Commit Graph

165713 Commits

Author SHA1 Message Date
Kevin Enderby c030848f8f Tweak the MCExternalSymbolizer to not use the SymbolLookUp() call back
to not guess at a symbol name in some cases.

The problem is that in object files assembled starting at address 0, when
trying to symbolicate something that starts like this:

% cat x.s
_t1:
	vpshufd	$0x0, %xmm1, %xmm0

the symbolic disassembly can end up like this:

% otool -tV x.o 
x.o:
(__TEXT,__text) section
_t1:
0000000000000000	vpshufd	$_t1, %xmm1, %xmm0

Which is in this case produced incorrect symbolication.

But it is useful in some cases to use the SymbolLookUp() call back
to guess at some immediate values.  For example one like this
that does not have an external relocation entry:

% cat y.s
_t1:
	movl	$_d1, %eax
.data
_d1:	.long	0

% clang -c -arch i386 y.s

% otool -tV y.o 
y.o:
(__TEXT,__text) section
_t1:
0000000000000000	movl	$_d1, %eax

% otool -rv y.o 
y.o:
Relocation information (__TEXT,__text) 1 entries
address  pcrel length extern type    scattered symbolnum/value
00000001 False long   False  VANILLA False     2 (__DATA,__data)

So the change is based on it is not likely that an immediate Value
coming from an instruction field of a width of 1 byte, other than branches
and items with relocation, are not likely symbol addresses.

With the change the first case above simply becomes:

% otool -tV x.o 
x.o:
(__TEXT,__text) section
_t1:
0000000000000000	vpshufd	$0x0, %xmm1, %xmm0

and the second case continues to work as expected.

rdar://14863405

llvm-svn: 199698
2014-01-21 00:23:17 +00:00
Kevin Enderby debfea62d4 To allow the X86 verbose assembly to print its informative comments
when used with symbolic disassembly, add a check that the operand
is an immediate and has not been symbolicated to MCExpr operand.

I’m trying to enable the ‘C’ disassembly API option
LLVMDisassembler_Option_SetInstrComments for darwin’s
otool(1) that uses the llvm disassembler API.  The problem is
that the disassembler API can change an immediate operand to
an MCExpr operand if it symbolicates it with the call backs.
And if it does the code in llvm::EmitAnyX86InstComments()
will crash when it assumes these operands are immediates.

The fix for this is very straight forward to just protect the call
to getImm() with a check of isImm().  So if the immediate for
an instruction is symbolicated it simply doesn’t get the X86
verbose assembly comments:

% otool -tV test_asm.o
test_asm.o:
(__TEXT,__text) section
_t1:
0000000000000000	vpshufd	$_t1, %xmm1, %xmm0
0000000000000005	retq
0000000000000006	nopw	%cs:_t1(%rax,%rax)
_t2:
0000000000000010	vpshufd	$-0x1, %xmm0, %xmm0     ## xmm0 = xmm0[3,3,3,3]
0000000000000015	retq
0000000000000016	nopw	%cs:_t1(%rax,%rax)
_t3:
0000000000000020	vpshufd	$_t1, %xmm1, %xmm0
0000000000000025	retq
0000000000000026	nopw	%cs:_t1(%rax,%rax)
_t4:
0000000000000030	vpshufd	$0x2d, %xmm0, %xmm0     ## xmm0 = xmm0[1,3,2,0]
0000000000000035	retq

The fact that the immediate $0x0 is being symbolicated at
all in this case is a different problem which my next patch
will address.

rdar://10989286

llvm-svn: 199697
2014-01-21 00:18:51 +00:00
Bob Wilson 74b6cd15ed Stop linking with crt1.o etc. for the iOS simulator. <rdar://problem/11800973>
Recent versions of the iOS simulator no longer require linking with the
crt1.o, dylib1.o, or bundle1.o files. The relevant code is now included in
libSystem for the simulator.

llvm-svn: 199696
2014-01-21 00:17:10 +00:00
Marshall Clow d724d60fff Removed extra space; thanks to thakis_'s eagle eye
llvm-svn: 199695
2014-01-21 00:03:44 +00:00
Marshall Clow 683268b2d8 Fixed test failure in is_iec559.pass.cpp on darwin-ppc32. Thanks to David Fang for the report (and suggested fix)
llvm-svn: 199694
2014-01-20 23:57:16 +00:00
Alp Toker f747ff1f54 Build fix following clang r199686
llvm-svn: 199689
2014-01-20 21:14:57 +00:00
Rafael Espindola d19f80a0b4 Give explicit sections for string constants used in NSStrings.
Without them they can be merged with non unnamed_addr constants during LTO.
The resulting constant is not unnamed_addr and goes in a different section,
which causes ld64 to crash.

A testcase that would crash before:

* file1.mm:
void g(id notification) {
  [notification valueForKey:@"name"];
}

* file2.cpp:
extern const char js_name_str[] = "name";

* file3.cpp
extern bool JS_GetProperty(const char *name);
extern const char js_name_str[];
bool js_ReportUncaughtException() { JS_GetProperty(js_name_str); }

run

clang file1.mm  -o file1.o -c -w -emit-llvm
clang file2.cpp -o file2.o -c -w -emit-llvm
clang file3.cpp -o file3.o -c -w

ld -dylib -o XUL file1.o file2.o file3.o -undefined dynamic_lookup.

llvm-svn: 199688
2014-01-20 20:33:18 +00:00
Tom Stellard 8a63b15b3c Fix build broken by LLVM commit r199279
Patch by: Udo van den Heuvel

Tom Stellard:
  - Added ifdef and error handling

llvm-svn: 199687
2014-01-20 20:28:48 +00:00
Alp Toker 9cacbabd33 Rename FunctionProtoType accessors from 'arguments' to 'parameters'
Fix a perennial source of confusion in the clang type system: Declarations and
function prototypes have parameters to which arguments are supplied, so calling
these 'arguments' was a stretch even in C mode, let alone C++ where default
arguments, templates and overloading make the distinction important to get
right.

Readability win across the board, especially in the casting, ADL and
overloading implementations which make a lot more sense at a glance now.

Will keep an eye on the builders and update dependent projects shortly.

No functional change.

llvm-svn: 199686
2014-01-20 20:26:09 +00:00
Rafael Espindola 8ff1610f06 Simplify further.
Thanks to David Blaikie for the push.

llvm-svn: 199685
2014-01-20 20:13:11 +00:00
Hal Finkel a69e5b8b9d Update StackProtector when coloring merges stack slots
StackProtector keeps a ValueMap of alloca instructions to layout kind tags for
use by PEI and other later passes. When stack coloring replaces one alloca with
a bitcast to another one, the key replacement in this map does not work.
Instead, provide an interface to manage this updating directly. This seems like
an improvement over the old behavior, where the layout map would not get
updated at all when the stack slots were merged. In practice, however, there is
likely no observable difference because PEI only did anything special with
'large array' kinds, and if one large array is merged with another, than the
replacement should already have been a large array.

This is an attempt to unbreak the clang-x86_64-darwin11-RA builder.

llvm-svn: 199684
2014-01-20 19:49:14 +00:00
Andrea Di Biagio 450d1661be [X86] Teach how to combine a vselect into a movss/movsd
Add target specific rules for combining vselect dag nodes into movss/movsd
when possible.

If the vector type of the vselect dag node in input is either MVT::v4i13 or
MVT::v4f32, then try to fold according to rules:

  1) fold (vselect (build_vector (0, -1, -1, -1)), A, B) -> (movss A, B)
  2) fold (vselect (build_vector (-1, 0, 0, 0)), A, B) -> (movss B, A)

If the vector type of the vselect dag node in input is either MVT::v2i64 or
MVT::v2f64 (and we have SSE2), then try to fold according to rules:

  3) fold (vselect (build_vector (0, -1)), A, B) -> (movsd A, B)
  4) fold (vselect (build_vector (-1, 0)), A, B) -> (movsd B, A)

llvm-svn: 199683
2014-01-20 19:35:22 +00:00
Fariborz Jahanian 15f60cb2ba ObjectiveC driver. reinstate -fno-objc-legacy-dispatch behavior
when the deployment target is 10.5. // rdar://15852259

llvm-svn: 199682
2014-01-20 19:32:33 +00:00
Adrian Prantl 671af5ca4b Debug info: On ARM ensure that all __TEXT sections come before the
optional DWARF sections, so compiling with -g does not result in
different code being generated for PC-relative loads.

This is reapplying a diet r197922 (__TEXT-only).

llvm-svn: 199681
2014-01-20 19:15:59 +00:00
Adrian Prantl 1a89924d84 Revert "Debug info: On ARM ensure that the data sections come before the"
Cut back on the cargo cult. The order of __DATA sections doesn't affect
generated code.

This reverts commit r197922.

llvm-svn: 199680
2014-01-20 19:15:55 +00:00
Aaron Ballman b39a9b8b0c Adding a bit of documentation that was missed with r198883 (when ParseArgumentsAsUnevaluated was added).
llvm-svn: 199679
2014-01-20 18:42:10 +00:00
Owen Anderson fb00d5bc7c Allow SMUL_LOHI and UMUL_LOHI to be narrow to MUL on targets where MUL is Custom rather than Legal. Even if the target is doing some kind of expansion for MUL, it's pretty much guaranteed to be more efficent than whatever it does for SMUL_LOHI or UMUL_LOHI!
llvm-svn: 199678
2014-01-20 18:41:34 +00:00
Aaron Ballman b9023ed0f6 Exposed a declarative way to specify that an attribute can be duplicated when merging attributes on a declaration. This replaces some hard-coded functionality from Sema.
llvm-svn: 199677
2014-01-20 18:07:09 +00:00
Aaron Ballman 9a99e0da5a Remove some hard-coded specialness for thread-safety attributes from the parser, and made it more declarative. If an attribute is allowed to appear on a function definition when late parsed, it can now use the FunctionDefinition attribute subject. It's treated as a FunctionDecl for most purposes, except it also gets exposed on the AttributeList so that it can be used while parsing.
llvm-svn: 199676
2014-01-20 17:18:35 +00:00
James Molloy 43ccae1bb4 Remove the useless pseudo instructions VDUPfdf and VDUPfqf, replacing them with patterns to match VDUPLN.
llvm-svn: 199675
2014-01-20 17:14:48 +00:00
NAKAMURA Takumi db2a4af37f [CMake] LLVMProcessSources.cmake: Add include(CMakeParseArguments).
I didn't realize that cmake_parse_arguments() would require explicit inclusion.

llvm-svn: 199674
2014-01-20 17:05:49 +00:00
Daniel Jasper 929b1db276 clang-format: Properly format custom options in protocol buffer definitions.
Before:
  option(my_option) = "abc";

After:
  option (my_option) = "abc";

llvm-svn: 199672
2014-01-20 16:47:22 +00:00
Aaron Ballman 3f5f3e7949 Formatting cleanups; no functional changes.
llvm-svn: 199671
2014-01-20 16:15:55 +00:00
Rafael Espindola 4f5f0b78a1 Add a triple. Should fix the 64 bit bots.
llvm-svn: 199670
2014-01-20 16:15:48 +00:00
Rafael Espindola a69024aa2a Make the test more strict.
llvm-svn: 199669
2014-01-20 15:58:59 +00:00
NAKAMURA Takumi 6515aef0f7 Whitespace.
llvm-svn: 199667
2014-01-20 15:47:15 +00:00
Aaron Ballman f28e49931a Fixing a typo (turned out to be harmless since the default priority values are the same between the two attributes).
llvm-svn: 199666
2014-01-20 15:22:57 +00:00
Rafael Espindola 63582d605e Remove virtual methods that were added in 2009 and still had 1 implementation.
llvm-svn: 199665
2014-01-20 15:19:43 +00:00
Aaron Ballman 05a63787a2 Since the diagnostics engine understands Attr objects, this code is no longer required -- we can just pass in the attribute directly.
llvm-svn: 199664
2014-01-20 15:06:09 +00:00
Aaron Ballman fc1951c505 Making some minor improvements to r199626.
llvm-svn: 199663
2014-01-20 14:19:44 +00:00
Aaron Ballman 9e013515aa HasFunctionProto is a more strict version of FunctionLike. Since attribute subjects are inclusive (passing a single subject test means no subject-related diagnostic will fire), these two subjects should not be combined.
llvm-svn: 199662
2014-01-20 14:18:40 +00:00
Hal Finkel 9ff54e1fcb Fix misched-aa-colored.ll to require asserts (trying again)
Perhaps it needs to be in caps.

llvm-svn: 199661
2014-01-20 14:15:28 +00:00
Daniel Jasper a55544a658 clang-format: Leave 2 empty lines in Google's JavaScript style.
As per the style guide, two lines are required between top-level
elements.

llvm-svn: 199660
2014-01-20 14:10:30 +00:00
Hal Finkel a6bcadeb4a Fix misched-aa-colored.ll to require asserts.
-misched=shuffle is NDEBUG only. Maybe we should change that.

llvm-svn: 199659
2014-01-20 14:09:34 +00:00
Hal Finkel cd9569c19e Update IR when merging slots in stack coloring
The way that stack coloring updated MMOs when merging stack slots, while
correct, is suboptimal, and is incompatible with the use of AA during
instruction scheduling. The solution, which involves the use of const_cast (and
more importantly, updating the IR from within an MI-level pass), obviously
requires some explanation:

When the stack coloring pass was originally committed, the code in
ScheduleDAGInstrs::buildSchedGraph tracked possible alias sets by using
GetUnderlyingObject, and all load/store and store/store memory control
dependencies where added between SUs at the object level (where only one
object, that returned by GetUnderlyingObject, was used to identify the object
associated with each MMO). When stack coloring merged stack slots, it would
replace MMOs derived from the remapped alloca with the alloca with which the
remapped alloca was being replaced. Because ScheduleDAGInstrs only used single
objects, and tracked alias sets at the object level, this was a fine solution.

In r169744, (Andy and) I updated the code in ScheduleDAGInstrs to use
GetUnderlyingObjects, and track alias sets using, potentially, multiple
underlying objects for each MMO. This was done, primarily, to provide the
ability to look through PHIs, and provide better scheduling for
induction-variable-dependent loads and stores inside loops. At this point, the
MMO-updating code in stack coloring became suboptimal, because it would clear
the MMOs for (i.e. completely pessimize) all instructions for which r169744
might help in scheduling. Updating the IR directly is the simplest fix for this
(and the one with, by far, the least compile-time impact), but others are
possible (we could give each MMO a small vector of potential values, or make
use of a remapping table, constructed from MFI, inside ScheduleDAGInstrs).

Unfortunately, replacing all MMO values derived from the remapped alloca with
the base replacement alloca fundamentally breaks our ability to use AA during
instruction scheduling (which is critical to performance on some targets). The
reason is that the original MMO might have had an offset (either constant or
dynamic) from the base remapped alloca, and that offset is not present in the
updated MMO. One possible way around this would be to use
GetPointerBaseWithConstantOffset, and update not only the MMO's value, but also
its offset based on the original offset. Unfortunately, this solution would
only handle constant offsets, and for safety (because AA is not completely
restricted to deducing relationships with constant offsets), we would need to
clear all MMOs without constant offsets over the entire function. This would be
an even worse pessimization than the current single-object restriction. Any
other solution would involve passing around a vector of remapped allocas, and
teaching AA to use it, introducing additional complexity and overhead into AA.

Instead, when remapping an alloca, we replace all IR uses of that alloca as
well (optionally inserting a bitcast as necessary). This is even more efficient
that the old MMO-updating code in the stack coloring pass (because it removes
the need to call GetUnderlyingObject on all MMO values), removes the
single-object pessimization in the default configuration, and enables the
correct use of AA during instruction scheduling (all without any additional
overhead).

LLVM now no longer miscompiles itself on x86_64 when using -enable-misched
-enable-aa-sched-mi -misched-bottomup=0 -misched-topdown=0 -misched=shuffle!
Fixed PR18497.

Because the alloca replacement is now done at the IR level, unless the MMO
directly refers to the remapped alloca, the change cannot be seen at the MI
level. As a result, there is no good way to fix test/CodeGen/X86/pr14090.ll.

llvm-svn: 199658
2014-01-20 14:03:16 +00:00
Hal Finkel a228a8187b Track multiple stores per object when using AA in ScheduleDAGInstrs
When using AA to break false chain dependencies, we need to track multiple
stores per object in ScheduleDAGInstrs. Historically, we tracked potential alias
chains at the object level, and so all loads of an object would retain
dependencies on any store to that object. With AA, however, this is not
sufficient: non-overlapping stores and loads to the same object all need to be
tested for dependencies separately, we cannot only test all loads to an object
against only the last store (see PR18497 for an explicit example).

To mitigate any unwelcome compile-time impact when not using AA, only one store
is kept in the list per object when not using AA.

This, along with a stack coloring change to come shortly, will provide a test
case, fix PR18497 (and allow LLVM to compile itself using -enable-aa-sched-mi
on x86-64).

llvm-svn: 199657
2014-01-20 14:03:02 +00:00
Evgeniy Stepanov e98f9099a9 [msandr] Access app TLS directly in native exec mode.
In optimized hybrid execution we do not use DynamoRIO private loader, which
mangles TLS access, so we can access the application's TLS directly.

Patch by Qin Zhao.

llvm-svn: 199655
2014-01-20 13:09:29 +00:00
David Woodhouse caaa2850c0 [x86] Fix disassembly of MOV16ao16 et al.
The addition of IC_OPSIZE_ADSIZE in r198759 wasn't quite complete. It
also turns out to have been unnecessary. The disassembler handles the
AdSize prefix for itself, and doesn't care about the difference between
(e.g.) MOV8ao8 and MOB8ao8_16 definitions. So just let them coexist and
don't worry about it.

llvm-svn: 199654
2014-01-20 12:02:53 +00:00
David Woodhouse 9c74fdb8b9 [x86] Fix 16-bit disassembly of JCXZ/JECXZ
llvm-svn: 199653
2014-01-20 12:02:48 +00:00
David Woodhouse 3442f3429e [x86] Rename MOVSD/STOSD/LODSD/OUTSD to MOVSL/STOSL/LODSL/OUTSL
The disassembler has a special case for 'L' vs. 'W' in its heuristic for
checking for 32-bit and 16-bit equivalents. We could expand the heuristic,
but better just to be consistent in using the 'L' suffix.

llvm-svn: 199652
2014-01-20 12:02:44 +00:00
David Woodhouse 70ced3e0b2 [x86] Fix disassembly of callw instruction
Not quite sure why this was marked isAsmParserOnly, but it means that the
disassembler can't see it either.

llvm-svn: 199651
2014-01-20 12:02:40 +00:00
David Woodhouse 5cf4c6750d [x86] Fix 16-bit handling of OpSize bit
When disassembling in 16-bit mode the meaning of the OpSize bit is
inverted. Instructions found in the IC_OPSIZE context will actually
*not* have the 0x66 prefix, and instructions in the IC context will
have the 0x66 prefix. Make use of the existing special-case handling
for the 0x66 prefix being in the wrong place, to cope with this.

llvm-svn: 199650
2014-01-20 12:02:35 +00:00
David Woodhouse 7dd218245c [x86] Infer disassembler mode from SubtargetInfo feature bits
Aside from cleaning up the code, this also adds support for the -code16
environment and actually enables the MODE_16BIT mode that was previously
not accessible.

There is no point adding any testing for 16-bit yet though; basically
nothing will work because we aren't handling the OpSize prefix correctly
for 16-bit mode.

llvm-svn: 199649
2014-01-20 12:02:31 +00:00
David Woodhouse 71d15edaf3 [x86] Support i386-*-*-code16 triple for emitting 16-bit code
llvm-svn: 199648
2014-01-20 12:02:25 +00:00
Chandler Carruth 4d35631a6c [PM] Wire up the Verifier for the new pass manager and connect it to the
various opt verifier commandline options.

Mostly mechanical wiring of the verifier to the new pass manager.
Exercises one of the more unusual aspects of it -- a pass can be either
a module or function pass interchangably. If this is ever problematic,
we can make things more constrained, but for things like the verifier
where there is an "obvious" applicability at both levels, it seems
convenient.

This is the next-to-last piece of basic functionality left to make the
opt commandline driving of the new pass manager minimally functional for
testing and further development. There is still a lot to be done there
(notably the factoring into .def files to kill the current boilerplate
code) but it is relatively uninteresting. The only interesting bit left
for minimal functionality is supporting the registration of analyses.
I'm planning on doing that on top of the .def file switch mostly because
the boilerplate for the analyses would be significantly worse.

llvm-svn: 199646
2014-01-20 11:34:08 +00:00
Kai Nacke e51c813859 ARM: add tlsldo relocation
Add support for the symbol(tlsldo) relocation. This is required in order to 
solve PR18554.

Reviewed by R. Golin, A. Korobeynikov.

llvm-svn: 199644
2014-01-20 11:00:40 +00:00
Bradley Smith 0f28f0cfc1 [ARM] Add ACLE enum/wchar size predefines
llvm-svn: 199642
2014-01-20 10:52:00 +00:00
NAKAMURA Takumi bf6d1efbba [CMake] Apply ADDITIONAL_HEADERS introduced in r199639.
llvm-svn: 199640
2014-01-20 10:24:05 +00:00
NAKAMURA Takumi 6acf320a99 [CMake] llvm_process_sources: Introduce a parameter, ADDITIONAL_HEADERS.
ADDITIONAL_HEADERS is intended to add header files for IDEs as hint.

For example:
  add_llvm_library(LLVMSupport
    Host.cpp
    ADDITIONAL_HEADERS
      Unix/Host.inc
      Windows/Host.inc
    )

llvm-svn: 199639
2014-01-20 10:20:23 +00:00
Artyom Skrobov 10e76a4eda [ARM] Do not generate Tag_DIV_use=AllowDIVExt when hardware div is non-optional: it should have the default value of AllowDIVIfExists
llvm-svn: 199638
2014-01-20 10:18:42 +00:00