Michael J. Spencer [Mon, 14 May 2012 22:43:21 +0000 (22:43 +0000)]
[Support/COFF] Make the order of members in symbol match the standard.
llvm-svn: 156785
Anna Zaks [Mon, 14 May 2012 22:38:24 +0000 (22:38 +0000)]
[analyzer] Fix a crash in templated code which uses blocks.
We should investigate why signature info is not set in this case.
llvm-svn: 156784
Richard Smith [Mon, 14 May 2012 22:06:02 +0000 (22:06 +0000)]
Slightly generalize FileCheck patterns to unbreak -Asserts builds.
llvm-svn: 156783
Argyrios Kyrtzidis [Mon, 14 May 2012 22:01:53 +0000 (22:01 +0000)]
[objcmt] When rewriting to array/dictionary literals, add an explicit
cast to 'id' for any argument that requires it.
Part of rdar://
11438360.
llvm-svn: 156782
Richard Smith [Mon, 14 May 2012 21:57:21 +0000 (21:57 +0000)]
Implement IRGen for C++11's "T{1, 2, 3}", where T is an aggregate and the
expression is treated as an lvalue.
llvm-svn: 156781
David Blaikie [Mon, 14 May 2012 21:48:19 +0000 (21:48 +0000)]
Fix use of uninitialized variable.
Found by GCC's maybe-uninitialized.
llvm-svn: 156780
Enrico Granata [Mon, 14 May 2012 21:39:38 +0000 (21:39 +0000)]
Fixing a typo in the lldb-gdb webpage
llvm-svn: 156779
Jakob Stoklund Olesen [Mon, 14 May 2012 21:30:58 +0000 (21:30 +0000)]
Don't access MO reference after invalidating operand list.
This should unbreak llvm-x86_64-linux.
llvm-svn: 156778
Jakob Stoklund Olesen [Mon, 14 May 2012 21:10:25 +0000 (21:10 +0000)]
Fix PR12821.
RAFast must add an <imp-def> operand when it is rewriting a sub-register
def that isn't a read-modify-write.
llvm-svn: 156777
Chad Rosier [Mon, 14 May 2012 20:35:04 +0000 (20:35 +0000)]
Move the capture analysis from MemoryDependencyAnalysis to a more general place
so that it can be reused in MemCpyOptimizer. This analysis is needed to remove
an unnecessary memcpy when returning a struct into a local variable.
rdar://
11341081
PR12686
llvm-svn: 156776
Brendon Cahoon [Mon, 14 May 2012 19:35:42 +0000 (19:35 +0000)]
Revert 156634 upon request until code improvement changes are made.
llvm-svn: 156775
Dan Gohman [Mon, 14 May 2012 18:58:10 +0000 (18:58 +0000)]
Rename @llvm.debugger to @llvm.debugtrap.
llvm-svn: 156774
Johnny Chen [Mon, 14 May 2012 18:44:23 +0000 (18:44 +0000)]
Fix missing NEON registers for the 'register read' command with the lldb debugserver which supports the 'qRegisterInfo' packet
that dynamically discovers remote register context information.
o GDBRemoteRegisterContext.h:
Change the prototype of HardcodeARMRegisters() to take a boolean flag, which now becomes
void
HardcodeARMRegisters(bool from_scratch);
o GDBRemoteRegisterContext.cpp:
HardcodeARMRegisters() now checks the from_scratch flag and decides whether to add composite registers to the already
existing primordial registers based on a table called g_composites which describes the composite registers.
o ProcessGDBRemote.cpp:
Modify the logic of ProcessGDBRemote::BuildDynamicRegisterInfo() to call m_register_info.HardcodeARMRegisters()
with the newly introduced 'bool from_scrach' flag.
rdar://problem/
10652076
llvm-svn: 156773
Akira Hatanaka [Mon, 14 May 2012 18:40:07 +0000 (18:40 +0000)]
Release notes for MIPS backend.
llvm-svn: 156772
Chandler Carruth [Mon, 14 May 2012 18:31:18 +0000 (18:31 +0000)]
Teach the driver on Linux to respect -nodefaultlibs.
Patch from Andrew C. Morrow.
llvm-svn: 156771
Andrew Trick [Mon, 14 May 2012 18:03:19 +0000 (18:03 +0000)]
Remove a stale forward declaration.
llvm-svn: 156770
Jordy Rose [Mon, 14 May 2012 17:58:35 +0000 (17:58 +0000)]
[analyzer] strncpy: Special-case a length of 0 to avoid an incorrect warning.
We check the address of the last element accessed, but with 0 calculating that
address results in element -1. This patch bails out early (and avoids a bunch
of other work at that).
Fixes PR12807.
llvm-svn: 156769
Jakob Stoklund Olesen [Mon, 14 May 2012 15:46:27 +0000 (15:46 +0000)]
Remove the expensive BitVector::operator~().
Returning a temporary BitVector is very expensive. If you must, create
the temporary explicitly: Use BitVector(A).flip() instead of ~A.
llvm-svn: 156768
Jakob Stoklund Olesen [Mon, 14 May 2012 15:37:25 +0000 (15:37 +0000)]
Remove BitVector binops.
These operators were crazy slow, calling malloc to return a temporary
result. At the same time, they look very innocent when used in code.
If you need temporary BitVectors to compute your thing, create them
explicitly, and use the inplace logical operators. This makes the high
cost explicit in the code.
llvm-svn: 156767
Dmitry Vyukov [Mon, 14 May 2012 15:33:00 +0000 (15:33 +0000)]
tsan: add more atomics to public interface (fetch_or/and/xor + 1-,2-byte versions)
llvm-svn: 156766
Dmitry Vyukov [Mon, 14 May 2012 15:28:03 +0000 (15:28 +0000)]
tsan: enabled report suppression for signal-unsafe reports
llvm-svn: 156765
Dmitry Vyukov [Mon, 14 May 2012 15:25:35 +0000 (15:25 +0000)]
tsan: fix output_tests script to actually verify tests results
llvm-svn: 156764
Jakob Stoklund Olesen [Mon, 14 May 2012 15:20:39 +0000 (15:20 +0000)]
Consider ad hoc aliasing when building RegUnits.
Register units can be used to compute if two registers overlap:
A overlaps B iff units(A) intersects units(B).
With this change, the above holds true even on targets that use ad hoc
aliasing (currently only ARM). This means that register units can be
used to implement regsOverlap() more efficiently, and the register
allocator can use the concept to model interference.
When there is no ad hoc aliasing, the register units correspond to the
maximal cliques in the register overlap graph. This is optimal, no other
register unit assignment can have fewer units.
With ad hoc aliasing, weird things are possible, and we don't try too
hard to compute the maximal cliques. The current approach is always
correct, and it works very well (probably optimally) as long as the ad
hoc aliasing doesn't have cliques larger than pairs. It seems unlikely
that any target would need more.
llvm-svn: 156763
Jakob Stoklund Olesen [Mon, 14 May 2012 15:12:37 +0000 (15:12 +0000)]
Record the ad hoc aliasing graph in CodeGenRegister.
The ad hoc aliasing specified in the 'Aliases' list in .td files is
currently only used by computeOverlaps(). It will soon be needed to
build accurate register units as well, so build the undirected graph in
CodeGenRegister::buildObjectGraph() instead.
Aliasing is a symmetric relationship with only one direction specified
in the .td files. Make sure both directions are represented in
getExplicitAliases().
llvm-svn: 156762
Jakob Stoklund Olesen [Mon, 14 May 2012 15:10:07 +0000 (15:10 +0000)]
Compute topological signatures of registers.
TableGen creates new register classes and sub-register indices based on
the sub-register structure present in the register bank. So far, it has
been doing that on a per-register basis, but that is not very efficient.
This patch teaches TableGen to compute topological signatures for
registers, and use that to reduce the amount of redundant computation.
Registers get the same TopoSig if they have identical sub-register
structure.
TopoSigs are not currently exposed outside TableGen.
llvm-svn: 156761
Jakob Stoklund Olesen [Mon, 14 May 2012 15:01:19 +0000 (15:01 +0000)]
Add BitVector::anyCommon().
The existing operation (A & B).any() is very slow.
llvm-svn: 156760
Dmitry Vyukov [Mon, 14 May 2012 14:38:29 +0000 (14:38 +0000)]
tsan: make addr2line symbolizer understand dynamic libraries
Collect info about all dynamic libraries in the process (name, base, size).
Determine to what dyn lib the address relates, route request to addr2line instance for the lib.
llvm-svn: 156759
Dmitry Vyukov [Mon, 14 May 2012 14:00:07 +0000 (14:00 +0000)]
tsan: update output tests to race on heap memory
Races on stack of main thread are problematic for COMPAT mapping, because it's not 1-to-1 and race addr is not properly mapped from shadow back to application memory.
Update output tests to race heap memory.
llvm-svn: 156758
Stepan Dyatkovskiy [Mon, 14 May 2012 08:26:31 +0000 (08:26 +0000)]
SwitchInst cosmetics: renamed "Hash" method to "hash"
llvm-svn: 156757
Bill Wendling [Mon, 14 May 2012 08:11:53 +0000 (08:11 +0000)]
Formatting changes. Remove the '...' placeholders.
llvm-svn: 156756
Bill Wendling [Mon, 14 May 2012 07:53:40 +0000 (07:53 +0000)]
Use ArrayRef instead of an explicit vector type.
llvm-svn: 156755
Bill Wendling [Mon, 14 May 2012 06:23:51 +0000 (06:23 +0000)]
Add blurb about Julia.
llvm-svn: 156754
Gregory Szorc [Mon, 14 May 2012 03:56:33 +0000 (03:56 +0000)]
[clang.py] Implement Cursor.canonical
llvm-svn: 156753
Gregory Szorc [Mon, 14 May 2012 03:53:29 +0000 (03:53 +0000)]
[clang.py] Implement Cursor.result_type
llvm-svn: 156752
Jordy Rose [Sun, 13 May 2012 20:36:47 +0000 (20:36 +0000)]
Test for -fdebugger-objc-literal: missing methods should be generated like missing classes.
llvm-svn: 156746
Justin Holewinski [Sun, 13 May 2012 17:32:35 +0000 (17:32 +0000)]
ReleaseNotes: Add info on PTX back-end
llvm-svn: 156745
Benjamin Kramer [Sun, 13 May 2012 15:13:22 +0000 (15:13 +0000)]
Hexagon: Initialize TBB to 0.
Found by valgrind.
llvm-svn: 156744
Jean-Daniel Dupas [Sun, 13 May 2012 14:36:15 +0000 (14:36 +0000)]
Fix Xcode case (Upper X, lower c)
llvm-svn: 156743
Jean-Daniel Dupas [Sun, 13 May 2012 14:32:11 +0000 (14:32 +0000)]
Fix Xcode case (Upper X, lower c)
llvm-svn: 156742
Dmitry Vyukov [Sun, 13 May 2012 14:24:46 +0000 (14:24 +0000)]
tsan: fix mmap fd
This is mostly to test my commit access.
Fixes fd passed to mmap(), -1 is the proper invalid fd.
llvm-svn: 156741
Benjamin Kramer [Sun, 13 May 2012 13:10:35 +0000 (13:10 +0000)]
ReleaseNotes: Add a note about zero_undef on llvm.cttz/ctlz. Extend x86 section. Add a bullet for dwarf access tables.
llvm-svn: 156740
Benjamin Kramer [Sun, 13 May 2012 12:01:16 +0000 (12:01 +0000)]
ReleaseNotes: Add a blurb about llvm-mc -g and move inliner changes into the optimizer sections. Verbosify some bullets.
llvm-svn: 156739
Benjamin Kramer [Sun, 13 May 2012 11:46:05 +0000 (11:46 +0000)]
ReleaseNotes: Document that LLVM was rewritten in python.
^~~~
llvm-build
llvm-svn: 156738
Benjamin Kramer [Sun, 13 May 2012 11:28:46 +0000 (11:28 +0000)]
ReleaseNotes: Add bullets for removed targets. Extend the note about llvm-ld removal.
llvm-svn: 156737
Benjamin Kramer [Sun, 13 May 2012 10:40:08 +0000 (10:40 +0000)]
Outline some clang 3.1 highlights off the top of my head.
llvm-svn: 156736
Benjamin Kramer [Sun, 13 May 2012 10:21:51 +0000 (10:21 +0000)]
Add a link for every project.
llvm-svn: 156735
Bill Wendling [Sun, 13 May 2012 10:00:58 +0000 (10:00 +0000)]
Add blurb for LLVM D Compiler.
llvm-svn: 156733
Bill Wendling [Sun, 13 May 2012 09:59:27 +0000 (09:59 +0000)]
Add blurbs for pocl and TCE.
llvm-svn: 156732
Bill Wendling [Sun, 13 May 2012 09:55:24 +0000 (09:55 +0000)]
Add OSL blurb.
llvm-svn: 156731
Bill Wendling [Sun, 13 May 2012 09:52:48 +0000 (09:52 +0000)]
Add FAUST blurb.
llvm-svn: 156730
Nadav Rotem [Sun, 13 May 2012 05:52:56 +0000 (05:52 +0000)]
Fix the tool documentation
llvm-svn: 156729
Rafael Espindola [Sun, 13 May 2012 03:25:18 +0000 (03:25 +0000)]
Refactor all the
if (Inherited)
Attr->setInherited(true);
To a central location.
llvm-svn: 156728
Rafael Espindola [Sun, 13 May 2012 02:42:42 +0000 (02:42 +0000)]
Produce a warning for mismatched section attributes. Completest pr9356.
llvm-svn: 156727
Rafael Espindola [Sun, 13 May 2012 01:56:10 +0000 (01:56 +0000)]
We were already passing the case in pr9356. Add it to attributes.c to make
sure we don't regress. I will add a warning in a sec.
llvm-svn: 156726
Gregory Szorc [Sat, 12 May 2012 21:12:22 +0000 (21:12 +0000)]
Document Python bindings in release notes
llvm-svn: 156724
Benjamin Kramer [Sat, 12 May 2012 21:10:52 +0000 (21:10 +0000)]
Add support for __attribute__((hot)) and __attribute__((cold)).
Currently cold functions are marked with the "optsize" attribute in CodeGen
so they are always optimized for size. The hot attribute is just ignored,
LLVM doesn't have a way to express hotness at the moment.
llvm-svn: 156723
Gregory Szorc [Sat, 12 May 2012 20:49:13 +0000 (20:49 +0000)]
[clang.py] Followup to TranslationUnit refactor
* Document index argument in TranslationUnit.from_source
* Add numeric error code to TranslationUnitSaveError string representation
* Use None instead of [] for default argument value in
TranslationUnit.codeComplete
llvm-svn: 156722
Gregory Szorc [Sat, 12 May 2012 20:45:56 +0000 (20:45 +0000)]
Add Python binding info to release notes
llvm-svn: 156721
Jordy Rose [Sat, 12 May 2012 17:32:59 +0000 (17:32 +0000)]
[analyzer] Test case: p->x is the same as p[0].x. (PR7297)
llvm-svn: 156720
Jordy Rose [Sat, 12 May 2012 17:32:56 +0000 (17:32 +0000)]
Only check NSArray/NSDictionary boxing method params once.
Once we've found a "good" method, we don't need to check its argument types
again. (Even if we might have later found a "bad" method, we were already
caching the method we first looked up.)
llvm-svn: 156719
Jordy Rose [Sat, 12 May 2012 17:32:52 +0000 (17:32 +0000)]
80-col violations and minor reformatting. No functionality change.
llvm-svn: 156718
Jordy Rose [Sat, 12 May 2012 17:32:44 +0000 (17:32 +0000)]
Clean up ObjC boxing method checks by reducing duplicated code.
llvm-svn: 156717
Benjamin Kramer [Sat, 12 May 2012 16:52:21 +0000 (16:52 +0000)]
Fix spacing after if.
llvm-svn: 156716
Rafael Espindola [Sat, 12 May 2012 16:35:00 +0000 (16:35 +0000)]
Remove unused variable. Thanks to Dmitri Gribenko for noticing it.
llvm-svn: 156715
Rafael Espindola [Sat, 12 May 2012 16:31:10 +0000 (16:31 +0000)]
Add support for the .rept directive. Patch by Vladmir Sorokin. I added support
for nesting.
llvm-svn: 156714
Jordy Rose [Sat, 12 May 2012 15:53:41 +0000 (15:53 +0000)]
Don't crash on boxed strings when +stringWithUTF8String: is missing.
Also, unify some diagnostics for boxed expressions that have the same form.
Fixes PR12804.
llvm-svn: 156713
Benjamin Kramer [Sat, 12 May 2012 14:30:47 +0000 (14:30 +0000)]
ELF: Add support for the asm .version directive.
llvm-svn: 156712
Alexander Potapenko [Sat, 12 May 2012 12:33:41 +0000 (12:33 +0000)]
Increase error_message_buffer_size to 64K (16K is insufficient for large programs)
llvm-svn: 156711
Evgeniy Stepanov [Sat, 12 May 2012 12:33:10 +0000 (12:33 +0000)]
Fix GetFrameNameByAddr hitting stale stack guards.
In the current implementation AsanThread::GetFrameNameByAddr scans the
stack for a magic guard value to locate base address of the stack
frame. This is not reliable, especially on ARM, where the code that
stores this magic value has to construct it in a register from two
small intermediates; this register can then end up stored in a random
stack location in the prologue of another function.
With this change, GetFrameNameByAddr scans the shadow memory for the
signature of a left stack redzone instead. It is now possible to
remove the magic from the instrumentation pass for additional
performance gain. We keep it there for now just to make sure the new
algorithm does not fail in some corner case.
llvm-svn: 156710
Benjamin Kramer [Sat, 12 May 2012 11:21:46 +0000 (11:21 +0000)]
AsmParser: Add support for the .purgem directive.
Based on a patch by Team PaX.
llvm-svn: 156709
Benjamin Kramer [Sat, 12 May 2012 11:19:04 +0000 (11:19 +0000)]
AsmParser: Give a nice error message for .code16gcc, which is currently unsupported.
Patch by Team PaX!
llvm-svn: 156708
Benjamin Kramer [Sat, 12 May 2012 11:18:59 +0000 (11:18 +0000)]
AsmParser: ignore the .extern directive.
llvm-svn: 156707
Benjamin Kramer [Sat, 12 May 2012 11:18:51 +0000 (11:18 +0000)]
AsmParser: Add support for .ifc and .ifnc directives.
Based on a patch from PaX Team.
llvm-svn: 156706
Benjamin Kramer [Sat, 12 May 2012 11:18:42 +0000 (11:18 +0000)]
AsmParser: Add support for .ifb and .ifnb directives.
Based on a patch from PaX Team.
llvm-svn: 156705
Stepan Dyatkovskiy [Sat, 12 May 2012 10:48:17 +0000 (10:48 +0000)]
Recommited r156374 with critical fixes in BitcodeReader/Writer:
Ordinary patch for PR1255.
Added new case-ranges orientated methods for adding/removing cases in SwitchInst. After this patch cases will internally representated as ConstantArray-s instead of ConstantInt, externally cases wrapped within the ConstantRangesSet object.
Old methods of SwitchInst are also works well, but marked as deprecated. So on this stage we have no side effects except that I added support for case ranges in BitcodeReader/Writer, of course test for Bitcode is also added. Old "switch" format is also supported.
llvm-svn: 156704
Jay Foad [Sat, 12 May 2012 08:30:16 +0000 (08:30 +0000)]
Teach Function::hasAddressTaken that BlockAddress doesn't really take
the address of a function.
llvm-svn: 156703
Sirish Pande [Sat, 12 May 2012 05:54:15 +0000 (05:54 +0000)]
Make sure new value jump is enabled for Hexagon V5 as well.
llvm-svn: 156700
Jordy Rose [Sat, 12 May 2012 05:10:43 +0000 (05:10 +0000)]
[analyzer] RetainCountChecker: track ObjC boxed expression objects.
llvm-svn: 156699
Sirish Pande [Sat, 12 May 2012 05:10:30 +0000 (05:10 +0000)]
Support for Hexagon feature, New Value Jump.
llvm-svn: 156698
Akira Hatanaka [Sat, 12 May 2012 03:25:16 +0000 (03:25 +0000)]
Fix test cases.
llvm-svn: 156697
Akira Hatanaka [Sat, 12 May 2012 03:24:03 +0000 (03:24 +0000)]
Remove MipsEmitGPRestore.cpp.
llvm-svn: 156696
Akira Hatanaka [Sat, 12 May 2012 03:22:13 +0000 (03:22 +0000)]
Delete all functions that are no longer needed in MipsFunctionInfo, including
the ones that get or set the frame index for the $gp save slot.
Remove the piece of code in MipsFunctionInfo::getGlobalBaseReg() which returns
GP. This function should always return a virtual register.
llvm-svn: 156695
Akira Hatanaka [Sat, 12 May 2012 03:21:18 +0000 (03:21 +0000)]
Stop reserving register $gp. Do not call isGPFI to check whether a frame object
is the $gp save slot.
llvm-svn: 156694
Akira Hatanaka [Sat, 12 May 2012 03:19:51 +0000 (03:19 +0000)]
Do not add the pass which restores $gp after every function call.
llvm-svn: 156693
Akira Hatanaka [Sat, 12 May 2012 03:19:04 +0000 (03:19 +0000)]
Make the following changes in MipsISelLowering.cpp:
- Stop creating stack frame objects needed for saving $gp.
- Insert a node that copies the global pointer register to register $gp
before the call node. This will ensure $gp is valid at the entry of the
called function.
llvm-svn: 156692
Akira Hatanaka [Sat, 12 May 2012 03:18:00 +0000 (03:18 +0000)]
Make the following changes in MipsFrameLowering.cpp:
- Stop emitting instructions needed to initialize the global pointer register.
- Stop emitting .cprestore directive.
- Do not take into account the $gp save slot when computing stack size.
llvm-svn: 156691
Jakob Stoklund Olesen [Sat, 12 May 2012 02:02:26 +0000 (02:02 +0000)]
Speed up computeComposites() by using the new SubReg -> SubIdx map.
TableGen doesn't need to search through the SubRegs map to find an
inverse entry.
llvm-svn: 156690
Akira Hatanaka [Sat, 12 May 2012 00:48:43 +0000 (00:48 +0000)]
Make the following changes in MipsAsmPrinter.cpp:
- Remove code which lowers pseudo SETGP01.
- Fix LowerSETGP01. The first two of the three instructions that are emitted to
initialize the global pointer register now use register $2.
- Stop emitting .cpload directive.
llvm-svn: 156689
Chad Rosier [Sat, 12 May 2012 00:43:40 +0000 (00:43 +0000)]
Hoist simpler checks above llvm::PointerMayBeCaptured. No functional change intended.
llvm-svn: 156687
Jim Ingham [Sat, 12 May 2012 00:38:30 +0000 (00:38 +0000)]
Add a useful error message to "log enable" with the wrong number of arguments, and reject "log enable lldb" which just silently did nothing before.
llvm-svn: 156686
Jakob Stoklund Olesen [Sat, 12 May 2012 00:33:28 +0000 (00:33 +0000)]
Don't look for empty live ranges in the unions.
Empty live ranges represent undef and still get allocated, but they
won't appear in LiveIntervalUnions.
Patch by Patrik Hägglund!
llvm-svn: 156685
Greg Clayton [Sat, 12 May 2012 00:26:42 +0000 (00:26 +0000)]
<rdar://problem/
11439169>
"lldb -a i386" doesn't set the calculator mode correctly if run on a 64 bit system.
The previous logic always used the current host architecture, not the default architecture. The default arch gets set into a static varaible in lldb_private::Target when an arch is set from the command line:
lldb -a i386
We now use the default arch correctly.
llvm-svn: 156680
Akira Hatanaka [Sat, 12 May 2012 00:17:17 +0000 (00:17 +0000)]
Insert instructions to the entry basic block which initializes the global
pointer register.
This is the first of the series of patches which clean up the way global pointer
register is used. The patches will make the following improvements:
- Make $gp an allocatable temporary register rather than reserving it.
- Use a virtual register as the global pointer register and let the register
allocator decide which register to assign to it or whether spill/reloads are
needed.
- Make sure $gp is valid at the entry of a called function, which is necessary
for functions using lazy binding.
- Remove the need for emitting .cprestore and .cpload directives.
llvm-svn: 156671
Greg Clayton [Sat, 12 May 2012 00:01:21 +0000 (00:01 +0000)]
<rdar://problem/
11439022>
Restore expressions with no target.
llvm-svn: 156669
Jim Ingham [Fri, 11 May 2012 23:49:49 +0000 (23:49 +0000)]
Fix a comment I had incorrectly altered.
llvm-svn: 156668
Jim Ingham [Fri, 11 May 2012 23:47:32 +0000 (23:47 +0000)]
Found one more place where the OkayToDiscard needs to be consulted.
Also changed the defaults for SBThread::Step* to not delete extant plans.
Also added some test cases to test more complex stepping scenarios.
llvm-svn: 156667
Douglas Gregor [Fri, 11 May 2012 23:37:49 +0000 (23:37 +0000)]
Teach the parser to deal with multiple spellings for the same
attribute, rather than requiring multiple cases in consumers of this
information.
llvm-svn: 156666
Michael J. Spencer [Fri, 11 May 2012 23:34:39 +0000 (23:34 +0000)]
Add doxygen comments.
llvm-svn: 156665
Anna Zaks [Fri, 11 May 2012 23:34:14 +0000 (23:34 +0000)]
[analyzer] Test variable modified types.
llvm-svn: 156664
Akira Hatanaka [Fri, 11 May 2012 23:22:18 +0000 (23:22 +0000)]
Do not replace operands of pseudo instructions with register $zero.
llvm-svn: 156663