- After moving logic recognizing vector shift with scalar amount from
DAG combining into DAG lowering, we declare to customize all vector
shifts even vector shift on AVX is legal. As a result, the cost model
needs special tuning to identify these legal cases.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@177586 91177308-0d34-0410-b5e6-96231b3b80d8
- Move SRA/SRL/SHL lowering support from DAG combination to DAG lowering
to support extended 256-bit integer in AVX but not AVX2.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@177478 91177308-0d34-0410-b5e6-96231b3b80d8
Add a new WriteZero SchedWrite type for the common dependency-breaking
instructions that clear a register.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@177442 91177308-0d34-0410-b5e6-96231b3b80d8
an X86Operand, but also performs a Sema lookup and adds the sizing directive
when appropriate. Use this when parsing a bracketed statement. This is
necessary to get the instruction matching correct as well. Test case coming
on clang side.
rdar://13455408
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@177439 91177308-0d34-0410-b5e6-96231b3b80d8
def : Pat<(load (i64 (X86Wrapper tglobaltlsaddr :$dst))),
(MOV64rm tglobaltlsaddr :$dst)>;
This pattern is invalid because the MOV64rm instruction expects a
source operand of type "i64mem", which is a subclass of X86MemOperand
and thus actually consists of five MI operands, but the Pat provides
only a single MI operand ("tglobaltlsaddr" matches an SDnode of
type ISD::TargetGlobalTLSAddress and provides a single output).
Thus, if the pattern were ever matched, subsequent uses of the MOV64rm
instruction pattern would access uninitialized memory. In addition,
with the TableGen patch I'm about to check in, this would actually be
reported as a build-time error.
Fortunately, the pattern does in fact never match, for at least two
independent reasons.
First, the code generator actually never generates a pattern of the
form (load (X86Wrapper (tglobaltlsaddr))). For most combinations of
TLS and code models, (tglobaltlsaddr) represents just an offset that
needs to be added to some base register, so it is never directly
dereferenced. The only exception is the initial-exec model, where
(tglobaltlsaddr) refers to the (pc-relative) address of a GOT slot,
which *is* in fact directly dereferenced: but in that case, the
X86WrapperRIP node is used, not X86Wrapper, so the Pat doesn't match.
Second, even if some patterns along those lines *were* ever generated,
we should not need an extra Pat pattern to match it. Instead, the
original MOV64rm instruction pattern ought to match directly, since
it uses an "addr" operand, which is implemented via the SelectAddr
C++ routine; this routine is supposed to accept the full range of
input DAGs that may be implemented by a single mov instruction,
including those cases involving ISD::TargetGlobalTLSAddress (and
actually does so e.g. in the initial-exec case as above).
To avoid build breaks (due to the above-mentioned error) after the
TableGen patch is checked in, I'm removing this Pat here.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@177426 91177308-0d34-0410-b5e6-96231b3b80d8
We hitch a ride with the existing OpndItins class that was used to add
instruction itinerary classes in the many multiclasses in this file.
Use the link provided by the X86FoldableSchedWrite.Folded to find the
right SchedWrite for folded loads.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@177326 91177308-0d34-0410-b5e6-96231b3b80d8
This new-style scheduling information is going to replace the
instruction iteneraries.
This also serves as a test case for Andy's fix in r177317.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@177323 91177308-0d34-0410-b5e6-96231b3b80d8
MinGW is almost completely compatible to MSVC, with the exception of the _tls_array global not being available.
Patch by David Nadlinger!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@177257 91177308-0d34-0410-b5e6-96231b3b80d8
Since almost all X86 instructions can fold loads, use a multiclass to
define register/memory pairs of SchedWrites.
An X86FoldableSchedWrite represents the register version of an
instruction. It holds a reference to the SchedWrite to use when the
instruction folds a load.
This will be used inside multiclasses that define rr and rm instruction
versions together.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@177210 91177308-0d34-0410-b5e6-96231b3b80d8
The new InstrSchedModel is easier to use than the instruction
itineraries. It will be used to model instruction latency and throughput
in modern Intel microarchitectures like Sandy Bridge.
InstrSchedModel should be able to coexist with instruction itinerary
classes, but for cleanliness we should switch the Atom processor model
to the new InstrSchedModel as well.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@177122 91177308-0d34-0410-b5e6-96231b3b80d8
LegalizeDAG.cpp uses the value of the comparison operands when checking
the legality of BR_CC, so DAGCombiner should do the same.
v2:
- Expand more BR_CC value types for NVPTX
v3:
- Expand correct BR_CC value types for Hexagon, Mips, and XCore.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176694 91177308-0d34-0410-b5e6-96231b3b80d8
That can usually be lowered efficiently and is common in sandybridge code.
It would be nice to do this in DAGCombiner but we can't insert arbitrary
BUILD_VECTORs this late.
Fixes PR15462.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176634 91177308-0d34-0410-b5e6-96231b3b80d8
- Phi nodes should be replaced/updated after lowering CMOV into branch
because 'mainMBB' updating operand in Phi node is changed.
- Add EFLAGS in livein before lowering the 2nd CMOV. It's necessary as
we will reuse the EFLAGS generated before the 1st lowered CMOV, which
won't clobber EFLAGS. However, we need explicitly specify that.
- '-attr=-cmov' test case are added.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176598 91177308-0d34-0410-b5e6-96231b3b80d8
- Clear 'mayStore' flag when loading from the atomic variable before the
spin loop
- Clear kill flag from one use to multiple use in registers forming the
address to that atomic variable
- don't use a physical register as live-in register in BB (neither entry
nor landing pad.) by copying it into virtual register
(patch by Cameron Zwarich)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176538 91177308-0d34-0410-b5e6-96231b3b80d8
one-byte NOPs. If the processor actually executes those NOPs, as it sometimes
does with aligned bundling, this can have a performance impact. From my
micro-benchmarks run on my one machine, a 15-byte NOP followed by twelve
one-byte NOPs is about 20% worse than a 15 followed by a 12. This patch
changes NOP emission to emit as many 15-byte (the maximum) as possible followed
by at most one shorter NOP.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176464 91177308-0d34-0410-b5e6-96231b3b80d8
* Only apply divide bypass optimization when not optimizing for size.
* Fixed bug caused by constant for 0 value of type Int32,
used dividend type to generate the constant instead.
* For atom x86-64 apply the divide bypass to use 16-bit divides instead of
64-bit divides when operand values are small enough.
* Added lit tests for 64-bit divide bypass.
Patch by Tyler Nowicki!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176442 91177308-0d34-0410-b5e6-96231b3b80d8
This matters for example in following matrix multiply:
int **mmult(int rows, int cols, int **m1, int **m2, int **m3) {
int i, j, k, val;
for (i=0; i<rows; i++) {
for (j=0; j<cols; j++) {
val = 0;
for (k=0; k<cols; k++) {
val += m1[i][k] * m2[k][j];
}
m3[i][j] = val;
}
}
return(m3);
}
Taken from the test-suite benchmark Shootout.
We estimate the cost of the multiply to be 2 while we generate 9 instructions
for it and end up being quite a bit slower than the scalar version (48% on my
machine).
Also, properly differentiate between avx1 and avx2. On avx-1 we still split the
vector into 2 128bits and handle the subvector muls like above with 9
instructions.
Only on avx-2 will we have a cost of 9 for v4i64.
I changed the test case in test/Transforms/LoopVectorize/X86/avx1.ll to use an
add instead of a mul because with a mul we now no longer vectorize. I did
verify that the mul would be indeed more expensive when vectorized with 3
kernels:
for (i ...)
r += a[i] * 3;
for (i ...)
m1[i] = m1[i] * 3; // This matches the test case in avx1.ll
and a matrix multiply.
In each case the vectorized version was considerably slower.
radar://13304919
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176403 91177308-0d34-0410-b5e6-96231b3b80d8
- ISD::SHL/SRL/SRA must have either both scalar or both vector operands
but TLI.getShiftAmountTy() so far only return scalar type. As a
result, backend logic assuming that breaks.
- Rename the original TLI.getShiftAmountTy() to
TLI.getScalarShiftAmountTy() and re-define TLI.getShiftAmountTy() to
return target-specificed scalar type or the same vector type as the
1st operand.
- Fix most TICG logic assuming TLI.getShiftAmountTy() a simple scalar
type.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176364 91177308-0d34-0410-b5e6-96231b3b80d8
- Check whether SSE is available before lowering all 1s vector building with
PCMPEQD, which is only available from SSE2
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176058 91177308-0d34-0410-b5e6-96231b3b80d8
fewer scalar integer (i32 or i64) arguments. It completely eliminates the need
for SDISel for trivial functions.
Also, add the new llc -fast-isel-abort-args option, which is similar to
-fast-isel-abort option, but for formal argument lowering.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176052 91177308-0d34-0410-b5e6-96231b3b80d8
to TargetFrameLowering, where it belongs. Incidentally, this allows us
to delete some duplicated (and slightly different!) code in TRI.
There are potentially other layering problems that can be cleaned up
as a result, or in a similar manner.
The refactoring was OK'd by Anton Korobeynikov on llvmdev.
Note: this touches the target interfaces, so out-of-tree targets may
be affected.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175788 91177308-0d34-0410-b5e6-96231b3b80d8
exists solely to enable it to call itself for i8 with some registers.
The proposed patch simplifies the function somewhat to make the High
bit only meaningful for the i8 mode, which makes sense. No functional
difference (getX86SubSuperRegister is not getting called from anywhere
outside with i64 and High=true).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175762 91177308-0d34-0410-b5e6-96231b3b80d8
sext <4 x i1> to <4 x i64>
sext <4 x i8> to <4 x i64>
sext <4 x i16> to <4 x i64>
I'm running Combine on SIGN_EXTEND_IN_REG and revert SEXT patterns:
(sext_in_reg (v4i64 anyext (v4i32 x )), ExtraVT) -> (v4i64 sext (v4i32 sext_in_reg (v4i32 x , ExtraVT)))
The sext_in_reg (v4i32 x) may be lowered to shl+sar operations.
The "sar" does not exist on 64-bit operation, so lowering sext_in_reg (v4i64 x) has no vector solution.
I also added a cost of this operations to the AVX costs table.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175619 91177308-0d34-0410-b5e6-96231b3b80d8
MS-style inline assembly.
This is a follow-on to r175334. Forcing a FP to be emitted doesn't ensure it
will be used. Therefore, force the base pointer as well. We now treat MS
inline assembly in the same way we treat functions with dynamic stack
realignment and VLAs. This guarantees the BP will be used to reference
parameters and locals.
rdar://13218191
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175576 91177308-0d34-0410-b5e6-96231b3b80d8
If the frame pointer is omitted, and any stack changes occur in the inline
assembly, e.g.: "pusha", then any C local variable or C argument references
will be incorrect.
I pass no judgement on anyone who would do such a thing. ;)
rdar://13218191
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175334 91177308-0d34-0410-b5e6-96231b3b80d8
When we're recalculating the feature set of the subtarget, we need to have the
ivars in their initial state.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175320 91177308-0d34-0410-b5e6-96231b3b80d8
If two functions require different features (e.g., `-mno-sse' vs. `-msse') then
we want to honor that, especially during LTO. We can do that by resetting the
subtarget's features depending upon the 'target-feature' attribute.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175314 91177308-0d34-0410-b5e6-96231b3b80d8
than we need to and some ELF linkers complain about directly accessing symbols
with default visibility.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175268 91177308-0d34-0410-b5e6-96231b3b80d8
blocks. We still don't have consensus if we should try to change clang or
the standard, but llvm should work with compilers that implement the current
standard and mangle those functions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175267 91177308-0d34-0410-b5e6-96231b3b80d8
This happens when there is both stack realignment and a dynamic alloca in the
function. If we overwrite %esi (rep;movsl uses fixed registers) we'll lose the
base pointer and the next register spill will write into oblivion.
Fixes PR15249 and unbreaks firefox on i386/freebsd. Mozilla uses dynamic allocas
and freebsd a 4 byte stack alignment.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175057 91177308-0d34-0410-b5e6-96231b3b80d8
Fixed decode of existing 3dNow prefetchw instruction
Intel is scheduled to add a compatible prefetchw (same encoding) to future CPUs
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174920 91177308-0d34-0410-b5e6-96231b3b80d8
account. Atoms use LEA for updating SP in prologs/epilogs, and the
exact LEA opcode depends on the data model.
Also reapplying the test case which was added and then reverted
(because of Atom failures), this time specifying explicitly the CPU in
addition to the triple. The test case now checks all variations (data
mode, cpu Atom vs. Core).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174542 91177308-0d34-0410-b5e6-96231b3b80d8
