All of the complexity is in cleanupret, and it mostly follows the same
codepaths as catchret, except it doesn't take a return value in RAX.
This small example now compiles and executes successfully on win32:
extern "C" int printf(const char *, ...) noexcept;
struct Dtor {
~Dtor() { printf("~Dtor\n"); }
};
void has_cleanup() {
Dtor o;
throw 42;
}
int main() {
try {
has_cleanup();
} catch (int) {
printf("caught it\n");
}
}
Don't try to put the cleanup in the same function as the catch, or Bad
Things will happen.
llvm-svn: 247219
The 32-bit tables don't actually contain PC range data, so emitting them
is incredibly simple.
The 64-bit tables, on the other hand, use the same table for state
numbering as well as label ranges. This makes things more difficult, so
it will be implemented later.
llvm-svn: 247192
With subregister liveness enabled we can detect the case where only
parts of a register are live in, this is expressed as a 32bit lanemask.
The current code only keeps registers in the live-in list and therefore
enumerated all subregisters affected by the lanemask. This turned out to
be too conservative as the subregister may also cover additional parts
of the lanemask which are not live. Expressing a given lanemask by
enumerating a minimum set of subregisters is computationally expensive
so the best solution is to simply change the live-in list to store the
lanemasks as well. This will reduce memory usage for targets using
subregister liveness and slightly increase it for other targets
Differential Revision: http://reviews.llvm.org/D12442
llvm-svn: 247171
with the new pass manager, and no longer relying on analysis groups.
This builds essentially a ground-up new AA infrastructure stack for
LLVM. The core ideas are the same that are used throughout the new pass
manager: type erased polymorphism and direct composition. The design is
as follows:
- FunctionAAResults is a type-erasing alias analysis results aggregation
interface to walk a single query across a range of results from
different alias analyses. Currently this is function-specific as we
always assume that aliasing queries are *within* a function.
- AAResultBase is a CRTP utility providing stub implementations of
various parts of the alias analysis result concept, notably in several
cases in terms of other more general parts of the interface. This can
be used to implement only a narrow part of the interface rather than
the entire interface. This isn't really ideal, this logic should be
hoisted into FunctionAAResults as currently it will cause
a significant amount of redundant work, but it faithfully models the
behavior of the prior infrastructure.
- All the alias analysis passes are ported to be wrapper passes for the
legacy PM and new-style analysis passes for the new PM with a shared
result object. In some cases (most notably CFL), this is an extremely
naive approach that we should revisit when we can specialize for the
new pass manager.
- BasicAA has been restructured to reflect that it is much more
fundamentally a function analysis because it uses dominator trees and
loop info that need to be constructed for each function.
All of the references to getting alias analysis results have been
updated to use the new aggregation interface. All the preservation and
other pass management code has been updated accordingly.
The way the FunctionAAResultsWrapperPass works is to detect the
available alias analyses when run, and add them to the results object.
This means that we should be able to continue to respect when various
passes are added to the pipeline, for example adding CFL or adding TBAA
passes should just cause their results to be available and to get folded
into this. The exception to this rule is BasicAA which really needs to
be a function pass due to using dominator trees and loop info. As
a consequence, the FunctionAAResultsWrapperPass directly depends on
BasicAA and always includes it in the aggregation.
This has significant implications for preserving analyses. Generally,
most passes shouldn't bother preserving FunctionAAResultsWrapperPass
because rebuilding the results just updates the set of known AA passes.
The exception to this rule are LoopPass instances which need to preserve
all the function analyses that the loop pass manager will end up
needing. This means preserving both BasicAAWrapperPass and the
aggregating FunctionAAResultsWrapperPass.
Now, when preserving an alias analysis, you do so by directly preserving
that analysis. This is only necessary for non-immutable-pass-provided
alias analyses though, and there are only three of interest: BasicAA,
GlobalsAA (formerly GlobalsModRef), and SCEVAA. Usually BasicAA is
preserved when needed because it (like DominatorTree and LoopInfo) is
marked as a CFG-only pass. I've expanded GlobalsAA into the preserved
set everywhere we previously were preserving all of AliasAnalysis, and
I've added SCEVAA in the intersection of that with where we preserve
SCEV itself.
One significant challenge to all of this is that the CGSCC passes were
actually using the alias analysis implementations by taking advantage of
a pretty amazing set of loop holes in the old pass manager's analysis
management code which allowed analysis groups to slide through in many
cases. Moving away from analysis groups makes this problem much more
obvious. To fix it, I've leveraged the flexibility the design of the new
PM components provides to just directly construct the relevant alias
analyses for the relevant functions in the IPO passes that need them.
This is a bit hacky, but should go away with the new pass manager, and
is already in many ways cleaner than the prior state.
Another significant challenge is that various facilities of the old
alias analysis infrastructure just don't fit any more. The most
significant of these is the alias analysis 'counter' pass. That pass
relied on the ability to snoop on AA queries at different points in the
analysis group chain. Instead, I'm planning to build printing
functionality directly into the aggregation layer. I've not included
that in this patch merely to keep it smaller.
Note that all of this needs a nearly complete rewrite of the AA
documentation. I'm planning to do that, but I'd like to make sure the
new design settles, and to flesh out a bit more of what it looks like in
the new pass manager first.
Differential Revision: http://reviews.llvm.org/D12080
llvm-svn: 247167
Instead of extracting both 32-bit components from the 128-bit
register. This produces fewer copies and is easier for
the copy peephole optimizer to understand and see the actual uses
as extracts from a reg_sequence.
This avoids needing to handle subregister composing in the
PeepholeOptimizer's ValueTracker for this case.
llvm-svn: 247162
Summary:
This helps mostly when we use add instructions for address calculations
that contain immediates.
Reviewers: arsenm
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D12256
llvm-svn: 247157
Summary:
We are not scalarizing the wide selects in codegen for i16 and i32 and
therefore we can remove the amortization factor. We still have issues
with i64 vectors in codegen though.
Reviewers: mcrosier
Subscribers: mcrosier, aemerson, llvm-commits, rengolin
Differential Revision: http://reviews.llvm.org/D12724
llvm-svn: 247156
Currently this hits an assert that extload should
always be supported, which assumes integer extloads.
This moves a hack out of SI's argument lowering and
is covered by existing tests.
llvm-svn: 247113
Summary:
32-bit funclets have short prologues that allocate enough stack for the
largest call in the whole function. The runtime saves CSRs for the
funclet. It doesn't restore CSRs after we finally transfer control back
to the parent funciton via a CATCHRET, but that's a separate issue.
32-bit funclets also have to adjust the incoming EBP value, which is
what llvm.x86.seh.recoverframe does in the old model.
64-bit funclets need to spill CSRs as normal. For simplicity, this just
spills the same set of CSRs as the parent function, rather than trying
to compute different CSR sets for the parent function and each funclet.
64-bit funclets also allocate enough stack space for the largest
outgoing call frame, like 32-bit.
Reviewers: majnemer
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D12546
llvm-svn: 247092
Adds vcc to output string input for e32. Allows option
of using e64 encoding with assembler.
Also fixes these instructions not implicitly reading exec.
llvm-svn: 247074
The pass is needed to remove __nvvm_reflect calls when we link in
libdevice bitcode that comes with CUDA.
Differential Revision: http://reviews.llvm.org/D11663
llvm-svn: 247072
Summary: This patch modifies X86TargetLowering::LowerVASTART so that
struct va_list is initialized with 32 bit pointers in x32. It also
includes tests that call @llvm.va_start() for x32.
Patch by João Porto
Subscribers: llvm-commits, hjl.tools
Differential Revision: http://reviews.llvm.org/D12346
llvm-svn: 247069
This allows backends which don't use a traditional register allocator,
but do need PHI lowering and other passes, to use the default
TargetPassConfig::addFastRegAlloc and
TargetPassConfig::addOptimizedRegAlloc implementations.
Differential Revision: http://reviews.llvm.org/D12691
llvm-svn: 247065
These were marked as WriteSALU, which is low latency.
I'm guessing at the value to use, but it should probably
be considered the highest latency instruction.
I'm not sure this has any actual effect since hasSideEffects
probably is preventing any moving of these.
llvm-svn: 247060
The old implementation assumed LP64 which is broken for x32. Specifically, the
MOVE8rm_NOREX and MOVE8mr_NOREX, when selected, would cause a 'Cannot emit
physreg copy instruction' error message to be reported.
This patch also enable the h-register*ll tests for x32.
Differential Revision: http://reviews.llvm.org/D12336
Patch by João Porto
llvm-svn: 247058
sub C, x - > add (sub 0, x), C for DS offsets.
This is mostly to fix regressions that show up when
SeparateConstOffsetFromGEP is enabled.
llvm-svn: 247054
Summary: And define them to have noop casts with address spaces 0-255.
Reviewers: pekka.jaaskelainen
Subscribers: pekka.jaaskelainen, llvm-commits
Differential Revision: http://reviews.llvm.org/D12678
llvm-svn: 246990
SelectT2ShifterOperandReg has identical behaviour to SelectImmShifterOperand,
so get rid of it and use SelectImmShifterOperand instead.
Differential Revision: http://reviews.llvm.org/D12195
llvm-svn: 246962
To commute a trivial rlwimi instructions (meaning one with a full mask and zero
shift), we'd need to ability to form an all-zero mask (instead of an all-one
mask) using rlwimi. We can't represent this, however, and we'll miscompile code
if we try.
The code quality problem that this highlights (that SDAG simplification can
lead to us generating an ISD::OR node with a constant zero LHS) will be fixed
as a follow-up.
Fixes PR24719.
llvm-svn: 246937
PPCISelDAGToDAG has a transformation that generates a rlwimi instruction from
an input pattern that looks like this:
and(or(x, c1), c2)
but the associated logic does not work if there are bits that are 1 in c1 but 0
in c2 (these are normally canonicalized away, but that can't happen if the 'or'
has other users. Make sure we abort the transformation if such bits are
discovered.
Fixes PR24704.
llvm-svn: 246900
This adds a basic cost model for interleaved-access vectorization (and a better
default for shuffles), and enables interleaved-access vectorization by default.
The relevant difference from the default cost model for interleaved-access
vectorization, is that on PPC, the shuffles that end up being used are *much*
cheaper than modeling the process with insert/extract pairs (which are
quite expensive, especially on older cores).
llvm-svn: 246824
On the A2, with an eye toward QPX unaligned-load merging, we should always use
aggressive interleaving. It is generally superior to only using concatenation
unrolling.
llvm-svn: 246819
When forming permutation-based unaligned vector loads, we need to know whether
it is valid to read ahead of the requested address by a full vector length.
Doing so is more efficient (and allows for more CSE with later loads), but
could trigger a page fault if invalid. To determine validity, we look for other
loads in the same block that access the relevant address range.
The relevant point here is that we need to do this as part of the process of
forming permutation-based vector loads, and this happens quite early in the
SDAG pipeline - specifically before many of the address calculations are fully
canonicalized. As a result, we need to try harder to recognize base+offset
address computations, because they still might appear as chain of adds
(base+offset+offset, for example). To account for this, we'll look through
chains of adds, accumulating the constant offsets.
llvm-svn: 246813
Pre-P8, when we generate code for unaligned vector loads (for Altivec and QPX
types), even when accounting for the combining that takes place for multiple
consecutive such loads, there is at least one load instructions and one
permutation for each load. Make sure the cost reported reflects the cost of the
permutes as well.
llvm-svn: 246807
If you compute the MMO offset using unsigned arithmetic, you end up with a
large positive offset instead of a small negative one. In theory, this could
cause bad instruction-scheduling decisions later.
I noticed this by inspection from the debug output, and using that for the
regression test is the best I can do right now.
llvm-svn: 246805
This prevents MC clients from getting COFF.h, which conflicts with
winnt.h macros. Also a minor IWYU cleanup. Now the only public headers
including COFF.h are in Object, and they actually need it.
llvm-svn: 246784
Use and check the 'IsFast' optional parameter to TLI.allowsMemoryAccess() any time
we have a merged access candidate. Without this patch, we were generating unaligned
16-byte (SSE) memops for x86 targets where those accesses are slow.
This change was mentioned in:
http://reviews.llvm.org/D10662 and
http://reviews.llvm.org/D10905
and will help solve PR21711.
Differential Revision: http://reviews.llvm.org/D12573
llvm-svn: 246771
This patch allows the mixing of scaled and unscaled load/stores to form
load/store pairs.
PR24465
http://reviews.llvm.org/D12116
Many thanks to Ahmed and Michael for fixes and code review.
llvm-svn: 246769
We used to accept (and even test, and generate) 16-byte alignment
for 32-byte nontemporal stores, but they require 32-byte alignment,
per SDM. Found by inspection.
Instead of hardcoding 16 in the patfrag, check for natural alignment.
Also fix the autoupgrade and the various tests.
Also, use explicit -mattr instead of -mcpu: I stared at the output
several minutes wondering why I get 2x movntps for the unaligned
case (which is the ideal output, but needs some work: see FIXME),
until I remembered corei7-avx implies +slow-unaligned-mem-32.
llvm-svn: 246733
We can chain other fragments to avoid repeating conditions.
This also fixes a potential bug (that realistically can't happen),
where we would match indexed nontemporal stores for i32/i64.
llvm-svn: 246719
I'm adding a regression test to better cover code generation for unaligned
vector loads and stores, but there's no functional change to the code
generation here. There is an improvement to the cost model for unaligned vector
loads and stores, mostly for QPX (for which we were not previously accounting
for the permutation-based loads), and the cost model implementation is cleaner.
llvm-svn: 246712
Some of the instructions use ' ', which drives OCD-me nuts.
Let's put an end to this.
NFC-ish: hopefully nobody cares about whitespace.
llvm-svn: 246686
LowerVECTOR_SHUFFLE needs to decide whether to pass a vector shuffle off to the
TableGen-generated matching code, and it does this by testing the same
predicates used by the TableGen files. Unfortunately, when we added new
P8Altivec-only predicates, we started universally testing them in
LowerVECTOR_SHUFFLE, and if then matched when targeting a system prior to a P8,
we'd end up with a selection failure.
llvm-svn: 246675
This is a continuation of the fix from:
http://reviews.llvm.org/D10662
and discussion in:
http://reviews.llvm.org/D12154
Here, we distinguish slow unaligned SSE (128-bit) accesses from slow unaligned
scalar (64-bit and under) accesses. Other lowering (eg, getOptimalMemOpType)
assumes that unaligned scalar accesses are always ok, so this changes
allowsMisalignedMemoryAccesses() to match that behavior.
Differential Revision: http://reviews.llvm.org/D12543
llvm-svn: 246658
The code introduced in r244314 assumed that EXTRACT_VECTOR_ELT only
takes constant indices, but it does accept variables.
Bail out for those: we can't use them, as the shuffles we want to
reconstruct do require constant masks.
llvm-svn: 246594
This matches the ARM behavior. In both cases, the register is part
of the optional Performance Monitors extension, so, add the feature,
and enable it for the A-class processors we support.
Differential Revision: http://reviews.llvm.org/D12425
llvm-svn: 246555
Summary:
This change turns on by default interleaved access vectorization
for AArch64.
We also clean up some tests which were spedifically enabling this
behaviour.
Reviewers: rengolin
Subscribers: aemerson, llvm-commits, rengolin
Differential Revision: http://reviews.llvm.org/D12149
llvm-svn: 246542
Summary:
This change turns on by default interleaved access vectorization on ARM,
as it has shown to be beneficial on ARM.
Reviewers: rengolin
Subscribers: aemerson, llvm-commits, rengolin
Differential Revision: http://reviews.llvm.org/D12146
llvm-svn: 246541
Summary: This handles all load/store operations that WebAssembly defines, and handles those necessary for C++ such as i1. I left a FIXME for outstanding features which aren't required for now.
Reviewers: sunfish
Subscribers: jfb, llvm-commits, dschuff
llvm-svn: 246500
Also delete and simplify a lot of MachineModuleInfo code that used to be
needed to handle personalities on landingpads. Now that the personality
is on the LLVM Function, we no longer need to track it this way on MMI.
Certainly it should not live on LandingPadInfo.
llvm-svn: 246478
The ISelLowering code turned insertion turned the element for the
lowest lane of a BUILD_VECTOR into an INSERT_SUBREG, this prohibited
the patterns for SCALAR_TO_VECTOR(Load) to match later. Restrict this
to cases without a load argument.
Reported in rdar://22223823
Differential Revision: http://reviews.llvm.org/D12467
llvm-svn: 246462
X86FastISel has been using the wrong register class for VBLENDVPS which
produces a VR128 and needs an extra copy to the target register. The
problem was already hit by the existing test cases when using
> llvm-lit -Dllc="llc -verify-machineinstr"
llvm-svn: 246461
There were really two problems here. The first was that we had the truth tables
for signed i1 comparisons backward. I imagine these are not very common, but if
you have:
setcc i1 x, y, LT
this has the '0 1' and the '1 0' results flipped compared to:
setcc i1 x, y, ULT
because, in the signed case, '1 0' is really '-1 0', and the answer is not the
same as in the unsigned case.
The second problem was that we did not have patterns (at all) for the unsigned
comparisons select_cc nodes for i1 comparison operands. This was the specific
cause of PR24552. These had to be added (and a missing Altivec promotion added
as well) to make sure these function for all types. I've added a bunch more
test cases for these patterns, and there are a few FIXMEs in the test case
regarding code-quality.
Fixes PR24552.
llvm-svn: 246400
Add support for MIR serialization of PowerPC-specific operand target flags
(based on the generic infrastructure added in r244185 and r245383).
I won't even pretend that this is good test coverage, but this includes the
regression test associated with r246372. Adding an MIR test for that fix is far
superior to adding an IR-level test because particular instruction-scheduling
decisions are necessary in order to expose the bug, and using an MIR test we
can start the pipeline post-scheduling.
llvm-svn: 246373
Even through ADDISdtprelHA generally has r3 as its source register, it is
possible for the instruction scheduler to move things around such that some
other register is the source. We need to print the actual source register, not
always r3. Fixes PR24394.
The test case will come in a follow-up commit because it depends on MIR
target-flags parsing.
llvm-svn: 246372
This is especially visible in softfp mode, for example in the implementation of libm fabs/fneg functions. If we have:
%1 = vmovdrr r0, r1
%2 = fabs %1
then move the fabs before the vmovdrr:
%1 = and r1, #0x7FFFFFFF
%2 = vmovdrr r0, r1
This is never a lose, and could be a serious win because the vmovdrr may be followed by a vmovrrd, which would enable us to remove the conversion into FPRs completely.
We already do this for f32, but not for f64. Tests are added for both.
llvm-svn: 246360
The VOP3 encoding of these allows any SGPR pair for the i1
output, but this was forced before to always use vcc.
This doesn't yet try to use this, but does add the operand
to the definitions so the main change is adding vcc to the
output of the VOP2 encoding.
llvm-svn: 246358
Without a memory operand, mayLoad or mayStore instructions
are treated as hasUnorderedMemRef, which results in much worse
scheduling.
We really should have a verifier check that any
non-side effecting mayLoad or mayStore has a memory operand.
There are a few instructions (interp and images) which I'm
not sure what / where to add these.
llvm-svn: 246356
Summary:
We were assuming tha if the use operand had a sub-register that
the immediate was 64-bits, but this was breaking the case of
folding a 64-bit immediate into another 64-bit instruction.
Reviewers: arsenm
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D12255
llvm-svn: 246354
This has been causing the prologue_end to be incorrectly positioned.
Patch by Vladimir Radosavljevic.
Differential Revision: http://reviews.llvm.org/D11293
llvm-svn: 246309
For targets that didn't support this, this will let us respect the
langref instead of failing to select.
Note that we don't need to change the 32-bit x86/PPC lowerings (to
account for the result type/# difference) because they're both
custom and bypass type legalization.
llvm-svn: 246258
more than 2 instructions.
I introduced this regression a while back and did not noticed it because I
somehow forgot to push the initial test cases for the pass!
Fix that as well!
llvm-svn: 246239
We can now run 32-bit programs with empty catch bodies. The next step
is to change PEI so that we get funclet prologues and epilogues.
llvm-svn: 246235
Summary:
Let NVPTX backend detect integer min and max patterns during isel and emit intrinsics that enable hardware support.
Reviewers: jholewinski, meheff, jingyue
Subscribers: arsenm, llvm-commits, meheff, jingyue, eliben, jholewinski
Differential Revision: http://reviews.llvm.org/D12377
llvm-svn: 246107
Previously in isProfitableToIfCvt() in ARMBaseInstrInfo.cpp, the multiplication between an integer and a branch probability is done manually in an unsafe way that may lead to overflow. This patch corrects those cases by using BranchProbability's member function scale() to avoid overflow (which stores the intermediate result in int64).
Differential Revision: http://reviews.llvm.org/D12295
llvm-svn: 246106
Things of note:
- Other linkage types aren't handled yet. We'll figure it out with dynamic linking.
- Special LLVM globals are either ignored, or error out for now.
- TLS isn't supported yet (WebAssembly will have threads later).
- There currently isn't a syntax for alignment, I left it in a comment so it's easy to hook up.
- Undef is convereted to whatever the type's appropriate null value is.
- assert versus report_fatal_error: follow what other AsmPrinters do, and assert only on what should have been caught elsewhere.
llvm-svn: 246092
A corresponding clang change will make it so that clang can consume part
of an assembler token. The assembler treats '.' as an identifier
character while clang does not, so it's view of the token stream is a
little different.
llvm-svn: 246089
We removed access to the DataLayout on the TargetMachine and
deprecated the C API function LLVMGetTargetMachineData() in r243114.
However the way I tried to be backward compatible was broken: I
changed the wrapper of the TargetMachine to be a structure that
includes the DataLayout as well. However the TargetMachine is also
wrapped by the ExecutionEngine, in the more classic way. A client
using the TargetMachine wrapped by the ExecutionEngine and trying
to get the DataLayout would break.
It seems tricky to solve the problem completely in the C API
implementation. This patch tries to address this backward
compatibility in a more lighter way in the C++ API. The C API is
restored in its original state and the removed C++ API is
reintroduced, but privately. The C API is friended to the
TargetMachine and should be the only consumer for this API.
Reviewers: ributzka
Differential Revision: http://reviews.llvm.org/D12263
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 246082
There is no context where s_mov_b64 is emitted
and could potentially be moved to the VALU.
It is currently only emitted for materializing
immediates, which can't be dependent on vector sources.
The immediate splitting is already done when selecting
constants. I'm not sure what contexts if any the register
splitting would have been used before.
Also clean up using s_mov_b64 in place of v_mov_b64_pseudo,
although this isn't required and just skips the extra step
of eliminating the copy from the SReg_64.
llvm-svn: 246080
When splitting 64-bit operations, create the correct
VALU instructions immediately.
This was splitting things like s_or_b64 into the two
s_or_b32s and then pushing the new instructions
onto the worklist. There's no reason we need
to do this intermediate step.
llvm-svn: 246077
This takes the existing static function hasLiveCondCodeDef and makes it a member function of the X86InstrInfo class. This is a useful utility function that an upcoming change would like to use. NFC.
Patch by: Kevin B. Smith
Differential Revision: http://reviews.llvm.org/D12371
llvm-svn: 246073
We removed access to the DataLayout on the TargetMachine and
deprecated the C API function LLVMGetTargetMachineData() in r243114.
However the way I tried to be backward compatible was broken: I
changed the wrapper of the TargetMachine to be a structure that
includes the DataLayout as well. However the TargetMachine is also
wrapped by the ExecutionEngine, in the more classic way. A client
using the TargetMachine wrapped by the ExecutionEngine and trying
to get the DataLayout would break.
It seems tricky to solve the problem completely in the C API
implementation. This patch tries to address this backward
compatibility in a more lighter way in the C++ API. The C API is
restored in its original state and the removed C++ API is
reintroduced, but privately. The C API is friended to the
TargetMachine and should be the only consumer for this API.
Reviewers: ributzka
Differential Revision: http://reviews.llvm.org/D12263
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 246052
We removed access to the DataLayout on the TargetMachine and
deprecated the C API function LLVMGetTargetMachineData() in r243114.
However the way I tried to be backward compatible was broken: I
changed the wrapper of the TargetMachine to be a structure that
includes the DataLayout as well. However the TargetMachine is also
wrapped by the ExecutionEngine, in the more classic way. A client
using the TargetMachine wrapped by the ExecutionEngine and trying
to get the DataLayout would break.
It seems tricky to solve the problem completely in the C API
implementation. This patch tries to address this backward
compatibility in a more lighter way in the C++ API. The C API is
restored in its original state and the removed C++ API is
reintroduced, but privately. The C API is friended to the
TargetMachine and should be the only consumer for this API.
Reviewers: ributzka
Differential Revision: http://reviews.llvm.org/D12263
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 246044
If you're going to realign %sp to get object alignment properly (which
the code does), and stack offsets and alignments are calculated going
down from %fp (which they are), then the total stack size had better
be a multiple of the alignment. LLVM did indeed ensure that.
And then, after aligning, the sparc frame code added 96 (for sparcv8)
to the frame size, making any requested alignment of 64-bytes or
higher *guaranteed* to be misaligned. The test case added with r245668
even tests this exact scenario, and asserted the incorrect behavior,
which I somehow failed to notice. D'oh.
This change fixes the frame lowering code to align the stack size
*after* adding the spill area, instead.
Differential Revision: http://reviews.llvm.org/D12349
llvm-svn: 246042
This is a fix for disassembling unusual instruction sequences in 64-bit
mode w.r.t the CALL rel16 instruction. It might be desirable to move the
check somewhere else, but it essentially mimics the special case
handling with JCXZ in 16-bit mode.
The current behavior accepts the opcode size prefix and causes the
call's immediate to stop disassembling after 2 bytes. When debugging
sequences of instructions with this pattern, the disassembler output
becomes extremely unreliable and essentially useless (if you jump midway
into what lldb thinks is a unified instruction, you'll lose %rip). So we
ignore the prefix and consume all 4 bytes when disassembling a 64-bit
mode binary.
Note: in Vol. 2A 3-99 the Intel spec states that CALL rel16 is N.S. N.S.
is defined as:
Indicates an instruction syntax that requires an address override
prefix in 64-bit mode and is not supported. Using an address
override prefix in 64-bit mode may result in model-specific
execution behavior. (Vol. 2A 3-7)
Since 0x66 is an operand override prefix we should be OK (although we
may want to warn about 0x67 prefixes to 0xe8). On the CPUs I tested
with, they all ignore the 0x66 prefix in 64-bit mode.
Patch by Matthew Barney!
Differential Revision: http://reviews.llvm.org/D9573
llvm-svn: 246038
Summary:
This change lowers the aarch64 integer vector min/max intrinsic nodes to
generic min/max nodes and replaces the intrinsic selection patterns with
the generic ones.
There should already be testing in place for this, so no further tests
were added.
Reviewers: jmolloy
Subscribers: aemerson, llvm-commits, rengolin
Differential Revision: http://reviews.llvm.org/D12276
llvm-svn: 246030
This should be no functional change but for the record: For three cases
in X86FastISel this will change the order in which the FalseMBB and
TrueMBB of a conditional branch is addedd to the successor/predecessor
lists.
llvm-svn: 245997
Summary:
This change makes the variable argument intrinsics, `llvm.va_start` and
`llvm.va_copy`, and the `va_arg` instruction behave as they do on Windows
inside a `CallingConv::X86_64_Win64` function. It's needed for a Clang patch
I have to add support for GCC's `__builtin_ms_va_list` constructs.
Reviewers: nadav, asl, eugenis
CC: llvm-commits
Differential Revision: http://llvm-reviews.chandlerc.com/D1622
llvm-svn: 245990
This is a follow-on from the discussion in http://reviews.llvm.org/D12154.
This change allows memset/memcpy to use SSE or AVX memory accesses for any chip that has
generally fast unaligned memory ops.
A motivating use case for this change is a clang invocation that doesn't explicitly set
the CPU, but does target a feature that we know only exists on a CPU that supports fast
unaligned memops. For example:
$ clang -O1 foo.c -mavx
This resolves a difference in lowering noted in PR24449:
https://llvm.org/bugs/show_bug.cgi?id=24449
Before this patch, we used different store types depending on whether the example can be
lowered as a memset or not.
Differential Revision: http://reviews.llvm.org/D12288
llvm-svn: 245950
This fixes two issues in x86 fptoui lowering.
1) Makes conversions from f80 go through the right path on AVX-512.
2) Implements an inline sequence for fptoui i64 instead of a library
call. This improves performance by 6X on SSE3+ and 3X otherwise.
Incidentally, it also removes the use of ftol2 for fptoui, which was
wrong to begin with, as ftol2 converts to a signed i64, producing
wrong results for values >= 2^63.
Patch by: mitch.l.bodart@intel.com
Differential Revision: http://reviews.llvm.org/D11316
llvm-svn: 245924
This reverts commit 433bfd94e4b7e3cc3f8b08f8513ce47817941b0c.
Broke some bot, have to see why it passed locally.
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 245917
We removed access to the DataLayout on the TargetMachine and
deprecated the C API function LLVMGetTargetMachineData() in r243114.
However the way I tried to be backward compatible was broken: I
changed the wrapper of the TargetMachine to be a structure that
includes the DataLayout as well. However the TargetMachine is also
wrapped by the ExecutionEngine, in the more classic way. A client
using the TargetMachine wrapped by the ExecutionEngine and trying
to get the DataLayout would break.
It seems tricky to solve the problem completely in the C API
implementation. This patch tries to address this backward
compatibility in a more lighter way in the C++ API. The C API is
restored in its original state and the removed C++ API is
reintroduced, but privately. The C API is friended to the
TargetMachine and should be the only consumer for this API.
Reviewers: ributzka
Differential Revision: http://reviews.llvm.org/D12263
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 245916
We might end up with a trivial copy as the addend, and if so, we should ignore
the corresponding FMA instruction. The trivial copy can be coalesced away later,
so there's nothing to do here. We should not, however, assert. Fixes PR24544.
llvm-svn: 245907
Summary: I forgot to squash git commits before doing an svn dcommit of D12219. Reverting, and re-submitting.
Subscribers: jfb, llvm-commits
Differential Revision: http://reviews.llvm.org/D12298
llvm-svn: 245886
This patch fixes PR24546, which demonstrates a segfault during the VSX
swap removal pass. The problem is that debug value instructions were
not excluded from the list of instructions to be analyzed for webs of
related computation. I've added the test case from the PR as a crash
test in test/CodeGen/PowerPC.
llvm-svn: 245862
Summary:
__shared__ variable may now emit undef value as initializer, do not
throw error on that.
Test Plan: test/CodeGen/NVPTX/global-addrspace.ll
Patch by Xuetian Weng
Reviewers: jholewinski, tra, jingyue
Subscribers: llvm-commits, jholewinski
Differential Revision: http://reviews.llvm.org/D12242
llvm-svn: 245785
Although the basic s_load_* instructions happen to use the same
opcode, some of the special case SMRD instructions have
different opcodes.
llvm-svn: 245775
We can wait on either VM, EXP or LGKM.
The waits are independent.
Without this patch, a wait inserted because of one of them
would also wait for all the previous others.
This patch makes s_wait only wait for the ones we need for the next
instruction.
Here's an example of subtle perf reduction this patch solves:
This is without the patch:
buffer_load_format_xyzw v[8:11], v0, s[44:47], 0 idxen
buffer_load_format_xyzw v[12:15], v0, s[48:51], 0 idxen
s_load_dwordx4 s[44:47], s[8:9], 0xc
s_waitcnt lgkmcnt(0)
buffer_load_format_xyzw v[16:19], v0, s[52:55], 0 idxen
s_load_dwordx4 s[48:51], s[8:9], 0x10
s_waitcnt vmcnt(1)
buffer_load_format_xyzw v[20:23], v0, s[44:47], 0 idxen
The s_waitcnt vmcnt(1) is useless.
The reason it is added is because the last
buffer_load_format_xyzw needs s[44:47], which was issued
by the first s_load_dwordx4. It waits for all VM
before that call to have finished.
Internally after every instruction, 3 counters (for VM, EXP and LGTM)
are updated after every instruction. For example buffer_load_format_xyzw
will
increase the VM counter, and s_load_dwordx4 the LGKM one.
Without the patch, for every defined register,
the current 3 counters are stored, and are used to know
how long to wait when an instruction needs the register.
Because of that, the s[44:47] counter includes that to use the register
you need to wait for the previous buffer_load_format_xyzw.
Instead this patch stores only the counters that matter for the
register,
and puts zero for the other ones, since we don't need any wait for them.
Patch by: Axel Davy
Differential Revision: http://reviews.llvm.org/D11883
llvm-svn: 245755
When PPCVSXFMAMutate would look at the input addend register, it would get its
input value number. This would fail, however, if the register was undef,
causing a segfault. Don't segfault (just skip such FMA instructions).
Fixes the test case from PR24542 (although that may have been over-reduced).
llvm-svn: 245741
See discussion in D12154 ( http://reviews.llvm.org/D12154 ), AMD Software
Optimization Guides for 10H/12H/15H/16H, and Agner Fog's experimental data.
llvm-svn: 245733
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.
Differential Revision: http://reviews.llvm.org/D12154
llvm-svn: 245729
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...)
Differential Revision: http://reviews.llvm.org/D12208
llvm-svn: 245668
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.
This fixes http://llvm.org/PR24459
llvm-svn: 245641
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.
llvm-svn: 245640
Fixes PR23464: one way to use the broadcast intrinsics is:
_mm256_broadcastw_epi16(_mm_cvtsi32_si128(*(int*)src));
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.
Differential Revision: http://reviews.llvm.org/D10557
llvm-svn: 245613
Summary:
Add an LSR test that exercises isTruncateFree. Without this change, LSR creates
another indvar representing the truncated value.
Reviewers: jholewinski, eliben
Subscribers: jholewinski, llvm-commits
Differential Revision: http://reviews.llvm.org/D12058
llvm-svn: 245611
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.
Differential Revision: http://reviews.llvm.org/D10555
llvm-svn: 245606
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.
llvm-svn: 245577
COMISD should receive QWORD because it is defined as
(V)COMISD xmm1, xmm2/m64
COMISS should receive DWORD because it is defined as
(V)COMISS xmm1, xmm2/m32
Differential Revision: http://reviews.llvm.org/D11712
llvm-svn: 245551
XVCMPEQDP is used for VSX v2f64 equality comparisons, but the value type needs
to be v2i64 (as that's the corresponding SETCC type).
Fixes PR24225.
llvm-svn: 245535