oota-llvm.git
10 years ago[mips] Splitting up class definition from implementation.
Matheus Almeida [Thu, 20 Mar 2014 09:29:54 +0000 (09:29 +0000)]
[mips] Splitting up class definition from implementation.

Also removed some unnecessary #includes.

No functional changes.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204320 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoAdd llvm_unreachable after fully-covered switches to appease GCC
Alexey Samsonov [Thu, 20 Mar 2014 07:30:40 +0000 (07:30 +0000)]
Add llvm_unreachable after fully-covered switches to appease GCC

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204318 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoTest case for r204305.
Craig Topper [Thu, 20 Mar 2014 06:45:10 +0000 (06:45 +0000)]
Test case for r204305.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204316 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoObject: Output .file symbols properly
David Majnemer [Thu, 20 Mar 2014 06:29:02 +0000 (06:29 +0000)]
Object: Output .file symbols properly

obj2yaml would emit the NUL bytes padding the auxiliary file symbol
records.  Trimming them looks nicer.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204314 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoObject: Abstract out the determination of function line symbols
David Majnemer [Thu, 20 Mar 2014 06:28:55 +0000 (06:28 +0000)]
Object: Abstract out the determination of function line symbols

No functionality change.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204313 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoObject: Don't double-escape empty hexdata
David Majnemer [Thu, 20 Mar 2014 06:28:52 +0000 (06:28 +0000)]
Object: Don't double-escape empty hexdata

We would emit a pair of double quotes inside a pair of single quotes.
Just use a pair of single quotes.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204312 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoReapply 'ARM IAS: support .thumb_set'
Saleem Abdulrasool [Thu, 20 Mar 2014 06:05:33 +0000 (06:05 +0000)]
Reapply 'ARM IAS: support .thumb_set'

Re-apply the change after it was reverted to do conflicts due to another change
being reverted.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204306 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[X86] Check return value of readSIB in disassembler so errors propagate. In particula...
Craig Topper [Thu, 20 Mar 2014 05:56:00 +0000 (05:56 +0000)]
[X86] Check return value of readSIB in disassembler so errors propagate. In particular this makes a too short instruction with a missing SIB byte fail.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204305 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[ARM]Fix an assertion failure in A15SDOptimizer about DPair reg class by treating...
Hao Liu [Thu, 20 Mar 2014 05:36:59 +0000 (05:36 +0000)]
[ARM]Fix an assertion failure in A15SDOptimizer about DPair reg class by treating DPair as QPR.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204304 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoLook through variables when computing relocations.
Rafael Espindola [Thu, 20 Mar 2014 02:12:01 +0000 (02:12 +0000)]
Look through variables when computing relocations.

Given

bar = foo + 4
.long bar

MC would eat the 4. GNU as includes it in the relocation. The rule seems to be
that a variable that defines a symbol is used in the relocation and one that
does not define a symbol is evaluated and the result included in the relocation.

Fixing this unfortunately required some other changes:

* Since the variable is now evaluated, it would prevent the ELF writer from
  noticing the weakref marker the elf streamer uses. This patch then replaces
  that with a VariantKind in MCSymbolRefExpr.

* Using VariantKind then requires us to look past other VariantKind to see

.weakref bar,foo
call bar@PLT

  doing this also fixes

zed = foo +2
call zed@PLT

  so that is a good thing.

* Looking past VariantKind means that the relocation selection has to use
  the fixup instead of the target.

This is a reboot of the previous fixes for MC. I will watch the sanitizer
buildbot and wait for a build before adding back the previous fixes.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204294 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoRevert "Use the range machinery for DW_AT_ranges and DW_AT_high/lo_pc."
Eric Christopher [Thu, 20 Mar 2014 00:12:06 +0000 (00:12 +0000)]
Revert "Use the range machinery for DW_AT_ranges and DW_AT_high/lo_pc."

This appears to trigger failures with optimization and function arguments somehow.

This reverts commit r204277.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204286 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoSupport: Add postincrement and include guards to LineIterator
Justin Bogner [Wed, 19 Mar 2014 22:58:31 +0000 (22:58 +0000)]
Support: Add postincrement and include guards to LineIterator

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204279 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoUse the range machinery for DW_AT_ranges and DW_AT_high/lo_pc.
Eric Christopher [Wed, 19 Mar 2014 22:42:36 +0000 (22:42 +0000)]
Use the range machinery for DW_AT_ranges and DW_AT_high/lo_pc.

This commit moves us from a single range per subprogram to extending
ranges if we are:

a) In the same section, and
b) In the same enclosing CU.

This means we have more fine grained ranges for compile units, and fewer
ranges overall when we have multiple functions in the same CU
adjacent to each other in the object file.

Also remove all of the earlier hacks around this functionality for
function sections etc. Also update all of the testcases to take into
account the merging functionality.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204277 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoR600/SI: Add unused LDS 2 form instructions.
Matt Arsenault [Wed, 19 Mar 2014 22:19:56 +0000 (22:19 +0000)]
R600/SI: Add unused LDS 2 form instructions.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204275 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoR600/SI: Add support for 64-bit LDS writes
Matt Arsenault [Wed, 19 Mar 2014 22:19:54 +0000 (22:19 +0000)]
R600/SI: Add support for 64-bit LDS writes

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204274 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoR600/SI: Add support for 64-bit LDS loads.
Matt Arsenault [Wed, 19 Mar 2014 22:19:52 +0000 (22:19 +0000)]
R600/SI: Add support for 64-bit LDS loads.

v2:
  -Use correct opcode for DS_READ_64

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204273 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoR600/SI: Match i16 immediate offset of LDS instructions.
Matt Arsenault [Wed, 19 Mar 2014 22:19:49 +0000 (22:19 +0000)]
R600/SI: Match i16 immediate offset of LDS instructions.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204272 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoR600/SI: Fix test checking wrong instruction operand.
Matt Arsenault [Wed, 19 Mar 2014 22:19:45 +0000 (22:19 +0000)]
R600/SI: Fix test checking wrong instruction operand.

The source and destination happen to be the same register.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204271 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoR600/SI: Don't display the GDS bit.
Matt Arsenault [Wed, 19 Mar 2014 22:19:43 +0000 (22:19 +0000)]
R600/SI: Don't display the GDS bit.

It isn't actually used now, and probably never will be, plus it makes
tests less annoying. I also think SC prints GDS instructions as a
separate instruction name.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204270 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoR600/SI: Merge offset0 and offset1 fields for single address DS instructions v2
Matt Arsenault [Wed, 19 Mar 2014 22:19:39 +0000 (22:19 +0000)]
R600/SI: Merge offset0 and offset1 fields for single address DS instructions v2

Also remove unused data fields from the DS_Load_Helper class.

v2:
  - Merge fields for DS_WRITE

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204269 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoCOFF: Fix a typo
David Majnemer [Wed, 19 Mar 2014 20:16:23 +0000 (20:16 +0000)]
COFF: Fix a typo

Introduced when I rebased my changes to use isReservedSectionNumber.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204260 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoFix comment (PR19188)
Hans Wennborg [Wed, 19 Mar 2014 18:41:38 +0000 (18:41 +0000)]
Fix comment (PR19188)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204256 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[mips] 80-column.
Matheus Almeida [Wed, 19 Mar 2014 16:29:06 +0000 (16:29 +0000)]
[mips] 80-column.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204252 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[mips] Making sure that a '.set noreorder' directive is correctly parsed
Matheus Almeida [Wed, 19 Mar 2014 16:20:19 +0000 (16:20 +0000)]
[mips] Making sure that a '.set noreorder' directive is correctly parsed
and emitted and that no NOPs are emitted in a 'noreorder section'.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204250 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoSet debug info for instructions inserted in SplitBlockAndInsertIfThen.
Evgeniy Stepanov [Wed, 19 Mar 2014 12:56:38 +0000 (12:56 +0000)]
Set debug info for instructions inserted in SplitBlockAndInsertIfThen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204230 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoPrune includes in X86 target.
Craig Topper [Wed, 19 Mar 2014 06:53:25 +0000 (06:53 +0000)]
Prune includes in X86 target.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204216 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoObject: Provide a richer means of describing auxiliary symbols
David Majnemer [Wed, 19 Mar 2014 04:47:47 +0000 (04:47 +0000)]
Object: Provide a richer means of describing auxiliary symbols

The current state of affairs has auxiliary symbols described as a big
bag of bytes. This is less than satisfying, it detracts from the YAML
file as being human readable.

Instead, allow for symbols to optionally contain their auxiliary data.
This allows us to have a much higher level way of describing things like
weak symbols, function definitions and section definitions.

This depends on D3105.

Differential Revision: http://llvm-reviews.chandlerc.com/D3092

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204214 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoObject: Move auxiliary symbol definitions from llvm-readobj
David Majnemer [Wed, 19 Mar 2014 04:33:27 +0000 (04:33 +0000)]
Object: Move auxiliary symbol definitions from llvm-readobj

Summary: These definitions are useful to other aspects of LLVM, move them out.

Reviewers: rafael, nrieck, ruiu

CC: llvm-commits
Differential Revision: http://llvm-reviews.chandlerc.com/D3105

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204213 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoObject: Clean up COFF.h
David Majnemer [Wed, 19 Mar 2014 02:37:19 +0000 (02:37 +0000)]
Object: Clean up COFF.h

The file violated the coding standard.  Make it conform.

No functionality change.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204212 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agollvm-profdata: Make "merge" into a subcommand.
Justin Bogner [Wed, 19 Mar 2014 02:20:46 +0000 (02:20 +0000)]
llvm-profdata: Make "merge" into a subcommand.

We'll be adding a few more subcommands in the near future.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204211 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agollvm-profdata: Update to use the naive text format with function hash
Justin Bogner [Wed, 19 Mar 2014 02:20:42 +0000 (02:20 +0000)]
llvm-profdata: Update to use the naive text format with function hash

This also uses line_iterator to simplify the parsing logic.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204210 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoMake the test harder by using a non-zero offset.
Rafael Espindola [Wed, 19 Mar 2014 00:26:58 +0000 (00:26 +0000)]
Make the test harder by using a non-zero offset.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204205 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoWhen destroying a StringMap, just iterate over the map and destroy the contained...
Pete Cooper [Wed, 19 Mar 2014 00:23:30 +0000 (00:23 +0000)]
When destroying a StringMap, just iterate over the map and destroy the contained elements.  Don't reset them back to 0 as their values aren't needed any more.  This results in ~StringMap() being mostly empty for POD types in BumpPtrAllocators

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204204 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoRevert "Add back r203962, r204028 and r204059."
Rafael Espindola [Wed, 19 Mar 2014 00:13:43 +0000 (00:13 +0000)]
Revert "Add back r203962, r204028 and r204059."

This reverts commit r204178.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204203 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoDebugInfo: Use the comp_dir of the referencing type units when building debug_line.dwo
David Blaikie [Wed, 19 Mar 2014 00:11:28 +0000 (00:11 +0000)]
DebugInfo: Use the comp_dir of the referencing type units when building debug_line.dwo

This isn't a complete fix - it falls back to non-comp_dir when multiple
compile units are in play. Adding a map of comp_dir to table is part of
the more general solution, but I gave up (in the short term) when I
realized I'd also have to calculate the size of each type unit so as to
produce correct DW_AT_stmt_list attributes.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204202 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoExpose "noduplicate" attribute as a property for intrinsics.
Eli Bendersky [Tue, 18 Mar 2014 23:51:07 +0000 (23:51 +0000)]
Expose "noduplicate" attribute as a property for intrinsics.

The "noduplicate" function attribute exists to prevent certain optimizations
from duplicating calls to the function. This is important on platforms where
certain function call duplications are unsafe (for example execution barriers
for CUDA and OpenCL).

This patch makes it possible to specify intrinsics as "noduplicate" and
translates that to the appropriate function attribute.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204200 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoObject/COFF: Add function to check if section number is reserved one.
Rui Ueyama [Tue, 18 Mar 2014 23:37:53 +0000 (23:37 +0000)]
Object/COFF: Add function to check if section number is reserved one.

Differential Revision: http://llvm-reviews.chandlerc.com/D3103

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204199 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoMove yet another test that requires ARM to an ARM test directory.
NAKAMURA Takumi [Tue, 18 Mar 2014 23:12:09 +0000 (23:12 +0000)]
Move yet another test that requires ARM to an ARM test directory.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204198 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoMove tests that require ARM to an ARM test directory.
Jim Grosbach [Tue, 18 Mar 2014 22:43:59 +0000 (22:43 +0000)]
Move tests that require ARM to an ARM test directory.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204197 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoFix use_iterator crash in ObjCArc from r203364
Duncan P. N. Exon Smith [Tue, 18 Mar 2014 22:32:43 +0000 (22:32 +0000)]
Fix use_iterator crash in ObjCArc from r203364

The use_iterator redesign in r203364 introduced an increment past the
end of a range in -objc-arc-contract.  Added an explicit check for the
end of the range.

<rdar://problem/16333235>

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204195 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoC++ style comments
Jim Grosbach [Tue, 18 Mar 2014 22:13:13 +0000 (22:13 +0000)]
C++ style comments

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204194 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoTidy up. Update per naming conventions.
Jim Grosbach [Tue, 18 Mar 2014 22:09:10 +0000 (22:09 +0000)]
Tidy up. Update per naming conventions.

No functional change.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204192 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoMachO: Emit a version-min load command when possible.
Jim Grosbach [Tue, 18 Mar 2014 22:09:08 +0000 (22:09 +0000)]
MachO: Emit a version-min load command when possible.

When deployment target version information is available, emit it to the
target streamer for inclusion in the object file.

rdar://11337778

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204191 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoDarwin: Add assembler directives to create version-min load commands.
Jim Grosbach [Tue, 18 Mar 2014 22:09:05 +0000 (22:09 +0000)]
Darwin: Add assembler directives to create version-min load commands.

Allow object files to be tagged with a version-min load command for iOS
or MacOSX.

Teach macho-dump to understand the version-min load commands for
testcases.

rdar://11337778

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204190 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoMachO: Comment fields in the version_min load command.
Jim Grosbach [Tue, 18 Mar 2014 22:08:58 +0000 (22:08 +0000)]
MachO: Comment fields in the version_min load command.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204189 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[LV] While I'm here, use range based for loops which are so much cleaner
Chandler Carruth [Tue, 18 Mar 2014 22:00:32 +0000 (22:00 +0000)]
[LV] While I'm here, use range based for loops which are so much cleaner
for this kind of walk.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204188 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[LV] The actual change I intended to commit in r204148. Sorry for the
Chandler Carruth [Tue, 18 Mar 2014 21:58:38 +0000 (21:58 +0000)]
[LV] The actual change I intended to commit in r204148. Sorry for the
noise.

Original commit log:
Replace some dead code with an assert. When I first ported this pass
from a loop pass to a function pass I did so in the naive, recursive
way. It doesn't actually work, we need a worklist instead. When
I switched to the worklist I didn't delete the naive recursion. That
recursion was also buggy because it was dead and never really exercised.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204187 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[LV] Replace some dead code with an assert. When I first ported this
Chandler Carruth [Tue, 18 Mar 2014 21:51:46 +0000 (21:51 +0000)]
[LV] Replace some dead code with an assert. When I first ported this
pass from a loop pass to a function pass I did so in the naive,
recursive way. It doesn't actually work, we need a worklist instead.
When I switched to the worklist I didn't delete the naive recursion.
That recursion was also buggy because it was dead and never really
exercised.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204184 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoMore header and forward declaration cleanup.
Eric Christopher [Tue, 18 Mar 2014 21:44:12 +0000 (21:44 +0000)]
More header and forward declaration cleanup.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204183 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoAdd back the headers we're getting via (likely) transitive includes.
Eric Christopher [Tue, 18 Mar 2014 20:58:35 +0000 (20:58 +0000)]
Add back the headers we're getting via (likely) transitive includes.
We really do use these things in the header.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204180 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoAdd back r203962, r204028 and r204059.
Rafael Espindola [Tue, 18 Mar 2014 20:40:38 +0000 (20:40 +0000)]
Add back r203962, r204028 and r204059.

This reverts commit r204137.

This includes a fix for handling aliases of aliases.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204178 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoFix for coding style.
Eric Christopher [Tue, 18 Mar 2014 20:39:54 +0000 (20:39 +0000)]
Fix for coding style.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204177 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoRemove a bunch of unnecessary includes and forward declarations.
Eric Christopher [Tue, 18 Mar 2014 20:37:10 +0000 (20:37 +0000)]
Remove a bunch of unnecessary includes and forward declarations.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204176 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoX86 memcpy lowering: use "rep movs" even when esi is used as base pointer
Hans Wennborg [Tue, 18 Mar 2014 20:04:34 +0000 (20:04 +0000)]
X86 memcpy lowering: use "rep movs" even when esi is used as base pointer

For functions where esi is used as base pointer, we would previously fall back
from lowering memcpy with "rep movs" because that clobbers esi.

With this patch, we just store esi in another physical register, and restore
it afterwards. This adds a little bit of register preassure, but the more
efficient memcpy should be worth it.

Differential Revision: http://llvm-reviews.chandlerc.com/D2968

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204174 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoUse early returns to reduce nesting.
Rui Ueyama [Tue, 18 Mar 2014 18:58:51 +0000 (18:58 +0000)]
Use early returns to reduce nesting.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204171 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoFix test lsr-normalization.ll broken in r204161.
Michael Zolotukhin [Tue, 18 Mar 2014 18:17:59 +0000 (18:17 +0000)]
Fix test lsr-normalization.ll broken in r204161.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204166 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoAdd support for scalarizing/splitting vector bswap.
Raul E. Silvera [Tue, 18 Mar 2014 17:49:12 +0000 (17:49 +0000)]
Add support for scalarizing/splitting vector bswap.

Summary:
  SLP Vectorization of intrinsics (r203707) has exposed cases where the
  expansion of vector bswap is failing (PR19151).

Reviewers: hfinkel

CC: chandlerc
Differential Revision: http://llvm-reviews.chandlerc.com/D3104

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204163 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoDebug info: Remove OdrMemberMap from DwarfDebug, it's not necessary.
Adrian Prantl [Tue, 18 Mar 2014 17:41:15 +0000 (17:41 +0000)]
Debug info: Remove OdrMemberMap from DwarfDebug, it's not necessary.
Follow-up to r203982.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204162 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoAdd stride normalization to SCEV Normalize/Denormalize transformation.
Michael Zolotukhin [Tue, 18 Mar 2014 17:34:03 +0000 (17:34 +0000)]
Add stride normalization to SCEV Normalize/Denormalize transformation.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204161 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[DAGCombiner] teach how to simplify xor/and/or nodes according to the following rules:
Andrea Di Biagio [Tue, 18 Mar 2014 17:12:59 +0000 (17:12 +0000)]
[DAGCombiner] teach how to simplify xor/and/or nodes according to the following rules:

 1)  (AND (shuf (A, C, Mask), shuf (B, C, Mask)) -> shuf (AND (A, B), C, Mask)
 2)  (OR  (shuf (A, C, Mask), shuf (B, C, Mask)) -> shuf (OR  (A, B), C, Mask)
 3)  (XOR (shuf (A, C, Mask), shuf (B, C, Mask)) -> shuf (XOR (A, B), V_0, Mask)

 4)  (AND (shuf (C, A, Mask), shuf (C, B, Mask)) -> shuf (C, AND (A, B), Mask)
 5)  (OR  (shuf (C, A, Mask), shuf (C, B, Mask)) -> shuf (C, OR  (A, B), Mask)
 6)  (XOR (shuf (C, A, Mask), shuf (C, B, Mask)) -> shuf (V_0, XOR (A, B), Mask)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204160 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoX86: Use enums for memory operand decoding instead of integer literals.
Manuel Jacob [Tue, 18 Mar 2014 16:14:11 +0000 (16:14 +0000)]
X86: Use enums for memory operand decoding instead of integer literals.

Summary:
X86BaseInfo.h defines an enum for the offset of each operand in a memory operand
sequence.  Some code uses it and some does not.  This patch replaces (hopefully)
all remaining locations where an integer literal was used instead of this enum.
No functionality change intended.

Reviewers: nadav

CC: llvm-commits, t.p.northover
Differential Revision: http://llvm-reviews.chandlerc.com/D3108

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204158 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoEnable CFI on Hexagon.
Krzysztof Parzyszek [Tue, 18 Mar 2014 16:02:37 +0000 (16:02 +0000)]
Enable CFI on Hexagon.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204157 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoFix PR19144: Incorrect offset generated for int-to-fp conversion at -O0.
Bill Schmidt [Tue, 18 Mar 2014 14:32:50 +0000 (14:32 +0000)]
Fix PR19144: Incorrect offset generated for int-to-fp conversion at -O0.

When converting a signed 32-bit integer to double-precision floating point on
hardware without a lfiwax instruction, we have to instead use a lfd followed
by fcfid.  We were erroneously offsetting the address by 4 bytes in
preparation for either a lfiwax or lfiwzx when generating the lfd.  This fixes
that silly error.

This was not caught in the test suite since the conversion tests were run with
-mcpu=pwr7, which implies availability of lfiwax.  I've added another test
case for older hardware that checks the code we expect in the absence of
lfiwax and other flavors of fcfid.  There are fewer tests in this test case
because we punt to DAG selection in more cases on older hardware.  (We must
generate complex fiddly sequences in those cases, and there is marginal
benefit in duplicating that logic in fast-isel.)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204155 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[msan] Origin tracking with history.
Evgeniy Stepanov [Tue, 18 Mar 2014 13:30:56 +0000 (13:30 +0000)]
[msan] Origin tracking with history.

LLVM part of MSan implementation of advanced origin tracking,
when we record not only creation point, but all locations where
an uninitialized value was stored to memory, too.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204151 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoTolerate unmangled names in sample profiles.
Diego Novillo [Tue, 18 Mar 2014 12:03:12 +0000 (12:03 +0000)]
Tolerate unmangled names in sample profiles.

Summary:
The compiler does not always generate linkage names. If a function
has been inlined and its body elided, its linkage name may not be
generated.

When the binary executes, the profiler will use its unmangled name
when attributing samples. This results in unmangled names in the
input profile.

We are currently failing hard when this happens. However, in this case
all that happens is that we fail to attribute samples to the inlined
function. While this means fewer optimization opportunities, it should
not cause a compilation failure.

This patch accepts all valid function names, regardless of whether
they were mangled or not.

Reviewers: chandlerc

CC: llvm-commits
Differential Revision: http://llvm-reviews.chandlerc.com/D3087

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204142 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoRevert r203962 and two revisions depending on it: r204028 and r204059.
Alexander Kornienko [Tue, 18 Mar 2014 10:36:11 +0000 (10:36 +0000)]
Revert r203962 and two revisions depending on it: r204028 and r204059.
The revision I'm reverting breaks handling of transitive aliases. This blocks us
and breaks sanitizer bootstrap:
http://lab.llvm.org:8011/builders/sanitizer-x86_64-linux-bootstrap/builds/2651
(and checked locally by Alexey).

This revision is the result of:

  svn merge -r204059:204058 -r204028:204027 -r203962:203961 .

+ the regression test added to test/MC/ELF/alias.s

Another way to reproduce the regression with clang:
  $ cat q.c
  void a1();
  void a2() __attribute__((alias("a1")));
  void a3() __attribute__((alias("a2")));
  void a1() {}

  $ ~/work/llvm-build/bin/clang-3.5-good -c q.c && mv q.o good.o && \
      ~/work/llvm-build/bin/clang-3.5-bad -c q.c && mv q.o bad.o && \
      objdump -t good.o bad.o

    good.o:     file format elf64-x86-64

    SYMBOL TABLE:
    0000000000000000 l    df *ABS*  0000000000000000 q.c
    0000000000000000 l    d  .text  0000000000000000 .text
    0000000000000000 l    d  .data  0000000000000000 .data
    0000000000000000 l    d  .bss   0000000000000000 .bss
    0000000000000000 l    d  .comment       0000000000000000 .comment
    0000000000000000 l    d  .note.GNU-stack        0000000000000000 .note.GNU-stack
    0000000000000000 l    d  .eh_frame      0000000000000000 .eh_frame
    0000000000000000 g     F .text  0000000000000006 a1
    0000000000000000 g     F .text  0000000000000006 a2
    0000000000000000 g     F .text  0000000000000006 a3

    bad.o:     file format elf64-x86-64

    SYMBOL TABLE:
    0000000000000000 l    df *ABS*  0000000000000000 q.c
    0000000000000000 l    d  .text  0000000000000000 .text
    0000000000000000 l    d  .data  0000000000000000 .data
    0000000000000000 l    d  .bss   0000000000000000 .bss
    0000000000000000 l    d  .comment       0000000000000000 .comment
    0000000000000000 l    d  .note.GNU-stack        0000000000000000 .note.GNU-stack
    0000000000000000 l    d  .eh_frame      0000000000000000 .eh_frame
    0000000000000000 g     F .text  0000000000000006 a1
    0000000000000000 g     F .text  0000000000000006 a2
    0000000000000000 g       .text  0000000000000000 a3

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204137 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoAdd IAS/EHABI changes to release notes
Renato Golin [Tue, 18 Mar 2014 10:16:15 +0000 (10:16 +0000)]
Add IAS/EHABI changes to release notes

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204134 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoDebugIRTests: Fixup for r204130.
NAKAMURA Takumi [Tue, 18 Mar 2014 09:58:28 +0000 (09:58 +0000)]
DebugIRTests: Fixup for r204130.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204132 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[msan] Kill -msan-store-clean-origin flag.
Evgeniy Stepanov [Tue, 18 Mar 2014 09:47:06 +0000 (09:47 +0000)]
[msan] Kill -msan-store-clean-origin flag.

Not only is it slower than the alternative, but also subtly broken.
This commit does not change the default behavior.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204131 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[C++11] Change DebugInfoFinder to use range-based loops
Alon Mishne [Tue, 18 Mar 2014 09:41:07 +0000 (09:41 +0000)]
[C++11] Change DebugInfoFinder to use range-based loops

Also changes the iterators to return actual DI type over MDNode.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204130 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[msan] Remove unused code.
Evgeniy Stepanov [Tue, 18 Mar 2014 08:29:42 +0000 (08:29 +0000)]
[msan] Remove unused code.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204125 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[C++11] Mark the target fast isel classes as 'final' so that the compiler can de...
Craig Topper [Tue, 18 Mar 2014 07:27:13 +0000 (07:27 +0000)]
[C++11] Mark the target fast isel classes as 'final' so that the compiler can de-virtualize some of the internal calls.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204123 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[C++11] Change the interface of getCOFF{Section,Relocation,Symbol} to make it work...
Alexey Samsonov [Tue, 18 Mar 2014 06:53:02 +0000 (06:53 +0000)]
[C++11] Change the interface of getCOFF{Section,Relocation,Symbol} to make it work with range-based for loops.

Reviewers: ruiu

Reviewed By: ruiu

CC: llvm-commits
Differential Revision: http://llvm-reviews.chandlerc.com/D3097

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204120 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoCodeGen/R600/v_cndmask.ll: Relax an expression to unbreak msvcrt.
NAKAMURA Takumi [Tue, 18 Mar 2014 06:17:22 +0000 (06:17 +0000)]
CodeGen/R600/v_cndmask.ll: Relax an expression to unbreak msvcrt.

  V_CNDMASK_B32_e64 v0, v0, -1.#QNAN0e+00, s[2:3], 0, 0, 0, 0

FIXME: We really need to implement our formatter...

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204118 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoDebugInfo/lto-comp-dir.ll: Tweak for dos path.
NAKAMURA Takumi [Tue, 18 Mar 2014 06:01:14 +0000 (06:01 +0000)]
DebugInfo/lto-comp-dir.ll: Tweak for dos path.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204117 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoARM: add an assertion
Saleem Abdulrasool [Tue, 18 Mar 2014 05:26:55 +0000 (05:26 +0000)]
ARM: add an assertion

Add an assertion that a valid section is referenced.  The potential NULL pointer
dereference was identified by the clang static analyzer.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204114 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoDwarfDebug: Only unique retained types instead of all types.
Adrian Prantl [Tue, 18 Mar 2014 02:35:03 +0000 (02:35 +0000)]
DwarfDebug: Only unique retained types instead of all types.
This is a follow-up to r203983 based on feedback from dblaikie and mren (Thanks!)
No functionality change.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204107 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoSwitch the type field in DIVariable and DIGlobalVariable over to DITypeRefs.
Adrian Prantl [Tue, 18 Mar 2014 02:34:58 +0000 (02:34 +0000)]
Switch the type field in DIVariable and DIGlobalVariable over to DITypeRefs.
This allows us to catch more opportunities for ODR-based type uniquing
during LTO.
Paired commit with CFE which updates some testcases to verify the new
DIBuilder behavior.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204106 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoReplace unnecessary #include directive with forward declarations.
Adrian Prantl [Tue, 18 Mar 2014 02:34:52 +0000 (02:34 +0000)]
Replace unnecessary #include directive with forward declarations.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204104 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoAdd explanatory comment.
Eric Christopher [Tue, 18 Mar 2014 02:18:27 +0000 (02:18 +0000)]
Add explanatory comment.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204103 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoShorten DotDebugLocEntry to just DebugLocEntry and reformat.
Eric Christopher [Tue, 18 Mar 2014 02:18:24 +0000 (02:18 +0000)]
Shorten DotDebugLocEntry to just DebugLocEntry and reformat.
No functional change.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204102 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoDebugInfo: Avoid emitting standard opcode lengths in debug_line.dwo headers where...
David Blaikie [Tue, 18 Mar 2014 02:13:23 +0000 (02:13 +0000)]
DebugInfo: Avoid emitting standard opcode lengths in debug_line.dwo headers where opcodes are never used anyway

Introduce a slightly tighter wrapper around the header structure that
handles this use case. (MCDwarfDwoLineTable)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204101 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoDebugInfo: Implement debug_line.dwo for file names used in type units during -gsplit...
David Blaikie [Tue, 18 Mar 2014 01:17:26 +0000 (01:17 +0000)]
DebugInfo: Implement debug_line.dwo for file names used in type units during -gsplit-dwarf

This removes an attribute (and more importantly, a relocation) from
skeleton type units and removes some unnecessary file names from the
debug_line section that remains in the .o (and linked executable) file.

There's still a few places we could shave off some more space here:

* use compilation dir of the underlying compilation unit (since all the
  type units share that compilation dir - though this would be more
  complicated in LTO cases where they don't (keep a map of compilation
  dir->line table header?))

* Remove some of the unnecessary header fields from the line table since
  they're not needed in this situation (about 12 bytes per table).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204099 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoDebugInfo: Flag test as requiring object emission support
David Blaikie [Tue, 18 Mar 2014 00:12:25 +0000 (00:12 +0000)]
DebugInfo: Flag test as requiring object emission support

Cleans up buildbot failures on R600 and similar.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204095 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoDebugInfo: Do not rely on the compilation dir (index 0) for files in line tables...
David Blaikie [Tue, 18 Mar 2014 00:11:48 +0000 (00:11 +0000)]
DebugInfo: Do not rely on the compilation dir (index 0) for files in line tables shared between compilation units

When emitting assembly there's no support for emitting separate line
tables for each compilation unit - so LLVM emits .loc directives
producing a single line table.

Line tables have an implicit directory (index 0) equal to the
compilation directory (DW_AT_comp_dir) of the compilation unit that
references them.

If multiple compilation units (with possibly disparate compilation
directories) reference the same line table, we must avoid relying on
this ambiguous directory.

Achieve this my simply not setting the compilation directory on the line
table when we're in this situation (multiple units while emitting
assembly).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204094 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoDebugInfo: Move getOrCreateSourceID from DwarfDebug to DwarfCompileUnit
David Blaikie [Mon, 17 Mar 2014 23:53:25 +0000 (23:53 +0000)]
DebugInfo: Move getOrCreateSourceID from DwarfDebug to DwarfCompileUnit

We still do a few lookups into the line table mapping in MCContext that
could be factored out into a single lookup (rather than looking it up
once for the table label, once to set the compilation unit, once for
each time we need a file ID, etc... ) but assembly output complicates
that somewhat as we still need a virtual dispatch back to the
MCAsmStreamer in that case.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204092 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoDebugInfo: Move line table zero-directory-index (compilation dir) handling into MCDwarf
David Blaikie [Mon, 17 Mar 2014 23:29:40 +0000 (23:29 +0000)]
DebugInfo: Move line table zero-directory-index (compilation dir) handling into MCDwarf

Our handling of compilation directory in DwarfDebug was broken
(incorrectly using the 'last' compilation directory (that of the last
CU in the metadata list) for all function emission in any CU). By moving
this handling down into MCDwarf the issue is fixed as the compilation
dir is tracked correctly per line table.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204089 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoMake methods static
Matt Arsenault [Mon, 17 Mar 2014 22:23:09 +0000 (22:23 +0000)]
Make methods static

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204085 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoUse range metadata instead of introducing selects.
Dan Gohman [Mon, 17 Mar 2014 19:57:04 +0000 (19:57 +0000)]
Use range metadata instead of introducing selects.

When GlobalOpt has determined that a GlobalVariable only ever has two values,
it would convert the GlobalVariable to a boolean, and introduce SelectInsts
at every load, to choose between the two possible values. These SelectInsts
introduce overhead and other unpleasantness.

This patch makes GlobalOpt just add range metadata to loads from such
GlobalVariables instead. This enables the same main optimization (as seen in
test/Transforms/GlobalOpt/integer-bool.ll), without introducing selects.

The main downside is that it doesn't get the memory savings of shrinking such
GlobalVariables, but this is expected to be negligible.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204076 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoSwitch a number of loops in lib/CodeGen over to range-based for-loops, now that
Owen Anderson [Mon, 17 Mar 2014 19:36:09 +0000 (19:36 +0000)]
Switch a number of loops in lib/CodeGen over to range-based for-loops, now that
the MachineRegisterInfo iterators are compatible with it.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204075 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoReally REALLY finish adding const to the MachineRegisterInfo iterator range methods.
Owen Anderson [Mon, 17 Mar 2014 19:34:51 +0000 (19:34 +0000)]
Really REALLY finish adding const to the MachineRegisterInfo iterator range methods.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204074 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoMaking a guess to fix the test case with r204056 to get the build bot working.
Kevin Enderby [Mon, 17 Mar 2014 19:00:03 +0000 (19:00 +0000)]
Making a guess to fix the test case with r204056 to get the build bot working.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204073 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoR600: Match sign_extend_inreg to BFE instructions
Matt Arsenault [Mon, 17 Mar 2014 18:58:11 +0000 (18:58 +0000)]
R600: Match sign_extend_inreg to BFE instructions

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204072 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoMake DAGCombiner work on vector bitshifts with constant splat vectors.
Matt Arsenault [Mon, 17 Mar 2014 18:58:01 +0000 (18:58 +0000)]
Make DAGCombiner work on vector bitshifts with constant splat vectors.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204071 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoAdd one more const marker that I missed.
Owen Anderson [Mon, 17 Mar 2014 18:16:10 +0000 (18:16 +0000)]
Add one more const marker that I missed.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204070 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoDebugInfo: Use MC line table file entry uniquing for non-asm input as well.
David Blaikie [Mon, 17 Mar 2014 18:13:58 +0000 (18:13 +0000)]
DebugInfo: Use MC line table file entry uniquing for non-asm input as well.

See r204027 for the precursor to this that applied to asm debug info.

This required some non-obvious API changes to handle the case of asm
output (we never go asm->asm so this didn't come up in r204027): the
modification of the file/directory name by MCDwarfLineTableHeader needed
to be reflected in the MCAsmStreamer caller so it could print the
appropriate .file directive, so those StringRef parameters are now
non-const ref (in/out) parameters rather than just const.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204069 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoMark MachineRegisterInfo's iterator range methods as const.
Owen Anderson [Mon, 17 Mar 2014 18:05:05 +0000 (18:05 +0000)]
Mark MachineRegisterInfo's iterator range methods as const.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204067 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoAdd iterator range definitions for the MachineRegisterInfo iterators.
Owen Anderson [Mon, 17 Mar 2014 18:01:41 +0000 (18:01 +0000)]
Add iterator range definitions for the MachineRegisterInfo iterators.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204066 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[X86] Fix unused variable warning with NDEBUG from r204058
Adam Nemet [Mon, 17 Mar 2014 17:32:53 +0000 (17:32 +0000)]
[X86] Fix unused variable warning with NDEBUG from r204058

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204063 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoMC: change runtime check to an assertion
Saleem Abdulrasool [Mon, 17 Mar 2014 17:13:57 +0000 (17:13 +0000)]
MC: change runtime check to an assertion

It is unclear how it would be possible to get M to be NULL in normal scenarios.
Change this to an assert rather than a runtime check as per dblakie's
suggestion.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204060 91177308-0d34-0410-b5e6-96231b3b80d8