Alex Lorenz [Fri, 21 Aug 2015 21:12:44 +0000 (21:12 +0000)]
MIR Serialization: Print MCSymbol operands.
This commit allows the MIR printer to print the MCSymbol machine operands.
Unfortunately they can't be parsed at this time. I will create a bug that will
track the fact that the MCSymbol operands can't be parsed yet.
Sanjay Patel [Fri, 21 Aug 2015 20:17:26 +0000 (20:17 +0000)]
[x86] invert logic for attribute 'FeatureFastUAMem'
This is a 'no functional change intended' patch. It removes one FIXME, but adds several more.
Motivation: the FeatureFastUAMem attribute may be too general. It is used to determine if any
sized misaligned memory access under 32-bytes is 'fast'. From the added FIXME comments, however,
you can see that we're not consistent about this. Changing the name of the attribute makes it
clearer to see the logic holes.
Changing this to a 'slow' attribute also means we don't have to add an explicit 'fast' attribute
to new chips; fast unaligned accesses have been standard for several generations of CPUs now.
David Blaikie [Fri, 21 Aug 2015 20:16:51 +0000 (20:16 +0000)]
[opaque pointer type]: Pass explicit pointee type when building a constant GEP.
Gets a bit tricky in the ValueMapper, of course - not sure if we should
just expose a list of explicit types for each Value so that the
ValueMapper can be neutral to these special cases (it's OK for things
like load, where the explicit type is the result type - but when that's
not the case, it means plumbing through another "special" type... )
Dan Liew [Fri, 21 Aug 2015 18:10:57 +0000 (18:10 +0000)]
Filter libraries that are not installed out of CMake exports (currently
gtest and gtest_main) when generating ``Makefile.llvmbuild``.
Libraries that are not installed should not be exported because they
won't be available from an install tree. Rather than filtering out the
gtest libraries in cmake/modules/Makefile, simply teach llvm-build to
filter out libraries that will not be installed from its generated list
of exported libraries.
Note that LLVMBUILD_LIB_DEPS_* are used during our own CMake build
process so we cannot filter LLVMBUILD_LIB_DEPS_gtest* out in llvm-build.
We must leave this gtest filter logic in cmake/modules/Makefile.
Dan Liew [Fri, 21 Aug 2015 18:10:55 +0000 (18:10 +0000)]
llvm-build: Adopt generation of LLVM_LIBS_TO_EXPORT. Patch by
Brad King.
Move `LLVM_LIBS_TO_EXPORT` over to Makefile.llvmbuild and generate it
from `llvm-build` using the same logic used to export the dependencies
of these libraries. This avoids depending on `llvm-config`.
This refactoring was originally motivated by issue #24154 due to commit
r243297 (Fix `llvm-config` to emit the linker flag for the combined
shared object, 2015-07-27) changing the output of `llvm-config --libs`
to not have the individual libraries when we configure with
`--enable-shared`. That change was reverted by r244108 but this
refactoring makes sense on its own anyway.
Dan Liew [Fri, 21 Aug 2015 18:10:51 +0000 (18:10 +0000)]
llvm-build: Factor out duplicate cmake export listing. Patch by
Brad King.
The write_cmake_fragment and write_cmake_exports_fragment methods share
some logic for selecting libraries that CMake needs to know about.
Factor it out into a helper to avoid duplication.
Yaron Keren [Fri, 21 Aug 2015 17:31:03 +0000 (17:31 +0000)]
Disable Visual C++ 2013 Debug mode assert on null pointer in some STL algorithms,
such as std::equal on the third argument. This reverts previous workarounds.
Predefining _DEBUG_POINTER_IMPL disables Visual C++ 2013 headers from defining
it to a function performing the null pointer check. In practice, it's not that
bad since any function actually using the nullptr will seg fault. The other
iterator sanity checks remain enabled in the headers.
Reviewed by Aaron Ballmanþ and Duncan P. N. Exon Smith.
John Brawn [Fri, 21 Aug 2015 10:48:17 +0000 (10:48 +0000)]
[DAGCombiner] Fold together mul and shl when both are by a constant
This is intended to improve code generation for GEPs, as the index value is
shifted by the element size and in GEPs of multi-dimensional arrays the index
of higher dimensions is multiplied by the lower dimension size.
James Y Knight [Fri, 21 Aug 2015 04:17:56 +0000 (04:17 +0000)]
[Sparc] Support user-specified stack object overalignment.
Note: I do not implement a base pointer, so it's still impossible to
have dynamic realignment AND dynamic alloca in the same function.
This also moves the code for determining the frame index reference
into getFrameIndexReference, where it belongs, instead of inline in
eliminateFrameIndex.
[Begin long-winded screed]
Now, stack realignment for Sparc is actually a silly thing to support,
because the Sparc ABI has no need for it -- unlike the situation on
x86, the stack is ALWAYS aligned to the required alignment for the CPU
instructions: 8 bytes on sparcv8, and 16 bytes on sparcv9.
However, LLVM unfortunately implements user-specified overalignment
using stack realignment support, so for now, I'm going to go along
with that tradition. GCC instead treats objects which have alignment
specification greater than the maximum CPU-required alignment for the
target as a separate block of stack memory, with their own virtual
base pointer (which gets aligned). Doing it that way avoids needing to
implement per-target support for stack realignment, except for the
targets which *actually* have an ABI-specified stack alignment which
is too small for the CPU's requirements.
Further unfortunately in LLVM, the default canRealignStack for all
targets effectively returns true, despite that implementing that is
something a target needs to do specifically. So, the previous behavior
on Sparc was to silently ignore the user's specified stack
alignment. Ugh.
Yet MORE unfortunate, if a target actually does return false from
canRealignStack, that also causes the user-specified alignment to be
*silently ignored*, rather than emitting an error.
(I started looking into fixing that last, but it broke a bunch of
tests, because LLVM actually *depends* on having it silently ignored:
some architectures (e.g. non-linux i386) have smaller stack alignment
than spilled-register alignment. But, the fact that a register needs
spilling is not known until within the register allocator. And by that
point, the decision to not reserve the frame pointer has been frozen
in place. And without a frame pointer, stack realignment is not
possible. So, canRealignStack() returns false, and
needsStackRealignment() then returns false, assuming everyone can just
go on their merry way assuming the alignment requirements were
probably just suggestions after-all. Sigh...)
The module splitter splits a module into linkable partitions. It will
be used to implement parallel LTO code generation.
This initial version of the splitter does not attempt to deal with the
somewhat subtle symbol visibility issues around module splitting. These
will be dealt with in a future change.
Matthias Braun [Thu, 20 Aug 2015 23:33:34 +0000 (23:33 +0000)]
AArch64: Fix cmp;ccmp ordering
When producing conditional compare sequences for or operations we need
to negate the operands and the finally tested flags. The thing is if we negate
the finally tested flags this equals a logical negation of all previously
emitted expressions. There was a case missing where we have to order OR
expressions so they get emitted first.
Matthias Braun [Thu, 20 Aug 2015 23:33:31 +0000 (23:33 +0000)]
AArch64: Do not create CCMP on multiple users.
Create CMP;CCMP sequences from and/or trees does not gain us anything if
the and/or tree is materialized to a GP register anyway. While most of
the code already checked for hasOneUse() there was one important case
missing.
Adrian Prantl [Thu, 20 Aug 2015 22:00:30 +0000 (22:00 +0000)]
Rename Instruction::dropUnknownMetadata() to dropUnknownNonDebugMetadata()
and make it always preserve debug locations, since all callers wanted this
behavior anyway.
This is addressing a post-commit review feedback for r245589.
We don't currently fold this, but now that we use native IR for
the intrinsics (r245605), we can look through one bitcast to find
the broadcast scalar.
Ahmed Bougacha [Thu, 20 Aug 2015 20:36:19 +0000 (20:36 +0000)]
[X86] Replace avx2 broadcast intrinsics with native IR.
Since r245605, the clang headers don't use these anymore.
r245165 updated some of the tests already; update the others, add
an autoupgrade, remove the intrinsics, and cleanup the definitions.
This patch adds support for asan on aarch64-linux with 42-bit VMA
(current default config for 64K pagesize kernels). The support is
enabled by defining the SANITIZER_AARCH64_VMA to 42 at build time
for both clang/llvm and compiler-rt. The default VMA is 39 bits.
Adrian Prantl [Thu, 20 Aug 2015 18:23:56 +0000 (18:23 +0000)]
Fix a debug location handling bug in GVN.
Caught by the famous "DebugLoc describes the currect SubProgram" assertion.
When GVN is removing a nonlocal load it updates the debug location of the
SSA value it replaced the load with with the one of the load. In the
testcase this actually overwrites a valid debug location with an empty one.
In reality GVN has to make an arbitrary choice between two equally valid
debug locations. This patch changes to behavior to only update the
location if the value doesn't already have a debug location.
Adam Nemet [Thu, 20 Aug 2015 17:22:29 +0000 (17:22 +0000)]
[LVer] Fix FIXME: hide addPHINodes, NFC
Since Ashutosh made findDefsUsedOutsideOfLoop public, we can clean this
up.
Now clients that don't compute DefsUsedOutsideOfLoop can just call
versionLoop() and computing DefsUsedOutsideOfLoop will happen
implicitly. With that there is no reason to expose addPHINodes anymore.
Ashutosh, you can now drop the calls to findDefsUsedOutsideOfLoop and
addPHINodes in LVerLICM and things should just work.
James Molloy [Thu, 20 Aug 2015 16:33:44 +0000 (16:33 +0000)]
[ARM] Don't try and custom lower a vNi64 SETCC.
It won't go well. We've already marked 64-bit SETCCs as non-Custom, but it's just possible that a SETCC has a legal result type but an illegal operand type. If this happens, bail out before we create unselectable nodes.
Fixes PR24292. I tried to create a testcase but in 99% of cases we can't trigger this - not surprising that this bug has been latent since 2009.
Bjorn Steinbrink [Thu, 20 Aug 2015 08:25:28 +0000 (08:25 +0000)]
[DSE] Enable removal of lifetime intrinsics in terminating blocks
Usually DSE is not supposed to remove lifetime intrinsics, but it's
actually ok to remove them for dead objects in terminating blocks,
because they convey no extra information there. Until we hit a lifetime
start that cannot be removed, that is. Because from that point on the
lifetime intrinsics become interesting again, e.g. for stack coloring.
Chandler Carruth [Thu, 20 Aug 2015 08:06:03 +0000 (08:06 +0000)]
[ARC] Pull the ObjC ARC components that really serve the role of
analyses into LLVM's Analysis library rather than having them in
a Transforms library.
This is motivated by the need to have the core AliasAnalysis
infrastructure be aware of the ObjCARCAliasAnalysis. However, it also
seems like a nice and clean separation. Everything was very easy to move
and this doesn't create much clutter in the analysis library IMO.
Hal Finkel [Thu, 20 Aug 2015 01:18:20 +0000 (01:18 +0000)]
[PowerPC] Fix the int2fp(fp2int(x)) DAGCombine to ignore ppc_fp128
This DAGCombine was creating custom SDAG nodes with an illegal ppc_fp128
operand type because it was triggering on f64/f32 int2fp(fp2int(ppc_fp128 x)),
but shouldn't (it should only apply to f32/f64 types). The result was a crash.
Alex Lorenz [Thu, 20 Aug 2015 00:20:03 +0000 (00:20 +0000)]
MIR Serialization: Use the global value syntax for global value memory operands.
This commit modifies the serialization syntax so that the global IR values in
machine memory operands use the global value '@<name>' syntax instead of the
current '%ir.<name>' syntax.
The unnamed global IR values are handled by this commit as well, as the
existing global value parsing method can parse the unnamed globals already.
Alex Lorenz [Thu, 20 Aug 2015 00:12:57 +0000 (00:12 +0000)]
MIR Serialization: Change syntax for the call entry pseudo source values.
The global IR values in machine memory operands should use the global value
'@<name>' syntax instead of the current '%ir.<name>' syntax.
However, the global value call entry pseudo source values use the global value
syntax already. Therefore, the syntax for the call entry pseudo source values
has to be changed so that the global values and call entry global value PSVs
can be parsed without ambiguities.
David Blaikie [Wed, 19 Aug 2015 23:07:27 +0000 (23:07 +0000)]
Allow Optionals to be compared to None
This is something like nullopt in std::experimental::optional. Optional
could already be constructed from None, so this seems like an obvious
extension from there.
I have a use in a future patch for Clang, though it may not go that
way/end up used - so this seemed worth committing now regardless.
Juergen Ributzka [Wed, 19 Aug 2015 20:52:55 +0000 (20:52 +0000)]
[AArch64][FastISel] Don't fold shifts with UB.
We are already falling back to SelectionDAG when encountering an shift with UB.
This adds the same checks for shifts with UB that get folded into arithmetic or
logical operations.
Dan Gohman [Wed, 19 Aug 2015 20:30:20 +0000 (20:30 +0000)]
[WebAssembly] Use the default alignment for SIMD types.
Previously WebAssembly's datalayout string had -v128:8:128. This had been an
attempt to declare a certain level of support for unaligned SIMD accesses.
However, clang makes its own determinations for SIMD alignment that are
independent of the datalayout string, so this wasn't actually meaningful.
Simon Pilgrim [Wed, 19 Aug 2015 20:09:50 +0000 (20:09 +0000)]
[DAGCombiner] Fold CONCAT_VECTORS of EXTRACT_SUBVECTOR (or undef) to VECTOR_SHUFFLE.
Check to see if this is a CONCAT_VECTORS of a bunch of EXTRACT_SUBVECTOR operations. If so, and if the EXTRACT_SUBVECTOR vector inputs come from at most two distinct vectors the same size as the result, attempt to turn this into a legal shuffle.
Alex Lorenz [Wed, 19 Aug 2015 19:05:34 +0000 (19:05 +0000)]
MIR Serialization: Serialize instruction's register ties.
This commit serializes the machine instruction's register operand ties.
The ties are printed out only when the instructon has register ties that are
different from the ties that are specified in the instruction's description.
Nemanja Ivanovic [Wed, 19 Aug 2015 19:04:47 +0000 (19:04 +0000)]
Temporary fix for the self-host failures introduced by rL244921.
This revision has introduced an issue that only affects bootstrapped compiler
when it is printing the ASM. I am working on resolving the issue, but in the
meantime, I'm disabling the legalization of scalar_to_vector operation for v2i64
and the associated testing until I can get this fixed.
Alex Lorenz [Wed, 19 Aug 2015 18:55:47 +0000 (18:55 +0000)]
MIR Serialization: Serialize defined registers that require 'def' register flag.
The defined registers are already serialized - they are represented by placing
them before the '=' in a machine instruction. However, certain instructions like
INLINEASM can have defined register operands after the '=', so this commit
introduces the 'def' register flag for such operands.
[PeepholeOptimizer] Look through PHIs to find additional register sources
Reintroduce r245442. Remove an overly conservative assertion introduced
in r245442. We could replace the assertion to use `shareSameRegisterFile`
instead, but in that point in `insertPHI` we already lost the original
Def subreg to check against. So drop the assertion completely.
Original commit message:
- Teaches the ValueTracker in the PeepholeOptimizer to look through PHI
instructions.
- Add findNextSourceAndRewritePHI method to lookup into multiple sources
returnted by the ValueTracker and rewrite PHIs with new sources.
With these changes we can find more register sources and rewrite more
copies to allow coaslescing of bitcast instructions. Hence, we eliminate
unnecessary VR64 <-> GR64 copies in x86, but it could be extended to
other archs by marking "isBitcast" on target specific instructions. The
x86 example follows:
Ahmed Bougacha [Wed, 19 Aug 2015 17:40:19 +0000 (17:40 +0000)]
[AArch64] Improve short-form diags on long-form Match_InvalidOperand.
Since r244955, we try to use the short-form ErrorInfo when both
tries failed, and the long-form match failed on a suffix operand.
However, this means we sometimes mix ErrorInfo and MatchResult
(one manifestation of this being PR24498). Instead, restore both.
Hal Finkel [Wed, 19 Aug 2015 17:26:07 +0000 (17:26 +0000)]
[SCEV] Fix GCC 4.8.0 ICE in lambda function
Rewrite some code to not use a lambda function. The non-lambda code is just
about as clean as the original, and not any longer. The lambda function causes
an internal compiler error in GCC 4.8.0, and it is not worth breaking support
for that compiler over this. NFC.