Evan Cheng wrote:
On Nov 7, 2007, at 2:01 PM, Anton Korobeynikov wrote:
Dale,
Complete conditionalization of Dwarf reg numbers.
This looks pretty hackish. The numbers used before were definitely
correct for x86-32/linux. I dunno about darwin, but it looks
Author: johannes
Date: Thu Nov 8 18:47:10 2007
New Revision: 43918
URL: http://llvm.org/viewvc/llvm-project?rev=43918view=rev
Log:
Rewrite Dwarf number handling per review comments.
Modified:
llvm/trunk/lib/Target/X86/X86RegisterInfo.cpp
llvm/trunk/lib/Target/X86/X86RegisterInfo.td
Author: dpatel
Date: Thu Nov 8 19:02:00 2007
New Revision: 43920
URL: http://llvm.org/viewvc/llvm-project?rev=43920view=rev
Log:
Use appropriate type for null initializer.
Modified:
llvm-gcc-4.0/trunk/gcc/objc/objc-act.c
Modified: llvm-gcc-4.0/trunk/gcc/objc/objc-act.c
URL:
On Nov 8, 2007, at 2:48 PM, Chris Lattner wrote:
On Nov 8, 2007, at 2:40 PM, Tanya Lattner wrote:
On Nov 8, 2007, at 2:39 PM, Devang Patel wrote:
On Nov 8, 2007, at 2:35 PM, Dale Johannesen wrote:
The error message is clean,
but something like dg-require would be even better.
We do
On Nov 8, 2007, at 2:39 PM, Devang Patel wrote:
On Nov 8, 2007, at 2:35 PM, Dale Johannesen wrote:
The error message is clean,
but something like dg-require would be even better.
We do not have anything like dg-require.
What is dg-require?
If you can tell me how to get if llvm-gcc is
On Nov 8, 2007, at 2:35 PM, Dale Johannesen wrote:
The error message is clean,
but something like dg-require would be even better.
We do not have anything like dg-require.
-
Devang
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
On Nov 8, 2007, at 2:26 PM, Devang Patel wrote:
==
--- llvm/trunk/test/FrontendObjC/2007-09-25-EH.m (original)
+++ llvm/trunk/test/FrontendObjC/2007-09-25-EH.m Thu Nov 8
16:26:46 2007
@@ -1,5 +1,5 @@
// RUN:
Author: dpatel
Date: Thu Nov 8 16:26:46 2007
New Revision: 43912
URL: http://llvm.org/viewvc/llvm-project?rev=43912view=rev
Log:
Run this test only on darwin.
Modified:
llvm/trunk/test/FrontendObjC/2007-09-25-EH.m
Modified: llvm/trunk/test/FrontendObjC/2007-09-25-EH.m
URL:
Author: resistor
Date: Thu Nov 8 16:23:57 2007
New Revision: 43911
URL: http://llvm.org/viewvc/llvm-project?rev=43911view=rev
Log:
This preserves critical edge breaking.
Modified:
llvm/trunk/lib/CodeGen/TwoAddressInstructionPass.cpp
Modified:
Author: resistor
Date: Thu Nov 8 16:20:23 2007
New Revision: 43910
URL: http://llvm.org/viewvc/llvm-project?rev=43910view=rev
Log:
Make BreakCriticalMachineEdges available as a pass that can be depended on.
Modified:
llvm/trunk/include/llvm/CodeGen/Passes.h
On Nov 8, 2007, at 2:10 PM, Lauro Ramos Venancio wrote:
Author: laurov
Date: Thu Nov 8 16:10:12 2007
New Revision: 43909
URL: http://llvm.org/viewvc/llvm-project?rev=43909view=rev
Log:
Implement the '%@' inline asm ARM extension.
Hey Lauro,
in .td files, you can use ${:comment} to print
Author: laurov
Date: Thu Nov 8 16:10:12 2007
New Revision: 43909
URL: http://llvm.org/viewvc/llvm-project?rev=43909view=rev
Log:
Implement the '%@' inline asm ARM extension.
Modified:
llvm-gcc-4.2/trunk/gcc/config/arm/arm.h
llvm-gcc-4.2/trunk/gcc/llvm-convert.cpp
Modified:
Author: baldrick
Date: Thu Nov 8 15:12:44 2007
New Revision: 43907
URL: http://llvm.org/viewvc/llvm-project?rev=43907view=rev
Log:
Pass alignment and volatility along with the address in
the DestLoc parameter. Done by changing it from a Value*
to a MemRef*. Fixed/improved a bunch of
Author: evancheng
Date: Thu Nov 8 19:27:11 2007
New Revision: 43922
URL: http://llvm.org/viewvc/llvm-project?rev=43922view=rev
Log:
Bug fix. Passive nodes are not in SUnitMap.
Modified:
llvm/trunk/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
Author: evancheng
Date: Thu Nov 8 19:28:33 2007
New Revision: 43923
URL: http://llvm.org/viewvc/llvm-project?rev=43923view=rev
Log:
Didn't mean to check these in.
Modified:
llvm/trunk/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
llvm/trunk/lib/CodeGen/SelectionDAG/TargetLowering.cpp
Thanks Dale. I think this is somewhat more maintainable!
Evan
On Nov 8, 2007, at 3:26 PM, Dale Johannesen wrote:
On Nov 7, 2007, at 11:43 PM, Evan Cheng wrote:
On Nov 7, 2007, at 2:01 PM, Anton Korobeynikov wrote:
Dale,
Complete conditionalization of Dwarf reg numbers.
This looks
On Nov 8, 2007, at 2:40 PM, Tanya Lattner wrote:
On Nov 8, 2007, at 2:39 PM, Devang Patel wrote:
On Nov 8, 2007, at 2:35 PM, Dale Johannesen wrote:
The error message is clean,
but something like dg-require would be even better.
We do not have anything like dg-require.
What is
Author: baldrick
Date: Thu Nov 8 15:13:47 2007
New Revision: 43908
URL: http://llvm.org/viewvc/llvm-project?rev=43908view=rev
Log:
Make greater use of BitCastToType, as requested by
Chris. Compiles but otherwise untested.
Modified:
llvm-gcc-4.2/trunk/gcc/llvm-convert.cpp
Modified:
Author: kremenek
Date: Thu Nov 8 13:50:46 2007
New Revision: 43903
URL: http://llvm.org/viewvc/llvm-project?rev=43903view=rev
Log:
Added typedef SerializedPtrID to represent the pointer handle written to disk
instead of just using unsigned. This gives us more flexibility in changing
the
Author: dpatel
Date: Thu Nov 8 19:40:58 2007
New Revision: 43925
URL: http://llvm.org/viewvc/llvm-project?rev=43925view=rev
Log:
Use appropriate type for null initializer.
Modified:
llvm-gcc-4.2/trunk/gcc/objc/objc-act.c
Modified: llvm-gcc-4.2/trunk/gcc/objc/objc-act.c
URL:
On Nov 8, 2007, at 10:54 AM, Bill Wendling wrote:
Hi Anton,
llvm-gcc-4.2/trunk/libobjc/
I expect, that nobody will use objc fe on non-darwin, but anyway -
why
removing libobjc (and not just excluding from build)?
Good question. Right now, I'm just trying to sync our tree with
Author: evancheng
Date: Thu Nov 8 19:32:10 2007
New Revision: 43924
URL: http://llvm.org/viewvc/llvm-project?rev=43924view=rev
Log:
Much improved pic jumptable codegen:
Then:
callL1$pb
L1$pb:
popl%eax
...
LBB1_1: # entry
imull $4, %ecx, %ecx
Author: void
Date: Thu Nov 8 19:22:33 2007
New Revision: 43921
URL: http://llvm.org/viewvc/llvm-project?rev=43921view=rev
Log:
Don't configure targets we don't want to build on Darwin
Modified:
llvm-gcc-4.2/trunk/configure
llvm-gcc-4.2/trunk/configure.in
Modified:
Author: void
Date: Thu Nov 8 19:48:20 2007
New Revision: 43926
URL: http://llvm.org/viewvc/llvm-project?rev=43926view=rev
Log:
Change to reflect Duncan's fix.
Modified:
llvm-gcc-4.2/trunk/gcc/config/rs6000/llvm-rs6000.cpp
Modified: llvm-gcc-4.2/trunk/gcc/config/rs6000/llvm-rs6000.cpp
URL:
Please follow the model of libstdc++. Instead of deleting them, just
make them not build on darwin (like libstdc++).
Totally made it so. :-)
-bw
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
On Nov 8, 2007, at 1:12 PM, Duncan Sands wrote:
Author: baldrick
Date: Thu Nov 8 15:12:44 2007
New Revision: 43907
URL: http://llvm.org/viewvc/llvm-project?rev=43907view=rev
Log:
Pass alignment and volatility along with the address in
the DestLoc parameter. Done by changing it from a
Does it need to be a pass? Seems to me it should just be a function, no?
Evan
On Nov 8, 2007, at 2:20 PM, Owen Anderson wrote:
Author: resistor
Date: Thu Nov 8 16:20:23 2007
New Revision: 43910
URL: http://llvm.org/viewvc/llvm-project?rev=43910view=rev
Log:
Make
StrongPHIElimination will require it in the near future.
--Owen
On Nov 8, 2007, at 4:24 PM, Evan Cheng wrote:
Does it need to be a pass? Seems to me it should just be a function,
no?
Evan
smime.p7s
Description: S/MIME cryptographic signature
___
Author: kremenek
Date: Thu Nov 8 18:43:51 2007
New Revision: 43916
URL: http://llvm.org/viewvc/llvm-project?rev=43916view=rev
Log:
Updated Deserializer class to provide more information about the current
block that is being visited in the bitstream. The client can also now
skip blocks before
On Nov 8, 2007, at 2:35 PM, Dale Johannesen wrote:
The error message is clean,
but something like dg-require would be even better.
We do not have anything like dg-require.
What is dg-require?
It checks whether target supports particular feature or not before
running the test.
The gcc
Author: cfr
Date: Thu Nov 8 18:48:27 2007
New Revision: 43919
URL: http://llvm.org/viewvc/llvm-project?rev=43919view=rev
Log:
Adding some recently added files to the VStudio projects
Modified:
llvm/trunk/win32/Bitcode/Bitcode.vcproj
llvm/trunk/win32/CodeGen/CodeGen.vcproj
Modified:
On Nov 8, 2007, at 2:53 PM, Devang Patel wrote:
On Nov 8, 2007, at 2:48 PM, Chris Lattner wrote:
On Nov 8, 2007, at 2:40 PM, Tanya Lattner wrote:
On Nov 8, 2007, at 2:39 PM, Devang Patel wrote:
On Nov 8, 2007, at 2:35 PM, Dale Johannesen wrote:
The error message is clean,
but
I don't really feel strongly about it one way or another. I'm just
used to having passes, so I made it a pass. I do plan to modify it to
update dominator info.
--Owen
On Nov 8, 2007, at 5:56 PM, Devang Patel wrote:
If BreakCriticalMachineEdges is going to require analysis info (e.g.
dom
On Nov 8, 2007, at 7:06 PM, Owen Anderson [EMAIL PROTECTED] wrote:
I don't really feel strongly about it one way or another. I'm just
used to having passes, so I made it a pass. I do plan to modify it
to update dominator info.
Ok!
Evan
--Owen
On Nov 8, 2007, at 5:56 PM, Devang
Author: nicholas
Date: Thu Nov 8 23:45:51 2007
New Revision: 43928
URL: http://llvm.org/viewvc/llvm-project?rev=43928view=rev
Log:
Add a link to the tutorial.
Modified:
llvm/trunk/docs/index.html
Modified: llvm/trunk/docs/index.html
URL:
URL: http://llvm.org/viewvc/llvm-project?rev=43918view=rev
Log:
Rewrite Dwarf number handling per review comments.
Hi Dale,
I'm pretty uncomfortable with this, because it means that we're tying
down the register order.
How about something like:
if (Darwin32) {
switch (RegNo) {
case
On Nov 8, 2007, at 2:58 PM, Lauro Ramos Venancio wrote:
2007/11/8, Chris Lattner [EMAIL PROTECTED]:
On Nov 8, 2007, at 2:10 PM, Lauro Ramos Venancio wrote:
Author: laurov
Date: Thu Nov 8 16:10:12 2007
New Revision: 43909
URL: http://llvm.org/viewvc/llvm-project?rev=43909view=rev
Log:
Author: void
Date: Fri Nov 9 00:59:33 2007
New Revision: 43929
URL: http://llvm.org/viewvc/llvm-project?rev=43929view=rev
Log:
Initial commit of files that support building LLVM the Apple way.
Added:
llvm/trunk/utils/buildit/
llvm/trunk/utils/buildit/GNUmakefile
Author: void
Date: Fri Nov 9 01:19:47 2007
New Revision: 43930
URL: http://llvm.org/viewvc/llvm-project?rev=43930view=rev
Log:
Last (penultimate?) Apple GCC 4.2 updates.
Added:
llvm-gcc-4.2/trunk/gcc/config/arm/darwin-libgcc.10.4.ver
Author: baldrick
Date: Fri Nov 9 01:46:02 2007
New Revision: 43931
URL: http://llvm.org/viewvc/llvm-project?rev=43931view=rev
Log:
Check that accesses to the second short, t, have
an alignment of 2 rather than 4.
Added:
llvm/trunk/test/CFrontend/2007-11-07-ZeroAggregateAlign.c
Added:
Author: baldrick
Date: Fri Nov 9 01:47:29 2007
New Revision: 43932
URL: http://llvm.org/viewvc/llvm-project?rev=43932view=rev
Log:
Check that the first and third characters, s and u,
are accessed with an alignment of 2 not 1.
Added:
llvm/trunk/test/CFrontend/2007-11-07-CopyAggregateAlign.c
Hi,
This patch is for PR1582. As recent discuss on that pr, the C++ FE doesn't
track the restrict qualifier in the function declaration at all. When
llvm-gcc handling CALL_EXPR and emiting callInst instruction, it can't get
the NoAlias attribute, and hence insert a BitCast in CallInst. This
67 matches
Mail list logo