OpenCL 2.0 atomic builtin functions have a scope argument which is ideally
represented as synchronization scope argument in LLVM atomic instructions.

Clang supports translating Clang atomic builtin functions to LLVM atomic
instructions. However it currently does not support synchronization scope
of LLVM atomic instructions. Without this, users have to use LLVM assembly
code to implement OpenCL atomic builtin functions.

This patch adds OpenCL 2.0 atomic builtin functions as Clang builtin
functions, which supports generating LLVM atomic instructions with
synchronization scope operand.

Currently only constant memory scope argument is supported. Support of
non-constant memory scope argument will be added later.

Differential Revision: https://reviews.llvm.org/D28691


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@310082 91177308-0d34-0410-b5e6-96231b3b80d8

Summary:
`case:` and `default:` would normally parse as labels for a `switch` block.
However in TypeScript, they can be used in field declarations, e.g.:

    interface I {
      case: string;
    }

This change special cases parsing them in declaration lines to avoid wrapping
them.

Reviewers: djasper

Subscribers: klimek, cfe-commits

Differential Revision: https://reviews.llvm.org/D36148

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@310070 91177308-0d34-0410-b5e6-96231b3b80d8

This is causing failures when compiling clang with -O3
as one of the structures used by clang is passed by
value and uses the fastcc calling convention.

Faliures manifest for stage2 mips build.


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@310057 91177308-0d34-0410-b5e6-96231b3b80d8

Summary:
M-class profiles do not support ARM execution mode, so providing
-marm/-mno-thumb does not make sense in combination with -mcpu/-march
options that support the M-profile.

This is a follow-up patch to D35569 and it seemed pretty clear that we
should emit an error in the driver in this case.

We probably also should warn/error if the provided -mcpu/-march options
do not match, e.g. -mcpu=cortex-m0 -march=armv8-a is invalid, as
cortex-m0 does not support armv8-a. But that should be a separate patch
I think.


Reviewers: echristo, richard.barton.arm, rengolin, labrinea, charles.baylis

Reviewed By: rengolin

Subscribers: aemerson, javed.absar, kristof.beyls, cfe-commits

Differential Revision: https://reviews.llvm.org/D35826

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@310047 91177308-0d34-0410-b5e6-96231b3b80d8

The hasDeclaration documentation was outdated after r288366.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@310041 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@310038 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@310037 91177308-0d34-0410-b5e6-96231b3b80d8

It generates MODULE-XXXXXXXXXXXX-%%%%%%%%.pcm, then GlobalModuleIndex.cpp is confused with the suffix ".pcm"

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@310030 91177308-0d34-0410-b5e6-96231b3b80d8

This reverts commit r309995. It looks like it's responsible for breaking
the stage2 coverage build:

http://green.lab.llvm.org/green/job/clang-stage2-coverage-R_build/1402

The cfe-commits discussion re: r309995 has more context.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@310019 91177308-0d34-0410-b5e6-96231b3b80d8

The .gnu_hash format is superior, and all versions of the Fuchsia
dynamic linker support it.

Differential Revision: https://reviews.llvm.org/D36254

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@310017 91177308-0d34-0410-b5e6-96231b3b80d8

The current coverage implementation doesn't handle region termination
very precisely. Take for example an `if' statement with a `return':

  void f() {
    if (true) {
      return; // The `if' body's region is terminated here.
    }
    // This line gets the same coverage as the `if' condition.
  }

If the function `f' is called, the line containing the comment will be
marked as having executed once, which is not correct.

The solution here is to create a deferred region after terminating a
region. The deferred region is completed once the start location of the
next statement is known, and is then pushed onto the region stack.
In the cases where it's not possible to complete a deferred region, it
can safely be dropped.

Testing: lit test updates, a stage2 coverage-enabled build of clang

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@310010 91177308-0d34-0410-b5e6-96231b3b80d8

This commit fixes a bug where clang/llvm doesn't emit an unwind table
for a function when it is marked noexcept. Without this patch, the
following code terminates with an uncaught exception on ARM64:

int foo1() noexcept {
  try {
    throw 0;
  } catch (int i) {
    return 0;
  }
  return 1;
}

int main() {
  return foo1();
}

rdar://problem/32411865

Differential Revision: https://reviews.llvm.org/D35693

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@310006 91177308-0d34-0410-b5e6-96231b3b80d8

Patch by Roland McGrath

Differential Revision: https://reviews.llvm.org/D35922

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309999 91177308-0d34-0410-b5e6-96231b3b80d8

The code after a noreturn call doesn't execute.

The pattern in the testcase is pretty common in LLVM (a switch with
a default case that calls llvm_unreachable).

Differential Revision: https://reviews.llvm.org/D36250



git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309995 91177308-0d34-0410-b5e6-96231b3b80d8

This helps some tools that do things based on the output's extension.

For example, we got reports from users on Windows that have a tool that scan a
build output dir (but skip .obj files). The tool would keep the "foo.obj-12345"
file open, and then when clang tried to rename the temp file to the final
output filename, that would fail. By making the tempfile end in ".obj", tools
like this will now skip the temp files as well.

https://reviews.llvm.org/D36238


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309984 91177308-0d34-0410-b5e6-96231b3b80d8

This reverts commit r309942 & commit r309940.

A revert was requested following post commit review.


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309978 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309976 91177308-0d34-0410-b5e6-96231b3b80d8

Patch by ~paul (cynecx on phabricator)! Some test massaging by me.


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309975 91177308-0d34-0410-b5e6-96231b3b80d8

This change adds support for cross-file diagnostic paths in html output. If the
diagnostic path is not cross-file, there is no change in the output.

Patch by Vlad Tsyrklevich!

Differential Revision: https://reviews.llvm.org/D30406

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309968 91177308-0d34-0410-b5e6-96231b3b80d8

See discussion in https://reviews.llvm.org/D33900#824172


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309960 91177308-0d34-0410-b5e6-96231b3b80d8

particular diagnostic.

Differential Revision: https://reviews.llvm.org/D36252

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309955 91177308-0d34-0410-b5e6-96231b3b80d8

Patch by: Reka Nikolett Kovacs


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309948 91177308-0d34-0410-b5e6-96231b3b80d8

Needed a // REQUIRES: mips-registered-target


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309942 91177308-0d34-0410-b5e6-96231b3b80d8

Summary:
This patch fixes the indentation of the code pattern `key <...>`and `key {...}` in text protos.
Previously, such line would be alinged depending on the column of the previous
colon, which usually indents too much.

I'm gonna go ahead and commit this since it's a straightforward bugfix.

Reviewers: djasper, klimek

Subscribers: klimek, cfe-commits

Differential Revision: https://reviews.llvm.org/D36143

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309941 91177308-0d34-0410-b5e6-96231b3b80d8

This option when combined with -mgpopt and -membedded-data places all
uninitialized constant variables in the read-only section.

Reviewers: atanasyan, nitesh.jain

Differential Revision: https://reviews.llvm.org/D35917


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309940 91177308-0d34-0410-b5e6-96231b3b80d8

Summary:
This patch fixes the parsing of proto option fields like `option op = <...>`.
Previously the parser did not enter the right code path inside the angle braces,
causing the contents to be split into several unwrapped lines inside.

I'll just go ahead and commit this since it's a straightforward bugfix.

Reviewers: djasper, klimek

Subscribers: cfe-commits

Differential Revision: https://reviews.llvm.org/D36217

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309937 91177308-0d34-0410-b5e6-96231b3b80d8

Add support for the -membedded-data option which places constant data in
the .rodata section, rather than the .sdata section.

Reviewers: atanasyan, nitesh.jain

Differential Revision: https://reviews.llvm.org/D35914


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309935 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309912 91177308-0d34-0410-b5e6-96231b3b80d8

Summary:
The mis-compile is triggered by internal code, but I haven't reduced it to a small piece of code. Add a FIXME here, since a decent fix doesn't seem to be trivial.

The decent fix can be changing Decl::Init to PointerUnion<Stmt *, EvaluatedStmt *, ParamVarDecl *>, and make setUninstantiatedDefaultArg take a ParamVarDecl *, which contains the Expr * as the default argument. This way, getTemplateInstantiationArgs can take that ParamVarDecl and do the right thing.

Reviewers: rsmith

Subscribers: sanjoy, cfe-commits

Differential Revision: https://reviews.llvm.org/D36253

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309908 91177308-0d34-0410-b5e6-96231b3b80d8

We never overwrite the end location of a region, so we would end up with
an overly large region when we reused the switch's region.

It's possible this code will be substantially rewritten in the near
future to deal with fallthrough more accurately, but this seems like
an improvement on its own for now.

Differential Revision: https://reviews.llvm.org/D34801



git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309901 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309888 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309885 91177308-0d34-0410-b5e6-96231b3b80d8

Summary:
This is to provide a default blacklist filename for UBSan.

While UBSan is turned on, it's better that clang pick up a blacklist file (when exists), just as what ASan / MSan does, so we do not end up adding the "-fsanitize-blacklist" option to every command line.

Reviewers: chandlerc, echristo, vsk, eugenis

Reviewed By: vsk, eugenis

Subscribers: vsk, eugenis, echristo, cfe-commits

Differential Revision: https://reviews.llvm.org/D35849

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309873 91177308-0d34-0410-b5e6-96231b3b80d8

Bot failure:
http://lab.llvm.org:8011/builders/clang-sphinx-docs/builds/12043/steps/docs-clang-html/logs/stdio

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309852 91177308-0d34-0410-b5e6-96231b3b80d8

In r309007, I made -fsanitize=null a hard prerequisite for -fsanitize=vptr. I
did not see the need for the two checks to have separate null checking logic
for the same pointer. I expected the two checks to either always be enabled
together, or to be mutually compatible.

In the mailing list discussion re: r309007 it became clear that that isn't the
case. If a codebase is -fsanitize=vptr clean but not -fsanitize=null clean,
it's useful to have -fsanitize=vptr emit its own null check. That's what this
patch does: with it, -fsanitize=vptr can be used without -fsanitize=null.

Differential Revision: https://reviews.llvm.org/D36112

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309846 91177308-0d34-0410-b5e6-96231b3b80d8

Summary: This is required by the libc++ locale support.

Reviewers: jyknight

Subscribers: fedor.sergeev

Differential Revision: https://reviews.llvm.org/D36121

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309815 91177308-0d34-0410-b5e6-96231b3b80d8

This function will be used by the clang-refactor's rename actions


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309813 91177308-0d34-0410-b5e6-96231b3b80d8

Originally, we weren't able to match on Type nodes themselves (only QualType),
so the hasDeclaration matcher was initially written to give what we thought are
reasonable results for QualType matches.

When we chagned the matchers to allow matching on Type nodes, it turned out
that the hasDeclaration matcher was by chance written templated enough to now
allow hasDeclaration to also match on (some) Type nodes.

This patch change the hasDeclaration matcher to:
a) work the same on Type and QualType nodes,
b) be completely explicit about what nodes we can match instead of just allowing
   anything with a getDecl() to match,
c) explicitly control desugaring only one level in very specific instances.
d) adds hasSpecializedTemplate and tagType matchers to allow migrating
  existing use cases that now need more explicit matchers

Note: This patch breaks clang-tools-extra. The corresponding patch there
is approved and will land in a subsequent patch.

Differential Revision: https://reviews.llvm.org/D27104

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309809 91177308-0d34-0410-b5e6-96231b3b80d8

Example:

#define MACRO(C)   if (C) { static int x; .. }
void foo() {
	MACRO(0);
}

Differential Revision: https://reviews.llvm.org/D36141



git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309799 91177308-0d34-0410-b5e6-96231b3b80d8

Summary: It used to call into llvm::sys::fs::make_absolute.

Reviewers: akyrtzi, erikjv, bkramer, krasimir, klimek

Reviewed By: klimek

Subscribers: cfe-commits

Differential Revision: https://reviews.llvm.org/D36155

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@309795 91177308-0d34-0410-b5e6-96231b3b80d8