Alexey Samsonov [Mon, 18 Mar 2013 07:02:08 +0000 (07:02 +0000)]
[TSan] Use ThreadRegistry::FindThreadContextLocked() to find thread by its stack/tls address.
llvm-svn: 177248
Alexey Samsonov [Mon, 18 Mar 2013 07:00:36 +0000 (07:00 +0000)]
[TSan] Makefiles: allow configurable paths to clang and FileCheck. Add -fno-rtti flag.
llvm-svn: 177247
Alexey Samsonov [Mon, 18 Mar 2013 06:27:13 +0000 (06:27 +0000)]
[Sanitizer] Get rid of global mutexes in StopTheWorld implementation
llvm-svn: 177246
Alexey Samsonov [Mon, 18 Mar 2013 06:07:44 +0000 (06:07 +0000)]
[Sanitizer] Rely on template argument deduction in sanitizer_allocator_test
llvm-svn: 177245
Jakob Stoklund Olesen [Mon, 18 Mar 2013 04:08:07 +0000 (04:08 +0000)]
Extract a method.
This computes the type of an instruction operand or result based on the
records in the instruction's ins and outs lists.
llvm-svn: 177244
Craig Topper [Mon, 18 Mar 2013 03:34:55 +0000 (03:34 +0000)]
Post process ADC/SBB and use a shorter encoding if they use a sign extended immediate.
llvm-svn: 177243
Craig Topper [Mon, 18 Mar 2013 02:53:34 +0000 (02:53 +0000)]
Refactor some duplicated code into helper functions.
llvm-svn: 177242
David Blaikie [Sun, 17 Mar 2013 21:32:54 +0000 (21:32 +0000)]
Fix the build broken in r177239
Seems some accidental C++11 crept in there. Reported by the C++98 buildbots.
llvm-svn: 177241
Michael Gottesman [Sun, 17 Mar 2013 21:31:12 +0000 (21:31 +0000)]
Reduced dont-infinite-loop-during-block-escape-analysis.ll with bugpoint and moved it to retain-block-escape-analysis.ll.
*NOTE* I verified that the original bug behind
dont-infinite-loop-during-block-escape-analysis.ll occurs when using opt on
retain-block-escape-analysis.ll.
llvm-svn: 177240
David Blaikie [Sun, 17 Mar 2013 21:13:55 +0000 (21:13 +0000)]
Split out filename & directory from DIFile to start generalizing over DIScopes
This is the first step to making all DIScopes have a common metadata prefix (so
that things (using directives, for example) that can appear in any scope can be
added to that common prefix). DIFile is itself a DIScope so the common prefix
of all DIScopes cannot be a DIFile - instead it's the raw filename/directory
name pair.
llvm-svn: 177239
David Blaikie [Sun, 17 Mar 2013 21:08:22 +0000 (21:08 +0000)]
Generalize debug info test to be resilient to changes in metadata node numbering
llvm-svn: 177238
David Blaikie [Sun, 17 Mar 2013 20:29:22 +0000 (20:29 +0000)]
Generalize a few debug info test cases
Checking for the annotation comment rather than the metadata values makes these
tests resilient to a coming refactor that will pull these fields out into a
separate metadata node.
llvm-svn: 177237
David Blaikie [Sun, 17 Mar 2013 20:28:12 +0000 (20:28 +0000)]
Improve DIFile debug info annotation by letting it fallback to DIScope
llvm-svn: 177236
Jakob Stoklund Olesen [Sun, 17 Mar 2013 17:26:09 +0000 (17:26 +0000)]
Use ArrayRef<MVT::SimpleValueType> when possible.
Not passing vector references around makes it possible to use
SmallVector in most places.
llvm-svn: 177235
Sylvestre Ledru [Sun, 17 Mar 2013 12:40:42 +0000 (12:40 +0000)]
To avoid symbol clash, undefine PPC here. PPC may be predefined on some hosts.
llvm-svn: 177234
Rafael Espindola [Sun, 17 Mar 2013 12:01:05 +0000 (12:01 +0000)]
Build LLVMgold.so on FreeBSD using cmake.
Patch by Stephen Checkoway.
llvm-svn: 177233
Michael Gottesman [Sun, 17 Mar 2013 08:42:58 +0000 (08:42 +0000)]
The promised test case for r175939.
This test makes sure that the ObjCARC escape analysis looks at the uses of
instructions which copy the block pointer value by checking all four cases where
that can occur.
llvm-svn: 177232
Hal Finkel [Sun, 17 Mar 2013 04:43:44 +0000 (04:43 +0000)]
Improve PPC VR (Altivec) register spilling
This change cleans up two issues with Altivec register spilling:
1. The spilling code was inefficient (using two instructions, and add and a
load, when just one would do)
2. The code assumed that r0 would always be available (true for now, but this
will change)
The new code handles VR spilling just like GPR spills but forced into r+r mode.
As a result, when any VR spills are present, we must now always allocate the
register-scavenger spill slot.
llvm-svn: 177231
Hal Finkel [Sat, 16 Mar 2013 23:02:31 +0000 (23:02 +0000)]
Remove FIXMEs in PPC test cases related to unaligned loads/stores
As pointed out by Bill in response to r177160, these two FIXMEs
can also be removed.
llvm-svn: 177229
Hal Finkel [Sat, 16 Mar 2013 22:50:51 +0000 (22:50 +0000)]
Remove PPC avoidWriteAfterWrite callback
As a follow-up to r158719, remove PPCRegisterInfo::avoidWriteAfterWrite.
Jakob pointed out in response to r158719 that this callback is currently unused
and so this has no effect (and the speedups that I thought that I had observed
as a result of implementing this function must have been noise).
llvm-svn: 177228
Andrew Trick [Sat, 16 Mar 2013 18:58:57 +0000 (18:58 +0000)]
Change the default latency for implicit defs.
Implicit defs are not currently positional and not modeled by the
per-operand machine model. Unfortunately, we treat defs that are part
of the architectural instruction description, like flags, the same as
other implicit defs. Really, they should have a fixed MachineInstr
layout and probably shouldn't be "implicit" at all.
For now, we'll change the default latency to be the max operand
latency. That will give flag setting operands full latency for x86
folded loads. Other kinds of "fake" implicit defs don't occur prior to
regalloc anyway, and we would like them to go away postRegAlloc as
well.
llvm-svn: 177227
Andrew Trick [Sat, 16 Mar 2013 18:58:55 +0000 (18:58 +0000)]
Machine model. Allow mixed itinerary classes and SchedRW lists.
We always supported a mixture of the old itinerary model and new
per-operand model, but it required a level of indirection to map
itinerary classes to SchedRW lists. This was done for ARM A9.
Now we want to define x86 SchedRW lists, with the goal of removing its
itinerary classes, but still support the itineraries in the mean
time. When I original developed the model, Atom did not have
itineraries, so there was no reason to expect this requirement.
llvm-svn: 177226
Sylvestre Ledru [Sat, 16 Mar 2013 17:01:13 +0000 (17:01 +0000)]
When built with cmake, only call the string functions when the previous commands (utils/GetSourceVersion & utils/GetRepositoryPath successed). Otherwise, they fail when not launched from a repository
llvm-svn: 177225
Sean Silva [Sat, 16 Mar 2013 16:58:20 +0000 (16:58 +0000)]
[docs] Discuss a potential bug to be aware of.
llvm-svn: 177224
Aaron Ballman [Sat, 16 Mar 2013 15:00:51 +0000 (15:00 +0000)]
Test case for graceful handling of long file names on Windows. Patch thanks to Paul Robinson!
llvm-svn: 177223
Manman Ren [Sat, 16 Mar 2013 04:47:38 +0000 (04:47 +0000)]
revert r177211 due to its potential issues
llvm-svn: 177222
Craig Topper [Sat, 16 Mar 2013 03:44:31 +0000 (03:44 +0000)]
Add X86 code emitter support AVX encoded MRMDestReg instructions.
Previously we weren't skipping the VVVV encoded register. Based on patch by Michael Liao.
llvm-svn: 177221
Jordan Rose [Sat, 16 Mar 2013 02:14:06 +0000 (02:14 +0000)]
[analyzer] Model trivial copy/move assignment operators with a bind as well.
r175234 allowed the analyzer to model trivial copy/move constructors as
an aggregate bind. This commit extends that to trivial assignment
operators as well. Like the last commit, one of the motivating factors here
is not warning when the right-hand object is partially-initialized, which
can have legitimate uses.
<rdar://problem/
13405162>
llvm-svn: 177220
Enrico Granata [Sat, 16 Mar 2013 01:50:07 +0000 (01:50 +0000)]
C++ formatters for NSTimeZone and for CFBitVector
llvm-svn: 177219
Argyrios Kyrtzidis [Sat, 16 Mar 2013 01:40:35 +0000 (01:40 +0000)]
Remove -Wspellcheck and replace it with a diagnostic option.
Thanks to Richard S. for pointing out that the warning would show up
with -Weverything.
llvm-svn: 177218
Enrico Granata [Sat, 16 Mar 2013 01:18:00 +0000 (01:18 +0000)]
NS(Mutable)IndexSet formatter moves from Python to C++
llvm-svn: 177217
Jordan Rose [Sat, 16 Mar 2013 01:07:58 +0000 (01:07 +0000)]
[analyzer] Separate graph trimming from creating the single-path graph.
When we generate a path diagnostic for a bug report, we have to take the
full ExplodedGraph and limit it down to a single path. We do this in two
steps: "trimming", which limits the graph to all the paths that lead to
this particular bug, and "creating the report graph", which finds the
shortest path in the trimmed path to any error node.
With BugReporterVisitor false positive suppression, this becomes more
expensive: it's possible for some paths through the trimmed graph to be
invalid (i.e. likely false positives) but others to be valid. Therefore
we have to run the visitors over each path in the graph until we find one
that is valid, or until we've ruled them all out. This can become quite
expensive.
This commit separates out graph trimming from creating the report graph,
performing the first only once per bug equivalence class and the second
once per bug report. It also cleans up that portion of the code by
introducing some wrapper classes.
This seems to recover most of the performance regression described in my
last commit.
<rdar://problem/
13433687>
llvm-svn: 177216
Jordan Rose [Sat, 16 Mar 2013 01:07:53 +0000 (01:07 +0000)]
[analyzer] Eliminate InterExplodedGraphMap class and NodeBackMap typedef.
...in favor of this typedef:
typedef llvm::DenseMap<const ExplodedNode *, const ExplodedNode *>
InterExplodedGraphMap;
Use this everywhere the previous class and typedef were used.
Took the opportunity to ArrayRef-ize ExplodedGraph::trim while I'm at it.
No functionality change.
llvm-svn: 177215
Jordan Rose [Sat, 16 Mar 2013 01:07:47 +0000 (01:07 +0000)]
[analyzer] Don't repeat a bug equivalence class if every report is invalid.
I removed this check in the recursion->iteration commit, but forgot that
generatePathDiagnostic may be called multiple times if there are multiple
PathDiagnosticConsumers.
llvm-svn: 177214
Enrico Granata [Sat, 16 Mar 2013 00:50:25 +0000 (00:50 +0000)]
Converting more data formatters to C++ - NSBundle, CFBinaryHeap, NSMachPort and NSNotification
llvm-svn: 177213
Howard Hinnant [Sat, 16 Mar 2013 00:17:53 +0000 (00:17 +0000)]
This should be nothing but a load-time optimization. I'm trying to reduce load time initializers and this is a big one. No visible functionality change intended.
llvm-svn: 177212
Manman Ren [Sat, 16 Mar 2013 00:11:09 +0000 (00:11 +0000)]
Exploit this-return of a callsite in a this-return function.
For constructors/desctructors that return 'this', if there exists a callsite
that returns 'this' and is immediately before the return instruction, make
sure we are using the return value from the callsite.
We don't need to keep 'this' alive through the callsite. It also enables
optimizations in the backend, such as tail call optimization.
rdar://
12818789
llvm-svn: 177211
Jakob Stoklund Olesen [Sat, 16 Mar 2013 00:02:17 +0000 (00:02 +0000)]
Define more SchedWrites for annotating X86 instructions.
Since almost all X86 instructions can fold loads, use a multiclass to
define register/memory pairs of SchedWrites.
An X86FoldableSchedWrite represents the register version of an
instruction. It holds a reference to the SchedWrite to use when the
instruction folds a load.
This will be used inside multiclasses that define rr and rm instruction
versions together.
llvm-svn: 177210
Richard Trieu [Fri, 15 Mar 2013 23:55:09 +0000 (23:55 +0000)]
Improve template diffing handling of default integer values.
When the template argument is both default and value dependent, the expression
retrieved for the default argument cannot be evaluated, thus never matching
any argument value. To get the proper value, get the template argument
from the desugared template specialization. Also, output the original
expression to provide more information about the argument mismatch.
llvm-svn: 177209
Greg Clayton [Fri, 15 Mar 2013 23:54:07 +0000 (23:54 +0000)]
<rdar://problem/
13194155>
Variables view out of sync with lldb in Xcode is now fixed. Depending on what happened stack frames could get out of date and a stale shared pointer (one that is no longer a current frame in a thread) could end up being used.
Now we don't store a weak_ptr to a frame in the ExecutionContextRef class, we just store its stack ID and we always regrab the frame from the thread by stack ID.
llvm-svn: 177208
Anna Zaks [Fri, 15 Mar 2013 23:34:31 +0000 (23:34 +0000)]
[analyzer] Address a TODO in the StreamChecker; otherwise the output is non-deterministic.
llvm-svn: 177207
Anna Zaks [Fri, 15 Mar 2013 23:34:29 +0000 (23:34 +0000)]
[analyzer] Use isLiveRegion to determine when SymbolRegionValue is dead.
Fixes a FIXME, improves dead symbol collection, suppresses a false positive,
which resulted from reusing the same symbol twice for simulation of 2 calls to the same function.
Fixing this lead to 2 possible false negatives in CString checker. Since the checker is still alpha and
the solution will not require revert of this commit, move the tests to a FIXME section.
llvm-svn: 177206
Anna Zaks [Fri, 15 Mar 2013 23:34:25 +0000 (23:34 +0000)]
[analyzer] BugReporterVisitors: handle the case where a ternary operator is wrapped in a cast.
llvm-svn: 177205
Anna Zaks [Fri, 15 Mar 2013 23:34:22 +0000 (23:34 +0000)]
[analyzer] Address Jordan’s review of r177138 (a micro optimization)
llvm-svn: 177204
Enrico Granata [Fri, 15 Mar 2013 23:33:15 +0000 (23:33 +0000)]
Performance improvements to the IsObjCNil () - we only try to resolve the value if the variable under consideration truly is an “Objective-C thing”
This also changes the ClangASTContext to make sure that id is correctly marked as being such an ObjC thing
llvm-svn: 177203
Han Ming Ong [Fri, 15 Mar 2013 23:19:44 +0000 (23:19 +0000)]
<rdar://problem/
13100435>
Don't discount regions that vmmap is not discounting.
llvm-svn: 177202
Ted Kremenek [Fri, 15 Mar 2013 23:09:37 +0000 (23:09 +0000)]
Fix buffer underrun (invalid read) triggered during diagnostic rendering. The test would overflow when computing '0 - 1'.
I don't have a good testcase for this that does not depend on system headers.
It did not trigger with preprocessed output, and I had trouble reducing the example.
Fixes <rdar://problem/
13324594>.
Thanks to Michael Greiner for reporting this issue.
llvm-svn: 177201
Jim Ingham [Fri, 15 Mar 2013 23:09:19 +0000 (23:09 +0000)]
Convert the -a option in "image lookup" to a address-expression, and pass the context
in the -a address expression lookup in "image list" so that it actually works.
llvm-svn: 177200
Jakob Stoklund Olesen [Fri, 15 Mar 2013 22:51:13 +0000 (22:51 +0000)]
Add SchedRW as an Instruction field.
Don't require instructions to inherit Sched<...>. Sometimes it is more
convenient to say:
let SchedRW = ... in {
...
}
Which is now possible.
llvm-svn: 177199
Matt Kopec [Fri, 15 Mar 2013 22:49:52 +0000 (22:49 +0000)]
Rework TestChangedValueAPI.py to account for gcc setting multiple breakpoints.
llvm-svn: 177198
Argyrios Kyrtzidis [Fri, 15 Mar 2013 22:43:10 +0000 (22:43 +0000)]
[modules] Don't record the macros from the predefines buffer.
These will be available in the current translation unit anyway, for
modules they only waste space and deserialization time.
llvm-svn: 177197
Jim Ingham [Fri, 15 Mar 2013 22:19:27 +0000 (22:19 +0000)]
Make the conditional break test case a little more robust.
llvm-svn: 177196
Jim Ingham [Fri, 15 Mar 2013 22:18:26 +0000 (22:18 +0000)]
The regular expression command object was resetting the execution context of the
resolved command, which it should not do. It should adopt whatever context the
regular expression command was called with. This was causing regular expression
commands run inside breakpoint commands to adopt the currently selected context,
not the one coming from the breakpoint that we hit.
<rdar://problem/
13411771>
llvm-svn: 177195
Douglas Gregor [Fri, 15 Mar 2013 22:15:07 +0000 (22:15 +0000)]
<rdar://problem/
13426257> Introduce SDKSettings.plist as an input file dependency for PCH/modules.
When we're building a precompiled header or module against an SDK on
Darwin, there will be a file SDKSettings.plist in the sysroot. Since
stat()'ing every system header on which a module or PCH file depends
is performance suicide, we instead stat() just SDKSettings.plist. This
hack works well on Darwin; it's unclear how we want to handle this on
other platforms. If there is a canonical file, we should use it; if
not, we either have to take the performance hit of stat()'ing system
headers repeatedly or roll the dice by not checking anything.
llvm-svn: 177194
Ted Kremenek [Fri, 15 Mar 2013 22:02:46 +0000 (22:02 +0000)]
Simplify print logic, per feedback from Jordan Rose.
llvm-svn: 177193
Matt Kopec [Fri, 15 Mar 2013 21:55:13 +0000 (21:55 +0000)]
Fix Makefile.rules where no icc is installed and -C gcc is used.
Patch by Ashok Thirumurthi.
llvm-svn: 177192
Ted Kremenek [Fri, 15 Mar 2013 21:50:10 +0000 (21:50 +0000)]
Enhance -Wtautological-constant-out-of-range-compare to include the name of the enum constant.
This is QoI. Fixes <rdar://problem/
13076064>.
llvm-svn: 177190
Jordan Rose [Fri, 15 Mar 2013 21:41:55 +0000 (21:41 +0000)]
[analyzer] Make GRBugReporter::generatePathDiagnostic iterative, not recursive.
The previous generatePathDiagnostic() was intended to be tail-recursive,
restarting and trying again if a report was marked invalid. However:
(1) this leaked all the cloned visitors, which weren't being deleted, and
(2) this wasn't actually tail-recursive because some local variables had
non-trivial destructors.
This was causing us to overflow the stack on inputs with large numbers of
reports in the same equivalence class, such as sqlite3.c. Being iterative
at least prevents us from blowing out the stack, but doesn't solve the
performance issue: suppressing thousands (yes, thousands) of paths in the
same equivalence class is expensive. I'm looking into that now.
<rdar://problem/
13423498>
llvm-svn: 177189
Jordan Rose [Fri, 15 Mar 2013 21:41:53 +0000 (21:41 +0000)]
[analyzer] Collect stats on the max # of bug reports in an equivalence class.
We discovered that sqlite3.c currently has 2600 reports in a single
equivalence class; it would be good to know if this is a recent
development or what.
(For the curious, the different reports in an equivalence class represent
the same bug found along different paths. When we're suppressing false
positives, we need to go through /every/ path to make sure there isn't a
valid path to a bug. This is a flaw in our after-the-fact suppression,
made worse by the fact that that function isn't particularly optimized.)
llvm-svn: 177188
Jordan Rose [Fri, 15 Mar 2013 21:41:50 +0000 (21:41 +0000)]
[analyzer] Include opcode in dumping a SymSymExpr.
For debugging use only; no functionality change.
llvm-svn: 177187
Jordan Rose [Fri, 15 Mar 2013 21:41:46 +0000 (21:41 +0000)]
[analyzer] Look through ExprWhenCleanups when trying to track a NULL.
Silences a few false positives in LLVM.
llvm-svn: 177186
Jordan Rose [Fri, 15 Mar 2013 21:41:35 +0000 (21:41 +0000)]
Add some assertions to appease the static analyzer.
No functionality change.
llvm-svn: 177185
David Blaikie [Fri, 15 Mar 2013 21:12:54 +0000 (21:12 +0000)]
Remove unnecessary default in covered switch over enum
This cleans up the Clang -Werror build that was broken by r177180.
llvm-svn: 177184
Daniel Dunbar [Fri, 15 Mar 2013 20:55:27 +0000 (20:55 +0000)]
[AST] Add a fast path to ConstantArrayType::getNumAddressingBits().
- This fast path is almost 100% effective on real code, and lets us avoid
multiple allocations of 128-bit APSInt objects in the common case.
- As with any overflow-check-skipping-code, I'd appreciate someone double
checking my logic.
llvm-svn: 177183
David Blaikie [Fri, 15 Mar 2013 20:52:10 +0000 (20:52 +0000)]
Remove IR scenario tests.
These are a burden to maintain & use hand crafted IR to test particular DWARF
generation cases in LLVM. If someone thinks there's high value coverage here
they can rephrase these as IR+dwarfdump tests in the LLVM regression suite so
they'll be portable/easily executed by all developers.
llvm-svn: 177182
Fariborz Jahanian [Fri, 15 Mar 2013 20:47:07 +0000 (20:47 +0000)]
c: perform integer overflow check on all binary
operations. // rdar://
13423975
llvm-svn: 177181
Richard Trieu [Fri, 15 Mar 2013 20:35:18 +0000 (20:35 +0000)]
Refactor template diffing to store an enum that records which type of
difference is stored inside a DiffNode. This should not change any
diagnostic messages.
llvm-svn: 177180
Edwin Vane [Fri, 15 Mar 2013 20:18:08 +0000 (20:18 +0000)]
Don't include outer-most explicit cast in nullptr replacement
The outer-most explicit cast is now left alone by the Use-Nullptr transform to
maintain the type of the expression and avoid introducing ambiguities.
Fixes PR15395.
Author: Ariel J Bernal <ariel.j.bernal@intel.com>
llvm-svn: 177179
Daniel Dunbar [Fri, 15 Mar 2013 20:16:59 +0000 (20:16 +0000)]
[ADT] Fix StringSet::insert() to not allocate on every lookup.
- The previous implementation always constructed the StringMap entry, even if
the key was present in the set.
llvm-svn: 177178
Edwin Vane [Fri, 15 Mar 2013 20:14:01 +0000 (20:14 +0000)]
ClangTool output cleanup
Information messages sent to stdout by ClangTool now only happen when the
-debug flag is set.
Error messages that used to go to stdout now go to stderr.
Author: Ariel J Bernal <ariel.j.bernal@intel.com>
llvm-svn: 177177
Matt Kopec [Fri, 15 Mar 2013 20:00:39 +0000 (20:00 +0000)]
Rollback r177173. Some OSs may not have ptrace extensions which lldb expects when building. This needs to be accounted for.
llvm-svn: 177176
Michael J. Spencer [Fri, 15 Mar 2013 19:25:47 +0000 (19:25 +0000)]
[Support][Path][Windows] Fix dangling else. Don't call CloseHandle when CloseFD is false.
llvm-svn: 177175
Matt Kopec [Fri, 15 Mar 2013 19:10:12 +0000 (19:10 +0000)]
Add icc support to the test suite
-adds icc to the lit of compilers to run the tests
-adds icc test decorators
-skip TestAnonymous.py for icc
Patch by Ashok Thirumurthi.
llvm-svn: 177174
Matt Kopec [Fri, 15 Mar 2013 19:06:45 +0000 (19:06 +0000)]
Add ptrace extensions to query a register set.
Patch by Ashok Thirumurthi.
llvm-svn: 177173
Enrico Granata [Fri, 15 Mar 2013 18:55:30 +0000 (18:55 +0000)]
Porting the Objective-C Class data type’s summary from Python to C++
llvm-svn: 177172
Enrico Granata [Fri, 15 Mar 2013 18:44:08 +0000 (18:44 +0000)]
Moving CFBag and NSBundle summaries from Python to C++
Temporarily disabled non-code-running summaries for CF*Dictionary and NSCountedSet
llvm-svn: 177171
Arnold Schwaighofer [Fri, 15 Mar 2013 18:31:01 +0000 (18:31 +0000)]
ARM cost model: Fix costs for some vector selects
I was too pessimistic in r177105. Vector selects that fit into a legal register
type lower just fine. I was mislead by the code fragment that I was using. The
stores/loads that I saw in those cases came from lowering the conditional off
an address.
Changing the code fragment to:
%T0_3 = type <8 x i18>
%T1_3 = type <8 x i1>
define void @func_blend3(%T0_3* %loadaddr, %T0_3* %loadaddr2,
%T1_3* %blend, %T0_3* %storeaddr) {
%v0 = load %T0_3* %loadaddr
%v1 = load %T0_3* %loadaddr2
==> FROM:
;%c = load %T1_3* %blend
==> TO:
%c = icmp slt %T0_3 %v0, %v1
==> USE:
%r = select %T1_3 %c, %T0_3 %v0, %T0_3 %v1
store %T0_3 %r, %T0_3* %storeaddr
ret void
}
revealed this mistake.
radar://
13403975
llvm-svn: 177170
Silviu Baranga [Fri, 15 Mar 2013 18:28:25 +0000 (18:28 +0000)]
Adding an A15 specific optimization pass for interactions between S/D/Q registers. The pass handles all the required transformations pre-regalloc.
llvm-svn: 177169
Edwin Vane [Fri, 15 Mar 2013 18:10:07 +0000 (18:10 +0000)]
Prevent nullptr_t-typed exprs from being replaced
The Use-Nullptr transform was replacing nullptr_t-typed expressions because in
the AST such expressions have an implicit NullToPointer cast around them. Now
the transform ignores these expressions.
Fixes PR15414.
llvm-svn: 177168
Benjamin Kramer [Fri, 15 Mar 2013 17:27:39 +0000 (17:27 +0000)]
ARM: Fix an old refacto.
Fixes PR15520.
llvm-svn: 177167
Enrico Granata [Fri, 15 Mar 2013 17:25:04 +0000 (17:25 +0000)]
<rdar://problem/
13194155>
Fixing an issue where threads and frames could get out of sync and cause ValueObjects to fail to retrieve their values correctly
llvm-svn: 177166
Bob Wilson [Fri, 15 Mar 2013 17:12:43 +0000 (17:12 +0000)]
Revert "Remove a pointless assertion."
This reverts commit r177158.
I'm blindly reverting this because it appears to be breaking numerous
buildbots. I'll reapply if it doesn't turn out to be the culprit.
llvm-svn: 177165
Adrian Prantl [Fri, 15 Mar 2013 17:09:05 +0000 (17:09 +0000)]
Force column info only for direct inlined functions. This should strike
the balance between expected behavior and compatibility with the gdb
testsuite.
(GDB gets confused if we break an expression into multiple debug
stmts so we enable this behavior only for inlined functions. For the
full experience people can still use -gcolumn-info.)
llvm-svn: 177164
Fariborz Jahanian [Fri, 15 Mar 2013 17:03:56 +0000 (17:03 +0000)]
c: Also chek for integer overflow for '%' operator.
llvm-svn: 177163
Fariborz Jahanian [Fri, 15 Mar 2013 16:36:04 +0000 (16:36 +0000)]
c: add the missing binary operatory when checking
for integer overflow. // rdar://
13423975
llvm-svn: 177162
Sylvestre Ledru [Fri, 15 Mar 2013 16:22:43 +0000 (16:22 +0000)]
Take in account the triplet 'powerpc-linux-gnuspe' for PowerPC SPE. Done for the port of Debian on this arch. More information on: wiki.debian.org/PowerPCSPEPort Patch by Roland Stigge
llvm-svn: 177161
Hal Finkel [Fri, 15 Mar 2013 15:27:13 +0000 (15:27 +0000)]
Enable unaligned memory access on PPC for scalar types
Unaligned access is supported on PPC for non-vector types, and is generally
more efficient than manually expanding the loads and stores.
A few of the existing test cases were using expanded unaligned loads and stores
to test other features (like load/store with update), and for these test cases,
unaligned access remains disabled.
llvm-svn: 177160
Arnold Schwaighofer [Fri, 15 Mar 2013 15:10:47 +0000 (15:10 +0000)]
ARM cost model: Fix cost of fptrunc and fpext instructions
A vector fptrunc and fpext simply gets split into scalar instructions.
radar://
13192358
llvm-svn: 177159
Nico Weber [Fri, 15 Mar 2013 15:02:37 +0000 (15:02 +0000)]
Remove a pointless assertion.
FindNodeOrInsertPos() is called 10 lines earlier already, and the function
early-returns there if the result is != 0. InsertPos isn't recomputed after
that check, so this assert is always trivially true. (And it has nothing to
do with if T is canonical or not.)
llvm-svn: 177158
Daniel Jasper [Fri, 15 Mar 2013 14:57:30 +0000 (14:57 +0000)]
Improve formatting of chained calls.
clang-format already prevented sequences like:
...
SomeParameter).someFunction(
...
as those are quite confusing. This failed on:
...
SomeParameter).someFunction(otherFunction(
...
Fixed in this patch.
llvm-svn: 177157
Alexander Potapenko [Fri, 15 Mar 2013 14:37:21 +0000 (14:37 +0000)]
[libsanitizer] StopTheWorld in sanitizer_common
StopTheWorld puts the process in a suspended state before running the
user-supplied callback. To be used in TSan and in leak checking code.
Linux implementation provided.
Patch by Sergey Matveev (earthdok@google.com)
llvm-svn: 177156
Alexey Samsonov [Fri, 15 Mar 2013 14:02:21 +0000 (14:02 +0000)]
[Sanitizer] Don't intercept frexpf and frexpl on Windows. Add frexp declaration to asan_intercepted_functions.h
llvm-svn: 177155
Alexey Samsonov [Fri, 15 Mar 2013 13:48:44 +0000 (13:48 +0000)]
[TSan] Switch TSan runtime to use ThreadRegistry class from sanitizer_common
llvm-svn: 177154
Alexander Kornienko [Fri, 15 Mar 2013 13:42:02 +0000 (13:42 +0000)]
Indent all lines in a multi-line comment by the same amount.
Summary:
Do this to avoid spoling nicely formatted multi-line comments (e.g.
with code examples or similar stuff).
Reviewers: djasper
Reviewed By: djasper
CC: cfe-commits, klimek
Differential Revision: http://llvm-reviews.chandlerc.com/D544
llvm-svn: 177153
Hal Finkel [Fri, 15 Mar 2013 13:21:21 +0000 (13:21 +0000)]
Protect PPC Altivec patterns with a predicate
In preparation for the addition of other SIMD ISA extensions (such as QPX) we
need to make sure that all Altivec patterns are properly predicated on having
Altivec support.
No functionality change intended (one test case needed to be updated b/c it
assumed that Altivec intrinsics would be supported without enabling Altivec
support).
llvm-svn: 177152
Alexey Samsonov [Fri, 15 Mar 2013 12:54:01 +0000 (12:54 +0000)]
[Sanitizer] Don't link tests with -lpthread on Android.
llvm-svn: 177151
Kostya Serebryany [Fri, 15 Mar 2013 12:27:52 +0000 (12:27 +0000)]
[sanitizer] fix gcc build warnings
llvm-svn: 177149
Kostya Serebryany [Fri, 15 Mar 2013 11:39:41 +0000 (11:39 +0000)]
[asan] Add ForEachChunk() to sanitizer allocators. Patch by Sergey Matveev
llvm-svn: 177147
Alexey Samsonov [Fri, 15 Mar 2013 10:39:26 +0000 (10:39 +0000)]
[Sanitizer] Generalize compile/link flags for sanitizer_common tests on Android and on other platforms.
llvm-svn: 177146
Alexey Samsonov [Fri, 15 Mar 2013 07:29:58 +0000 (07:29 +0000)]
Fixup for r176934. More careful setup of path to llvm-symbolizer
llvm-svn: 177145