Eric Christopher [Tue, 10 Jun 2014 18:06:25 +0000 (18:06 +0000)]
Remove caching of the subtarget for AArch64SelectionDAGInfo.
llvm-svn: 210553
Eric Christopher [Tue, 10 Jun 2014 18:06:23 +0000 (18:06 +0000)]
Move DataLayout onto the AArch64 subtarget.
llvm-svn: 210552
Zachary Turner [Tue, 10 Jun 2014 18:03:04 +0000 (18:03 +0000)]
Test commit, wraps some lines to fit in 80 columns.
llvm-svn: 210551
Hans Wennborg [Tue, 10 Jun 2014 17:53:23 +0000 (17:53 +0000)]
Don't inherit dll attributes to deleted methods (PR19988)
We would previously end up with an error when instantiating the
following template:
template <typename> struct __declspec(dllimport) S {
void foo() = delete;
};
S<int> s;
error: attribute 'dllimport' cannot be applied to a deleted function
llvm-svn: 210550
Eric Christopher [Tue, 10 Jun 2014 17:44:12 +0000 (17:44 +0000)]
Move AArch64FrameLowering into the subtarget.
llvm-svn: 210549
Eric Christopher [Tue, 10 Jun 2014 17:33:39 +0000 (17:33 +0000)]
Remove the uses of AArch64TargetMachine and AArch64Subtarget from
AArch64FrameLowering.
llvm-svn: 210548
Fariborz Jahanian [Tue, 10 Jun 2014 17:08:04 +0000 (17:08 +0000)]
Objective-C IRGen. Patch to generate a weak symbol reference when
'super' dispatches a class method in category for OBJC_METACLASS.
This is when class is a weak_import class.
// rdar://
16529125
llvm-svn: 210547
Reed Kotler [Tue, 10 Jun 2014 16:45:44 +0000 (16:45 +0000)]
Do Materialize Floating Point in Mips Fast-Isel
Summary:
Implement materialize of floating point literals in Mips Fast-Isel
Reopened version of D3659
Test Plan: simplestorefp1.ll
Reviewers: dsanders
Reviewed By: dsanders
Differential Revision: http://reviews.llvm.org/D4071
llvm-svn: 210546
Andrea Di Biagio [Tue, 10 Jun 2014 16:42:57 +0000 (16:42 +0000)]
[X86] Improved target combine rules for selecting horizontal add/sub.
This patch slightly changes the algorithm introduced at revision 210477
to fix a problem where the algorithm was producing incorrect code for
the VEX.256 encoded versions of horizontal add/sub.
For these cases, we now try to split the two 256-bit vectors into
128-bit chunks before emitting horizontal add/sub dag nodes.
Added a new test case into haddsub-2.ll.
llvm-svn: 210545
Tom Stellard [Tue, 10 Jun 2014 16:42:41 +0000 (16:42 +0000)]
Hexagon: Expand i1 SELECT_CC
il is legal for Hexagon, so I should have marked this as Expand for
SELECT_CC when I removed setOperationAction(ISD::SELECT_CC, MVT::Other,
Expand); in r210541.
llvm-svn: 210544
Adam Nemet [Tue, 10 Jun 2014 16:39:53 +0000 (16:39 +0000)]
[X86] AVX512: Add vmovntdqa
Along with the corresponding intrinsic and tests.
llvm-svn: 210543
Renato Golin [Tue, 10 Jun 2014 16:39:21 +0000 (16:39 +0000)]
Fix a bug in the Thumb1 ARM Load/Store optimizer
Previously, the basic block was searched for future uses of the base register,
and if necessary any writeback to the base register was reset using a SUB
instruction (e.g. before calling a function) just before such a use. However,
this step happened *before* the merged LDM/STM instruction was built. So if
there was (e.g.) a function call directly after the not-yet-formed LDM/STM,
the pass would first insert a SUB instruction to reset the base register,
and then (at the same location, incorrectly) insert the LDM/STM itself.
This patch fixes PR19972. Patch by Moritz Roth.
llvm-svn: 210542
Tom Stellard [Tue, 10 Jun 2014 16:01:29 +0000 (16:01 +0000)]
SelectionDAG: Don't use MVT::Other to determine legality of ISD::SELECT_CC
The SelectionDAG bad a special case for ISD::SELECT_CC, where it would
allow targets to specify:
setOperationAction(ISD::SELECT_CC, MVT::Other, Expand);
to indicate that they wanted to expand ISD::SELECT_CC for all types.
This wasn't applied correctly everywhere, and it makes writing new
DAG patterns with ISD::SELECT_CC difficult.
llvm-svn: 210541
Tom Stellard [Tue, 10 Jun 2014 16:01:25 +0000 (16:01 +0000)]
SelectionDAG: Enable (and (setcc x), (setcc y)) -> (setcc (and x, y)) for vectors
This prevents a future commit from regressing:
test/CodeGen/R600/setcc-equivalent.ll
llvm-svn: 210540
Tom Stellard [Tue, 10 Jun 2014 16:01:22 +0000 (16:01 +0000)]
SelectionDAG: Expand SELECT_CC to SELECT + SETCC
This consolidates code from the Hexagon, R600, and XCore targets.
No functionality change intended.
llvm-svn: 210539
Tobias Grosser [Tue, 10 Jun 2014 14:48:17 +0000 (14:48 +0000)]
Test delinearization of 2D diagonal matrix
llvm-svn: 210538
Daniel Jasper [Tue, 10 Jun 2014 14:44:02 +0000 (14:44 +0000)]
clang-format: [JS] Treat dict literals similar to objc method exprs.
Before:
return {
link:
function() {
f(); //
}
};
return {
a: a,
link: function() {
f(); //
}
}
After:
return {
link: function() {
f(); //
}
};
return {
a: a,
link: function() {
f(); //
}
};
llvm-svn: 210537
Bill Schmidt [Tue, 10 Jun 2014 14:35:01 +0000 (14:35 +0000)]
[PPC64LE] Recognize shufflevector patterns for little endian
Various masks on shufflevector instructions are recognizable as
specific PowerPC instructions (vector pack, vector merge, etc.).
There is existing code in PPCISelLowering.cpp to recognize the correct
patterns for big endian code. The masks for these instructions are
different for little endian code due to the big-endian numbering
employed by these instructions. This patch adds the recognition code
for little endian.
I've added a new test case test/CodeGen/PowerPC/vec_shuffle_le.ll for
this. The existing recognizer test (vec_shuffle.ll) is unnecessarily
verbose and difficult to read, so I felt it was better to add a new
test rather than modify the old one.
llvm-svn: 210536
Chad Rosier [Tue, 10 Jun 2014 14:32:08 +0000 (14:32 +0000)]
[AArch64] Emit .ident compiler version attribute.
Patch by Ana Pazos<apazos@codeaurora.org>!
llvm-svn: 210535
Alexander Potapenko [Tue, 10 Jun 2014 14:22:00 +0000 (14:22 +0000)]
Add detection of OS X relocatable SDK to compiler-rt as a lit.util function
Clang's lit cfg already detects the currently selected SDK via
"xcrun --show-sdk-path". The same thing should be done for compiler-rt tests,
to make them work on recent OS X versions. Instead of duplicating the detection
code, this patch extracts the detection function into a lit.util method.
Patch by Kuba Brecka (kuba.brecka@gmail.com),
reviewed at http://reviews.llvm.org/D4072
llvm-svn: 210534
Evgeniy Stepanov [Tue, 10 Jun 2014 13:54:15 +0000 (13:54 +0000)]
[sanitizer] Relax sanity checks in ioctl decoding.
Standard KVM ioctls don't pass currect ioctl_decode().
llvm-svn: 210533
Simon Atanasyan [Tue, 10 Jun 2014 13:43:24 +0000 (13:43 +0000)]
[ReaderWriter] Simplify initialization of class fields.
No functional changes.
llvm-svn: 210532
Simon Atanasyan [Tue, 10 Jun 2014 13:43:19 +0000 (13:43 +0000)]
[ELF] Fix typo in the type name of the range-based loop item.
No functional changes.
llvm-svn: 210531
Simon Atanasyan [Tue, 10 Jun 2014 13:43:13 +0000 (13:43 +0000)]
[ELF] Fix the bug -- LLD crashes if the --whole-archive option
is around non-archive.
llvm-svn: 210530
Daniel Jasper [Tue, 10 Jun 2014 13:27:57 +0000 (13:27 +0000)]
Artyom Skrobov [Tue, 10 Jun 2014 13:11:35 +0000 (13:11 +0000)]
Condition codes AL and NV are invalid in the aliases that use
inverted condition codes (CINC, CINV, CNEG, CSET, and CSETM).
Matching aliases based on "immediate classes", when disassembling,
wasn't previously supported, hence adding MCOperandPredicate
into class Operand, and implementing the support for it
in AsmWriterEmitter.
The parsing for those aliases was already custom, so just adding
the missing condition into AArch64AsmParser::parseCondCode.
llvm-svn: 210528
Artyom Skrobov [Tue, 10 Jun 2014 12:47:23 +0000 (12:47 +0000)]
Refactoring in AsmWriterEmitter::EmitPrintAliasInstruction()
llvm-svn: 210527
Artyom Skrobov [Tue, 10 Jun 2014 12:41:14 +0000 (12:41 +0000)]
Anonymous definitions in foreach blocks triggered a 'def already exists'
llvm-svn: 210526
Tim Northover [Tue, 10 Jun 2014 10:50:24 +0000 (10:50 +0000)]
AArch64: disallow x30 & x29 as the destination for indirect tail calls
As Ana Pazos pointed out, these have to be restored to their incoming values
before a function returns; i.e. before the tail call. So they can't be used
correctly as the destination register.
llvm-svn: 210525
Tim Northover [Tue, 10 Jun 2014 10:50:11 +0000 (10:50 +0000)]
Revert "X86: elide comparisons after cmpxchg instructions."
This reverts commit r210523. It was committed prematurely without waiting for
review.
llvm-svn: 210524
Tim Northover [Tue, 10 Jun 2014 10:49:07 +0000 (10:49 +0000)]
X86: elide comparisons after cmpxchg instructions.
The C++ and C semantics of the compare_and_swap operations actually
require us to return a boolean "success" value. In LLVM terms this
means a second comparison of the output of "cmpxchg" against the input
desired value.
However, x86's "cmpxchg" instruction sets all flags for the comparison
formed, so we can skip any secondary comparison. (N.b. this isn't true
for cmpxchg8b/16b, which only set ZF).
rdar://problem/
13201607
llvm-svn: 210523
Daniel Jasper [Tue, 10 Jun 2014 10:42:26 +0000 (10:42 +0000)]
clang-format: Fix enum formatting with specific comment.
Before:
enum Fruit { //
APPLE,
PEAR };
After:
enum Fruit { //
APPLE,
PEAR
};
llvm-svn: 210522
Alp Toker [Tue, 10 Jun 2014 09:58:45 +0000 (09:58 +0000)]
Prospective build fix following clang r210518
llvm-svn: 210521
Tim Northover [Tue, 10 Jun 2014 09:52:44 +0000 (09:52 +0000)]
AArch64: teach FastISel how to handle offset FrameIndices
Previously we were abandonning the attempt, leading to some combination of
extra work (when selection of a load/store fails completely) and inferior code
(when this leads to a real memcpy call instead of inlining).
rdar://problem/
17187463
llvm-svn: 210520
Tim Northover [Tue, 10 Jun 2014 09:52:40 +0000 (09:52 +0000)]
AArch64: make FastISel memcpy emission more robust.
We were hitting an assert if FastISel couldn't create the load or store we
requested. Currently this happens for large frame-local addresses, though
CodeGen could be improved there.
rdar://problem/
17187463
llvm-svn: 210519
Alp Toker [Tue, 10 Jun 2014 09:31:37 +0000 (09:31 +0000)]
Improve diagnostic mapping terminology
Diagnostic mappings are used to calculate the final severity of diagnostic
instances.
Detangle the implementation to reflect the terminology used in documentation
and bindings.
No change in functionality.
llvm-svn: 210518
Evgeniy Stepanov [Tue, 10 Jun 2014 08:32:01 +0000 (08:32 +0000)]
[asan] Disable a flaky test.
llvm-svn: 210517
Eric Christopher [Tue, 10 Jun 2014 08:03:42 +0000 (08:03 +0000)]
Delete X86JITInfo in the subtarget destructor.
llvm-svn: 210516
Alp Toker [Tue, 10 Jun 2014 07:03:25 +0000 (07:03 +0000)]
Don't suppress backend diagnostics in system headers
Doing so would be inconsistent with the common fallback case where backend
diagnostics without source locations are emitted unconditionally.
llvm-svn: 210515
Daniel Jasper [Tue, 10 Jun 2014 06:39:03 +0000 (06:39 +0000)]
clang-format: Support variadic lambda captures.
Before:
return [ i, args... ]{};
After:
return [i, args...] {};
llvm-svn: 210514
Daniel Jasper [Tue, 10 Jun 2014 06:27:23 +0000 (06:27 +0000)]
clang-format: Handle multiline strings inside ternary expressions.
With AlwaysSplitBeforeMultilineStrings, clang-format would not find any
valid solution.
llvm-svn: 210513
Alp Toker [Tue, 10 Jun 2014 06:09:00 +0000 (06:09 +0000)]
Show -Wdate-time in system headers
Anyone enabling this warning would expect to hear about all occurrences
including those in system headers that can cause non-reproducible builds.
To achieve this, rework ShowInSystemHeader to remove broken unused mapping code
that didn't make sense with a simpler and correct scheme.
llvm-svn: 210512
Alp Toker [Tue, 10 Jun 2014 06:08:51 +0000 (06:08 +0000)]
Implement -Wdate-time preprocessor warning
This GCC warning is useful for validating reproducible builds
and might help when tracking down issues with modules too.
llvm-svn: 210511
Alp Toker [Tue, 10 Jun 2014 06:08:41 +0000 (06:08 +0000)]
Improve specificity in the diag-mapping2.c test
Make sure grep catches the correct diagnostic, otherwise this test fails to
detect broken functionality.
llvm-svn: 210510
Simon Atanasyan [Tue, 10 Jun 2014 05:59:15 +0000 (05:59 +0000)]
[llvm-readobj][ELF] Factor out the code retrieve ELF symbol information
(section name, section index, full name) into the separate functions.
No functional changes.
llvm-svn: 210509
Craig Topper [Tue, 10 Jun 2014 04:50:50 +0000 (04:50 +0000)]
Simplify code. No functional change.
llvm-svn: 210508
NAKAMURA Takumi [Tue, 10 Jun 2014 04:06:56 +0000 (04:06 +0000)]
SmallVectorTest.cpp: Use LLVM_DELETED_FUNCTION.
llvm-svn: 210507
Nick Kledzik [Tue, 10 Jun 2014 01:50:00 +0000 (01:50 +0000)]
[mach-o] refactor mach-o output section selection to be table driven.
In -r mode the sections use the table used to parse .o files. Otherwise
use final exectuable table. No functionality change.
llvm-svn: 210506
Hans Wennborg [Tue, 10 Jun 2014 00:55:51 +0000 (00:55 +0000)]
Improve checking for dynamic initializers of dllimport fields in template instantiation
We would previously assert if the initializer was dependent. I also think that
checking isConstantInitializer is more correct here than checkInitIsICE.
llvm-svn: 210505
Juergen Ributzka [Tue, 10 Jun 2014 00:32:29 +0000 (00:32 +0000)]
[ConstantHoisting][X86] Improve the cost model for small constants with large types (i64 and above).
This improves the X86 cost model for small constants with large types. Before
this commit we would even hoist trivial constants such as i96 2.
This is related to <rdar://problem/
17070936>
llvm-svn: 210504
Alp Toker [Mon, 9 Jun 2014 23:59:38 +0000 (23:59 +0000)]
Make '-Werror=frame-larger-than=' and associated diagnostic pragmas GCC-compatible
It turns out the trailing '=' really is part of the option name spelling and
treating it as such gets us compatible with GCC's -Werror= and pragmas.
(GCC doesn't appear to support any -Wno- form for this diagnostic but we do.)
llvm-svn: 210503
Nick Kledzik [Mon, 9 Jun 2014 23:35:37 +0000 (23:35 +0000)]
[mach-o] parse multiple symbols on same address into aliases
llvm-svn: 210502
Reid Kleckner [Mon, 9 Jun 2014 23:32:20 +0000 (23:32 +0000)]
Reorder Value and User fields to save 8 bytes of padding on 64-bit
Reviewered by: rafael
Differential Revision: http://reviews.llvm.org/D4073
llvm-svn: 210501
Marshall Clow [Mon, 9 Jun 2014 23:27:27 +0000 (23:27 +0000)]
Fixed damaged row in issues table; thanks to STL for the catch
llvm-svn: 210500
Reid Kleckner [Mon, 9 Jun 2014 23:16:24 +0000 (23:16 +0000)]
Reduce indentation in ActOnIdExpression, NFC
llvm-svn: 210499
Richard Trieu [Mon, 9 Jun 2014 22:53:25 +0000 (22:53 +0000)]
Removing an "if (this == nullptr)" check from two print methods. The condition
will never be true in a well-defined context. The checking for null pointers
has been moved into the caller logic so it does not rely on undefined behavior.
llvm-svn: 210498
Richard Trieu [Mon, 9 Jun 2014 22:53:16 +0000 (22:53 +0000)]
Removing an "if (!this)" check from two print methods. The condition will
never be true in a well-defined context. The checking for null pointers
has been moved into the caller logic so it does not rely on undefined behavior.
llvm-svn: 210497
Alp Toker [Mon, 9 Jun 2014 22:42:55 +0000 (22:42 +0000)]
Reduce verbiage of lit.local.cfg files
We can just split targets_to_build in one place and make it immutable.
llvm-svn: 210496
David Blaikie [Mon, 9 Jun 2014 22:26:20 +0000 (22:26 +0000)]
SmallVector: support resize(N) with move-only types
Unfortunately there's no way to elegantly do this with pre-canned
algorithms. Using a generating iterator doesn't work because you default
construct for each element, then move construct into the actual slot
(bad for copy but non-movable types, and a little unneeded overhead even
in the move-only case), so just write it out manually.
This solution isn't exception safe (if one of the element's ctors calls
we don't fall back, destroy the constructed elements, and throw on -
which std::uninitialized_fill does do) but SmallVector (and LLVM) isn't
exception safe anyway.
llvm-svn: 210495
Alexey Samsonov [Mon, 9 Jun 2014 22:02:14 +0000 (22:02 +0000)]
[TSan] Exclude blacklist tests from manual test runner
llvm-svn: 210494
Bill Schmidt [Mon, 9 Jun 2014 22:00:52 +0000 (22:00 +0000)]
[PPC64LE] Generate correct code for unaligned little-endian vector loads
The code in PPCTargetLowering::PerformDAGCombine() that handles
unaligned Altivec vector loads generates a lvsl followed by a vperm.
As we've seen in numerous other places, the vperm instruction has a
big-endian bias, and this is fixed for little endian by complementing
the permute control vector and swapping the input operands. In this
case the lvsl is providing the permute control vector. Rather than
generating an lvsl and a complement operation, it is sufficient to
generate an lvsr instruction instead. Thus for LE code generation we
will generate an lvsr rather than an lvsl, and swap the other input
arguments on the vperm.
The existing test/CodeGen/PowerPC/vec_misalign.ll is updated to test
the code generation for PPC64 and PPC64LE, in addition to the existing
PPC32/G5 testing.
llvm-svn: 210493
Alexey Samsonov [Mon, 9 Jun 2014 21:53:47 +0000 (21:53 +0000)]
Generate better location ranges for some register-described variables.
Don't terminate location ranges for register-described variables
at the end of machine basic block if this register is never modified
in the function body, except for the prologue and epilogue. Prologue
location is guessed by FrameSetup flags on MachineInstructions, while
epilogue location is deduced from debug locations of instructions
in the basic blocks ending with return instructions.
This patch is mostly targeted to fix non-trivial debug locations for
variables addressed via stack and frame pointers.
It is not really a generic fix. We can still produce poor debug info
for register-described variables if this register *is* modified somewhere
in the function, but in unrelated places. This might be the case for the debug
info in optimized binaries (e.g. for local variables in inlined functions).
LiveDebugVariables pass in CodeGen attempts to fix this problem by adjusting
DBG_VALUE instructions, but this pass is tied to greedy register allocator,
which is used in optimized builds only. Proper fix would likely involve
generalizing LiveDebugVariables to all register allocators. See more discussion
in http://reviews.llvm.org/D3933 review thread.
I'm proceeding with this patch to fix immediate severe problems and
important cases, e.g. fix completely broken debug info with AddressSanitizer
and fix PR19307 (missing debug info for by-value std::string arguments).
llvm-svn: 210492
Fariborz Jahanian [Mon, 9 Jun 2014 21:42:01 +0000 (21:42 +0000)]
Objective-C. Consider block pointer as NSObject as well as conforming to
'NSCopying' protocol when diagnosing block to ObjC pointer conversion.
// rdar://
16739120
llvm-svn: 210491
Rafael Espindola [Mon, 9 Jun 2014 21:21:33 +0000 (21:21 +0000)]
Update langref for unnamed_addr being allowed in aliases.
Thanks to Duncan P. N. Exon Smith and Owen Anderson for noticing.
llvm-svn: 210490
Saleem Abdulrasool [Mon, 9 Jun 2014 20:18:42 +0000 (20:18 +0000)]
ARM: add VLA extension for WoA Itanium ABI
The armv7-windows-itanium environment is nearly identical to the MSVC ABI. It
has a few divergences, mostly revolving around the use of the Itanium ABI for
C++. VLA support is one of the extensions that are amongst the set of the
extensions.
This adds support for proper VLA emission for this environment. This is
somewhat similar to the handling for __chkstk emission on X86 and the large
stack frame emission for ARM. The invocation style for chkstk is still
controlled via the -mcmodel flag to clang.
Make an explicit note that this is an extension.
llvm-svn: 210489
Matt Arsenault [Mon, 9 Jun 2014 19:20:29 +0000 (19:20 +0000)]
Look through addrspacecasts when turning ptr comparisons into
index comparisons.
llvm-svn: 210488
Alp Toker [Mon, 9 Jun 2014 19:09:28 +0000 (19:09 +0000)]
Disallow multiple inclusion of clang/Config/config.h
Internal config.h headers are only meant to be included from the main file.
llvm-svn: 210487
Stepan Dyatkovskiy [Mon, 9 Jun 2014 19:03:02 +0000 (19:03 +0000)]
Added functions cross-reference test.
Originally this similar was initiated by Björn Steinbrink here:
http://reviews.llvm.org/D3437
Bug itself has been fixed by principal changes in MergeFunctions. Though
special checks for functions merging are still actual. And the test has
been accepted with slight modifications.
llvm-svn: 210486
Alp Toker [Mon, 9 Jun 2014 19:00:52 +0000 (19:00 +0000)]
Remove old fenv.h workaround for a historic clang driver bug
Tested and works fine with clang using libstdc++.
All indications are that this was fixed some time ago and isn't a problem with
any clang version we support.
I've added a note in PR6907 which is still open for some reason.
llvm-svn: 210485
Hans Wennborg [Mon, 9 Jun 2014 18:30:28 +0000 (18:30 +0000)]
Allow definition of dllimport static fields in partial specializations (PR19956)
This expands the logic from r210141 to cover partial specializations too.
llvm-svn: 210484
Alp Toker [Mon, 9 Jun 2014 18:28:53 +0000 (18:28 +0000)]
Fold FEnv.h into the implementation
Support headers shouldn't use config.h definitions, and they should never be
undefined like this.
ConstantFolding.cpp was the only user of this facility and already includes
config.h for other math features, so it makes sense to move the checks there at
point of use.
(The implicit config.h was also quite dangerous -- removing the FEnv.h include
would have silently disabled math constant folding without causing any tests to
fail. Need to investigate -Wundef once the cleanup is done.)
This eliminates the last config.h include from LLVM headers, paving the way for
more consistent configuration checks.
llvm-svn: 210483
Peter Zotov [Mon, 9 Jun 2014 18:28:47 +0000 (18:28 +0000)]
[OCaml] Add more Llvm_target tests
Patch by Jacques-Pascal Deplaix
llvm-svn: 210482
Todd Fiala [Mon, 9 Jun 2014 17:46:37 +0000 (17:46 +0000)]
Added gdb-remote test for s packet, single stepping.
llvm-svn: 210481
Peter Zotov [Mon, 9 Jun 2014 17:34:34 +0000 (17:34 +0000)]
[OCaml] Unbreak Llvm_target.TargetMachine.set_verbose_asm
Patch by Jacques-Pascal Deplaix
llvm-svn: 210480
Eric Christopher [Mon, 9 Jun 2014 17:08:19 +0000 (17:08 +0000)]
Move all of the x86 subtarget initialized variables down into the x86 subtarget
from the x86 target machine. Should be no functional change.
llvm-svn: 210479
Matt Arsenault [Mon, 9 Jun 2014 17:00:46 +0000 (17:00 +0000)]
R600/SI: Rename VOP3 helper class to be more general
It has other uses besides shift instructions.
llvm-svn: 210478
Andrea Di Biagio [Mon, 9 Jun 2014 16:54:41 +0000 (16:54 +0000)]
[X86] Add target combine rules for horizontal add/sub.
This patch adds new target specific combine rules to identify horizontal
add/sub idioms from BUILD_VECTOR dag nodes.
This patch also teaches the DAGCombiner how to canonicalize sequences of
insert_vector_elt dag nodes according to the following rule:
(insert_vector_elt (insert_vector_elt A, I0), I1) ->
(insert_vecto_elt (insert_vector_elt A, I1), I0)
This new canonicalization rule only triggers if the inner insert_vector
dag node has exactly one use; also, both indices must be known constants,
and I1 < I0.
This last rule made it possible to write a simpler algorithm to identify
horizontal add/sub patterns because now we don't have to worry about the
ordering of insert_vector_elt dag nodes.
llvm-svn: 210477
Matt Arsenault [Mon, 9 Jun 2014 16:36:31 +0000 (16:36 +0000)]
R600/SI: Keep 64-bit not on SALU
llvm-svn: 210476
Matt Arsenault [Mon, 9 Jun 2014 16:20:25 +0000 (16:20 +0000)]
R600: Fix selection failure for vector bswap
llvm-svn: 210475
Bill Schmidt [Mon, 9 Jun 2014 16:06:29 +0000 (16:06 +0000)]
[PPC64LE] Generate correct little-endian code for v16i8 multiply
The existing code in PPCTargetLowering::LowerMUL() for multiplying two
v16i8 values assumes that vector elements are numbered in big-endian
order. For little-endian targets, the vector element numbering is
reversed, but the vmuleub, vmuloub, and vperm instructions still
assume big-endian numbering. To account for this, we must adjust the
permute control vector and reverse the order of the input registers on
the vperm instruction.
The existing test/CodeGen/PowerPC/vec_mul.ll is updated to be executed
on powerpc64 and powerpc64le targets as well as the original powerpc
(32-bit) target.
llvm-svn: 210474
Evgeniy Stepanov [Mon, 9 Jun 2014 14:48:53 +0000 (14:48 +0000)]
Fix test in r210472.
llvm-svn: 210473
Evgeniy Stepanov [Mon, 9 Jun 2014 14:29:34 +0000 (14:29 +0000)]
[msan] Workaround for invalid origins in shufflevector.
Makes origin propagation ignore literal undef operands, and,
in general, any operand we don't have origin for.
https://code.google.com/p/memory-sanitizer/issues/detail?id=56
llvm-svn: 210472
NAKAMURA Takumi [Mon, 9 Jun 2014 14:20:23 +0000 (14:20 +0000)]
llvm/test/CodeGen/X86/2014-05-29-factorial.ll: Relax an expression to match Win32 x64.
llvm-svn: 210471
Sasa Stankovic [Mon, 9 Jun 2014 14:09:28 +0000 (14:09 +0000)]
[mips] Fix a bug for NaCl target - Don't report the error when non-dangerous
load/store is in branch delay slot.
Differential Revision: http://llvm-reviews.chandlerc.com/D4048
llvm-svn: 210470
Simon Atanasyan [Mon, 9 Jun 2014 13:45:58 +0000 (13:45 +0000)]
[Mips] Make dt-textrel.test test case independent from external input files.
llvm-svn: 210469
Andrea Di Biagio [Mon, 9 Jun 2014 12:34:50 +0000 (12:34 +0000)]
[X86] Avoid emitting unnecessary test instructions.
This patch teaches the backend how to check for the 'NoSignedWrap' flag on
binary operations to improve the emission of 'test' instructions.
If the result of a binary operation is known not to overflow we know that
resetting the Overflow flag is unnecessary and so we can avoid emitting
the test instruction.
Patch by Marcello Maggioni.
llvm-svn: 210468
Andrea Di Biagio [Mon, 9 Jun 2014 12:32:53 +0000 (12:32 +0000)]
[DAG] Expose NoSignedWrap, NoUnsignedWrap and Exact flags to SelectionDAG.
This patch modifies SelectionDAGBuilder to construct SDNodes with associated
NoSignedWrap, NoUnsignedWrap and Exact flags coming from IR BinaryOperator
instructions.
Added a new SDNode type called 'BinaryWithFlagsSDNode' to allow accessing
nsw/nuw/exact flags during codegen.
Patch by Marcello Maggioni.
llvm-svn: 210467
Alexey Volkov [Mon, 9 Jun 2014 11:40:41 +0000 (11:40 +0000)]
[X86] Use ADD/SUB instead of INC/DEC for Silvermont
According to Intel Software Optimization Manual
on Silvermont INC or DEC instructions require
an additional uop to merge the flags.
As a result, a branch instruction depending
on an INC or a DEC instruction incurs a 1 cycle penalty.
Differential Revision: http://reviews.llvm.org/D3990
llvm-svn: 210466
Evgeniy Stepanov [Mon, 9 Jun 2014 11:17:37 +0000 (11:17 +0000)]
[asan] Remove dependency of tests on runtime library in standalone build.
llvm-svn: 210465
Artyom Skrobov [Mon, 9 Jun 2014 11:10:14 +0000 (11:10 +0000)]
[AArch64] Missing aliases for CMP/CMN [W]SP with no shift
llvm-svn: 210464
Evgeniy Stepanov [Mon, 9 Jun 2014 10:41:22 +0000 (10:41 +0000)]
[msan] Intercept __strto*_internal.
This should fix strtoimax/strtoumax on newer glibc.
https://code.google.com/p/memory-sanitizer/issues/detail?id=36
llvm-svn: 210463
Jeroen Ketema [Mon, 9 Jun 2014 10:12:29 +0000 (10:12 +0000)]
[docs] Fix typo, align comments, fix syntax highlighting
llvm-svn: 210462
Rui Ueyama [Mon, 9 Jun 2014 09:58:53 +0000 (09:58 +0000)]
Remove dead parameter.
llvm-svn: 210461
Zoran Jovanovic [Mon, 9 Jun 2014 09:49:51 +0000 (09:49 +0000)]
[mips][mips64r6] Add LDPC instruction
Differential Revision: http://reviews.llvm.org/D3822
llvm-svn: 210460
Evgeniy Stepanov [Mon, 9 Jun 2014 09:09:19 +0000 (09:09 +0000)]
Fix line numbers for code inlined from __nodebug__ functions.
Instructions from __nodebug__ functions don't have file:line
information even when inlined into no-nodebug functions. As a result,
intrinsics (SSE and other) from <*intrin.h> clang headers _never_
have file:line information.
With this change, an instruction without !dbg metadata gets one from
the call instruction when inlined.
Fixes PR19001.
llvm-svn: 210459
Evgeniy Stepanov [Mon, 9 Jun 2014 08:58:41 +0000 (08:58 +0000)]
[msan] Add a test for mmx.packuswb.
llvm-svn: 210458
Evgeniy Stepanov [Mon, 9 Jun 2014 08:57:40 +0000 (08:57 +0000)]
[msan] Simplify tests.
llvm-svn: 210457
Rui Ueyama [Mon, 9 Jun 2014 08:57:37 +0000 (08:57 +0000)]
Re-commit r210425.
llvm-svn: 210456
Rui Ueyama [Mon, 9 Jun 2014 08:42:38 +0000 (08:42 +0000)]
Add missing dependency for check-lld.
llvm-svn: 210455
Evgeniy Stepanov [Mon, 9 Jun 2014 08:40:16 +0000 (08:40 +0000)]
[msan] Fix vector pack intrinsic handling.
This fixes a crash on MMX intrinsics, as well as a corner case in handling of
all unsigned pack intrinsics.
PR19953.
llvm-svn: 210454