OSDN Git Service
Richard Sandiford [Wed, 15 May 2013 12:53:31 +0000 (12:53 +0000)]
[SystemZ] Add more future work items to the README
Based on an analysis by Ulrich Weigand.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181882
91177308-0d34-0410-b5e6-
96231b3b80d8
Richard Sandiford [Wed, 15 May 2013 11:00:31 +0000 (11:00 +0000)]
[SystemZ] Consolidate disassembler tests for valid input into 2 big tests
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181879
91177308-0d34-0410-b5e6-
96231b3b80d8
Richard Sandiford [Wed, 15 May 2013 09:58:19 +0000 (09:58 +0000)]
[SystemZ] Consolidate assembler tests into 4 big tests
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181878
91177308-0d34-0410-b5e6-
96231b3b80d8
Timur Iskhodzhanov [Wed, 15 May 2013 09:00:30 +0000 (09:00 +0000)]
Fix build on Windows
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181873
91177308-0d34-0410-b5e6-
96231b3b80d8
David Blaikie [Wed, 15 May 2013 07:36:59 +0000 (07:36 +0000)]
Use only explicit bool conversion operators
BitVector/SmallBitVector::reference::operator bool remain implicit since
they model more exactly a bool, rather than something else that can be
boolean tested.
The most common (non-buggy) case are where such objects are used as
return expressions in bool-returning functions or as boolean function
arguments. In those cases I've used (& added if necessary) a named
function to provide the equivalent (or sometimes negative, depending on
convenient wording) test.
One behavior change (YAMLParser) was made, though no test case is
included as I'm not sure how to reach that code path. Essentially any
comparison of llvm::yaml::document_iterators would be invalid if neither
iterator was at the end.
This helped uncover a couple of bugs in Clang - test cases provided for
those in a separate commit along with similar changes to `operator bool`
instances in Clang.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181868
91177308-0d34-0410-b5e6-
96231b3b80d8
NAKAMURA Takumi [Wed, 15 May 2013 02:16:23 +0000 (02:16 +0000)]
ELFRelocationEntry::operator<(): Try to stabilize the order. r_offset was insufficient to sort Relocs.
It should fix llvm/test/CodeGen/ARM/ehabi-mc-compact-pr*.ll on some hosts.
RELOCATION RECORDS FOR [.ARM.exidx]:
0 R_ARM_PREL31 .text
0 R_ARM_NONE __aeabi_unwind_cpp_pr0
FIXME: I am not sure of the directions of extra comparators, in Type and Index.
For now, they are different from the direction in r_offset.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181864
91177308-0d34-0410-b5e6-
96231b3b80d8
Arnold Schwaighofer [Wed, 15 May 2013 02:02:45 +0000 (02:02 +0000)]
LoopVectorize: Fix comments
No functionality change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181862
91177308-0d34-0410-b5e6-
96231b3b80d8
Arnold Schwaighofer [Wed, 15 May 2013 01:44:30 +0000 (01:44 +0000)]
LoopVectorize: Hoist conditional loads if possible
InstCombine can be uncooperative to vectorization and sink loads into
conditional blocks. This prevents vectorization.
Undo this optimization if there are unconditional memory accesses to the same
addresses in the loop.
radar://
13815763
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181860
91177308-0d34-0410-b5e6-
96231b3b80d8
Jakob Stoklund Olesen [Tue, 14 May 2013 23:45:56 +0000 (23:45 +0000)]
Speed up Value::isUsedInBasicBlock() for long use lists.
This is expanding Ben's original heuristic for short basic blocks to
also work for longer basic blocks and huge use lists.
Scan the basic block and the use list in parallel, terminating the
search when the shorter list ends. In almost all cases, either the basic
block or the use list is short, and the function returns quickly.
In one crazy test case with very long use chains, CodeGenPrepare runs
400x faster. When compiling ARMDisassembler.cpp it is 5x faster.
<rdar://problem/
13840497>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181851
91177308-0d34-0410-b5e6-
96231b3b80d8
Sylvestre Ledru [Tue, 14 May 2013 23:36:24 +0000 (23:36 +0000)]
Fix two typo
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181848
91177308-0d34-0410-b5e6-
96231b3b80d8
NAKAMURA Takumi [Tue, 14 May 2013 23:05:00 +0000 (23:05 +0000)]
ExceptionDemo: Corresponding to r181820, SectionMemoryManager should belong to RTDyldMemoryManager.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181844
91177308-0d34-0410-b5e6-
96231b3b80d8
Ahmed Bougacha [Tue, 14 May 2013 22:41:29 +0000 (22:41 +0000)]
Object: Fix Mach-O relocation printing.
There were two problems that made llvm-objdump -r crash:
- for non-scattered relocations, the symbol/section index is actually in the
(aptly named) symbolnum field.
- sections are 1-indexed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181843
91177308-0d34-0410-b5e6-
96231b3b80d8
Arnold Schwaighofer [Tue, 14 May 2013 22:33:24 +0000 (22:33 +0000)]
ARM ISel: Don't create illegal types during LowerMUL
The transformation happening here is that we want to turn a
"mul(ext(X), ext(X))" into a "vmull(X, X)", stripping off the extension. We have
to make sure that X still has a valid vector type - possibly recreate an
extension to a smaller type. In case of a extload of a memory type smaller than
64 bit we used create a ext(load()). The problem with doing this - instead of
recreating an extload - is that an illegal type is exposed.
This patch fixes this by creating extloads instead of ext(load()) sequences.
Fixes PR15970.
radar://
13871383
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181842
91177308-0d34-0410-b5e6-
96231b3b80d8
Manman Ren [Tue, 14 May 2013 21:52:44 +0000 (21:52 +0000)]
GlobalOpt: fix an issue where CXAAtExitFn points to a deleted function.
CXAAtExitFn was set outside a loop and before optimizations where functions
can be deleted. This patch will set CXAAtExitFn inside the loop and after
optimizations.
Seg fault when running LTO because of accesses to a deleted function.
rdar://problem/
13838828
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181838
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Tue, 14 May 2013 21:52:01 +0000 (21:52 +0000)]
Revert previous patch, it's actually on under Wall.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181837
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Tue, 14 May 2013 21:49:38 +0000 (21:49 +0000)]
Add -Wreorder to the list of C++ warnings.
This built clean with clang, but if we see false positives on the bots
then we'll revert and turn it into a compiler specific check.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181836
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Tue, 14 May 2013 21:33:10 +0000 (21:33 +0000)]
Make getCompileUnit non-const and return the current DIE if it
happens to be a compile unit. Noticed on inspection and tested
via calling on a newly created compile unit. No functional change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181835
91177308-0d34-0410-b5e6-
96231b3b80d8
Michael Liao [Tue, 14 May 2013 20:34:12 +0000 (20:34 +0000)]
Add 'CHECK-DAG' support
Refer to 'FileCheck.rst'f for details of 'CHECK-DAG'.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181827
91177308-0d34-0410-b5e6-
96231b3b80d8
Michael Liao [Tue, 14 May 2013 20:29:52 +0000 (20:29 +0000)]
Refactor string checking. No functionality change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181824
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Schmidt [Tue, 14 May 2013 19:35:45 +0000 (19:35 +0000)]
Implement the PowerPC system call (sc) instruction.
Instruction added at request of Roman Divacky. Tested via asm-parser.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181821
91177308-0d34-0410-b5e6-
96231b3b80d8
Filip Pizlo [Tue, 14 May 2013 19:29:00 +0000 (19:29 +0000)]
SectionMemoryManager shouldn't be a JITMemoryManager. Previously, the
EngineBuilder interface required a JITMemoryManager even if it was being used
to construct an MCJIT. But the MCJIT actually wants a RTDyldMemoryManager.
Consequently, the SectionMemoryManager, which is meant for MCJIT, derived
from the JITMemoryManager and then stubbed out a bunch of JITMemoryManager
methods that weren't relevant to the MCJIT.
This patch fixes the situation: it teaches the EngineBuilder that
RTDyldMemoryManager is a supertype of JITMemoryManager, and that it's
appropriate to pass a RTDyldMemoryManager instead of a JITMemoryManager if
we're using the MCJIT. This allows us to remove the stub methods from
SectionMemoryManager, and make SectionMemoryManager a direct subtype of
RTDyldMemoryManager.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181820
91177308-0d34-0410-b5e6-
96231b3b80d8
Jyotsna Verma [Tue, 14 May 2013 18:54:06 +0000 (18:54 +0000)]
Hexagon: Pass to replace tranfer/copy instructions into combine instruction
where possible.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181817
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Tue, 14 May 2013 18:33:40 +0000 (18:33 +0000)]
Reapply "Subtract isn't commutative, fix this for MMX psub." with
a somewhat randomly chosen cpu that will minimize cpu specific
differences on bots.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181814
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Tue, 14 May 2013 18:20:42 +0000 (18:20 +0000)]
Temporarily revert "Subtract isn't commutative, fix this for MMX psub."
It's causing failures on the atom bot.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181812
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Tue, 14 May 2013 18:06:14 +0000 (18:06 +0000)]
Fix __clear_cache declaration.
This fixes the build with gcc in gnu++98 and gnu++11 mode.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181811
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Tue, 14 May 2013 17:52:05 +0000 (17:52 +0000)]
Subtract isn't commutative, fix this for MMX psub.
Patch by Andrea DiBiagio.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181809
91177308-0d34-0410-b5e6-
96231b3b80d8
Jakob Stoklund Olesen [Tue, 14 May 2013 17:47:27 +0000 (17:47 +0000)]
Recognize sparc64 as an alias for sparcv9 triples.
Patch by Brad Smith!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181808
91177308-0d34-0410-b5e6-
96231b3b80d8
Jyotsna Verma [Tue, 14 May 2013 17:16:38 +0000 (17:16 +0000)]
Hexagon: Add patterns to generate 'combine' instructions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181805
91177308-0d34-0410-b5e6-
96231b3b80d8
Jyotsna Verma [Tue, 14 May 2013 16:36:34 +0000 (16:36 +0000)]
Hexagon: ArePredicatesComplement should not restrict itself to TFRs.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181803
91177308-0d34-0410-b5e6-
96231b3b80d8
Kai Nacke [Tue, 14 May 2013 16:30:51 +0000 (16:30 +0000)]
Add bitcast to store of personality function.
The personality function is user defined and may have an arbitrary result type.
The code assumes always i8*. This results in an assertion failure if a different
type is used. A bitcast to i8* is added to prevent this failure.
Reviewed by: Renato Golin, Bob Wilson
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181802
91177308-0d34-0410-b5e6-
96231b3b80d8
Derek Schuff [Tue, 14 May 2013 16:26:38 +0000 (16:26 +0000)]
Fix ARM FastISel tests, as a first step to enabling ARM FastISel
ARM FastISel is currently only enabled for iOS non-Thumb1, and I'm working on
enabling it for other targets. As a first step I've fixed some of the tests.
Changes to ARM FastISel tests:
- Different triples don't generate the same relocations (especially
movw/movt versus constant pool loads). Use a regex to allow either.
- Mangling is different. Use a regex to allow either.
- The reserved registers are sometimes different, so registers get
allocated in a different order. Capture the names only where this
occurs.
- Add -verify-machineinstrs to some tests where it works. It doesn't
work everywhere it should yet.
- Add -fast-isel-abort to many tests that didn't have it before.
- Split out the VarArg test from fast-isel-call.ll into its own
test. This simplifies test setup because of --check-prefix.
Patch by JF Bastien
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181801
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Schmidt [Tue, 14 May 2013 16:08:32 +0000 (16:08 +0000)]
PPC32: Fix stack collision between FP and CR save areas.
The changes to CR spill handling missed a case for 32-bit PowerPC.
The code in PPCFrameLowering::processFunctionBeforeFrameFinalized()
checks whether CR spill has occurred using a flag in the function
info. This flag is only set by storeRegToStackSlot and
loadRegFromStackSlot. spillCalleeSavedRegisters does not call
storeRegToStackSlot, but instead produces MI directly. Thus we don't
see the CR is spilled when assigning frame offsets, and the CR spill
ends up colliding with some other location (generally the FP slot).
This patch sets the flag in spillCalleeSavedRegisters for PPC32 so
that the CR spill is properly detected and gets its own slot in the
stack frame.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181800
91177308-0d34-0410-b5e6-
96231b3b80d8
Jyotsna Verma [Tue, 14 May 2013 15:50:49 +0000 (15:50 +0000)]
Hexagon: Test case to check if branch probabilities are properly reflected in
the jump instructions in the form of taken/not-taken hint.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181799
91177308-0d34-0410-b5e6-
96231b3b80d8
Jyotsna Verma [Tue, 14 May 2013 15:33:27 +0000 (15:33 +0000)]
Hexagon: Remove dead-code after unconditional return from addPreSched2.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181797
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Tue, 14 May 2013 14:42:56 +0000 (14:42 +0000)]
R600/SI: Add processor type for Hainan asic
Patch by: Alex Deucher
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
NOTE: This is a candidate for the 3.3 branch.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181792
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Tue, 14 May 2013 13:29:16 +0000 (13:29 +0000)]
Get the unittests compiling when building with cmake and the setting
-DLLVM_ENABLE_THREADS=false.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181788
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Tue, 14 May 2013 13:02:37 +0000 (13:02 +0000)]
Declare __clear_cache.
GCC declares __clear_cache in the gnu modes (-std=gnu++98,
-std=gnu++11), but not in the strict modes (-std=c++98, -std=c++11). This patch
declares it and therefore fixes the build when using one of the strict modes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181785
91177308-0d34-0410-b5e6-
96231b3b80d8
Richard Sandiford [Tue, 14 May 2013 10:17:52 +0000 (10:17 +0000)]
[SystemZ] Add disassembler support
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181777
91177308-0d34-0410-b5e6-
96231b3b80d8
Michel Danzer [Tue, 14 May 2013 09:53:30 +0000 (09:53 +0000)]
R600/SI: Add lit test coverage for the remaining patterns added recently
Reviewed-by: Christian König <christian.koenig@amd.com>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181775
91177308-0d34-0410-b5e6-
96231b3b80d8
Richard Sandiford [Tue, 14 May 2013 09:49:11 +0000 (09:49 +0000)]
[SystemZ] Add extra testscases for r181773
Forgot to svn add these...
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181774
91177308-0d34-0410-b5e6-
96231b3b80d8
Richard Sandiford [Tue, 14 May 2013 09:47:26 +0000 (09:47 +0000)]
[SystemZ] Rework handling of constant PC-relative operands
The GNU assembler treats things like:
brasl %r14, 100
in the same way as:
brasl %r14, .+100
rather than as a branch to absolute address 100. We implemented this in
LLVM by creating an immediate operand rather than the usual expr operand,
and by handling immediate operands specially in the code emitter.
This was undesirable for (at least) three reasons:
- the specialness of immediate operands was exposed to the backend MC code,
rather than being limited to the assembler parser.
- in disassembly, an immediate operand really is an absolute address.
(Note that this means reassembling printed disassembly can't recreate
the original code.)
- it would interfere with any assembly manipulation that we might
try in future. E.g. operations like branch shortening can change
the relative position of instructions, but any code that updates
sym+offset addresses wouldn't update an immediate "100" operand
in the same way as an explicit ".+100" operand.
This patch changes the implementation so that the assembler creates
a "." label for immediate PC-relative operands, so that the operand
to the MCInst is always the absolute address. The patch also adds
some error checking of the offset.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181773
91177308-0d34-0410-b5e6-
96231b3b80d8
Richard Sandiford [Tue, 14 May 2013 09:38:07 +0000 (09:38 +0000)]
[SystemZ] Remove bogus isAsmParserOnly
Marking instructions as isAsmParserOnly stops them from being disassembled.
However, in cases where separate asm and codegen versions exist, we actually
want to disassemble to the asm ones.
No functional change intended.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181772
91177308-0d34-0410-b5e6-
96231b3b80d8
Richard Sandiford [Tue, 14 May 2013 09:36:44 +0000 (09:36 +0000)]
[SystemZ] Match operands to fields by name rather than by order
The SystemZ port currently relies on the order of the instruction operands
matching the order of the instruction field lists. This isn't desirable
for disassembly, where the two are matched only by name. E.g. the R1 and R2
fields of an RR instruction should have corresponding R1 and R2 operands.
The main complication is that addresses are compound operands,
and as far as I know there is no mechanism to allow individual
suboperands to be selected by name in "let Inst{...} = ..." assignments.
Luckily it doesn't really matter though. The SystemZ instruction
encoding groups all address fields together in a predictable order,
so it's just as valid to see the entire compound address operand as
a single field. That's the approach taken in this patch.
Matching by name in turn means that the operands to COPY SIGN and
CONVERT TO FIXED instructions can be given in natural order.
(It was easier to do this at the same time as the rename,
since otherwise the intermediate step was too confusing.)
No functional change intended.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181771
91177308-0d34-0410-b5e6-
96231b3b80d8
Richard Sandiford [Tue, 14 May 2013 09:28:21 +0000 (09:28 +0000)]
[SystemZ] Match operands to fields by name rather than by order
The SystemZ port currently relies on the order of the instruction operands
matching the order of the instruction field lists. This isn't desirable
for disassembly, where the two are matched only by name. E.g. the R1 and R2
fields of an RR instruction should have corresponding R1 and R2 operands.
The main complication is that addresses are compound operands,
and as far as I know there is no mechanism to allow individual
suboperands to be selected by name in "let Inst{...} = ..." assignments.
Luckily it doesn't really matter though. The SystemZ instruction
encoding groups all address fields together in a predictable order,
so it's just as valid to see the entire compound address operand as
a single field. That's the approach taken in this patch.
Matching by name in turn means that the operands to COPY SIGN and
CONVERT TO FIXED instructions can be given in natural order.
(It was easier to do this at the same time as the rename,
since otherwise the intermediate step was too confusing.)
No functional change intended.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181769
91177308-0d34-0410-b5e6-
96231b3b80d8
Michael Gottesman [Tue, 14 May 2013 06:40:10 +0000 (06:40 +0000)]
Removed trailing whitespace.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181760
91177308-0d34-0410-b5e6-
96231b3b80d8
Reed Kotler [Tue, 14 May 2013 06:00:01 +0000 (06:00 +0000)]
Fix typo.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181759
91177308-0d34-0410-b5e6-
96231b3b80d8
Reed Kotler [Tue, 14 May 2013 02:13:45 +0000 (02:13 +0000)]
Removed an unnamed namespace and forgot to make two of the functions inside
"static".
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181754
91177308-0d34-0410-b5e6-
96231b3b80d8
Reed Kotler [Tue, 14 May 2013 02:00:24 +0000 (02:00 +0000)]
This is the first of three patches which creates stubs used for
Mips16/32 floating point interoperability.
When Mips16 code calls external functions that would normally have some
of its parameters or return values passed in floating point registers,
it needs (Mips32) helper functions to do this because while in Mips16 mode
there is no ability to access the floating point registers.
In Pic mode, this is done with a set of predefined functions in libc.
This case is already handled in llvm for Mips16.
In static relocation mode, for efficiency reasons, the compiler generates
stubs that the linker will use if it turns out that the external function
is a Mips32 function. (If it's Mips16, then it does not need the helper
stubs).
These stubs are identically named and the linker knows about these tricks
and will not create multiple copies and will delete them if they are not
needed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181753
91177308-0d34-0410-b5e6-
96231b3b80d8
Akira Hatanaka [Tue, 14 May 2013 01:42:44 +0000 (01:42 +0000)]
StackColoring: don't clear an instruction's mem operand if the underlying
object is a PseudoSourceValue and PseudoSourceValue::isConstant returns true (i.e.,
points to memory that has a constant value).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181751
91177308-0d34-0410-b5e6-
96231b3b80d8
David Blaikie [Tue, 14 May 2013 00:35:19 +0000 (00:35 +0000)]
Assert that DIEEntries are constructed with non-null DIEs
This just brings a crash a little further forward from DWARF emission to
DIE construction to make errors easier to diagnose.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181748
91177308-0d34-0410-b5e6-
96231b3b80d8
Arnold Schwaighofer [Tue, 14 May 2013 00:21:18 +0000 (00:21 +0000)]
LoopVectorize: Handle loops with multiple forward inductions
We used to give up if we saw two integer inductions. After this patch, we base
further induction variables on the chosen one like we do in the reverse
induction and pointer induction case.
Fixes PR15720.
radar://
13851975
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181746
91177308-0d34-0410-b5e6-
96231b3b80d8
Michael Gottesman [Tue, 14 May 2013 00:08:09 +0000 (00:08 +0000)]
[objc-arc-opts] Added debug statements when we set and unset whether a pointer is known positive.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181745
91177308-0d34-0410-b5e6-
96231b3b80d8
Michael Gottesman [Mon, 13 May 2013 23:49:42 +0000 (23:49 +0000)]
[objc-arc-opts] In the presense of an alloca unconditionally remove RR pairs if and only if we are both KnownSafeBU/KnownSafeTD rather than just either or.
In the presense of a block being initialized, the frontend will emit the
objc_retain on the original pointer and the release on the pointer loaded from
the alloca. The optimizer will through the provenance analysis realize that the
two are related (albiet different), but since we only require KnownSafe in one
direction, will match the inner retain on the original pointer with the guard
release on the original pointer. This is fixed by ensuring that in the presense
of allocas we only unconditionally remove pointers if both our retain and our
release are KnownSafe (i.e. we are KnownSafe in both directions) since we must
deal with the possibility that the frontend will emit what (to the optimizer)
appears to be unbalanced retain/releases.
An example of the miscompile is:
%A = alloca
retain(%x)
retain(%x) <--- Inner Retain
store %x, %A
%y = load %A
... DO STUFF ...
release(%y)
call void @use(%x)
release(%x) <--- Guarding Release
getting optimized to:
%A = alloca
retain(%x)
store %x, %A
%y = load %A
... DO STUFF ...
release(%y)
call void @use(%x)
rdar://
13750319
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181743
91177308-0d34-0410-b5e6-
96231b3b80d8
Matt Beaumont-Gay [Mon, 13 May 2013 21:10:49 +0000 (21:10 +0000)]
Move a couple more statistics inside '#ifndef NDEBUG'.
Suppresses an unused-variable warning in -Asserts builds.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181733
91177308-0d34-0410-b5e6-
96231b3b80d8
Jack Carter [Mon, 13 May 2013 20:26:46 +0000 (20:26 +0000)]
Mips assembler: Assembler macro ADDIU $rs,imm
This patch adds alias for addiu instruction which enables following syntax:
addiu $rs,imm
The macro is translated as:
addiu $rs,$rs,imm
Contributer: Vladimir Medic
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181729
91177308-0d34-0410-b5e6-
96231b3b80d8
Michael Gottesman [Mon, 13 May 2013 19:40:39 +0000 (19:40 +0000)]
[objc-arc-opts] Add comment to BBState making it clear that get{TopDown,BottomUp}PtrState will create a new PtrState object if it does not find a PtrState for Arg.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181726
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Schmidt [Mon, 13 May 2013 19:40:36 +0000 (19:40 +0000)]
Fix goofy commentary in PPCTargetObjectFile.cpp.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181725
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Schmidt [Mon, 13 May 2013 19:34:37 +0000 (19:34 +0000)]
PPC64: Constant initializers with dynamic relocations go in .data.rel.ro.
This fixes warning messages observed in the oggenc application test in
projects/test-suite. Special handling is needed for the 64-bit
PowerPC SVR4 ABI when a constant is initialized with a pointer to a
function in a shared library. Because a function address is
implemented as the address of a function descriptor, the use of copy
relocations can lead to problems with initialization. GNU ld
therefore replaces copy relocations with dynamic relocations to be
resolved by the dynamic linker. This means the constant cannot reside
in the read-only data section, but instead belongs in .data.rel.ro,
which is designed for constants containing dynamic relocations.
The implementation creates a class PPC64LinuxTargetObjectFile
inheriting from TargetLoweringObjectFileELF, which behaves like its
parent except to place constants of this sort into .data.rel.ro.
The test case is reduced from the oggenc application.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181723
91177308-0d34-0410-b5e6-
96231b3b80d8
Bob Wilson [Mon, 13 May 2013 19:02:31 +0000 (19:02 +0000)]
Remove redundant variable introduced by r181682.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181721
91177308-0d34-0410-b5e6-
96231b3b80d8
Michael Gottesman [Mon, 13 May 2013 18:29:07 +0000 (18:29 +0000)]
[objc-arc] Move the before optimization statistics gathering phase out of OptimizeIndividualCalls.
This makes the statistics gathering completely independent of the actual
optimization occuring, preventing any sort of bleeding over from occuring.
Additionally, it simplifies a switch statement in the non-statistic gathering case.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181719
91177308-0d34-0410-b5e6-
96231b3b80d8
Akira Hatanaka [Mon, 13 May 2013 18:23:35 +0000 (18:23 +0000)]
[mips] Add option -mno-ldc1-sdc1.
This option is used when the user wants to avoid emitting double precision FP
loads and stores. Double precision FP loads and stores are expanded to single
precision instructions after register allocation.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181718
91177308-0d34-0410-b5e6-
96231b3b80d8
Shuxin Yang [Mon, 13 May 2013 18:03:12 +0000 (18:03 +0000)]
Fix a bug that APFloat::fusedMultiplyAdd() mistakenly evaluate "14.5f * -14.5f + 225.0f" to 225.0f.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181715
91177308-0d34-0410-b5e6-
96231b3b80d8
Akira Hatanaka [Mon, 13 May 2013 17:57:42 +0000 (17:57 +0000)]
[mips] Define a helper function which creates an instruction with the same
operands as the prototype instruction but with a different opcode.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181714
91177308-0d34-0410-b5e6-
96231b3b80d8
Akira Hatanaka [Mon, 13 May 2013 17:43:19 +0000 (17:43 +0000)]
[mips] Rename functions. No functionality changes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181713
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Mon, 13 May 2013 14:34:48 +0000 (14:34 +0000)]
Remove unused fields and arguments.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181706
91177308-0d34-0410-b5e6-
96231b3b80d8
Mihai Popa [Mon, 13 May 2013 14:10:04 +0000 (14:10 +0000)]
The purpose of the patch is to fix the syntax of ARM mrc and mrc2 instructions when they are used to write to the APSR. In this case, the destination operand should be APSR_nzcv, and the encoding of the target should be 0b1111 (same as for PC). In pre-UAL syntax, this form used the PC register as a textual target. This is still allowed for backward compatibility.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181705
91177308-0d34-0410-b5e6-
96231b3b80d8
Lang Hames [Mon, 13 May 2013 10:21:19 +0000 (10:21 +0000)]
Correctly preserve the input chain for potential tailcall nodes whose
return values are bitcasts.
The chain had previously been being clobbered with the entry node to
the dag, which sometimes caused other code in the function to be
erroneously deleted when tailcall optimization kicked in.
<rdar://problem/
13827621>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181696
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 13 May 2013 07:50:47 +0000 (07:50 +0000)]
Suppress GCC compiler warnings in release builds about variables that are only
read in asserts.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181689
91177308-0d34-0410-b5e6-
96231b3b80d8
Nadav Rotem [Mon, 13 May 2013 05:13:13 +0000 (05:13 +0000)]
SLPVectorizer: Swap LHS and RHS. No functionality change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181684
91177308-0d34-0410-b5e6-
96231b3b80d8
Hao Liu [Mon, 13 May 2013 02:07:05 +0000 (02:07 +0000)]
Fix PR15950 A bug in DAG Combiner about undef mask
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181682
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Mon, 13 May 2013 01:16:13 +0000 (01:16 +0000)]
Remove the MachineMove class.
It was just a less powerful and more confusing version of
MCCFIInstruction. A side effect is that, since MCCFIInstruction uses
dwarf register numbers, calls to getDwarfRegNum are pushed out, which
should allow further simplifications.
I left the MachineModuleInfo::addFrameMove interface unchanged since
this patch was already fairly big.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181680
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Mon, 13 May 2013 00:18:24 +0000 (00:18 +0000)]
XFAIL this test for mingw too.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181678
91177308-0d34-0410-b5e6-
96231b3b80d8
Nadav Rotem [Sun, 12 May 2013 22:58:45 +0000 (22:58 +0000)]
SLPVectorizer: Fix a bug in the code that generates extracts for values with multiple users.
The external user does not have to be in lane #0. We have to save the lane for each scalar so that we know which vector lane to extract.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181674
91177308-0d34-0410-b5e6-
96231b3b80d8
Nadav Rotem [Sun, 12 May 2013 22:55:57 +0000 (22:55 +0000)]
SLPVectorizer: Clear the map that maps between scalars to vectors after each round of vectorization.
Testcase in the next commit.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181673
91177308-0d34-0410-b5e6-
96231b3b80d8
David Majnemer [Sun, 12 May 2013 00:07:05 +0000 (00:07 +0000)]
InstCombine: Flip the order of two urem transforms
There are two transforms in visitUrem that conflict with each other.
*) One, if a divisor is a power of two, subtracts one from the divisor
and turns it into a bitwise-and.
*) The other unwraps both operands if they are surrounded by zext
instructions.
Flipping the order allows the subtraction to go beneath the sign
extension.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181668
91177308-0d34-0410-b5e6-
96231b3b80d8
Arnold Schwaighofer [Sat, 11 May 2013 23:04:28 +0000 (23:04 +0000)]
LoopVectorize: Use the widest induction variable type
Use the widest induction type encountered for the cannonical induction variable.
We used to turn the following loop into an empty loop because we used i8 as
induction variable type and truncated 1024 to 0 as trip count.
int a[1024];
void fail() {
int reverse_induction = 1023;
unsigned char forward_induction = 0;
while ((reverse_induction) >= 0) {
forward_induction++;
a[reverse_induction] = forward_induction;
--reverse_induction;
}
}
radar://
13862901
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181667
91177308-0d34-0410-b5e6-
96231b3b80d8
Arnold Schwaighofer [Sat, 11 May 2013 23:04:26 +0000 (23:04 +0000)]
LoopVectorize: Use variable instead of repeated function call
No functionality change intended.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181666
91177308-0d34-0410-b5e6-
96231b3b80d8
Arnold Schwaighofer [Sat, 11 May 2013 23:04:24 +0000 (23:04 +0000)]
LoopVectorize: Use IRBuilder interface in more places
No functionality change intended.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181665
91177308-0d34-0410-b5e6-
96231b3b80d8
David Blaikie [Sat, 11 May 2013 19:14:10 +0000 (19:14 +0000)]
Correct parameter name in doc comment to match declaration.
(review feedback on r181632 from Dmitri)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181664
91177308-0d34-0410-b5e6-
96231b3b80d8
Benjamin Kramer [Sat, 11 May 2013 18:24:28 +0000 (18:24 +0000)]
StringRefize some debug accel table bits.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181663
91177308-0d34-0410-b5e6-
96231b3b80d8
David Majnemer [Sat, 11 May 2013 09:01:28 +0000 (09:01 +0000)]
InstCombine: Turn urem to bitwise-and more often
Use isKnownToBeAPowerOfTwo in visitUrem so that we may more aggressively
fold away urem instructions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181661
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Sat, 11 May 2013 02:38:11 +0000 (02:38 +0000)]
Change getFrameMoves to return a const reference.
To add a frame now there is a dedicated addFrameMove which also takes
care of constructing the move itself.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181657
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Sat, 11 May 2013 02:24:41 +0000 (02:24 +0000)]
Remove more dead code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181656
91177308-0d34-0410-b5e6-
96231b3b80d8
Reed Kotler [Sat, 11 May 2013 01:02:20 +0000 (01:02 +0000)]
Add -mtriple=mipsel-linux-gnu to the test so that the compiler does
not think it can support small data sections.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181654
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Fri, 10 May 2013 23:34:51 +0000 (23:34 +0000)]
Remove dead code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181649
91177308-0d34-0410-b5e6-
96231b3b80d8
Nadav Rotem [Fri, 10 May 2013 22:59:33 +0000 (22:59 +0000)]
SLPVectorizer: Add support for trees with external users.
For example:
bar() {
int a = A[i];
int b = A[i+1];
B[i] = a;
B[i+1] = b;
foo(a); <--- a is used outside the vectorized expression.
}
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181648
91177308-0d34-0410-b5e6-
96231b3b80d8
Nadav Rotem [Fri, 10 May 2013 22:56:18 +0000 (22:56 +0000)]
Add a debug print
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181647
91177308-0d34-0410-b5e6-
96231b3b80d8
Nadav Rotem [Fri, 10 May 2013 22:55:44 +0000 (22:55 +0000)]
Add an additional testcase for PR15882.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181646
91177308-0d34-0410-b5e6-
96231b3b80d8
Reed Kotler [Fri, 10 May 2013 22:25:39 +0000 (22:25 +0000)]
Checkin in of first of several patches to finish implementation of
mips16/mips32 floating point interoperability.
This patch fixes returns from mips16 functions so that if the function
was in fact called by a mips32 hard float routine, then values
that would have been returned in floating point registers are so returned.
Mips16 mode has no floating point instructions so there is no way to
load values into floating point registers.
This is needed when returning float, double, single complex, double complex
in the Mips ABI.
Helper functions in libc for mips16 are available to do this.
For efficiency purposes, these helper functions have a different calling
convention from normal Mips calls.
Registers v0,v1,a0,a1 are used to pass parameters instead of
a0,a1,a2,a3.
This is because v0,v1,a0,a1 are the natural registers used to return
floating point values in soft float. These values can then be moved
to the appropriate floating point registers with no extra cost.
The only register that is modified is ra in this call.
The helper functions make sure that the return values are in the floating
point registers that they would be in if soft float was not in effect
(which it is for mips16, though the soft float is implemented using a mips32
library that uses hard float).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181641
91177308-0d34-0410-b5e6-
96231b3b80d8
Jordan Rose [Fri, 10 May 2013 22:17:10 +0000 (22:17 +0000)]
Micro-optimization: don't shift an entire bitcode record over to get the code.
Previously, BitstreamCursor read an abbreviated record by splatting the
whole thing into a data vector, then extracting and removing the /first/
element. Now, it reads the first element--the record code--separately from
the actual field values.
No (intended) functionality change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181639
91177308-0d34-0410-b5e6-
96231b3b80d8
David Blaikie [Fri, 10 May 2013 22:14:39 +0000 (22:14 +0000)]
Give the test from r181632 a target triple.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181637
91177308-0d34-0410-b5e6-
96231b3b80d8
David Blaikie [Fri, 10 May 2013 21:52:07 +0000 (21:52 +0000)]
PR14492: Debug Info: Support for values of non-integer non-type template parameters.
This is only tested for global variables at the moment (& includes tests
for the unnamed parameter case, since apparently this entire function
was completely untested previously)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181632
91177308-0d34-0410-b5e6-
96231b3b80d8
Jyotsna Verma [Fri, 10 May 2013 21:44:02 +0000 (21:44 +0000)]
Fix unused variable error.
Earlier, this variable was used in an assert and was causing failure on
darwin.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181630
91177308-0d34-0410-b5e6-
96231b3b80d8
Jyotsna Verma [Fri, 10 May 2013 20:58:11 +0000 (20:58 +0000)]
Hexagon: Fix switch statements in GetDotOldOp and IsNewifyStore.
No functionality change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181628
91177308-0d34-0410-b5e6-
96231b3b80d8
Jyotsna Verma [Fri, 10 May 2013 20:27:34 +0000 (20:27 +0000)]
Hexagon: Fix switch cases in HexagonVLIWPacketizer.cpp.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181624
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Fri, 10 May 2013 18:31:42 +0000 (18:31 +0000)]
Fix the R600 build.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181621
91177308-0d34-0410-b5e6-
96231b3b80d8
Chad Rosier [Fri, 10 May 2013 18:24:17 +0000 (18:24 +0000)]
[ms-inline asm] Fix a crasher when we fail on a direct match.
The issue was that the MatchingInlineAsm and VariantID args to the
MatchInstructionImpl function weren't being set properly. Specifically, when
parsing intel syntax, the parser thought it was parsing inline assembly in the
at&t dialect; that will never be the case.
The crash was caused when the emitter tried to emit the instruction, but the
operands weren't set. When parsing inline assembly we only set the opcode, not
the operands, which is used to lookup the instruction descriptor.
rdar://
13854391 and PR15945
Also, this commit reverts r176036. Now that we're correctly parsing the intel
syntax the pushad/popad don't match properly. I've reimplemented that fix using
a MnemonicAlias.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181620
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Fri, 10 May 2013 18:16:59 +0000 (18:16 +0000)]
Remove unused argument.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181618
91177308-0d34-0410-b5e6-
96231b3b80d8
Andrew Kaylor [Fri, 10 May 2013 17:58:41 +0000 (17:58 +0000)]
Fix MCJITCAPITest.cpp unit test on Windows.
MCJIT on Windows requires an explicit target triple with "-elf" appended to generate objects in ELF format. The common test framework was setting up this triple, but it wasn't passed to the C API in the test.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181614
91177308-0d34-0410-b5e6-
96231b3b80d8
Alexander Kornienko [Fri, 10 May 2013 17:15:51 +0000 (17:15 +0000)]
Better output for long help strings for command-line options.
Summary:
This patch allows using \n inside long help strings for command-line
options, so that all lines are equally indented. This is not a perfect solution,
as we don't (and probably don't want to) know about terminal width, but it
allows to format long help strings somehow readable without manually padding
them with spaces. A motivating example is -help output from clang-format (source
code in tools/clang-format/ClangFormat.cpp, see cl options offset, length,
style, and dump-config).
Reviewers: atrick, alexfh
Reviewed By: alexfh
CC: llvm-commits, rafael
Differential Revision: http://llvm-reviews.chandlerc.com/D779
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181608
91177308-0d34-0410-b5e6-
96231b3b80d8