Craig Topper [Fri, 2 Sep 2016 06:11:31 +0000 (06:11 +0000)]
[AVX-512] Move tests for masked floating point logical operations to avx512dqvl-intrinsics-upgrade.ll since they have now been autoupgraded.
llvm-svn: 280467
Craig Topper [Fri, 2 Sep 2016 05:29:17 +0000 (05:29 +0000)]
[AVX-512] Remove floating point logical operation instrinsics and replace them with native IR.
llvm-svn: 280466
Craig Topper [Fri, 2 Sep 2016 05:29:13 +0000 (05:29 +0000)]
[AVX-512] Add more patterns for masked and broadcasted logical operations where the select or broadcast has a floating point type.
These are needed in order to remove the masked floating point logical operation intrinsics and use native IR.
llvm-svn: 280465
Craig Topper [Fri, 2 Sep 2016 05:29:09 +0000 (05:29 +0000)]
[AVX-512] Add execution domain fixing for logical operations with broadcast loads. This builds on the handling of masked ops since we need to keep element size the same.
llvm-svn: 280464
Craig Topper [Fri, 2 Sep 2016 04:25:33 +0000 (04:25 +0000)]
[X86] Strengthen some SDNode type constraints.
llvm-svn: 280463
Craig Topper [Fri, 2 Sep 2016 04:25:30 +0000 (04:25 +0000)]
[AVX-512] Add NoVLX Predicates to some patterns so they don't rely on pattern ordering to be lower priority than their equivalent VLX pattern.
llvm-svn: 280462
Lang Hames [Fri, 2 Sep 2016 03:50:50 +0000 (03:50 +0000)]
[Docs] Fix another typo in the Error/Expected docs.
llvm-svn: 280461
Lang Hames [Fri, 2 Sep 2016 03:46:08 +0000 (03:46 +0000)]
[Docs] Fix a couple of typos in the Error/Expected docs.
llvm-svn: 280460
Lang Hames [Fri, 2 Sep 2016 03:45:44 +0000 (03:45 +0000)]
[ORC] Fix some missing fields in OrcRemoteTargetClient's move constructor.
llvm-svn: 280459
George Burgess IV [Fri, 2 Sep 2016 03:38:43 +0000 (03:38 +0000)]
Add missing &. NFC.
llvm-svn: 280458
Hal Finkel [Fri, 2 Sep 2016 02:58:25 +0000 (02:58 +0000)]
[PowerPC] hasAndNotCompare should return true
As Sanjay suggested when he added the hook, PPC should return true from
hasAndNotCompare. We have an efficient negated 'and' on PPC (which can feed a
compare).
Fixes PR27203.
llvm-svn: 280457
Alexander Shaposhnikov [Fri, 2 Sep 2016 02:56:07 +0000 (02:56 +0000)]
Add clang-reorder-fields to clang-tools-extra
This diff adds v0 of clang-reorder-fields tool to clang/tools/extra.
The main idea behind this tool is to simplify and make less error-prone refactoring of large codebases when
someone needs to change the order fields of a struct/class (for example to remove excessive padding).
Differential revision: https://reviews.llvm.org/D23279
llvm-svn: 280456
Greg Parker [Fri, 2 Sep 2016 02:44:07 +0000 (02:44 +0000)]
[lit] Fail testing if a googletest executable crashes during test discovery
googletest formatted tests are discovered by running the test executable.
Previously testing would silently succeed if the test executable crashed
during the discovery process. Now testing fails with "error: unable to
discover google-tests ..." if the test executable exits with a non-zero status.
llvm-svn: 280455
Hal Finkel [Fri, 2 Sep 2016 02:34:44 +0000 (02:34 +0000)]
[PowerPC] Add a pattern for a runtime bit check
Following a suggestion by Sanjay, we should lower:
%shl = shl i32 1, %y
%and = and i32 %x, %shl
%cmp = icmp eq i32 %and, %shl
ret i1 %cmp
into:
subfic r4, r4, 32
rlwnm r3, r3, r4, 31, 31
Add this pattern and some associated patterns for the 64-bit case and the
not-equal case. Fixes PR27356.
llvm-svn: 280454
Dehao Chen [Fri, 2 Sep 2016 01:59:27 +0000 (01:59 +0000)]
revert r280429 and r280425:
r280425 | dehao | 2016-09-01 16:15:50 -0700 (Thu, 01 Sep 2016) | 9 lines
Refactor LICM pass in preparation for LoopSink pass.
Summary: LoopSink pass uses some common function in LICM. This patch refactor the LICM code to make it usable by LoopSink pass (https://reviews.llvm.org/D22778).
r280429 | dehao | 2016-09-01 16:31:25 -0700 (Thu, 01 Sep 2016) | 9 lines
Refactor LICM to expose canSinkOrHoistInst to LoopSink pass.
Summary: LoopSink pass shares the same canSinkOrHoistInst functionality with LICM pass. This patch exposes this function in preparation of https://reviews.llvm.org/D22778
llvm-svn: 280453
Dehao Chen [Fri, 2 Sep 2016 01:47:13 +0000 (01:47 +0000)]
revert r280432:
r280432 | dehao | 2016-09-01 16:51:37 -0700 (Thu, 01 Sep 2016) | 9 lines
Explicitly require DominatorTreeAnalysis pass for instsimplify pass.
Summary: DominatorTreeAnalysis is always required by instsimplify.
llvm-svn: 280452
NAKAMURA Takumi [Fri, 2 Sep 2016 01:33:00 +0000 (01:33 +0000)]
llvm/test/Transforms/GCOVProfiling/three-element-mdnode.ll: Use %/T instead of %T, not to emit backslashes.
llvm-svn: 280451
Justin Bogner [Fri, 2 Sep 2016 01:21:37 +0000 (01:21 +0000)]
bugpoint: clang-format all of bugpoint. NFC
I'm going to clean up the APIs here a bit and touch many many lines
anyway.
llvm-svn: 280450
NAKAMURA Takumi [Fri, 2 Sep 2016 01:20:18 +0000 (01:20 +0000)]
raw_pwrite_stream_test.cpp: _putenv_s() may be assumed as win32-generic.
llvm-svn: 280449
Kyle Butt [Fri, 2 Sep 2016 01:20:06 +0000 (01:20 +0000)]
IfConversion: Don't count branches in # of duplicates.
If the entire blocks match, we would count the branch instructions
toward the number of duplicated instructions. This doesn't match what we
do elsewhere, and was causing a bug.
llvm-svn: 280448
Chandler Carruth [Fri, 2 Sep 2016 01:16:27 +0000 (01:16 +0000)]
[PM] Add a unittest for invalidating module analyses with an SCC pass.
This wasn't really well explicitly tested with a nice unittest before.
It seems good to have reasonably broken out unittests for this kind of
functionality as I'm workin go other invalidation features to make sure
none of the existing ones regress.
This still has too much duplicated code, I plan to factor that out in
a subsequent commit to use common helpers for repeated parts of this.
llvm-svn: 280447
Chandler Carruth [Fri, 2 Sep 2016 01:14:05 +0000 (01:14 +0000)]
[PM] (NFC) Split the IR parsing into a fixture so that I can split out
more testing into other test routines while using the same core module.
llvm-svn: 280446
Reid Kleckner [Fri, 2 Sep 2016 01:10:53 +0000 (01:10 +0000)]
Fix a real temp file leak in FileOutputBuffer
If we failed to commit the buffer but did not die to a signal, the temp
file would remain on disk on Windows. Having an open file mapping and
file handle prevents the file from being deleted. I am choosing not to
add an assertion of success on the temp file removal, since virus
scanners and other environmental things can often cause removal to fail
in real world tools.
Also fix more temp file leaks in unit tests.
llvm-svn: 280445
Chandler Carruth [Fri, 2 Sep 2016 01:08:04 +0000 (01:08 +0000)]
[PM] (NFC) Refactor the CGSCC pass manager tests to use lambda-based
passes.
This simplifies the test some and makes it more focused and clear what
is being tested. It will also make it much easier to extend with further
testing of different pass behaviors.
I've also replaced a pointless module pass with running the requires
pass directly as that is all that it was really doing.
llvm-svn: 280444
Reid Kleckner [Fri, 2 Sep 2016 00:51:34 +0000 (00:51 +0000)]
Try to fix some temp file leaks in SupportTests, PR18335
llvm-svn: 280443
Adam Nemet [Fri, 2 Sep 2016 00:28:26 +0000 (00:28 +0000)]
[CFGPrinter] Display branch weight on the edges
Summary:
This is pretty useful especially in connection with
BFI's -view-block-freq-propagation-dags. It helped me to track down the
bug that is being fixed in D24118.
While -view-block-freq-propagation-dags displays the high-level
information with static heuristics included (and block frequencies), the
new thing only shows the raw weight as presented by PGO without any of
the static estimates. This helps to distinguished what has been
measured vs. estimated.
For the sample loop in D24118, -view-block-freq-propagation-dags=integer
looks like this:
https://reviews.llvm.org/
F2381352
While with -view-cfg-only you can see the underlying branch weights:
https://reviews.llvm.org/
F2392296
Reviewers: dexonsmith, bogner, davidxl
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D24144
llvm-svn: 280442
Hal Finkel [Fri, 2 Sep 2016 00:28:20 +0000 (00:28 +0000)]
[PowerPC] Don't apply the PPC64 address-formation peephole for offsets greater than 7
When applying our address-formation PPC64 peephole, we are reusing the @ha TOC
addis value with the low parts associated with different offsets (i.e.
different effective symbol addends). We were assuming this was okay so long as
the offsets were less than the alignment of the global variable being accessed.
This ignored the fact, however, that the TOC base pointer itself need only be
8-byte aligned. As a result, what we were doing is legal only for offsets less
than 8 regardless of the alignment of the object being accessed.
Fixes PR28727.
llvm-svn: 280441
Hal Finkel [Fri, 2 Sep 2016 00:27:50 +0000 (00:27 +0000)]
[PowerPC] Don't consider fusion in PPC64 address-formation peephole
The logic in this function assumes that the P8 supports fusion of addis/addi,
but it does not. As a result, there is no advantage to restricting our peephole
application, merging addi instructions into dependent memory accesses, even
when the addi has multiple users, regardless of whether or not we're optimizing
for size.
We might need something like this again for the P9; I suspect we'll revisit
this code when we work on P9 tuning.
llvm-svn: 280440
Jason Henline [Fri, 2 Sep 2016 00:25:52 +0000 (00:25 +0000)]
[StreamExecutor] Pass device memory by ref
Summary:
Step 3 of getting GlobalDeviceMemory to own its handle.
Since GlobalDeviceMemory will no longer by copy-constructible, we must
pass instances by reference rather than by value.
Reviewers: jlebar
Subscribers: jprice, parallel_libs-commits
Differential Revision: https://reviews.llvm.org/D24172
llvm-svn: 280439
Alexander Shaposhnikov [Fri, 2 Sep 2016 00:24:06 +0000 (00:24 +0000)]
Revert https://reviews.llvm.org/D23279 because the tests have failed on several platforms
llvm-svn: 280438
Jason Henline [Fri, 2 Sep 2016 00:22:05 +0000 (00:22 +0000)]
[SE] Make Kernel movable
Summary:
Kernel is basically just a smart pointer to the underlying
implementation, so making it movable prevents having to store a
std::unique_ptr to it.
Reviewers: jlebar
Subscribers: jprice, parallel_libs-commits
Differential Revision: https://reviews.llvm.org/D24150
llvm-svn: 280437
Richard Smith [Fri, 2 Sep 2016 00:18:05 +0000 (00:18 +0000)]
Clean up handling of reading module files from stdin. Don't bother trying to
look for a corresponding file, since we're not going to read it anyway.
No observable behavior change (though we now avoid pointlessly trying to stat
or open a file named "-").
llvm-svn: 280436
Richard Smith [Fri, 2 Sep 2016 00:10:28 +0000 (00:10 +0000)]
Refactor to avoid holding a reference to a container element that could go away
during this function, and to avoid rolling back changes to the module manager's
data structures. Instead, we defer registering the module file until after we
have successfully finished loading it.
llvm-svn: 280434
Dehao Chen [Thu, 1 Sep 2016 23:51:37 +0000 (23:51 +0000)]
Explicitly require DominatorTreeAnalysis pass for instsimplify pass.
Summary: DominatorTreeAnalysis is always required by instsimplify.
Reviewers: davidxl, danielcdh
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D24173
llvm-svn: 280432
Alexander Shaposhnikov [Thu, 1 Sep 2016 23:49:48 +0000 (23:49 +0000)]
Add clang-reorder-fields to clang-tools-extra
This diff adds v0 of clang-reorder-fields tool to clang/tools/extra.
The main idea behind this tool is to simplify and make less error-prone refactoring of large codebases when
someone needs to change the order fields of a struct/class (for example to remove excess padding).
Differential revision: https://reviews.llvm.org/D23279
llvm-svn: 280431
Aditya Kumar [Thu, 1 Sep 2016 23:35:26 +0000 (23:35 +0000)]
[SelectionDAGBuilder] Add const to relevant places
Reviewers: hans, evandro, sebpop
Differential Revision: https://reviews.llvm.org/D24112
llvm-svn: 280430
Dehao Chen [Thu, 1 Sep 2016 23:31:25 +0000 (23:31 +0000)]
Refactor LICM to expose canSinkOrHoistInst to LoopSink pass.
Summary: LoopSink pass shares the same canSinkOrHoistInst functionality with LICM pass. This patch exposes this function in preparation of https://reviews.llvm.org/D22778
Reviewers: chandlerc, davidxl, danielcdh
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D24171
llvm-svn: 280429
Jason Henline [Thu, 1 Sep 2016 23:27:39 +0000 (23:27 +0000)]
[StreamExecutor] Read dev array directly in test
Summary:
Step 2 of getting GlobalDeviceMemory to own its handle.
Use the SimpleHostPlatformDevice allocate methods to create device
arrays for tests, and check for successful copies by dereferncing the
device array handle directly because we know it is really a host
pointer.
Reviewers: jlebar
Subscribers: jprice, parallel_libs-commits
Differential Revision: https://reviews.llvm.org/D24148
llvm-svn: 280428
Dehao Chen [Thu, 1 Sep 2016 23:26:48 +0000 (23:26 +0000)]
Refactor replaceDominatedUsesWith to have a flag to control whether to replace uses in BB itself.
Summary: This is in preparation for LoopSink pass which calls replaceDominatedUsesWith to update after sinking.
Reviewers: chandlerc, davidxl, danielcdh
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D24170
llvm-svn: 280427
Alexander Shaposhnikov [Thu, 1 Sep 2016 23:18:00 +0000 (23:18 +0000)]
Remove excessive padding from MismatchingNewDeleteDetector
The class MismatchingNewDeleteDetector is in
lib/Sema/SemaExprCXX.cpp inside the anonymous namespace.
This diff reorders the fields and removes the excessive padding.
Test plan: make -j8 check-clang
Differential revision: https://reviews.llvm.org/D23898
llvm-svn: 280426
Dehao Chen [Thu, 1 Sep 2016 23:15:50 +0000 (23:15 +0000)]
Refactor LICM pass in preparation for LoopSink pass.
Summary: LoopSink pass uses some common function in LICM. This patch refactor the LICM code to make it usable by LoopSink pass (https://reviews.llvm.org/D22778).
Reviewers: chandlerc, davidxl, danielcdh
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D24168
llvm-svn: 280425
Michael Kuperstein [Thu, 1 Sep 2016 23:02:32 +0000 (23:02 +0000)]
[Legalizer] Don't throw away false low half when expanding GT/LT SETCC
When expanding a SETCC for which the low half is known to evaluate to false,
we can only throw it away for LT/GT comparisons, not LE/GE.
This fixes PR29170.
Differential Revision: https://reviews.llvm.org/D24151
llvm-svn: 280424
Rui Ueyama [Thu, 1 Sep 2016 22:48:05 +0000 (22:48 +0000)]
Add comments.
llvm-svn: 280423
Reid Kleckner [Thu, 1 Sep 2016 22:46:26 +0000 (22:46 +0000)]
Un-XFAIL cfi/stats.cpp, it passes since fixing llvm-symbolizer to look for dwarf again
llvm-svn: 280422
Petr Hosek [Thu, 1 Sep 2016 22:43:03 +0000 (22:43 +0000)]
[ELF] Do not omit debug sections when computing build-id
The primary use of build-id is in debugging, hence omitting debug
sections when computing it significantly reduces its usability as
changes in debug section content wouldn't alter the build-id.
Differential Revision: https://reviews.llvm.org/D24120
llvm-svn: 280421
Derek Schuff [Thu, 1 Sep 2016 22:38:37 +0000 (22:38 +0000)]
[WebAssembly] Change wasm SizeType to match asmjs
Summary:
We want wasm and asmjs to have matching ABIs, and right now asmjs uses
unsigned int for its size_t. This causes exported symbols in libcxx to
not match and can cause weird breakage where libcxx doesn't get linked
as a result. Long-term we probably want wasm32, wasm64, and asmjs to
all use unsigned long, but that would cause unnecessary ABI churn for
asmjs so defer that until we can make all the ABI changes at once.
Patch by Jacob Gravelle
Differential Revision: https://reviews.llvm.org/D24134
llvm-svn: 280420
Chandler Carruth [Thu, 1 Sep 2016 22:18:25 +0000 (22:18 +0000)]
Make the coding standards a bit more clear that we prefer the fancy new
auto-brief format for doxygen comments. Most notable is switching to
that in the example doxygen comment. I've also tweaked the wording but
am happy to tweak it further if others have suggestions here.
Mostly doing this to capture something I and others have been writing
consistently and repeatedly in code reviews.
llvm-svn: 280419
Michael Kuperstein [Thu, 1 Sep 2016 21:32:09 +0000 (21:32 +0000)]
[SelectionDAG] Generate vector_shuffle nodes for undersized result vector sizes
Prior to this, we could generate a vector_shuffle from an IR shuffle when the
size of the result was exactly the sum of the sizes of the input vectors.
If the output vector was narrower - e.g. a <12 x i8> being formed by a shuffle
with two <8 x i8> inputs - we would lower the shuffle to a sequence of extracts
and inserts.
Instead, we can form a larger vector_shuffle, and then extract a subvector
of the right size - e.g. shuffle the two <8 x i8> inputs into a <16 x i8>
and then extract a <12 x i8>.
This also includes a target-specific X86 combine that in the presence of
AVX2 combines:
(vector_shuffle <mask> (concat_vectors t1, undef)
(concat_vectors t2, undef))
into:
(vector_shuffle <mask> (concat_vectors t1, t2), undef)
in cases where this allows us to form VPERMD/VPERMQ.
(This is not a separate commit, as that pattern does not appear without
the DAGBuilder change.)
llvm-svn: 280418
Saleem Abdulrasool [Thu, 1 Sep 2016 21:09:19 +0000 (21:09 +0000)]
cstdio: limit gets to CRT versions below 14
Microsoft removed gets from the CRT in Visual Studio 2015 onwards [1].
Attempting to reference it when targeting CRT versions 14 and above will cause
compile errors.
[1] https://msdn.microsoft.com/en-us/library/
2029ea5f.aspx
Patch by Shoaib Meenai!
llvm-svn: 280417
Chris Bieneman [Thu, 1 Sep 2016 21:05:49 +0000 (21:05 +0000)]
[CMake] Add option (defaulted off) to exclude atomic.c
The atomic builtin source is problematic when cross-compiling builtins because it requires a variable and sometimes large set of system headers. This option allows users to optionally prevent it from being built.
llvm-svn: 280416
Heejin Ahn [Thu, 1 Sep 2016 21:05:15 +0000 (21:05 +0000)]
[WebAssembly] Add asm.js-style setjmp/longjmp handling for wasm (reland r280302)
Summary: This patch adds asm.js-style setjmp/longjmp handling support for WebAssembly. It also uses JavaScript's try and catch mechanism.
Reviewers: jpp, dschuff
Subscribers: jfb, dschuff
Differential Revision: https://reviews.llvm.org/D24121
llvm-svn: 280415
Justin Bogner [Thu, 1 Sep 2016 21:04:36 +0000 (21:04 +0000)]
bugpoint: clang-format and modernize comments in ListReducer. NFC
llvm-svn: 280414
Saleem Abdulrasool [Thu, 1 Sep 2016 20:49:54 +0000 (20:49 +0000)]
build: don't add -fPIC on Windows
`-fPIC` doesn't make much sense for Windows, since Windows DLLs aren't compiled
position independent and are instead relocated at runtime.
Patch by Shoaib Meenai!
llvm-svn: 280413
Tim Northover [Thu, 1 Sep 2016 20:45:41 +0000 (20:45 +0000)]
GlobalISel: add a G_PHI instruction to give phis a type.
They're another source of generic vregs, which are going to need a type on the
definition when we remove the register width from MachineRegisterInfo.
llvm-svn: 280412
Evgeniy Stepanov [Thu, 1 Sep 2016 20:43:59 +0000 (20:43 +0000)]
[tsan] Bump stack frame size limit.
llvm-svn: 280411
Reid Kleckner [Thu, 1 Sep 2016 20:28:59 +0000 (20:28 +0000)]
Fix the ASan fuse-lld.cc test after LLD r280012
With that change, images built with 'lld-link /debug' always have a
debug directory. If no PDB filename was passed on the command line, then
the filename in the executable is empty.
PDB information would never work anyway if the PDB file name is empty,
so go ahead and try DWARF in that case.
llvm-svn: 280410
Richard Smith [Thu, 1 Sep 2016 20:15:25 +0000 (20:15 +0000)]
When we reach the end of a #include of a header of a local submodule that we
textually included, create an ImportDecl just as we would if we reached a
#include of any other modular header. This is necessary in order to correctly
determine the set of variables to initialize for an imported module.
This should hopefully make the modules selfhost buildbot green again.
llvm-svn: 280409
Michael Kruse [Thu, 1 Sep 2016 19:53:31 +0000 (19:53 +0000)]
Allow mapping scalar MemoryAccesses to array elements.
Change the code around setNewAccessRelation to allow to use a an existing array
element for memory instead of an ad-hoc alloca. This facility will be used for
DeLICM/DeGVN to convert scalar dependencies into regular ones.
The changes necessary include:
- Make the code generator use the implicit locations instead of the alloca ones.
- A test case
- Make the JScop importer accept changes of scalar accesses for that test case.
- Adapt the MemoryAccess interface to the fact that the MemoryKind can change.
They are named (get|is)OriginalXXX() to get the status of the memory access
before any change by setNewAccessRelation() (some properties such as
getIncoming() do not change even if the kind is changed and are still
required). To get the modified properties, there is (get|is)LatestXXX(). The
old accessors without Original|Latest become synonyms of the
(get|is)OriginalXXX() to not make functional changes in unrelated code.
Differential Revision: https://reviews.llvm.org/D23962
llvm-svn: 280408
Chris Bieneman [Thu, 1 Sep 2016 19:47:39 +0000 (19:47 +0000)]
[CMake] NFC. Missed the other uses in r280406
Doh! Obviously need to slow down.
llvm-svn: 280407
Chris Bieneman [Thu, 1 Sep 2016 19:46:02 +0000 (19:46 +0000)]
[CMake] NFC. Fixing spelling from r280400
Doh!
llvm-svn: 280406
Matthew Simpson [Thu, 1 Sep 2016 19:40:19 +0000 (19:40 +0000)]
[LV] Use ScalarParts for ad-hoc pointer IV scalarization (NFCI)
We can now maintain scalar values in VectorLoopValueMap. Thus, we no longer
have to create temporary vectors with insertelement instructions when handling
pointer induction variables. This case was mistakenly missed from r279649 when
refactoring the other scalarization code.
llvm-svn: 280405
Michael Kruse [Thu, 1 Sep 2016 19:16:58 +0000 (19:16 +0000)]
Check validity of new access relations. NFC.
There are some constraints on maps that can be access relations. In builds with assertions enabled, verify
- The access domain is the same space as the statement's domain (modulo parameters).
- Whether an access is defined for every instance of the statement. (codegen does not yet support partial access relations)
- Whether the access range links to an array, represented by a ScopArrayInfo.
- The number of access dimensions equals the dimensions of the array.
- The array is not an indirect access. (also not supported by codegen)
Differential Revision: https://reviews.llvm.org/D23916
llvm-svn: 280404
Sanjay Patel [Thu, 1 Sep 2016 19:14:19 +0000 (19:14 +0000)]
[InstCombine] add tests to show potential shuffle+insert folds
llvm-svn: 280403
Andrey Turetskiy [Thu, 1 Sep 2016 18:50:02 +0000 (18:50 +0000)]
[X86] Loosen memory folding requirements for cvtdq2pd and cvtps2pd instructions.
According to spec cvtdq2pd and cvtps2pd instructions don't require memory operand to be aligned
to 16 bytes. This patch removes this requirement from the memory folding table.
Differential Revision: https://reviews.llvm.org/D23919
llvm-svn: 280402
Jason Henline [Thu, 1 Sep 2016 18:48:21 +0000 (18:48 +0000)]
[StreamExecutor] Dev handles in platform interface
Summary:
This is the first in a series of patches that will convert
GlobalDeviceMemory to own its device memory handle. The first step is to
remove GlobalDeviceMemoryBase from the PlatformInterface interfaces and
use raw handles there instead. This is useful because
GlobalDeviceMemoryBase is going to lose its importance in this process.
Reviewers: jlebar
Subscribers: jprice, parallel_libs-commits
Differential Revision: https://reviews.llvm.org/D24114
llvm-svn: 280401
Chris Bieneman [Thu, 1 Sep 2016 18:46:53 +0000 (18:46 +0000)]
[CMake] Don't copy headers on fully standalone
In r280108 I tried to make the headers copy relative to LLVM_BINARY_DIR, and the intent was that it would only happen on in-tree builds or runtimes directory builds. It didn't actually work that way.
This patch adds a check for CMAKE_SOURCE_DIR being equal to CMAKE_CURRENT_SOURCE_DIR. In this case we set a variable LIBCXX_USING_INSTLLED_LLVM. This doesn't necessarily mean the LLVM is installed (it could be a build directory), but it means we need to treat the LLVM directory as read-only.
llvm-svn: 280400
Yaxun Liu [Thu, 1 Sep 2016 18:46:49 +0000 (18:46 +0000)]
AMDGPU: Add runtime metadata for pointee alignment of argument.
Add runtime metdata for pointee alignment of pointer type kernel argument. The key is KeyArgPointeeAlign and the value is a 32 bit unsigned integer.
Differential Revision: https://reviews.llvm.org/D24145
llvm-svn: 280399
Filipe Cabecinhas [Thu, 1 Sep 2016 18:40:12 +0000 (18:40 +0000)]
Revert "Copy over most of the scariness_score test to the general tests"
This reverts commit r280361 until we have the proper change in place.
llvm-svn: 280398
Jason Henline [Thu, 1 Sep 2016 18:35:37 +0000 (18:35 +0000)]
[SE] Make Stream movable
Summary:
The example code makes it clear that this is a much better design
decision.
Reviewers: jlebar
Subscribers: jprice, parallel_libs-commits
Differential Revision: https://reviews.llvm.org/D24142
llvm-svn: 280397
Davide Italiano [Thu, 1 Sep 2016 18:34:47 +0000 (18:34 +0000)]
[lib/LTO] Simplify a bit. NFCI.
llvm-svn: 280396
Enrico Granata [Thu, 1 Sep 2016 18:31:40 +0000 (18:31 +0000)]
When updating a ValueObjectRegister, set the flag to remind yourself of whether it changed
llvm-svn: 280395
Chris Bieneman [Thu, 1 Sep 2016 18:28:49 +0000 (18:28 +0000)]
[CMake] Properly connecting Compiler-RT check and test-depends
This correctly connects compiler-rt-test-depends to test-depends and
check-compiler-rt to check-all.
Based on LLVM r280392, and Compiler-RT r280393.
llvm-svn: 280394
Chris Bieneman [Thu, 1 Sep 2016 18:26:51 +0000 (18:26 +0000)]
[CMake] Adding compiler-rt-test-depends target
This exposes a target for building the compiler-rt test dependencies and matches llvm-test-depends.
llvm-svn: 280393
Chris Bieneman [Thu, 1 Sep 2016 18:26:01 +0000 (18:26 +0000)]
[CMake] Connecting check-all and test-depends targets correctly
My previous attempt at this connected the sub-project check targets to the test-depends target instead of to the check-all target. That resulted in the tests running multiple times on bots that built "test-depends" and "check-all" in separate build invocations.
llvm-svn: 280392
Michael Kuperstein [Thu, 1 Sep 2016 18:24:42 +0000 (18:24 +0000)]
Rename some variables to have meaningful names. NFC.
llvm-svn: 280391
Matthew Simpson [Thu, 1 Sep 2016 18:14:27 +0000 (18:14 +0000)]
[LV] Move VectorParts allocation and mapping into PHI widening (NFC)
This patch moves the allocation of VectorParts for PHI nodes into the actual
PHI widening code. Previously, we allocated these VectorParts in
vectorizeBlockInLoop, and passed them by reference to widenPHIInstruction. Upon
returning, we would then map the VectorParts in VectorLoopValueMap. This
behavior is problematic for the cases where we only want to generate a scalar
version of a PHI node. For example, if in the future we only generate a scalar
version of an induction variable, we would end up inserting an empty vector
entry into the map once we return to vectorizeBlockInLoop. We now no longer
need to pass VectorParts to the various PHI widening functions, and we can keep
VectorParts allocation as close as possible to the point at which they are
actually mapped in VectorLoopValueMap.
llvm-svn: 280390
Enrico Granata [Thu, 1 Sep 2016 18:09:01 +0000 (18:09 +0000)]
Change the formula for tagged NSIndexPath data formatting
Fixes rdar://
25192935
llvm-svn: 280389
Zachary Turner [Thu, 1 Sep 2016 18:08:19 +0000 (18:08 +0000)]
[codeview] Properly propagate the TypeLeafKind through the pipeline.
llvm-svn: 280388
Reid Kleckner [Thu, 1 Sep 2016 18:08:05 +0000 (18:08 +0000)]
Fix dll_host.cc test after adding div and gep trace coverage points
llvm-svn: 280387
Michael Kuperstein [Thu, 1 Sep 2016 17:59:24 +0000 (17:59 +0000)]
[DAGCombine] Don't fold a trunc if it feeds an anyext
Legalization tends to create anyext(trunc) patterns. This should always be
combined - into either a single trunc, a single ext, or nothing if the
types match exactly. But if we happen to combine the trunc first, we may pull
the trunc away from the anyext or make it implicit (e.g. the truncate(extract)
-> extract(bitcast) fold).
To prevent this, we can avoid doing the fold, similarly to how we already handle
fpround(fpextend).
Differential Revision: https://reviews.llvm.org/D23893
llvm-svn: 280386
Changpeng Fang [Thu, 1 Sep 2016 17:54:54 +0000 (17:54 +0000)]
AMDGPU/SI: MIMG TD Refactoring.
Summary:
Created a new td file MIMGInstructions.td which contains all definitions
of MIMG related instructions.
Reviewed by:
kzhuravl, vpykhtin
Differential Revision:
http://reviews.llvm.org/D24106
llvm-svn: 280385
Jason Henline [Thu, 1 Sep 2016 17:47:17 +0000 (17:47 +0000)]
[SE] Docs use JAVADOC_AUTOBRIEF
That way we don't have to explicitly annotate each brief description as
\brief.
llvm-svn: 280384
Reid Kleckner [Thu, 1 Sep 2016 17:19:44 +0000 (17:19 +0000)]
[lit] Use multiprocessing by default on Windows
Apparently nobody evaluated multiprocessing on Windows since Daniel
enabled multiprocessing on Unix in r193279. It works so far as I can
tell.
Today this is worth about an 8x speedup (631.29s to 73.25s) on my 24
core Windows machine. Hopefully this will improve Windows buildbot cycle
time, where currently it takes more time to run check-all than it does
to self-host with assertions enabled:
http://lab.llvm.org:8011/builders/clang-x86-windows-msvc2015/builds/20
build stage 2 ninja all ( 28 mins, 22 secs )
ninja check 2 stage 2 ( 37 mins, 38 secs )
llvm-svn: 280382
Chris Bieneman [Thu, 1 Sep 2016 16:43:39 +0000 (16:43 +0000)]
[CMake] Revive LLVM_*_DIRS variables
This is a partial revert of r280013. Brad King pointed out these variable names are matching CMake conventions, so we should preserve them.
I've also added a direct mapping of the LLVM_*_DIR variables which we need to make projects support building in and out of tree.
llvm-svn: 280380
Geoff Berry [Thu, 1 Sep 2016 15:07:46 +0000 (15:07 +0000)]
[EarlyCSE] Change C API pass interface for EarlyCSE w/ MemorySSA
Previous change broke the C API for creating an EarlyCSE pass w/
MemorySSA by adding a bool parameter to control whether MemorySSA was
used or not. This broke the OCaml bindings. Instead, change the old C
API entry point back and add a new one to request an EarlyCSE pass with
MemorySSA.
llvm-svn: 280379
Saleem Abdulrasool [Thu, 1 Sep 2016 15:05:36 +0000 (15:05 +0000)]
build: profile is not a sanitizer
There was an invalid entry in the sanitizer list, remove it. This has no effect
on the building, just removes the definition of a cached variable.
llvm-svn: 280378
Simon Dardis [Thu, 1 Sep 2016 15:03:13 +0000 (15:03 +0000)]
[mips] Include missed file from previous commit
llvm-svn: 280377
Simon Pilgrim [Thu, 1 Sep 2016 14:59:20 +0000 (14:59 +0000)]
[X86][SSE] Dropped (V)CVTPD2PS intrinsic patterns now that its bound to X86vfpround
It now uses X86vfpround patterns directly instead.
Followup to D23797
llvm-svn: 280376
Rafael Espindola [Thu, 1 Sep 2016 14:59:11 +0000 (14:59 +0000)]
Add missing REQUIRES.
llvm-svn: 280375
Simon Dardis [Thu, 1 Sep 2016 14:53:53 +0000 (14:53 +0000)]
[mips] interAptiv based generic schedule model
This scheduler describes a processor which covers all MIPS ISAs based
around the interAptiv and P5600 timings.
Reviewers: vkalintiris, dsanders
Differential Revision: https://reviews.llvm.org/D23551
llvm-svn: 280374
Renato Golin [Thu, 1 Sep 2016 14:44:02 +0000 (14:44 +0000)]
Try to fix another profile test
Another CFG optimisation patch (280364) has broken bad profile tests, and this
is a similar attempt to fix the test without changing the semantics.
llvm-svn: 280373
Filipe Cabecinhas [Thu, 1 Sep 2016 14:41:17 +0000 (14:41 +0000)]
Remove extra scariness.Clear() call.
llvm-svn: 280372
Andrey Bokhanko [Thu, 1 Sep 2016 14:39:54 +0000 (14:39 +0000)]
[CMake] Fix LLVM_ENABLE_EH and LLVM_ENABLE_RTTI on MSVC
Patch by Johannes Sebastian Mueller-Roemer.
Differential Revision: https://reviews.llvm.org/D23645
llvm-svn: 280371
Sanjay Patel [Thu, 1 Sep 2016 14:20:43 +0000 (14:20 +0000)]
[InstCombine] remove fold of an icmp pattern that should never happen
While removing a scalar shackle from an icmp fold, I noticed that I couldn't find any tests to trigger
this code path.
The 'and' shrinking transform should be handled by InstCombiner::foldCastedBitwiseLogic()
or eliminated with InstSimplify. The icmp narrowing is part of InstCombiner::foldICmpWithCastAndCast().
Differential Revision: https://reviews.llvm.org/D24031
llvm-svn: 280370
Krzysztof Parzyszek [Thu, 1 Sep 2016 13:59:35 +0000 (13:59 +0000)]
[Hexagon] Deal with undefs when extending live intervals
Reapply r280275, since MSVC accepts r280358.
llvm-svn: 280369
Elena Demikhovsky [Thu, 1 Sep 2016 13:58:53 +0000 (13:58 +0000)]
Optimized FMA intrinsic + FNEG , like
-(a*b+c)
and FNEG + FMA, like
a*b-c or (-a)*b+c.
The bug description is here : https://llvm.org/bugs/show_bug.cgi?id=28892
Differential revision: https://reviews.llvm.org/D23313
llvm-svn: 280368
Aleksei Sidorin [Thu, 1 Sep 2016 13:55:38 +0000 (13:55 +0000)]
[analyzer] ExprEngine: remove second call to PreStmt<CastExpr>
This patch also introduces AnalysisOrderChecker which is intended for testing
of callback call correctness.
Differential Revision: https://reviews.llvm.org/D23804
llvm-svn: 280367
Rafael Espindola [Thu, 1 Sep 2016 13:52:52 +0000 (13:52 +0000)]
Sign extend a value before passing it to the Target.
This is what InputSectionBase<ELFT>::relocate does and we need to be
consistent. The other option would be to be more explicit about which
relocations are signed and which are not, and sign extend only when
appropriated. That would require extending the target interface.
llvm-svn: 280366
Renato Golin [Thu, 1 Sep 2016 13:45:47 +0000 (13:45 +0000)]
Fix profile test assuming dumb compiler
Commit r280364 has introduced some call-graph optmisations making a profiler
test "fail" due to not expecting the compiler to be "smart", and fold constants
across functions. This commit works around the issue, leaving the origial
semantics intact.
llvm-svn: 280365
James Molloy [Thu, 1 Sep 2016 12:58:13 +0000 (12:58 +0000)]
[SimplifyCFG] Handle tail-sinking of more than 2 incoming branches
This was a real restriction in the original version of SinkIfThenCodeToEnd. Now it's been rewritten, the restriction can be lifted.
As part of this, we handle a very common and useful case where one of the incoming branches is actually conditional. Consider:
if (a)
x(1);
else if (b)
x(2);
This produces the following CFG:
[if]
/ \
[x(1)] [if]
| | \
| | \
| [x(2)] |
\ | /
[ end ]
[end] has two unconditional predecessor arcs and one conditional. The conditional refers to the implicit empty 'else' arc. This same pattern can also be caused by an empty default block in a switch.
We can't sink the call to x() down to end because no call to x() happens on the third incoming arc (assume that x() has sideeffects for the sake of argument; if something is safe to speculate we could indeed sink nevertheless but this cannot happen in the general case and causes many extra selects).
We are now able to detect this case and split off the unconditional arcs to a common successor:
[if]
/ \
[x(1)] [if]
| | \
| | \
| [x(2)] |
\ / |
[sink.split] |
\ /
[ end ]
Now we can sink the call to x() into %sink.split. This can cause significant code simplification in many testcases.
llvm-svn: 280364