Commit Graph

120423 Commits

Author SHA1 Message Date
Chandler Carruth 4bb2eb15d4 FileCheck-ize a test, make it more specific to directly test the shift
removal desired.

llvm-svn: 148213
2012-01-15 09:32:57 +00:00
Chandler Carruth 5b15a9be6a Two variables had been added for an assert, but their values were
re-computed rather than the variables be re-used just after the assert.
Just use the variables since we have them already. Fixes an unused
variable warning.

Also fix an 80-column violation.

llvm-svn: 148212
2012-01-15 09:03:45 +00:00
Chandler Carruth da22f30e72 Remove SetWorkingDirectory from the Process interface. Nothing in LLVM
or Clang is using this, and it would be hard to use it correctly given
the thread hostility of the function. Also, it never checked the return
which is rather dangerous with chdir. If someone was in fact using this,
please let me know, as well as what the usecase actually is so that
I can add it back and make it more correct and secure to use. (That
said, it's never going to be "safe" per-se, but we could at least
document the risks...)

llvm-svn: 148211
2012-01-15 08:41:35 +00:00
Richard Smith bd1530949d decltype(e) is type-dependent if e is instantiation-dependent. Scary but true.
Don't consider decltype(e) for an instantiation-dependent, but not
type-dependent, e to be non-type-dependent but canonical(!).

llvm-svn: 148210
2012-01-15 06:24:57 +00:00
Richard Smith 2ec4061e39 Pedantic diagnostic correction: in C++, we have integral constant expressions,
not integer constant expressions. In passing, fix the 'folding is an extension'
diagnostic to not claim we're accepting the code, since that's not true in
-pedantic-errors mode, and add this diagnostic to -Wgnu.

llvm-svn: 148209
2012-01-15 03:51:30 +00:00
Richard Smith b19ac0d6ca constexpr: casts to void* are allowed in constant expressions, don't set the
designator invalid. (Since we can't read the value of such a pointer, this only
affects the quality of diagnostics.)

llvm-svn: 148208
2012-01-15 03:25:41 +00:00
Eli Friedman 839192fd29 Change linkage computation so it doesn't depend on FunctionDecl::isExternC or VarDecl::isExternC, and instead queries what it actually cares about: whether the given declaration is inside an extern "C" context. Fundamentally, figuring out whether a function/variable uses C linkage requires knowing the linkage, and the logic in FunctionDecl::isExternC and VarDecl::isExternC was getting it wrong. Given that, fix FunctionDecl::isExternC and VarDecl::isExternC to use much simpler implementations that depend on the fixed linkage computation.
Fixes a regression to test/SemaCXX/linkage.cpp caused by a new warning exposing the fact that the internal state was wrong.

llvm-svn: 148207
2012-01-15 01:23:58 +00:00
David Blaikie fdcd669bc6 Remove dead code.
llvm-svn: 148206
2012-01-15 01:09:13 +00:00
Craig Topper 201c1a3505 Truncate of undef is just undef of smaller size.
llvm-svn: 148205
2012-01-15 01:05:11 +00:00
Benjamin Kramer 490afa6b9d Use a smaller vector than SmallVector.
Shrinks OverloadCandidate from 208 to 168 bytes.

llvm-svn: 148204
2012-01-14 21:05:10 +00:00
Greg Clayton 907018f0ad Bumped xcode project versions for lldb-106 and debugserver-165
llvm-svn: 148202
2012-01-14 20:57:34 +00:00
Greg Clayton 32720b51e2 <rdar://problem/9731573>
Fixed two double "int close(int fd)" issues found by our file descriptor
interposing library on darwin:

The first is in SBDebugger::SetInputFileHandle (FILE *file, bool transfer_ownership)
where we would give our FILE * to a lldb_private::File object member variable and tell
it that it owned the file descriptor if "transfer_ownership" was true, and then we
would also give it to the communication plug-in that waits for stdin to come in and
tell it that it owned the FILE *. They would both try and close the file.

The seconds was when we use a file descriptor through ConnectionFileDescriptor 
where someone else is creating a connection with ConnectionFileDescriptor and a URL
like: "fd://123". We were always taking ownwership of the fd 123, when we shouldn't
be. There is a TODO in the comments that says we should allow URL options to be passed
to be able to specify this later (something like: "fd://123?transer_ownership=1"), but
we can get to this later.

llvm-svn: 148201
2012-01-14 20:47:38 +00:00
Benjamin Kramer 02b0843684 Clear ImplicitConversionSequence the obvious way which turns out to be less fragile.
llvm-svn: 148200
2012-01-14 20:16:52 +00:00
Sean Callanan 36fa92af46 I forgot to import a header file change. This
should fix builds.

llvm-svn: 148199
2012-01-14 20:16:24 +00:00
Benjamin Kramer 0b9c509c4e Give OverloadCandidateSet the responsibility for destroying the implicit conversion sequences so we don't get double frees when the vector reallocates.
llvm-svn: 148198
2012-01-14 19:31:39 +00:00
Fariborz Jahanian 52d02f68e5 objc: disallow __block attribute on method params.
llvm-svn: 148197
2012-01-14 18:44:35 +00:00
Craig Topper c10e1abaf3 Fix the memop type on a couple 256-bit AVX instructions that were using f128mem instead of f256mem.
llvm-svn: 148196
2012-01-14 18:29:57 +00:00
Greg Clayton 662125ae93 Added some fixes for the fd interposing dylib:
- don't strip too many frames from the backtrace when logging close backtraces
- cleanup some logging messages

llvm-svn: 148195
2012-01-14 18:18:31 +00:00
Craig Topper d78429f850 Add a bunch of AVX instructions to the folding tables. Also fixed the alignment on 256-bit AVX2 instructions.
llvm-svn: 148194
2012-01-14 18:14:53 +00:00
Benjamin Kramer 9059ff5056 Reorder initializers, make use of the whole array.
llvm-svn: 148193
2012-01-14 17:59:15 +00:00
Benjamin Kramer 1f8f8a3e64 Move the inline array to the end.
llvm-svn: 148192
2012-01-14 17:23:10 +00:00
Duncan Sands 90212bde1f Speculatively revert commit 148175 (rafael), to see if this fixes
non-determinism in the 32 bit dragonegg buildbot.  Original commit
message:
Only emit the Leh_func_endN symbol when needed.

llvm-svn: 148191
2012-01-14 17:16:48 +00:00
Benjamin Kramer da765a7322 Destroy OverloadCandidates before taking away the underlying memory.
llvm-svn: 148190
2012-01-14 17:10:59 +00:00
Greg Clayton ccbc08e6ae <rdar://problem/10684141>
When the lldb_private::Debugger goes away, it should cleanup all
of its targets.

llvm-svn: 148189
2012-01-14 17:04:19 +00:00
Douglas Gregor aefdb38352 Replace a using declararion with a typedef in the hope of satisfying GCC
llvm-svn: 148188
2012-01-14 16:48:12 +00:00
Douglas Gregor ec9fd13c77 De-virtualize getPreviousDecl() and getMostRecentDecl() when we know
we have a redeclarable type, and only use the new virtual versions
(getPreviousDeclImpl() and getMostRecentDeclImpl()) when we don't have
that type information. This keeps us from penalizing users with strict
type information (and is the moral equivalent of a "final" method).

Plus, settle on the names getPreviousDecl() and getMostRecentDecl()
throughout.

llvm-svn: 148187
2012-01-14 16:38:05 +00:00
Benjamin Kramer b009517ad4 Pool allocate ImplicitConversionSequences.
To avoid malloc thrashing give OverloadCandidateSet an inline capacity for conversion sequences.
We use the fact that OverloadCandidates never outlive the OverloadCandidateSet and have a fixed
amount of conversion sequences.

This eliminates the oversized SmallVector from OverloadCandidate shrinking it from 752 to 208 bytes.

On the test case from the "Why is CLANG++ so freaking slow" thread on llvmdev this avoids one gig
of vector reallocation (including memcpy) which translates into 5-10% speedup on Lion/x86_64.

Overload candidate computation is still the biggest malloc contributor when compiling templated
c++ code.

llvm-svn: 148186
2012-01-14 16:32:05 +00:00
Benjamin Kramer fb761ff544 OverloadCandidateSet: Stop exposing SmallVector internals
Replace push_back with addCandidate which will let us make use of the fixed size of
the conversion sequence vector soon.

llvm-svn: 148185
2012-01-14 16:31:55 +00:00
Douglas Gregor 0bc8a21dba Introduce Decl::getPreviousDecl() and Decl::getMostRecentDecl(),
virtual functions that provide previous/most recent redeclaration
information for any declaration. Use this to eliminate the redundant,
less efficient getPreviousDecl() functions.

llvm-svn: 148184
2012-01-14 15:55:47 +00:00
Douglas Gregor 9abeca16e5 Revert accidental commit
llvm-svn: 148183
2012-01-14 15:31:52 +00:00
Douglas Gregor 463c8e7070 Add a FIXME for mutation of the common pointer of a RedeclarableTemplateDecl. It is not clear that it's worth delaying the allocation of said pointer
llvm-svn: 148182
2012-01-14 15:30:55 +00:00
Douglas Gregor 68444de354 Reimplement RedeclarableTemplateDecl in terms of
Redeclarable<RedeclarableTemplateDecl>, eliminating a bunch of
redeclaration-chain logic both in RedeclarableTemplateDecl and
especially in its (de-)serialization.

As part of this, eliminate the RedeclarableTemplate<> class template,
which was an abstraction that didn't actually save anything.

llvm-svn: 148181
2012-01-14 15:13:49 +00:00
Howard Hinnant f55694af01 Two bug fixes, several clarifications, and a few comment updates.
llvm-svn: 148180
2012-01-14 14:15:43 +00:00
Zhongxing Xu 82846eae0a Remove a redundant word.
llvm-svn: 148179
2012-01-14 09:08:15 +00:00
Richard Smith dafff94759 constexpr irgen: Add irgen support for APValue::Struct, APValue::Union,
APValue::Array and APValue::MemberPointer. All APValue values can now be emitted
as constants.

Add new CGCXXABI entry point for emitting an APValue MemberPointer. The other
entrypoints dealing with constant member pointers are no longer necessary and
will be removed in a later change.

Switch codegen from using EvaluateAsRValue/EvaluateAsLValue to
VarDecl::evaluateValue. This performs caching and deals with the nasty cases in
C++11 where a non-const object's initializer can refer indirectly to
previously-initialized fields within the same object.

Building the intermediate APValue object incurs a measurable performance hit on
pathological testcases with huge initializer lists, so we continue to build IR
directly from the Expr nodes for array and record types outside of C++11.

llvm-svn: 148178
2012-01-14 04:30:29 +00:00
Andrew Trick 23ef0d6c40 Fix a corner case hit by redundant phi elimination running after LSR.
Fixes PR11761: bad IR w/ redundant Phi elim

llvm-svn: 148177
2012-01-14 03:17:23 +00:00
Anna Zaks 0244cd7450 [analyzer] Taint: add system and popen as undesirable sinks for taint
data.

llvm-svn: 148176
2012-01-14 02:48:40 +00:00
Rafael Espindola dfde7631fa Only emit the Leh_func_endN symbol when needed.
llvm-svn: 148175
2012-01-14 02:36:51 +00:00
Andrew Trick 59ac4fb706 misched: Initial code for building an MI level scheduling DAG
llvm-svn: 148174
2012-01-14 02:17:18 +00:00
Andrew Trick dbee9d8900 Move physreg dependency generation into aptly named addPhysRegDeps.
llvm-svn: 148173
2012-01-14 02:17:15 +00:00
Andrew Trick 1d028a364d misched: Added ScheduleDAGInstrs::IsPostRA
llvm-svn: 148172
2012-01-14 02:17:12 +00:00
Andrew Trick 7e120f4e66 misched: Invoke the DAG builder on each sequence of schedulable instructions.
llvm-svn: 148171
2012-01-14 02:17:09 +00:00
Andrew Trick 6344087e17 Move things around to make the file navigable, even though it will probably be split up later.
llvm-svn: 148170
2012-01-14 02:17:06 +00:00
Argyrios Kyrtzidis 9ef9486ab7 [libclang] If CXIndexOpt_IndexFunctionLocalSymbols is enabled, also
index parameters.

llvm-svn: 148169
2012-01-14 02:05:51 +00:00
Evan Cheng 6bb95253eb After r147827 and r147902, it's now possible for unallocatable registers to be
live across BBs before register allocation. This miscompiled 197.parser
when a cmp + b are optimized to a cbnz instruction even though the CPSR def
is live-in a successor.
        cbnz    r6, LBB89_12
...
LBB89_12:
        ble     LBB89_1

The fix consists of two parts. 1) Teach LiveVariables that some unallocatable
registers might be liveouts so don't mark their last use as kill if they are.
2) ARM constantpool island pass shouldn't form cbz / cbnz if the conditional
branch does not kill CPSR.

rdar://10676853

llvm-svn: 148168
2012-01-14 01:53:46 +00:00
Chad Rosier 71a185c5c6 Fix pasto from r146196.
llvm-svn: 148167
2012-01-14 01:50:21 +00:00
Chad Rosier d027d528b5 Cleanup test case by adding checks for test names.
llvm-svn: 148166
2012-01-14 01:46:51 +00:00
Jakob Stoklund Olesen b3c18a691b Add TRI::getCallPreservedMask() hook.
The hook returns a bit-mask of call-preserved registers that will
eventually replace the current list of implicit defs on call
instructions.  This will make it possible to support multiple calling
conventions without duplicating call instruction descriptors.

The call-preserved mask is slightly different from the list returned by
the getCalleeSavedRegs() hook, it includes all aliases that are
preserved by calls.

The hook takes a CallingConv::ID argument instead of a MachineFunction
pointer, so it can provide information about calls to extern functions,
and even indirect function calls.

TRI::getCalleeSavedRegs() returns information about the function
currently being compiled. TRI::getCallPreservedMask() returns
information about the functions it is calling.

llvm-svn: 148165
2012-01-14 01:45:25 +00:00
Dan Gohman 4cf362acc1 Fix an unused variable warning that Chad noticed.
llvm-svn: 148164
2012-01-14 00:47:44 +00:00
Rafael Espindola 2f869a3f4f Remember if a type has its visibility set explicitly or implicitly.
With that, centralize the way we merge visibility, always preferring explicit over
implicit and then picking the most restrictive one.
Fixes pr10113 and pr11690.

llvm-svn: 148163
2012-01-14 00:30:36 +00:00