Revert "[CodeGen] Fix crash with duplicated mangled name."
This reverts commit r234767, as it was breaking all ARM buildbots for two days and the
assert is not in the code, making it difficult to spot the error, which would keep the
bots red for a few more days. New errors were silently introduced because of this bug,
and we don't want this to escalate.
[OPENMP] Codegen for 'firstprivate' clause in 'for' directive.
Adds proper codegen for 'firstprivate' clause in for directive. Initially codegen for 'firstprivate' clause was implemented for 'parallel' directive only.
Also this patch emits sync point only after initialization of firstprivate variables, not all private variables. This sync point is not required for privates, lastprivates etc., only for initialization of firstprivate variables.
Differential Revision: http://reviews.llvm.org/D8660
Follow up to r234962, start respecting `-emit-llvm-uselists even for
LLVM assembly. Note that the driver never passes this flag; this is
just a interface convenience/consistency for those using `-cc1`
directly. This required LLVM r234969 (and predecessors).
Stop relying on `cl::opt` to pass along the driver's decision to
preserve use-lists. Create a new `-cc1` option called
`-emit-llvm-uselists` that does the right thing (when -emit-llvm-bc).
Note that despite its generic name, it *doesn't* do the right thing when
-emit-llvm (LLVM assembly) yet. I'll hook that up soon.
This doesn't really change the behaviour of the driver. The default is
still to preserve use-lists for `clang -emit-llvm` and `clang
-save-temps`, and nothing else. But it stops relying on global state
(and also is a nicer interface for hackers using `clang -cc1`).
uselistorder: Update for LLVM API change in r234959
Now that `addBitcodeWriterPass()` requires an explicit bit to preserve
use-list order, send it in from `clang`. It looks like I'll be able to
push this up to the `-cc1` options.
Change `clang` to set `-preserve-bc-uselistorder` for the driver options
`-emit-llvm` and `-save-temps`. The former is useful for reproducing
results from `clang` in `opt` or `llc`, while the latter prevents
`-save-temps` from affecting the output. This is part of PR5680.
`-preserve-bc-uselistorder=true` is currently on by default, but a
follow-up commit in LLVM will reverse it.
Anton Yartsev [Tue, 14 Apr 2015 14:18:04 +0000 (14:18 +0000)]
[analyzer] This implements potential undefbehavior.ZeroAllocDereference checker.
TODO: support realloc(). Currently it is not possible due to the present realloc() handling. Currently RegionState is not being attached to realloc() in case of a zero Size argument.
Petar Jovanovic [Tue, 14 Apr 2015 12:49:08 +0000 (12:49 +0000)]
[Mips] Generate warning for invalid '-mnan' and '-march' combinations
This patch generates a warning for invalid combination of '-mnan' and
'-march' options, it properly sets NaN encoding for a given '-march',
and it passes a proper NaN encoding to the assembler.
[OPENMP] Fixed codegen for arrays in 'copyprivate' clause.
Fixed a bug with codegen of variables with array types specified in 'copyprivate' clause of 'single' directive.
Differential Revision: http://reviews.llvm.org/D8914
Richard Trieu [Mon, 13 Apr 2015 22:08:55 +0000 (22:08 +0000)]
Add new warning -Wrange-loop-analysis to warn on copies during loops.
-Wrange-loop-analysis is a subgroup of -Wloop-analysis and will warn when
a range-based for-loop makes copies of the elements in the range. If possible,
suggest the proper type to prevent copies, or the non-reference to help
distinguish copy versus non-copy forms. Existing warnings in -Wloop-analysis
are moved to -Wfor-loop-analysis, also a subgroup of -Wloop-analysis.
Revert r234620 and others, "Actually check if lseek works instead of using a filename based heuristic." It was affected by r234615, which was reverted in r234721.
r234620, "Actually check if lseek works instead of using a filename based heuristic."
r234621, "Testcase for the previous commit."
r234718, "Suppress clang/test/PCH/emit-pth.c on win32, for now while investigating."
[OPENMP] Initial codegen for 'parallel for' directive.
Allows generation of combined 'parallel for' directive that represents 'parallel' region with internal implicit 'for' worksharing region.
Differential Revision: http://reviews.llvm.org/D8631
DebugInfo: Switch from replaceAllUsesWith() to replaceTemporary()
Stop using `DIDescriptor`'s wrapper around
`MDNode::replaceAllUsesWith()` (which is going away). The new home for
this logic is `DIBuilder::replaceTemporary()`, added in LLVM r234695.
Richard Trieu [Sat, 11 Apr 2015 01:53:13 +0000 (01:53 +0000)]
Improve the error message for assigning to read-only variables.
Previously, many error messages would simply be "read-only variable is not
assignable" This change provides more information about why the variable is
not assignable, as well as note to where the const is located.
Ahmed Bougacha [Sat, 11 Apr 2015 00:10:44 +0000 (00:10 +0000)]
[Driver] Properly support -mglobal-merge using explicit options.
Follow-up to r234666. With this, the -m[no-]global-merge options
have the expected behavior. Previously, -mglobal-merge was ignored,
and there was no way of enabling the optimization.
Summary:
tools::arm::getARMFloatABI() was falling back to guessing soft-float because
it wasn't seeing the GNUEABIHF environment from ComputeEffectivClangTriple
when it was called from gnutools::Assemble::ConstructJob.
Fix by using the effective clang triple in gnutools::Assemble, which now
matches the -triple flag used by cc1 and ClangAs jobs.
Ben Langmuir [Fri, 10 Apr 2015 22:25:42 +0000 (22:25 +0000)]
Workaround a performance issue with modules + PCH
More fallout from r228234; when looking up an identifier in a PCH that
imports the Cocoa module on Darwin, it was taking 2 to 5 seconds
because we were hammering the MapVector::erase() function, which is
O(n). For now, just clear() the contained SmallVector to get back to
0.25 - 0.5 seconds. This is probably not the long-term fix, because
without modules or without PCH the performance is more like 0.02
seconds.
Don't eagerly typo-correct to a keyword if the next token is a right paren.
Take advantage of the delayed typo no longer being eagerly corrected to
a keyword to filter out keyword corrections (and other things like
unresolved & overloaded expressions, which have placeholder types) when
correcting typos inside of a decltype().
[Frontend] Close open file handles before renaming output files
The placement of the 'delete' call that was removed in the unique_ptr
migration in r234597 was not an accident. The raw_ostream has to be
destroyed before you do the rename on Windows, otherwise you get
ERROR_ACCESS_DENIED. We can still use unique_ptr, we just need to do a
manual reset().
Remove threshold for inserting lifetime markers for named temporaries
Now that TailRecursionElimination has been fixed with r222354, the
threshold on size for lifetime marker insertion can be removed. This
only affects named temporary though, as the patch for unnamed temporaries
is still in progress.
My previous commit (r222993) was not handling debuginfo correctly, but
this could only be seen with some asan tests. Basically, lifetime markers
are just instrumentation for the compiler's usage and should not affect
debug information; however, the cleanup infrastructure was assuming it
contained only destructors, i.e. actual code to be executed, and was
setting the breakpoint for the end of the function to the closing '}', and
not the return statement, in order to show some destructors have been
called when leaving the function. This is wrong when the cleanups are only
lifetime markers, and this is now fixed.
David Majnemer [Fri, 10 Apr 2015 04:52:06 +0000 (04:52 +0000)]
[Sema] Don't assume that an initializer list has an initializer
Given something like 'int({}, 1)', we would try to emit a diagnostic
regarding the excess element in the scalar initializer. However, we
assumed that the initializer list had an element in it.
David Majnemer [Thu, 9 Apr 2015 19:53:25 +0000 (19:53 +0000)]
[Sema] Diagnose references to unbound arrays in function definitions
A [*] is only allowed in a declaration for a function, not in its
definition. We didn't correctly recurse on reference types while
looking for it, causing us to crash in CodeGen instead of rejecting it.
Properly implement warn_unused_result checking for classes/structs.
The previous implementation would copy the attribute from the class to
functions that have the class as their return type when the functions
are first declared. This proved to have two flaws:
1) if the class is forward-declared without the attribute and a
function or method with the class as a its return type is declared,
and afterward the class is defined with warn_unused_result, the
function or method would never inherit the attribute, and
2) the check simply failed for functions and methods that are part of
a template instantiation, regardless of whether the class with
warn_unused_result is part of a specific instantiation or part of
the template itself (presumably because those function/method
declaration does not hit the same code path as a non-template one
and so never inherits the attribute).
The new approach is to instead modify the two places where a function or
method call is checked for the warn_unused_result attribute on the decl
by extending the checks to also look for the attribute on the decl's
return type.
Additionally, the check for return types that have the warn_unused_result
now excludes pointers and references to such types, as such return types do
not necessarily imply a transfer of ownership for the underlying object
being referred to by the return value. This does not change the behavior
of functions that are directly given the warn_unused_result attribute.
Process the -freciprocal-math optimization flag (PR20912)
The driver currently accepts but ignores the -freciprocal-math flag.
This patch passes the flag through and enables 'arcp' fast-math-flag
generation in IR.
Note that this change does not actually enable the optimization for
any target. The reassociation optimization that this flag specifies
was implemented by http://reviews.llvm.org/D6334 :
http://llvm.org/viewvc/llvm-project?view=revision&revision=222510
Because the optimization is done in the backend rather than IR,
the backend must be modified to understand instruction-level
fast-math-flags or a new function-level attribute must be created.
Also note that -freciprocal-math is independent of any target-specific
usage of reciprocal estimate hardware instructions. That requires
its own flag ('-mrecip').
Adds ARM Cortex-R4 and R4F support and tests in Clang. Though Cortex-R4
support was present, the support for hwdiv in thumb-mode was not defined
or tested properly. This has also been added.