Commit Graph

21766 Commits

Author SHA1 Message Date
Rafael Espindola
3d8a2f07d5 Don't redeclare a pure virtual method.
I.E., there is no value is having

void foo() override = 0;

If it is override it is already present in a base class. Since it is pure,
some other class will have to implement it.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221537 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-07 18:07:04 +00:00
Daniel Sanders
b4b823941c [mips] Remove MipsCC::analyzeCallOperands in favour of CCState::AnalyzeCallOperands. NFC
Summary:
In addition to the usual f128 workaround, it was also necessary to provide
a means of accessing ArgListEntry::IsFixed.

Reviewers: theraven, vmedic

Reviewed By: vmedic

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D6111

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221518 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-07 11:43:49 +00:00
Jay Foad
222fcc59cb llvm-symbolizer: teach it about PowerPC64 ELF function descriptors
Summary:
Teach llvm-symbolizer about PowerPC64 ELF function descriptors. Symbols in the .opd section point to function descriptors, the first word of which is a pointer to the real function. For the purposes of symbolizing we pretend that the symbol points directly to the function.

This is enough to get decent function names in stack traces for unoptimized binaries, which fixes the sanitizer print-stack-trace test on PowerPC64 Linux.

Reviewers: kcc, willschm, samsonov

Reviewed By: samsonov

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D6110

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221514 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-07 09:08:39 +00:00
Justin Hibbits
893f22f882 Add Position-independent Code model Module API.
Summary:
This makes PIC levels a Module flag attribute, which can be queried by the
backend.  The flag is named `PIC Level`, and can have a value of:

  0 - Backend-default
  1 - Small-model (-fpic)
  2 - Large-model (-fPIC)

These match the `-pic-level' command line argument for clang, and the value of the
preprocessor macro `__PIC__'.

Test Plan:
New flags tests specific for the 'PIC Level' module flag.
Tests to be added as part of a future commit for PowerPC, which will use this new API.

Reviewers: rafael, echristo

Reviewed By: rafael, echristo

Subscribers: rafael, llvm-commits

Differential Revision: http://reviews.llvm.org/D5882

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221510 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-07 04:46:10 +00:00
Rafael Espindola
cce2cdd98a Don't repeat names in comments. NFC.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221495 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-06 22:57:10 +00:00
Simon Atanasyan
848edb1bfa [ELF][yaml2obj] Handle additional MIPS specific st_other field flags
The ELF symbol `st_other` field might contain additional flags besides
visibility ones. This patch implements support for some MIPS specific
flags.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221491 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-06 22:46:24 +00:00
Frederic Riss
6df2a3be33 Change DIBuilder::createImportedDeclaration from taking a DIScope to a DIDescriptor.
Imported declarations can be DIGlobalVariables which aren't a DIScope. Today
clang (unknowingly I believe) shoehorns these into a DIScope and it all works
just because we never access the fields.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221466 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-06 17:46:55 +00:00
Steven Wu
36c4749c19 Remove obsolete ARM intrinsics vclz and vcnt
Both of the intrinsics get autoupgraded to target independent
intrinsics.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221396 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-05 21:02:55 +00:00
Simon Atanasyan
5563be0f4f Add accessor to get 'visibility' part of st_other field
This new `getVisibility()` function will also be used in the LLD code.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221392 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-05 20:47:35 +00:00
Duncan P. N. Exon Smith
bad06b13ba IR: MDNode => Value: NamedMDNode::getOperator()
Change `NamedMDNode::getOperator()` from returning `MDNode *` to
returning `Value *`.  To reduce boilerplate at some call sites, add a
`getOperatorAsMDNode()` for named metadata that's expected to only
return `MDNode` -- for now, that's everything, but debug node named
metadata (such as llvm.dbg.cu and llvm.dbg.sp) will soon change.  This
is part of PR21433.

Note that there's a follow-up patch to clang for the API change.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221375 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-05 18:16:03 +00:00
Duncan P. N. Exon Smith
72ced73d45 IR: MDNode => Value: NamedMDNode::addOperand()
Change `NamedMDNode::addOperand()` to take a `Value *` instead of an
`MDNode *`.  This is part of PR21433.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221359 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-05 17:16:09 +00:00
Zoran Jovanovic
0f0a134d9d [mips][microMIPS] Mark symbols as microMIPS if necessary
Differential Revision: http://reviews.llvm.org/D6039


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221355 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-05 16:35:20 +00:00
David Majnemer
4787059b2f llvm-readobj: Add support for dumping the DOS header in PE files
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221333 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-05 06:24:35 +00:00
Jiangning Liu
07e1080b89 Revert 220932.
Commit 220932 caused crash when building clang-tblgen on aarch64 debian target,
so it's blocking all daily tests.

The std::call_once implementation in pthread has bug for aarch64 debian.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221331 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-05 04:44:31 +00:00
Matt Arsenault
1e288409dc Fix broken C++ mode comment
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221327 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-05 01:36:22 +00:00
Justin Bogner
d2d9ad0d66 ErrorOr: Be more explicit in the implicit conversion to bool docs
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221295 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-04 21:01:48 +00:00
Arnaud A. de Grandmaison
8025a39d11 [PBQP] Tweak spill costs and coalescing benefits
This patch improves how the different costs (register, interference, spill
and coalescing) relates together. The assumption is now that:
 - coalescing (or any other "side effect" of reg alloc) is negative, and
   instead of being derived from a spill cost, they use the block
   frequency info.
 - spill costs are in the [MinSpillCost:+inf( range
 - register or interference costs are in [0.0:MinSpillCost( or +inf

The current MinSpillCost is set to 10.0, which is a random value high
enough that the current constraint builders do not need to worry about
when settings costs. It would however be worth adding a normalization
step for register and interference costs as the last step in the
constraint builder chain to ensure they are not greater than SpillMinCost
(unless this has some sense for some architectures). This would work well
with the current builder pipeline, where all costs are tweaked relatively
to each others, but could grow above MinSpillCost if the pipeline is
deep enough.

The current heuristic is tuned to depend rather on the number of uses of
a live interval rather than a density of uses, as used by the greedy
allocator. This heuristic provides a few percent improvement on a number
of benchmarks (eembc, spec, ...) and will definitely need to change once
spill placement is implemented: the current spill placement is really
ineficient, so making the cost proportionnal to the number of use is a
clear win.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221292 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-04 20:51:24 +00:00
Rafael Espindola
d21956e2ec Remove unused DisableRedZone option.
Patch by Steve King.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221282 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-04 18:18:52 +00:00
Rafael Espindola
5dac5bd225 Remove FindProgramByName. NFC.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221258 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-04 12:35:47 +00:00
Michael J. Spencer
376a2e104f [Support][Program] Add findProgramByName(Name, OptionalPaths)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221220 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-04 01:29:29 +00:00
Reid Kleckner
9dfa407dc9 Use the LLVM_GNUC_PREREQ macro instead of the longhand version check
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221219 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-04 01:28:33 +00:00
Reid Kleckner
1bf7f442aa Remove the END_WITH_NULL macro now that Clang doesn't use it
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221218 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-04 01:15:53 +00:00
Reid Kleckner
d282ca71b0 Rename END_WITH_NULL to LLVM_END_WITH_NULL and move to Compiler.h
We shouldn't put this kind of attribute stuff in DataTypes.h.

Leave the END_WITH_NULL name for now so I can update clang without
making build spam.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221215 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-04 01:12:21 +00:00
Sanjoy Das
67bcf74b14 The patchpoint lowering logic would crash with live constants equal to
the tombstone or empty keys of a DenseMap<int64_t, T>.  This patch
fixes the issue (and adds a tests case).



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221214 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-04 00:59:21 +00:00
Kevin Enderby
a25032214d Add the code and test cases for 32-bit Intel to llvm-objdump’s Mach-O symbolizer.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221211 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-04 00:43:16 +00:00
Hal Finkel
cd9acde7f3 Use AA in LoadCombine
LoadCombine can be smarter about aborting when a writing instruction is
encountered, instead of aborting upon encountering any writing instruction, use
an AliasSetTracker, and only abort when encountering some write that might
alias with the loads that could potentially be combined.

This was originally motivated by comments made (and a test case provided) by
David Majnemer in response to PR21448. It turned out that LoadCombine was not
responsible for that PR, but LoadCombine should also be improved so that
unrelated stores (and @llvm.assume) don't interrupt load combining.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221203 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-03 23:19:16 +00:00
Reid Kleckner
cf7bb7fb68 Relax the LLVM_NOEXCEPT _MSC_VER version check back to 1900
Unconditional noexcept support was added in the VS 2013 Nov CTP. Given
that there have been three CTPs since then, I don't think we need
careful macro magic to target that specific tech preview. Instead,
target the major release version number of 1900, which corresponds to
the as-yet unreleased VS "14".

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221169 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-03 18:22:42 +00:00
Duncan P. N. Exon Smith
5e84760dde IR: MDNode => Value: Instruction::getAllMetadataOtherThanDebugLoc()
Change `Instruction::getAllMetadataOtherThanDebugLoc()` from a vector of
`MDNode` to one of `Value`.  Part of PR21433.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221167 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-03 18:13:57 +00:00
Oliver Stannard
25f4c35b22 Revert r221150, as it broke sanitizer tests
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221151 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-03 12:19:03 +00:00
Oliver Stannard
b63d71ef81 Emit .eh_frame with relocations to functions, rather than sections
When LLVM emits DWARF call frame information, it currently creates a local,
section-relative symbol in the code section, which is pointed to by a
relocation on the .eh_frame section. However, for C++ we emit some functions in
section groups, and the SysV ABI has some rules to make it easier to remove
these sections
(http://www.sco.com/developers/gabi/latest/ch4.sheader.html#section_group_rules):

  A symbol table entry with STB_LOCAL binding that is defined relative to one
  of a group's sections, and that is contained in a symbol table section that is
  not part of the group, must be discarded if the group members are discarded.
  References to this symbol table entry from outside the group are not allowed.

This means that we need to use the function symbol for the relocation, not a
temporary symbol.

There was a comment in the code claiming that the local symbol was used to
avoid creating a relocation, but a relocation must be created anyway as the
code and CFI are in different sections.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221150 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-03 12:02:51 +00:00
Charlie Turner
8e003cdb99 Test commit.
Fixes two typos.

Change-Id: I129f647de8933e1d8f0dc9941bcb91602edce7e2

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221148 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-03 10:58:05 +00:00
Diego Novillo
2b6bd7aaf5 Use ErrorOr for the ::create factory on instrumented and sample profilers.
Summary:
As discussed in
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20141027/242445.html,
the creation of reader and writer instances is better done using
ErrorOr. There are no functional changes, but several callers needed to
be adjusted.

Reviewers: bogner

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D6076

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221120 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-03 00:51:45 +00:00
Matt Arsenault
2220408e1a Support REG_SEQUENCE in tablegen.
The problem is mostly that variadic output instruction
aren't handled, so it is rejected for having an inconsistent
number of operands, and then the right number of operands
isn't emitted.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221117 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-02 23:46:51 +00:00
David Majnemer
ed297abb0a InstCombine: Don't assume that m_ZExt matches an Instruction
m_ZExt might bind against a ConstantExpr instead of an Instruction.
Assuming this, using cast<Instruction>, results in InstCombine crashing.

Instead, introduce ZExtOperator to bridge both Instruction and
ConstantExpr ZExts.

This fixes PR21445.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221069 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-01 23:46:05 +00:00
Daniel Sanders
cfe761c9e6 Renamed CCState members that appear to misspell 'Processed' as 'Proceed'. NFC.
Reviewers: rnk

Reviewed By: rnk

Subscribers: rnk, llvm-commits

Differential Revision: http://reviews.llvm.org/D5978

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221061 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-01 19:32:23 +00:00
Daniel Sanders
5e4b155521 [tablegen] Add CustomCallingConv and use it to tablegen-erate the outermost parts of the Mips O32 implementation
Summary:
CustomCallingConv is simply a CallingConv that tablegen should not generate the
implementation for. It allows regular CallingConv's to delegate to these custom
functions. This is (currently) necessary for Mips and we cannot use CCCustom
without having to adapt to the different API that CCCustom uses.

This brings us a bit closer to being able to remove
MipsCC::analyzeCallOperands and MipsCC::analyzeFormalArguments in favour of
the common implementation.

No functional change to the targets.

Depends on D3341

Reviewers: vmedic

Reviewed By: vmedic

Subscribers: vmedic, llvm-commits

Differential Revision: http://reviews.llvm.org/D5965

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221052 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-01 17:38:22 +00:00
Diego Novillo
9657de5f22 Add show and merge tools for sample PGO profiles.
Summary:
This patch extends the 'show' and 'merge' commands in llvm-profdata to handle
sample PGO formats. Using the 'merge' command it is now possible to convert
one sample PGO format to another.

The only format that is currently not working is 'gcc'. I still need to
implement support for it in lib/ProfileData.

The changes in the sample profile support classes are needed for the
merge operation.

Reviewers: bogner

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D6065

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221032 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-01 00:56:55 +00:00
Duncan P. N. Exon Smith
b2187ede9e IR: MDNode => Value: Instruction::getAllMetadata()
Change `Instruction::getAllMetadata()` to modify a vector of `Value`
instead of `MDNode` and update call sites.  This is part of PR21433.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221027 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-01 00:26:42 +00:00
Duncan P. N. Exon Smith
3a84a6377c IR: MDNode => Value: Instruction::getMetadata()
Change `Instruction::getMetadata()` to return `Value` as part of
PR21433.

Update most callers to use `Instruction::getMDNode()`, which wraps the
result in a `cast_or_null<MDNode>`.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221024 91177308-0d34-0410-b5e6-96231b3b80d8
2014-11-01 00:10:31 +00:00
Duncan P. N. Exon Smith
9ed44978ef IR: MDNode => Value: Add Instruction::getMDNode()
Add `Instruction::getMDNode()` that casts to `MDNode` before changing
`Instruction::getMetadata()` to return `Value`.  This avoids adding
`cast_or_null<MDNode>` boiler-plate throughout the code.

Part of PR21433.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221023 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-31 23:58:04 +00:00
Reid Kleckner
e1a4787d5d Work around bugs in MSVC "14" CTP 3's conversion logic
It appears to ignore or find ambiguous MachineInstrBuilder's conversion
operators that allow conversion to MachineInstr* and
MachineBasicBlock::bundle_iterator.

As a workaround, add an explicit way to get the MachineInstr.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221017 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-31 23:19:46 +00:00
Reid Kleckner
fc46c68310 Define LLVM_NOEXCEPT with MSVC 14 CTP 3 or newer
We have to use _MSC_FULL_VER here as CTP 2 and earlier didn't define
noexcept to my knowledge.

Fixes build error in lib/Support/Error.cpp when inheriting from
std::error_category, which has a noexcept virtual method.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221013 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-31 23:02:40 +00:00
Lang Hames
7b8ba81502 [Object] Modify OwningBinary's interface to separate inspection from ownership.
The getBinary and getBuffer method now return ordinary pointers of appropriate
const-ness. Ownership is transferred by calling takeBinary(), which returns a
pair of the Binary and a MemoryBuffer.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221003 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-31 21:37:49 +00:00
Duncan P. N. Exon Smith
815275f5de IR: MDNode => Value: Instruction::setMetadata()
Change `Instruction::setMetadata()` API to accept `Value` instead of
`MDNode`.  Part of PR21433.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220994 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-31 20:13:11 +00:00
Bill Schmidt
2d32816a45 [PowerPC] Initial VSX intrinsic support, with min/max for vector double
Now that we have initial support for VSX, we can begin adding
intrinsics for programmer access to VSX instructions.  This patch adds
basic support for VSX intrinsics in general, and tests it by
implementing intrinsics for minimum and maximum for the vector double
data type.

The LLVM portion of this is quite straightforward.  There is a
companion patch for Clang.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220988 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-31 19:19:07 +00:00
Quentin Colombet
9b6ca9304c [CodeGenPrepare] Move extractelement close to store if they can be combined.
This patch adds an optimization in CodeGenPrepare to move an extractelement
right before a store when the target can combine them.
The optimization may promote any scalar operations to vector operations in the
way to make that possible.


** Context **

Some targets use different register files for both vector and scalar operations.
This means that transitioning from one domain to another may incur copy from one
register file to another. These copies are not coalescable and may be expensive.
For example, according to the scheduling model, on cortex-A8 a vector to GPR
move is 20 cycles.


** Motivating Example **

Let us consider an example:
define void @foo(<2 x i32>* %addr1, i32* %dest) {
 %in1 = load <2 x i32>* %addr1, align 8
 %extract = extractelement <2 x i32> %in1, i32 1
 %out = or i32 %extract, 1
 store i32 %out, i32* %dest, align 4
 ret void
}

As it is, this IR generates the following assembly on armv7:
  vldr  d16, [r0]            @vector load  
  vmov.32 r0, d16[1]  @ cross-register-file copy: 20 cycles
  orr r0, r0, #1           @ scalar bitwise or
  str r0, [r1]               @ scalar store
  bx  lr

Whereas we could generate much faster code:
  vldr  d16, [r0]               @ vector load
  vorr.i32  d16, #0x1     @ vector bitwise or
  vst1.32 {d16[1]}, [r1:32] @ vector extract + store
  bx  lr

Half of the computation made in the vector is useless, but this allows to get
rid of the expensive cross-register-file copy.


** Proposed Solution **

To avoid this cross-register-copy penalty, we promote the scalar operations to
vector operations. The penalty will be removed if we manage to promote the whole
chain of computation in the vector domain.
Currently, we do that only when the chain of computation ends by a store and the
target is able to combine an extract with a store.

Stores are the most likely candidates, because other instructions produce values
that would need to be promoted and so, extracted as some point[1]. Moreover,
this is customary that targets feature stores that perform a vector extract (see
AArch64 and X86 for instance).

The proposed implementation relies on the TargetTransformInfo to decide whether
or not it is beneficial to promote a chain of computation in the vector domain.
Unfortunately, this interface is rather inaccurate for this level of details and
although this optimization may be beneficial for X86 and AArch64, the inaccuracy
will lead to the optimization being too aggressive.
Basically in TargetTransformInfo, everything that is legal has a cost of 1,
whereas, even if a vector type is legal, usually a vector operation is slightly
more expensive than its scalar counterpart. That will lead to too many
promotions that may not be counter balanced by the saving of the
cross-register-file copy. For instance, on AArch64 this penalty is just 4
cycles.

For now, the optimization is just enabled for ARM prior than v8, since those
processors have a larger penalty on cross-register-file copies, and the scope is
limited to basic blocks. Because of these two factors, we limit the effects of
the inaccuracy. Indeed, I did not want to build up a fancy cost model with block
frequency and everything on top of that.

[1] We can imagine targets that can combine an extractelement with  other
instructions than just stores. If we want to go into that direction, the current
interfaces must be augmented and, moreover, I think this becomes a global isel
problem.

Differential Revision: http://reviews.llvm.org/D5921

<rdar://problem/14170854>


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220978 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-31 17:52:53 +00:00
David Majnemer
a2715904e5 Object, COFF: Cleanup symbol type code, improve binutils compatibility
Do a better job classifying symbols.  This increases the consistency
between the COFF handling code and the ELF side of things.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220952 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-31 05:07:00 +00:00
NAKAMURA Takumi
714a629ad8 Threading.h: Give named parameters to llvm::call_once(flag,UserFn). [-Wdocumentation]
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220941 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-31 00:54:20 +00:00
Chris Bieneman
8744520b53 Removing the static initializer in ManagedStatic.cpp by using llvm_call_once to initialize the ManagedStatic mutex.
Summary:
This patch adds an llvm_call_once which is a wrapper around std::call_once on platforms where it is available and devoid of bugs. The patch also migrates the ManagedStatic mutex to be allocated using llvm_call_once.

These changes are philosophically equivalent to the changes added in r219638, which were reverted due to a hang on Win32 which was the result of a bug in the Windows implementation of std::call_once.

Reviewers: aaron.ballman, chapuni, chandlerc, rnk

Reviewed By: rnk

Subscribers: majnemer, llvm-commits

Differential Revision: http://reviews.llvm.org/D5922

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220932 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-30 22:07:09 +00:00
Diego Novillo
6f01d27a0e Fix comment spelling and tidy diagnostic call in profile reader.
No functional changes.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220922 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-30 20:19:19 +00:00