platform/upstream/llvm.git
6 years agoPreserve debug info when DAG-combinging (zext (truncate x)) -> (and x, mask).
Adrian Prantl [Thu, 9 Nov 2017 19:50:20 +0000 (19:50 +0000)]
Preserve debug info when DAG-combinging (zext (truncate x)) -> (and x, mask).

rdar://problem/27139077

llvm-svn: 317825

6 years ago[Support] Make llvm::Error and Expected faster.
Zachary Turner [Thu, 9 Nov 2017 19:31:52 +0000 (19:31 +0000)]
[Support] Make llvm::Error and Expected faster.

Whenever LLVM_ENABLE_ABI_BREAKING_CHECKS is enabled, which
is usually the case for example when asserts are enabled,
Error's destructor does some additional checking to make sure
that that it does not represent an error condition and that it
was checked.

However, this is -- by definition -- not the likely codepath.
Some profiling shows that at least with some compilers, simply
calling assertIsChecked -- in a release build with full
optimizations -- can account for up to 15% of the entire
runtime of the program, even though this function should almost
literally be a no-op.

The problem is that the assertIsChecked function can be considered
too big to inline depending on the compiler's inliner.  Since it's
unlikely to ever need to failure path though, we can move it out
of line and force it to not be inlined, so that the fast path
can be inlined.

In my test (using lld to link clang with CMAKE_BUILD_TYPE=Release
and LLVM_ENABLE_ASSERTIONS=ON), this reduces link time from 27
seconds to 23.5 seconds, which is a solid 15% gain.

llvm-svn: 317824

6 years agoRevert new ThinLTO ASAN test until lit support added
Teresa Johnson [Thu, 9 Nov 2017 19:26:54 +0000 (19:26 +0000)]
Revert new ThinLTO ASAN test until lit support added

This reverts commit r317723 and r317728. Will be re-added when
support for LTO/ThinLTO added to test/asan/lit.cfg.

llvm-svn: 317823

6 years ago[scudo] Make getNumberOfCPUs Fuchsia compliant
Kostya Kortchinsky [Thu, 9 Nov 2017 19:18:55 +0000 (19:18 +0000)]
[scudo] Make getNumberOfCPUs Fuchsia compliant

Summary: This change allows Fuchsia to boot properly using the Scudo allocator.

Reviewers: cryptoad, alekseyshl, krytarowski

Reviewed By: cryptoad, krytarowski

Subscribers: rnk, krytarowski, kubamracek, llvm-commits

Differential Revision: https://reviews.llvm.org/D39490

llvm-svn: 317822

6 years ago[SLP] Fix PR23510: Try to find best possible vectorizable stores.
Alexey Bataev [Thu, 9 Nov 2017 19:07:16 +0000 (19:07 +0000)]
[SLP] Fix PR23510: Try to find best possible vectorizable stores.

Summary:
The analysis of the store sequence goes in straight order - from the
first store to the last. Bu the best opportunity for vectorization will
happen if we're going to use reverse order - from last store to the
first. It may be best because usually users have some initialization
part + further processing and this first initialization may confuse
SLP vectorizer.

Reviewers: RKSimon, hfinkel, mkuper, spatel

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D39606

llvm-svn: 317821

6 years ago[analyzer] assume bitwise arithmetic axioms
George Karpenkov [Thu, 9 Nov 2017 19:06:22 +0000 (19:06 +0000)]
[analyzer] assume bitwise arithmetic axioms

Patches the solver to assume that bitwise OR of an unsigned value with a
constant always produces a value larger-or-equal than the constant, and
bitwise AND with a constant always produces a value less-or-equal than
the constant.

This patch is especially useful in the context of using bitwise
arithmetic for error code encoding: the analyzer would be able to state
that the error code produced using a bitwise OR is non-zero.

Differential Revision: https://reviews.llvm.org/D39707

llvm-svn: 317820

6 years ago[Reassociate] auto-generate test checks; NFC
Sanjay Patel [Thu, 9 Nov 2017 18:26:49 +0000 (18:26 +0000)]
[Reassociate] auto-generate test checks; NFC

llvm-svn: 317819

6 years ago[Reassociate] don't name values "tmp"; NFCI
Sanjay Patel [Thu, 9 Nov 2017 18:14:24 +0000 (18:14 +0000)]
[Reassociate] don't name values "tmp"; NFCI

The toxic stew of created values named 'tmp' and tests that already have
values named 'tmp' and CHECK lines looking for values named 'tmp' causes
bad things to happen in our test line auto-generation scripts because it
wants to use 'TMP' as a prefix for unnamed values. Use less 'tmp' to
avoid that.

llvm-svn: 317818

6 years ago[GlobalMerge] Stable sort GlobalSets to fix non-deterministic sort order
Mandeep Singh Grang [Thu, 9 Nov 2017 18:05:17 +0000 (18:05 +0000)]
[GlobalMerge] Stable sort GlobalSets to fix non-deterministic sort order

Summary: This fixes failure in CodeGen/AArch64/global-merge-group-by-use.ll uncovered by D39245.

Reviewers: ab, asl

Reviewed By: ab

Subscribers: aemerson, llvm-commits, kristof.beyls

Differential Revision: https://reviews.llvm.org/D39635

llvm-svn: 317817

6 years agoAdd _LIBCPP_INLINE_VISIBILITY to __compressed_pair_elem members
Alex Lorenz [Thu, 9 Nov 2017 17:54:49 +0000 (17:54 +0000)]
Add _LIBCPP_INLINE_VISIBILITY to __compressed_pair_elem members

The commit r300140 changed the implementation of compressed_pair, but didn't add
_LIBCPP_INLINE_VISIBILITY to the constructors and get members of the
compressed_pair_elem class. This patch adds the visibility annotation.

I didn't find a way to test this change with libc++ regression tests.

rdar://35352579

Differential Revision: https://reviews.llvm.org/D39751

llvm-svn: 317816

6 years agorevert r317812 [BasicAA] fix build break by converting the previously introduced...
Nuno Lopes [Thu, 9 Nov 2017 17:35:36 +0000 (17:35 +0000)]
revert r317812 [BasicAA] fix build break by converting the previously introduced assert into an if stmt

The code has a bug, but some tests regress.
I'll discuss this further on the mailing list.

llvm-svn: 317815

6 years ago[Builtins] Do not use tailcall for Thumb1
Weiming Zhao [Thu, 9 Nov 2017 17:32:57 +0000 (17:32 +0000)]
[Builtins] Do not use tailcall for Thumb1

Summary:
The `b` instruction in Thumb1 has limited range, which may cause link-time errors if the jump target is far away.
This patch guards the tailcalls for non-Thumb1

Reviewers: peter.smith, compnerd, rengolin, eli.friedman

Reviewed By: rengolin

Subscribers: joerg, dalias, javed.absar, llvm-commits

Differential Revision: https://reviews.llvm.org/D39700

llvm-svn: 317814

6 years ago[OPENMP] Codegen for `#pragma omp target parallel for simd`.
Alexey Bataev [Thu, 9 Nov 2017 17:32:15 +0000 (17:32 +0000)]
[OPENMP] Codegen for `#pragma omp target parallel for simd`.

Added codegen for `#pragma omp target parallel for simd` and clauses.

llvm-svn: 317813

6 years ago[BasicAA] fix build break by converting the previously introduced assert into an...
Nuno Lopes [Thu, 9 Nov 2017 17:06:42 +0000 (17:06 +0000)]
[BasicAA] fix build break by converting the previously introduced assert into an if stmt
Apparently V1Size == -1 doest imply V2Size == -1, which is a bit surprising to me.

llvm-svn: 317812

6 years ago[OPENMP] Treat '#pragma omp target parallel for simd' as simd directive.
Alexey Bataev [Thu, 9 Nov 2017 17:01:35 +0000 (17:01 +0000)]
[OPENMP] Treat '#pragma omp target parallel for simd' as simd directive.

`#pragma omp target parallel for simd` mistakenly was not treated as a
simd directive, fixed this problem.

llvm-svn: 317811

6 years agorevert r317809 - [Reassociate] regenerate test checks; NFC
Sanjay Patel [Thu, 9 Nov 2017 16:46:04 +0000 (16:46 +0000)]
revert r317809 - [Reassociate] regenerate test checks; NFC

The reassociate pass generates named values such as "%tmp2" which trips up the script's regex's
because the script uses a 'TMP' prefix for unnamed values (%2).

llvm-svn: 317810

6 years ago[Reassociate] regenerate test checks; NFC
Sanjay Patel [Thu, 9 Nov 2017 16:35:30 +0000 (16:35 +0000)]
[Reassociate] regenerate test checks; NFC

llvm-svn: 317809

6 years agoUpdate formatting to reflect change in clang-format. NFC.
Michael Kruse [Thu, 9 Nov 2017 16:33:29 +0000 (16:33 +0000)]
Update formatting to reflect change in clang-format. NFC.

clang-format has changed its algorithm
for sorting includes in r317794.

llvm-svn: 317808

6 years ago[SystemZ] Add support for the "o" inline asm constraint
Ulrich Weigand [Thu, 9 Nov 2017 16:31:57 +0000 (16:31 +0000)]
[SystemZ] Add support for the "o" inline asm constraint

We don't really need any special handling of "offsettable"
memory addresses, but since some existing code uses inline
asm statements with the "o" constraint, add support for this
constraint for compatibility purposes.

llvm-svn: 317807

6 years ago[Reassociate] regenerate test checks; NFC
Sanjay Patel [Thu, 9 Nov 2017 16:30:19 +0000 (16:30 +0000)]
[Reassociate] regenerate test checks; NFC

llvm-svn: 317806

6 years ago[Reassociate] add check lines; NFC
Sanjay Patel [Thu, 9 Nov 2017 16:25:35 +0000 (16:25 +0000)]
[Reassociate] add check lines; NFC

llvm-svn: 317805

6 years ago[Reassociate] add tests with 'reassoc' FMF and regenerate checks; NFC
Sanjay Patel [Thu, 9 Nov 2017 16:23:32 +0000 (16:23 +0000)]
[Reassociate] add tests with 'reassoc' FMF and regenerate checks; NFC

llvm-svn: 317804

6 years ago[BasicAA] add assertion for corner case in aliasGEP()
Nuno Lopes [Thu, 9 Nov 2017 16:16:46 +0000 (16:16 +0000)]
[BasicAA] add assertion for corner case in aliasGEP()

llvm-svn: 317803

6 years ago[PowerPC][msan] Update msan to handle changed memory layouts in newer kernels
Bill Seurer [Thu, 9 Nov 2017 16:14:57 +0000 (16:14 +0000)]
[PowerPC][msan] Update msan to handle changed memory layouts in newer kernels

In more recent Linux kernels (including those with 47 bit VMAs) the layout of
virtual memory for powerpc64 changed causing the memory sanitizer to not
work properly.  This patch adjusts the memory ranges in the tables for the
memory sanitizer to work on the newer kernels while continuing to work on the
older ones as well.

Tested on several 4.x and 3.x kernel releases.

llvm-svn: 317802

6 years ago[mips] Correct microMIP's jump and add unconditional branch pseudo
Simon Dardis [Thu, 9 Nov 2017 16:02:18 +0000 (16:02 +0000)]
[mips] Correct microMIP's jump and add unconditional branch pseudo

Correct the definition of 'j' as being unavailable for microMIPS32R6 and
provide the 'b' assembly idiom for codegen purposes for microMIPS32r3.

Provide the necessary 'br' pattern for microMIPS32R6 as it now longer
incorrectly uses the 'j' instruction.

Reviewers: atanasyan

Differential Revision: https://reviews.llvm.org/D39741

llvm-svn: 317801

6 years ago[VirtualFileSystem] InMemoryFileSystem::addFile(): Type and Perms
Ben Hamilton [Thu, 9 Nov 2017 16:01:16 +0000 (16:01 +0000)]
[VirtualFileSystem] InMemoryFileSystem::addFile(): Type and Perms

Summary:
This implements a FIXME in InMemoryFileSystem::addFile(), allowing
clients to specify User, Group, Type, and/or Perms when creating a
file in an in-memory filesystem.

New tests included. Ran tests with:

% ninja BasicTests && ./tools/clang/unittests/Basic/BasicTests

Fixes PR#35172 (https://bugs.llvm.org/show_bug.cgi?id=35172)

Reviewers: bkramer, hokein

Reviewed By: bkramer, hokein

Subscribers: alexfh

Differential Revision: https://reviews.llvm.org/D39572

llvm-svn: 317800

6 years ago[clang-format] Keep Sphinx happy after r317794
Krasimir Georgiev [Thu, 9 Nov 2017 15:54:59 +0000 (15:54 +0000)]
[clang-format] Keep Sphinx happy after r317794

llvm-svn: 317799

6 years agoAdd const to some variables to avoid const_casts
Jonas Hahnfeld [Thu, 9 Nov 2017 15:52:29 +0000 (15:52 +0000)]
Add const to some variables to avoid const_casts

In these places the const attribute seems correct and doesn't
need any other change, so let's do it.

Differential Revision: https://reviews.llvm.org/D39756

llvm-svn: 317798

6 years agoRemove const from variables with dynamic memory
Jonas Hahnfeld [Thu, 9 Nov 2017 15:52:25 +0000 (15:52 +0000)]
Remove const from variables with dynamic memory

Allocated memory is typically not 'const' if it needs to be freed.
This patch removes around 50 wrong const attributes, modifies the
corresponding functions and finally gets rid of some const_casts.
These have especially been strange for __kmp_str_fname_free() that
added a 'const' to call __kmp_str_free() which removed it again.

Two minor cleanups that I performed in this process:
 * __kmp_tool_libraries now lives in kmp_settings.cpp as it is
   used nowhere else.
 * __kmp_msg_empty was removed as it was never used and Clang
   now complained that it was assigned a string literal that
   is 'const char *'.

Differential Revision: https://reviews.llvm.org/D39755

llvm-svn: 317797

6 years ago[RISCV] Re-generate test/CodeGen/RISCV/alu32.ll using update_llc_test_checks.py
Alex Bradbury [Thu, 9 Nov 2017 15:45:42 +0000 (15:45 +0000)]
[RISCV] Re-generate test/CodeGen/RISCV/alu32.ll using update_llc_test_checks.py

No real change, but makes it marginally easier to merge the remainder of the
out-of-tree patches.

llvm-svn: 317796

6 years agollgs-tests: Replace the "log+return false" pattern with llvm::Error
Pavel Labath [Thu, 9 Nov 2017 15:45:09 +0000 (15:45 +0000)]
llgs-tests: Replace the "log+return false" pattern with llvm::Error

Summary:
These tests used to log the error message and return plain bool mainly
because at the time they we written, we did not have a nice way to
assert on llvm::Error values. That is no longer true, so replace this
pattern with a more idiomatic approach.

As a part of this patch, I also move the formatting of
GDBRemoteCommunication::PacketResult values out of the test code, as
that can be useful elsewhere.

Reviewers: zturner, eugene

Subscribers: mgorny, lldb-commits

Differential Revision: https://reviews.llvm.org/D39790

llvm-svn: 317795

6 years ago[clang-format] Sort using declarations by splitting on '::'
Krasimir Georgiev [Thu, 9 Nov 2017 15:41:23 +0000 (15:41 +0000)]
[clang-format] Sort using declarations by splitting on '::'

Summary: This patch improves using declarations sorting.

Reviewers: bkramer

Reviewed By: bkramer

Subscribers: cfe-commits, klimek

Differential Revision: https://reviews.llvm.org/D39786

llvm-svn: 317794

6 years ago[clang-format] Apply a clang-tidy suggestion, NFC
Krasimir Georgiev [Thu, 9 Nov 2017 15:12:17 +0000 (15:12 +0000)]
[clang-format] Apply a clang-tidy suggestion, NFC

llvm-svn: 317793

6 years agoAdd a unit test for ClangASTContext template arguments handling
Pavel Labath [Thu, 9 Nov 2017 15:06:31 +0000 (15:06 +0000)]
Add a unit test for ClangASTContext template arguments handling

I am planning to make changes to this piece of code, so I wrote this
test to add more coverage to it first.

llvm-svn: 317792

6 years ago[RISCV] MC layer support for the standard RV32A instruction set extension
Alex Bradbury [Thu, 9 Nov 2017 15:00:03 +0000 (15:00 +0000)]
[RISCV] MC layer support for the standard RV32A instruction set extension

llvm-svn: 317791

6 years agoFix 'not all control paths return a value' warning on MSVC builds
Simon Pilgrim [Thu, 9 Nov 2017 14:56:17 +0000 (14:56 +0000)]
Fix 'not all control paths return a value' warning on MSVC builds

llvm-svn: 317790

6 years agoReapply: Allow yaml2obj to order implicit sections for ELF
Dave Lee [Thu, 9 Nov 2017 14:53:43 +0000 (14:53 +0000)]
Reapply: Allow yaml2obj to order implicit sections for ELF

Summary:
This change allows yaml input to control the order of implicitly added sections
(`.symtab`, `.strtab`, `.shstrtab`). The order is controlled by adding a
placeholder section of the given name to the Sections field.

This change is to support changes in D39582, where it is desirable to control
the location of the `.dynsym` section.

This reapplied version fixes:
  1. use of a function call within an assert
  2. failing lld test which has an unnamed section
  3. incorrect section count when given an unnamed section

Additionally, one more test to cover the unnamed section failure.

Reviewers: compnerd, jakehehrlich

Reviewed By: jakehehrlich

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D39749

llvm-svn: 317789

6 years ago[RISCV] MC layer support for the standard RV32M instruction set extension
Alex Bradbury [Thu, 9 Nov 2017 14:46:30 +0000 (14:46 +0000)]
[RISCV] MC layer support for the standard RV32M instruction set extension

llvm-svn: 317788

6 years ago[OMPT] Fix test cancel_parallel.c
Jonas Hahnfeld [Thu, 9 Nov 2017 14:26:14 +0000 (14:26 +0000)]
[OMPT] Fix test cancel_parallel.c

If a parallel region is cancelled, execution resumes at the end
of the structured block. That is why this test cannot use the
"normal" macros that print right after inserting the label.
Instead it previously printed the addresses before the pragma
and swapped the checks compared to the other tests.

However, this does not work because FileChecks '*' is greedy
so that RETURN_ADDRESS always matched the second address. This
makes the test fail when an "overflow" occurrs and the first
address matches the value of codeptr_ra.

I discovered this on my MacBook but I'm unable to reproduce the
failure with the current version. Nevertheless we should fix this
problem to avoid that this test fails later after an unrelated change.

Differential Revision: https://reviews.llvm.org/D39708

llvm-svn: 317787

6 years ago[OMPT] Add support for testing return addresses on POWER
Jonas Hahnfeld [Thu, 9 Nov 2017 14:26:12 +0000 (14:26 +0000)]
[OMPT] Add support for testing return addresses on POWER

Return addresses are determined based on the address of a label
that is inserted directly after a pragma / API call. In some cases
the tests can assume a known number of instructions between the
addresses. However, the instructions and their encoded lengths
depend on the target that the test is compiled on.

Firstly, this patch refactors the macro print_current_address() to
allow such target dependent modifications and adds information for
the observed instructions on POWER. Secondly, it adapts the related
macro print_fuzzy_address() to reuse much of "hacky" code and fixes
the used formatting strings in the printf() call. Finally, it also
adds documentation about how these macros are intended to work.

Differential Revision: https://reviews.llvm.org/D39699

llvm-svn: 317786

6 years agoSched model improving on btver2: JFPU01 resource, vtestp* for xmm.
Andrew V. Tischenko [Thu, 9 Nov 2017 14:19:59 +0000 (14:19 +0000)]
Sched model improving on btver2: JFPU01 resource, vtestp* for xmm.
Differential Revision: https://reviews.llvm.org/D39802

llvm-svn: 317785

6 years ago[clang-format] Fix a clang-tidy finding, NFC
Krasimir Georgiev [Thu, 9 Nov 2017 13:22:03 +0000 (13:22 +0000)]
[clang-format] Fix a clang-tidy finding, NFC

llvm-svn: 317784

6 years ago[clang-format] Fix argument name comment, NFC
Krasimir Georgiev [Thu, 9 Nov 2017 13:19:14 +0000 (13:19 +0000)]
[clang-format] Fix argument name comment, NFC

llvm-svn: 317783

6 years agoAdd -print-schedule scheduling comments to inline asm.
Andrew V. Tischenko [Thu, 9 Nov 2017 12:45:40 +0000 (12:45 +0000)]
Add -print-schedule scheduling comments to inline asm.
Differential Revision: https://reviews.llvm.org/D39728

llvm-svn: 317782

6 years ago[MIPS] Fix calculation of the R_MICROMIPS_LO16 / HI16 relocations
Simon Atanasyan [Thu, 9 Nov 2017 12:10:14 +0000 (12:10 +0000)]
[MIPS] Fix calculation of the R_MICROMIPS_LO16 / HI16 relocations

llvm-svn: 317781

6 years ago[clangd] Add rename support.
Haojian Wu [Thu, 9 Nov 2017 11:30:04 +0000 (11:30 +0000)]
[clangd] Add rename support.

Summary:
Make clangd handle "textDocument/rename" request. The rename
functionality comes from the "local-rename" sub-tool of clang-refactor.

Currently clangd only supports local rename (only symbol occurrences in
the main file will be renamed).

Reviewers: sammccall, ilya-biryukov

Reviewed By: sammccall

Subscribers: cfe-commits, ioeric, arphaman, mgorny

Differential Revision: https://reviews.llvm.org/D39676

llvm-svn: 317780

6 years agoSimplify NativeProcessProtocol::GetArchitecture/GetByteOrder
Pavel Labath [Thu, 9 Nov 2017 10:43:16 +0000 (10:43 +0000)]
Simplify NativeProcessProtocol::GetArchitecture/GetByteOrder

Summary:
These functions used to return bool to signify whether they were able to
retrieve the data. This is redundant because the ArchSpec and ByteOrder
already have their own "invalid" states, *and* because both of the
current implementations (linux, netbsd) can always provide a valid
result.

This allows us to simplify bits of the code handling these values.

Reviewers: eugene, krytarowski

Subscribers: javed.absar, lldb-commits

Differential Revision: https://reviews.llvm.org/D39733

llvm-svn: 317779

6 years ago[MIPS] Setup less-significant bit in a symbol value in microMIPS thunks
Simon Atanasyan [Thu, 9 Nov 2017 10:42:22 +0000 (10:42 +0000)]
[MIPS] Setup less-significant bit in a symbol value in microMIPS thunks

The less-significant bit signals about microMIPS code for jump/branch instructions.

llvm-svn: 317778

6 years ago[Tooling] Use FixedCompilationDatabase when `compile_flags.txt` is found.
Sam McCall [Thu, 9 Nov 2017 10:37:39 +0000 (10:37 +0000)]
[Tooling] Use FixedCompilationDatabase when `compile_flags.txt` is found.

Summary:
This is an alternative to JSONCompilationDatabase for simple projects that
don't use a build system such as CMake.
(You can also drop one in ~, to make your tools use e.g. C++11 by default)

There's no facility for varying flags per-source-file or per-machine.
Possibly this could be accommodated backwards-compatibly using cpp, but even if
not the simplicity seems worthwhile for the cases that are addressed.

Tested with clangd, works great! (requires clangd restart)

Reviewers: klimek

Subscribers: ilya-biryukov, cfe-commits

Differential Revision: https://reviews.llvm.org/D39799

llvm-svn: 317777

6 years agoFix a bug with the use of __builtin_bzero in a conditional expression.
John McCall [Thu, 9 Nov 2017 09:32:32 +0000 (09:32 +0000)]
Fix a bug with the use of __builtin_bzero in a conditional expression.

Patch by Bharathi Seshadri!

llvm-svn: 317776

6 years ago[X86] Give priority to EVEX FMA instructions over FMA4 instructions.
Craig Topper [Thu, 9 Nov 2017 08:26:26 +0000 (08:26 +0000)]
[X86] Give priority to EVEX FMA instructions over FMA4 instructions.

No existing processor has both so it doesn't really matter what we do here. But we were previously just relying on pattern order which gave FMA4 priority.

llvm-svn: 317775

6 years ago[sanitizers] Rename GetStackTraceWithPcBpAndContext
Vitaly Buka [Thu, 9 Nov 2017 07:53:06 +0000 (07:53 +0000)]
[sanitizers] Rename GetStackTraceWithPcBpAndContext

Name does not need to enumerate arguments.

llvm-svn: 317774

6 years ago[msan] Add context argument into GetStackTrace
Vitaly Buka [Thu, 9 Nov 2017 07:48:53 +0000 (07:48 +0000)]
[msan] Add context argument into GetStackTrace

llvm-svn: 317773

6 years ago[lsan] Add "static" to internal function
Vitaly Buka [Thu, 9 Nov 2017 07:46:30 +0000 (07:46 +0000)]
[lsan] Add "static" to internal function

llvm-svn: 317772

6 years agoFix "default label in switch which covers all enumeration values" warning
Vitaly Buka [Thu, 9 Nov 2017 07:46:13 +0000 (07:46 +0000)]
Fix "default label in switch which covers all enumeration values" warning

llvm-svn: 317771

6 years ago[SectionMemoryManager] Abstract out mmap, munmap, mprotect even more ; NFC
Sanjoy Das [Thu, 9 Nov 2017 06:31:33 +0000 (06:31 +0000)]
[SectionMemoryManager] Abstract out mmap, munmap, mprotect even more ; NFC

Summary:
This will let ORC JIT clients plug in custom logic for the mmap, munmap and
mprotect paths.

Reviewers: loladiro, dblaikie

Subscribers: mcrosier, llvm-commits

Differential Revision: https://reviews.llvm.org/D39300

llvm-svn: 317770

6 years ago[X86] Make X86ISD::FMADDS3 isel patterns commutable.
Craig Topper [Thu, 9 Nov 2017 06:17:05 +0000 (06:17 +0000)]
[X86] Make X86ISD::FMADDS3 isel patterns commutable.

This was missed when FMADDS3 was split from X86ISD::FMADDS3_RND.

llvm-svn: 317769

6 years ago[GVN PRE] Patch the source for Phi node in PRE
Serguei Katkov [Thu, 9 Nov 2017 06:02:18 +0000 (06:02 +0000)]
[GVN PRE] Patch the source for Phi node in PRE

We must patch all existing incoming values of Phi node,
otherwise it is possible that we can see poison
where program does not expect to see it.

This is the similar what GVN does.

The added test test/Transforms/GVN/PRE/pre-jt-add.ll shows an
example of wrong optimization done by jump threading due to
GVN PRE did not patch existing incoming value.

Reviewers: mkazantsev, wmi, dberlin, davide
Reviewed By: dberlin
Subscribers: efriedma, llvm-commits
Differential Revision: https://reviews.llvm.org/D39637

llvm-svn: 317768

6 years ago[libFuzzer] allow merge to resume after being preempted
Kostya Serebryany [Thu, 9 Nov 2017 05:49:28 +0000 (05:49 +0000)]
[libFuzzer] allow merge to resume after being preempted

llvm-svn: 317767

6 years ago[X86] Rename the VEX scalar fma builtins to end with a '3' to match gcc
Craig Topper [Thu, 9 Nov 2017 04:10:46 +0000 (04:10 +0000)]
[X86] Rename the VEX scalar fma builtins to end with a '3' to match gcc

I think we need to use different builtins for the FMA4 instructions since those instructions zero the upper bits and FMA3 instructions pass the bits through.

So this moves the existing builtins to be the FMA3 versions. New versions will be added for FMA4.

llvm-svn: 317766

6 years ago[X86] Rename the VEX scalar fma builtins to end with a '3' to match gcc
Craig Topper [Thu, 9 Nov 2017 04:10:42 +0000 (04:10 +0000)]
[X86] Rename the VEX scalar fma builtins to end with a '3' to match gcc

I think we need to use different builtins for the FMA4 instructions since those instructions zero the upper bits and FMA3 instructions pass the bits through.

So this moves the existing builtins to be the FMA3 versions. New versions will be added for FMA4.

llvm-svn: 317765

6 years ago[llvm-cov] Fix more -path-equivalence test bugs
Vedant Kumar [Thu, 9 Nov 2017 02:50:24 +0000 (02:50 +0000)]
[llvm-cov] Fix more -path-equivalence test bugs

llvm-svn: 317764

6 years ago[llvm-cov] Fix a -path-equivalence bug in a test
Vedant Kumar [Thu, 9 Nov 2017 02:42:34 +0000 (02:42 +0000)]
[llvm-cov] Fix a -path-equivalence bug in a test

llvm-svn: 317763

6 years ago[llvm-cov] Don't render empty region marker lines
Vedant Kumar [Thu, 9 Nov 2017 02:33:44 +0000 (02:33 +0000)]
[llvm-cov] Don't render empty region marker lines

This fixes an issue where llvm-cov prints an empty line, thinking it
needs to display region markers, when it actually doesn't.

llvm-svn: 317762

6 years ago[Coverage] Use the wrapped segment when a line has entry segments
Vedant Kumar [Thu, 9 Nov 2017 02:33:43 +0000 (02:33 +0000)]
[Coverage] Use the wrapped segment when a line has entry segments

We've worked around bugs in the frontend by ignoring the count from
wrapped segments when a line has at least one region entry segment.
Those frontend bugs are now fixed, so it's time to regenerate the
checked-in covmapping files and remove the workaround.

llvm-svn: 317761

6 years ago[Coverage] Emit deferred regions in headers
Vedant Kumar [Thu, 9 Nov 2017 02:33:40 +0000 (02:33 +0000)]
[Coverage] Emit deferred regions in headers

There are some limitations with emitting regions in macro expansions
because we don't gather file IDs within the expansions. Fix the check
that prevents us from emitting deferred regions in expansions to make an
exception for headers, which is something we can handle.

rdar://35373009

llvm-svn: 317760

6 years ago[Coverage] Complete top-level deferred regions before labels
Vedant Kumar [Thu, 9 Nov 2017 02:33:39 +0000 (02:33 +0000)]
[Coverage] Complete top-level deferred regions before labels

The area immediately after a terminated region in the function top-level
should have the same count as the label it precedes.

This solves another problem with wrapped segments. Consider:

  1| a:
  2|   return 0;
  3| b:
  4|   return 1;

Without a gap area starting after the first return, the wrapped segment
from line 2 would make it look like line 3 is executed, when it's not.

rdar://35373009

llvm-svn: 317759

6 years ago[Coverage] Emit a gap area after if conditions
Vedant Kumar [Thu, 9 Nov 2017 02:33:38 +0000 (02:33 +0000)]
[Coverage] Emit a gap area after if conditions

The area immediately after the closing right-paren of an if condition
should have a count equal to the 'then' block's count. Use a gap region
to set this count, so that region highlighting for the 'then' block
remains precise.

This solves a problem we have with wrapped segments. Consider:

  1| if (false)
  2|   foo();

Without a gap area starting after the condition, the wrapped segment
from line 1 would make it look like line 2 is executed, when it's not.

rdar://35373009

llvm-svn: 317758

6 years agoubsan: Allow programs to use setenv to configure ubsan_standalone.
Peter Collingbourne [Thu, 9 Nov 2017 02:22:07 +0000 (02:22 +0000)]
ubsan: Allow programs to use setenv to configure ubsan_standalone.

Previously ubsan_standalone used the GetEnv function to read the
environment variables UBSAN_OPTIONS and UBSAN_SYMBOLIZER_PATH. The
problem with GetEnv is that it does not respect changes to the
environment variables made using the libc setenv function, which
prevents clients from setting environment variables to configure
ubsan before loading ubsan-instrumented libraries.

The reason why we have GetEnv is that some runtimes need to read
environment variables while they initialize using .preinit_array,
and getenv does not work while .preinit_array functions are being
called. However, it is unnecessary for ubsan_standalone to initialize
that early. So this change switches ubsan_standalone to using getenv
and removes the .preinit_array entry. The static version of the runtime
still ends up being initialized using a C++ constructor that exists
to support the shared runtime.

Differential Revision: https://reviews.llvm.org/D39827

llvm-svn: 317757

6 years ago[libFuzzer] mechanically simplify a test, NFC
Kostya Serebryany [Thu, 9 Nov 2017 02:13:43 +0000 (02:13 +0000)]
[libFuzzer] mechanically simplify a test, NFC

llvm-svn: 317756

6 years agoAMDGPU: Merge BUFFER_STORE_DWORD_OFFEN/OFFSET into x2, x4
Marek Olsak [Thu, 9 Nov 2017 01:52:55 +0000 (01:52 +0000)]
AMDGPU: Merge BUFFER_STORE_DWORD_OFFEN/OFFSET into x2, x4

Summary:
Only 56 shaders (out of 48486) are affected.

Totals from affected shaders (changed stats only):
SGPRS: 2420 -> 2460 (1.65 %)
Spilled VGPRs: 94 -> 112 (19.15 %)
Scratch size: 524 -> 528 (0.76 %) dwords per thread
Code Size: 187400 -> 184992 (-1.28 %) bytes

One DiRT Showdown shader spills 6 more VGPRs.
One Grid Autosport shader spills 12 more VGPRs.

The other 54 shaders only have a decrease in code size.
(I'm ignoring the SGPR noise)

Reviewers: arsenm, nhaehnle

Subscribers: kzhuravl, wdng, yaxunl, dstuttard, tpr, llvm-commits, t-tye

Differential Revision: https://reviews.llvm.org/D39012

llvm-svn: 317755

6 years agoAMDGPU: Lower buffer store and atomic intrinsics manually
Marek Olsak [Thu, 9 Nov 2017 01:52:48 +0000 (01:52 +0000)]
AMDGPU: Lower buffer store and atomic intrinsics manually

Summary:
Without this, SIMemoryLegalizer inserts s_waitcnt vmcnt(0) before every
buffer store and atomic instruction.

Reviewers: arsenm, nhaehnle

Subscribers: kzhuravl, wdng, yaxunl, dstuttard, tpr, llvm-commits, t-tye

Differential Revision: https://reviews.llvm.org/D39060

llvm-svn: 317754

6 years agoAMDGPU: Merge BUFFER_LOAD_DWORD_OFFSET into x2, x4
Marek Olsak [Thu, 9 Nov 2017 01:52:36 +0000 (01:52 +0000)]
AMDGPU: Merge BUFFER_LOAD_DWORD_OFFSET into x2, x4

Summary: Only 3 (out of 48486) shaders are affected.

Reviewers: arsenm, nhaehnle

Subscribers: kzhuravl, wdng, yaxunl, dstuttard, tpr, t-tye, llvm-commits

Differential Revision: https://reviews.llvm.org/D38951

llvm-svn: 317753

6 years agoAMDGPU: Merge BUFFER_LOAD_DWORD_OFFEN into x2, x4
Marek Olsak [Thu, 9 Nov 2017 01:52:30 +0000 (01:52 +0000)]
AMDGPU: Merge BUFFER_LOAD_DWORD_OFFEN into x2, x4

Summary:
-9.9% code size decrease in affected shaders.

Totals (changed stats only):
SGPRS: 2151462 -> 2170646 (0.89 %)
VGPRS: 1634612 -> 1640288 (0.35 %)
Spilled SGPRs: 8942 -> 8940 (-0.02 %)
Code Size: 52940672 -> 51727288 (-2.29 %) bytes
Max Waves: 373066 -> 371718 (-0.36 %)

Totals from affected shaders:
SGPRS: 283520 -> 302704 (6.77 %)
VGPRS: 227632 -> 233308 (2.49 %)
Spilled SGPRs: 3966 -> 3964 (-0.05 %)
Code Size: 12203080 -> 10989696 (-9.94 %) bytes
Max Waves: 44070 -> 42722 (-3.06 %)

Reviewers: arsenm, nhaehnle

Subscribers: kzhuravl, wdng, yaxunl, dstuttard, tpr, llvm-commits, t-tye

Differential Revision: https://reviews.llvm.org/D38950

llvm-svn: 317752

6 years agoAMDGPU: Merge S_BUFFER_LOAD_DWORD_IMM into x2, x4
Marek Olsak [Thu, 9 Nov 2017 01:52:23 +0000 (01:52 +0000)]
AMDGPU: Merge S_BUFFER_LOAD_DWORD_IMM into x2, x4

Summary:
Only constant offsets (*_IMM opcodes) are merged.
It reuses code for LDS load/store merging.
It relies on the scheduler to group loads.

The results are mixed, I think they are mostly positive. Most shaders are
affected, so here are total stats only:

 SGPRS: 2072198 -> 2151462 (3.83 %)
 VGPRS: 1628024 -> 1634612 (0.40 %)
 Spilled SGPRs: 7883 -> 8942 (13.43 %)
 Spilled VGPRs: 97 -> 101 (4.12 %)
 Scratch size: 1488 -> 1492 (0.27 %) dwords per thread
 Code Size: 60222620 -> 52940672 (-12.09 %) bytes
 Max Waves: 374337 -> 373066 (-0.34 %)

There is 13.4% increase in SGPR spilling, DiRT Showdown spills a few more
VGPRs (now 37), but 12% decrease in code size.

These are the new stats for SGPR spilling. We already spill a lot SGPRs,
so it's uncertain whether more spilling will make any difference since
SGPRs are always spilled to VGPRs:

 SGPR SPILLING APPS   Shaders SpillSGPR AvgPerSh
 alien_isolation         2938       100      0.0
 batman_arkham_origins    589         6      0.0
 bioshock-infinite       1769         4      0.0
 borderlands2            3968        22      0.0
 counter_strike_glob..   1142        60      0.1
 deus_ex_mankind_div..   1410        79      0.1
 dirt-showdown            533         4      0.0
 dirt_rally               364      1163      3.2
 divinity                1052         2      0.0
 dota2                   1747         7      0.0
 f1-2015                  776      1515      2.0
 grid_autosport          1767      1505      0.9
 hitman                  1413       273      0.2
 left_4_dead_2           1762         4      0.0
 life_is_strange         1296        26      0.0
 mad_max                  358        96      0.3
 metro_2033_redux        2670        60      0.0
 payday2                 1362        22      0.0
 portal                   474         3      0.0
 saints_row_iv           1704         8      0.0
 serious_sam_3_bfe        392      1348      3.4
 shadow_of_mordor        1418        12      0.0
 shadow_warrior          3956       239      0.1
 talos_principle          324      1735      5.4
 thea                     172        17      0.1
 tomb_raider             1449       215      0.1
 total_war_warhammer      242        56      0.2
 ue4_effects_cave         295        55      0.2
 ue4_elemental            572        12      0.0
 unigine_tropics          210        56      0.3
 unigine_valley           278       152      0.5
 victor_vran             1262        84      0.1
 yofrankie                 82         2      0.0

Reviewers: arsenm, nhaehnle

Subscribers: kzhuravl, wdng, yaxunl, dstuttard, tpr, llvm-commits, t-tye

Differential Revision: https://reviews.llvm.org/D38949

llvm-svn: 317751

6 years agoAMDGPU: Fold immediate offset into BUFFER_LOAD_DWORD lowered from SMEM
Marek Olsak [Thu, 9 Nov 2017 01:52:17 +0000 (01:52 +0000)]
AMDGPU: Fold immediate offset into BUFFER_LOAD_DWORD lowered from SMEM

Summary:
-5.3% code size in affected shaders.

Changed stats only:

48486 shaders in 30489 tests
Totals:
SGPRS: 2086406 -> 2072430 (-0.67 %)
VGPRS: 1626872 -> 1627960 (0.07 %)
Spilled SGPRs: 7865 -> 7912 (0.60 %)
Code Size: 60978060 -> 60188764 (-1.29 %) bytes
Max Waves: 374530 -> 374342 (-0.05 %)

Totals from affected shaders:
SGPRS: 299664 -> 285688 (-4.66 %)
VGPRS: 233844 -> 234932 (0.47 %)
Spilled SGPRs: 3959 -> 4006 (1.19 %)
Code Size: 14905272 -> 14115976 (-5.30 %) bytes
Max Waves: 46202 -> 46014 (-0.41 %)

Reviewers: arsenm, nhaehnle

Subscribers: kzhuravl, wdng, yaxunl, dstuttard, tpr, llvm-commits, t-tye

Differential Revision: https://reviews.llvm.org/D38915

llvm-svn: 317750

6 years ago[libFuzzer] fix a test (and hopefully, the bot)
Kostya Serebryany [Thu, 9 Nov 2017 01:45:59 +0000 (01:45 +0000)]
[libFuzzer] fix a test (and hopefully, the bot)

llvm-svn: 317749

6 years ago[X86] Make sure we don't read too many operands from X86ISD::FMADDS1/FMADDS3 nodes...
Craig Topper [Thu, 9 Nov 2017 01:06:47 +0000 (01:06 +0000)]
[X86] Make sure we don't read too many operands from X86ISD::FMADDS1/FMADDS3 nodes when doing FNEG combine.

r317453 added new ISD nodes without rounding modes that were added to an existing if/else chain. But all the previous nodes handled there included a rounding mode. The final code after this if/else chain expected an extra operand that isn't present for the new nodes.

llvm-svn: 317748

6 years ago[libFuzzer] allow user to specify the merge control file
Kostya Serebryany [Thu, 9 Nov 2017 01:05:29 +0000 (01:05 +0000)]
[libFuzzer] allow user to specify the merge control file

llvm-svn: 317747

6 years ago[CMake] Passthrough CMAKE_SYSROOT to external projects
Petr Hosek [Thu, 9 Nov 2017 00:21:29 +0000 (00:21 +0000)]
[CMake] Passthrough CMAKE_SYSROOT to external projects

Differential Revision: https://reviews.llvm.org/D39029

llvm-svn: 317744

6 years ago[cfi-verify] Adds blacklist blame behaviour to cfi-verify.
Mitch Phillips [Thu, 9 Nov 2017 00:18:31 +0000 (00:18 +0000)]
[cfi-verify] Adds blacklist blame behaviour to cfi-verify.

Adds the blacklist behaviour to llvm-cfi-verify. Now will calculate which lines caused expected failures in the blacklist and reports the number of affected indirect CF instructions for each blacklist entry.

Also moved DWARF checking after instruction analysis to improve performance significantly - unrolling the inlining stack is expensive.

Reviewers: vlad.tsyrklevich

Subscribers: aprantl, pcc, kcc, llvm-commits

Differential Revision: https://reviews.llvm.org/D39750

llvm-svn: 317743

6 years ago[CMake][runtimes] Fix the variable name
Petr Hosek [Wed, 8 Nov 2017 23:44:27 +0000 (23:44 +0000)]
[CMake][runtimes] Fix the variable name

This typo causes the llvm-lit path resolution to fail.

Differential Revision: https://reviews.llvm.org/D39811

llvm-svn: 317742

6 years ago[MIPS] Setup less-significant bit in the .got and .got.plt entries in case of microMI...
Simon Atanasyan [Wed, 8 Nov 2017 23:34:34 +0000 (23:34 +0000)]
[MIPS] Setup less-significant bit in the .got and .got.plt entries in case of microMIPS code

The less-significant bit signals about microMIPS code for jump/branch
instructions.

llvm-svn: 317741

6 years agoHandle "-" in tryCreateFile.
Rafael Espindola [Wed, 8 Nov 2017 23:07:32 +0000 (23:07 +0000)]
Handle "-" in tryCreateFile.

Otherwise we would fail with -M if the we didn't have write
permissions to the current directory.

llvm-svn: 317740

6 years ago[FileOutputBuffer] Move factory methods out of their classes.
Rui Ueyama [Wed, 8 Nov 2017 22:57:48 +0000 (22:57 +0000)]
[FileOutputBuffer] Move factory methods out of their classes.

InMemoryBuffer and OnDiskBuffer classes have both factory methods and
public constructors, and that looks a bit odd. This patch makes factory
methods non-member function to fix it.

Differential Revision: https://reviews.llvm.org/D39693

llvm-svn: 317739

6 years ago[Sanitizers, CMake] Also use version script for libclang_rt.asan-i386.so
Evgeniy Stepanov [Wed, 8 Nov 2017 22:51:09 +0000 (22:51 +0000)]
[Sanitizers, CMake] Also use version script for libclang_rt.asan-i386.so

When building LLVM on x86_64-pc-linux-gnu (Fedora 25) with the bundled gcc 6.4.1
which uses gld 2.26.1-1.fc25, the dynamic/Asan-i386-calls-Dynamic-Test and
dynamic/Asan-i386-inline-Dynamic-Test tests failed to link with

/usr/bin/ld: /var/scratch/gcc/llvm/dist/lib/clang/6.0.0/lib/linux/libclang_rt.asan-i386.so: fork: invalid version 21 (max 0)
/var/scratch/gcc/llvm/dist/lib/clang/6.0.0/lib/linux/libclang_rt.asan-i386.so: error adding symbols: Bad value

I tried building with a self-compiled gcc 7.1.0 using gld 2.28, but the error remained.

It seems the error has been hit before (cf. https://reviews.llvm.org/rL314085), but
no real explanation has been found.

However, the problem goes away when linking the i386 libclang_rt.asan with a version
script just like every other variant is. Not using the version script in this single case
dates back to the initial introduction of the version script in r236551, but this change
was just checked in without any explanation AFAICT.

Since I've not found any other workaround and no reason for not always using the
version script, I propose to do so.

Tested on x86_64-pc-linux-gnu.

Patch by Rainer Orth.

Differential Revision: https://reviews.llvm.org/D39795

llvm-svn: 317738

6 years agoRemove redundant copy-pasted comment in test file from r317736
Alex Lorenz [Wed, 8 Nov 2017 22:47:15 +0000 (22:47 +0000)]
Remove redundant copy-pasted comment in test file from r317736

llvm-svn: 317737

6 years ago[ObjC] Fix function signature handling for blocks literals with attributes
Alex Lorenz [Wed, 8 Nov 2017 22:44:34 +0000 (22:44 +0000)]
[ObjC] Fix function signature handling for blocks literals with attributes

Block literals can have a type with attributes in its signature, e.g.
ns_returns_retained. The code that inspected the type loc of the block when
declaring its parameters didn't account for this fact, and only looked through
paren type loc. This commit ensures that getAsAdjusted is used instead of
IgnoreParens to find the block's FunctionProtoTypeLoc. This ensures that
block parameters are declared correctly in the block and avoids the
'undeclared identifier' error.

rdar://35416160

llvm-svn: 317736

6 years agoCorrect atexit(3) support in TSan/NetBSD
Kamil Rytarowski [Wed, 8 Nov 2017 22:34:17 +0000 (22:34 +0000)]
Correct atexit(3) support in TSan/NetBSD

Summary:
The NetBSD specific implementation of cxa_atexit() does not
preserve the 2nd argument if dso is equal to NULL.

Changes:

 - Split paths of handling intercepted __cxa_atexit() and atexit(3).
   This affects all supported Operating Systems.
 - Add a local stack-like structure to hold the __cxa_atexit() context.
   atexit(3) is documented in the C standard as calling callback from the
   earliest to the oldest entry. This path also fixes potential ABI
   problem of passing an argument to a function from the atexit(3)
   callback mechanism.
 - Add new test to ensure LIFO style of atexit(3) callbacks: atexit3.cc

Proposal to change the behavior of __cxa_atexit() in NetBSD has been rejected.

With the above changes TSan/NetBSD with the current tsan_interceptors.cc
can bootstrap into operation.

Sponsored by <The NetBSD Foundation>

Reviewers: vitalybuka, dvyukov, joerg, kcc, eugenis

Reviewed By: dvyukov

Subscribers: kubamracek, llvm-commits, #sanitizers

Tags: #sanitizers

Differential Revision: https://reviews.llvm.org/D39619

llvm-svn: 317735

6 years ago[libcxx] Mark test cxa_deleted_virtual.pass.cpp as failing for previous libcxx versions.
Volodymyr Sapsai [Wed, 8 Nov 2017 22:30:29 +0000 (22:30 +0000)]
[libcxx] Mark test cxa_deleted_virtual.pass.cpp as failing for previous libcxx versions.

r313500 added a fix for undefined "___cxa_deleted_virtual" symbol.
Previous libcxx versions don't have the fix and corresponding test
should be failing.

rdar://problem/34521053

Reviewers: EricWF, mclow.lists, ahatanak

Reviewed By: ahatanak

Subscribers: mehdi_amini, cfe-commits

Differential Revision: https://reviews.llvm.org/D39776

llvm-svn: 317734

6 years ago[X86] X86MaskedGatherSDNode shouldn't inherit from MaskedGatherScatterSDNode
Craig Topper [Wed, 8 Nov 2017 22:26:41 +0000 (22:26 +0000)]
[X86] X86MaskedGatherSDNode shouldn't inherit from MaskedGatherScatterSDNode

The classof implementation in MaskedGatherScatterSDNode doesn't consider X86MaskedGatherSDNode so its misleading.

llvm-svn: 317733

6 years ago[X86] Preserve memory refs when folding loads into divides.
Craig Topper [Wed, 8 Nov 2017 22:26:39 +0000 (22:26 +0000)]
[X86] Preserve memory refs when folding loads into divides.

This is similar to what we already do for multiplies. Without this we can't unfold and hoist an invariant load.

llvm-svn: 317732

6 years ago[X86] Remove an if check on the result of a cast. NFC
Craig Topper [Wed, 8 Nov 2017 22:26:37 +0000 (22:26 +0000)]
[X86] Remove an if check on the result of a cast. NFC

cast takes a non-null input and produces a non-null output. So this if can never fail.

llvm-svn: 317731

6 years agoLet replaceVTableHolder accept any type.
Adrian Prantl [Wed, 8 Nov 2017 22:04:43 +0000 (22:04 +0000)]
Let replaceVTableHolder accept any type.

In Rust, a trait can be implemented for any type, and if a trait
object pointer is used for the type, then a virtual table will be
emitted for that trait/type combination.

We would like debuggers to be able to inspect trait objects, which
requires finding the concrete type associated with a given vtable.

This patch changes LLVM so that any type can be passed to
replaceVTableHolder. This allows the Rust compiler to emit the needed
debug info -- associating a vtable with the concrete type for which it
was emitted.

This is a DWARF extension: DWARF only specifies the meaning of
DW_AT_containing_type in one specific situation. This style of DWARF
extension is routine, though, and LLVM already has one such case for
DW_AT_containing_type.

Patch by Tom Tromey!

Differential Revision: https://reviews.llvm.org/D39503

llvm-svn: 317730

6 years agoAdd an @llvm.sideeffect intrinsic
Dan Gohman [Wed, 8 Nov 2017 21:59:51 +0000 (21:59 +0000)]
Add an @llvm.sideeffect intrinsic

This patch implements Chandler's idea [0] for supporting languages that
require support for infinite loops with side effects, such as Rust, providing
part of a solution to bug 965 [1].

Specifically, it adds an `llvm.sideeffect()` intrinsic, which has no actual
effect, but which appears to optimization passes to have obscure side effects,
such that they don't optimize away loops containing it. It also teaches
several optimization passes to ignore this intrinsic, so that it doesn't
significantly impact optimization in most cases.

As discussed on llvm-dev [2], this patch is the first of two major parts.
The second part, to change LLVM's semantics to have defined behavior
on infinite loops by default, with a function attribute for opting into
potential-undefined-behavior, will be implemented and posted for review in
a separate patch.

[0] http://lists.llvm.org/pipermail/llvm-dev/2015-July/088103.html
[1] https://bugs.llvm.org/show_bug.cgi?id=965
[2] http://lists.llvm.org/pipermail/llvm-dev/2017-October/118632.html

Differential Revision: https://reviews.llvm.org/D38336

llvm-svn: 317729

6 years ago[ThinLTO] New test needs to require LTO
Teresa Johnson [Wed, 8 Nov 2017 21:48:27 +0000 (21:48 +0000)]
[ThinLTO] New test needs to require LTO

Fix buildbot failures:
http://lab.llvm.org:8011/builders/sanitizer-x86_64-linux/builds/5262/steps/annotate/logs/stdio

llvm-svn: 317728

6 years ago[ObjC] Boxed strings should use the nullability from stringWithUTF8String's return...
Alex Lorenz [Wed, 8 Nov 2017 21:33:15 +0000 (21:33 +0000)]
[ObjC] Boxed strings should use the nullability from stringWithUTF8String's return type

Objective-C NSString has a class method stringWithUTF8String that creates a new
NSString from a C string. Objective-C box expression @(...) can be used to
create an NSString instead of invoking the stringWithUTF8String method directly
(The compiler lowers it down to the invocation though). This commit ensures that
the type of @(string-value) gets the same nullability attributes as the return
type of stringWithUTF8String to ensure that the diagnostics are consistent
between the two.

rdar://33847186

Differential Revision: https://reviews.llvm.org/D39762

llvm-svn: 317727

6 years agoRevert "Correct dwarf unwind information in function epilogue for X86"
Reid Kleckner [Wed, 8 Nov 2017 21:31:14 +0000 (21:31 +0000)]
Revert "Correct dwarf unwind information in function epilogue for X86"

This reverts r317579, originally committed as r317100.

There is a design issue with marking CFI instructions duplicatable. Not
all targets support the CFIInstrInserter pass, and targets like Darwin
can't cope with duplicated prologue setup CFI instructions. The compact
unwind info emission fails.

When the following code is compiled for arm64 on Mac at -O3, the CFI
instructions end up getting tail duplicated, which causes compact unwind
info emission to fail:
  int a, c, d, e, f, g, h, i, j, k, l, m;
  void n(int o, int *b) {
    if (g)
      f = 0;
    for (; f < o; f++) {
      m = a;
      if (l > j * k > i)
        j = i = k = d;
      h = b[c] - e;
    }
  }

We get assembly that looks like this:
; BB#1:                                 ; %if.then
Lloh3:
adrp x9, _f@GOTPAGE
Lloh4:
ldr x9, [x9, _f@GOTPAGEOFF]
mov  w8, wzr
Lloh5:
str wzr, [x9]
stp x20, x19, [sp, #-16]!   ; 8-byte Folded Spill
.cfi_def_cfa_offset 16
.cfi_offset w19, -8
.cfi_offset w20, -16
cmp w8, w0
b.lt LBB0_3
b LBB0_7
LBB0_2:                                 ; %entry.if.end_crit_edge
Lloh6:
adrp x8, _f@GOTPAGE
Lloh7:
ldr x8, [x8, _f@GOTPAGEOFF]
Lloh8:
ldr w8, [x8]
stp x20, x19, [sp, #-16]!   ; 8-byte Folded Spill
.cfi_def_cfa_offset 16
.cfi_offset w19, -8
.cfi_offset w20, -16
cmp w8, w0
b.ge LBB0_7
LBB0_3:                                 ; %for.body.lr.ph

Note the multiple .cfi_def* directives. Compact unwind info emission
can't handle that.

llvm-svn: 317726

6 years ago[cmake] Allow LLVM_BUILD_INSTRUMENTED to be set to IR or Frontend
Vedant Kumar [Wed, 8 Nov 2017 21:26:40 +0000 (21:26 +0000)]
[cmake] Allow LLVM_BUILD_INSTRUMENTED to be set to IR or Frontend

- This deprecates LLVM_ENABLE_IR_PGO but keeps it around for now.
- Errors out when LLVM_BUILD_INSTRUMENTED and LLVM_BUILD_INSTRUMENTED_COVERAGE
  are both set.

Motivated by bogner's post-commit review of r313770.

llvm-svn: 317725

6 years agoMake sure an error is always handled.
Rafael Espindola [Wed, 8 Nov 2017 21:15:21 +0000 (21:15 +0000)]
Make sure an error is always handled.

llvm-svn: 317724