a vector packed single/double fp operation followed by a vector insert.
The effect is that the backend coverts the packed fp instruction
followed by a vectro insert into a SSE or AVX scalar fp instruction.
For example, given the following code:
__m128 foo(__m128 A, __m128 B) {
__m128 C = A + B;
return (__m128) {c[0], a[1], a[2], a[3]};
}
previously we generated:
addps %xmm0, %xmm1
movss %xmm1, %xmm0
we now generate:
addss %xmm1, %xmm0
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197145 91177308-0d34-0410-b5e6-96231b3b80d8
Aside from a few minor latency corrections, the major change here is a new
hazard recognizer which focuses on better dispatch-group formation on the
POWER7. As with the PPC970's hazard recognizer, the most important thing it
does is avoid load-after-store hazards within the same dispatch group. It uses
the POWER7's special dispatch-group-terminating nop instruction (instead of
inserting multiple regular nop instructions). This new hazard recognizer makes
use of the scheduling dependency graph itself, built using AA information, to
robustly detect the possibility of load-after-store hazards.
significant test-suite performance changes (the error bars are 99.5% confidence
intervals based on 5 test-suite runs both with and without the change --
speedups are negative):
speedups:
MultiSource/Benchmarks/FreeBench/pcompress2/pcompress2
-0.55171% +/- 0.333168%
MultiSource/Benchmarks/TSVC/CrossingThresholds-dbl/CrossingThresholds-dbl
-17.5576% +/- 14.598%
MultiSource/Benchmarks/TSVC/Reductions-dbl/Reductions-dbl
-29.5708% +/- 7.09058%
MultiSource/Benchmarks/TSVC/Reductions-flt/Reductions-flt
-34.9471% +/- 11.4391%
SingleSource/Benchmarks/BenchmarkGame/puzzle
-25.1347% +/- 11.0104%
SingleSource/Benchmarks/Misc/flops-8
-17.7297% +/- 9.79061%
SingleSource/Benchmarks/Shootout-C++/ary3
-35.5018% +/- 23.9458%
SingleSource/Regression/C/uint64_to_float
-56.3165% +/- 25.4234%
SingleSource/UnitTests/Vectorizer/gcc-loops
-18.5309% +/- 6.8496%
regressions:
MultiSource/Benchmarks/ASCI_Purple/SMG2000/smg2000
18.351% +/- 12.156%
SingleSource/Benchmarks/Shootout-C++/methcall
27.3086% +/- 14.4733%
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197099 91177308-0d34-0410-b5e6-96231b3b80d8
The assertion was checking that the virtual register VReg used to represent the
physical register PReg uses the same register class as the one passed to
MachineFunction::addLiveIn.
This is over-constraining because it is sufficient to check that the register
class of VReg (VRegRC) is a subclass of the register class of PReg (PRegRC) and
that VRegRC contains PReg.
Indeed, if VReg gets constrained because of some operation constraints
between two calls of MachineFunction::addLiveIn, the original assertion
cannot match.
This fixes <rdar://problem/15633429>.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197097 91177308-0d34-0410-b5e6-96231b3b80d8
For one predicate to subsume another, they must both check the same condition
register. Failure to check this prerequisite was causing miscompiles.
Fixes PR18003.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197089 91177308-0d34-0410-b5e6-96231b3b80d8
The linkers on these systems don't have anything special to do with these
symbols. Since the intent is for them to be absent from the final object,
just treat them as private.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197080 91177308-0d34-0410-b5e6-96231b3b80d8
This reverts commit r197073.
The test seems to be failing on some buildbots for unknown reasons.
Reverting until I can figure that out. If anyone's got a reproduction
(.s and .o together would be great) - I'd really appreciate it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197079 91177308-0d34-0410-b5e6-96231b3b80d8
This commit does not complete the type units feature - there are issues
around fission support (skeletal type units, pubtypes/pubnames) and
hashing of some types including those containing references to types in
other type units.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197073 91177308-0d34-0410-b5e6-96231b3b80d8
floating-point reciprocal square root step LLVM AArch64 intrinsics to
use f32/f64 types, rather than their vector equivalents.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197067 91177308-0d34-0410-b5e6-96231b3b80d8
point reciprocal exponent, and floating-point reciprocal square root estimate
LLVM AArch64 intrinsics to use f32/f64 types, rather than their vector
equivalents.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197066 91177308-0d34-0410-b5e6-96231b3b80d8
The tests were no longer using fast-isel at all (MachO needs an "ios" rather
than "darwin" triple at the moment and Linux needs ARM mode). Once that was
corrected, the verifier complained about a t2ADDri created for the alloca.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197046 91177308-0d34-0410-b5e6-96231b3b80d8
I moved a test from avx512-vbroadcast-crash.ll to avx512-vbroadcast.ll
I defined HasAVX512 predicate as AssemblerPredicate. It means that you should invoke llvm-mc with "-mcpu=knl" to get encoding for AVX-512 instructions. I need this to let AsmMatcher to set different encoding for AVX and AVX-512 instructions that have the same mnemonic and operands (all scalar instructions).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197041 91177308-0d34-0410-b5e6-96231b3b80d8
In such cases it's often better to test the result of the negation instead,
since the negation also sets CC.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197032 91177308-0d34-0410-b5e6-96231b3b80d8
DAGCombiner could fold (truncate (load)) -> smaller load if the original
load was the width of the truncation result or wider. This patch extends
it to handle cases where the original load was narrower (and so the
extension type stays the same).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197030 91177308-0d34-0410-b5e6-96231b3b80d8
The combination of inline asm, stack realignment, and dynamic allocas
turns out to be too common to reject out of hand.
ASan inserts empy inline asm fragments and uses aligned allocas.
Compiling any trivial function containing a dynamic alloca with ASan is
enough to trigger the check.
XFAIL the test cases that would be miscompiled and add one that uses the
relevant functionality.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@196986 91177308-0d34-0410-b5e6-96231b3b80d8
This re-lands commit r196876, which was reverted in r196879.
The tests have been fixed to pass on platforms with a stack alignment
larger than 4.
Update to clang side tests will land shortly.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@196939 91177308-0d34-0410-b5e6-96231b3b80d8
immediately after SSE scalar fp instructions like addss or mulss.
Added patterns to select SSE scalar fp arithmetic instructions from a scalar
fp operation followed by a blend.
For example, given the following code:
__m128 foo(__m128 A, __m128 B) {
A[0] += B[0];
return A;
}
previously we generated:
addss %xmm0, %xmm1
movss %xmm1, %xmm0
now we generate:
addss %xmm1, %xmm0
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@196925 91177308-0d34-0410-b5e6-96231b3b80d8
Save S2(reg 18) only when we are calling floating point stubs that
have a return value of float or complex. Some more work to make this
better but this is the first step.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@196921 91177308-0d34-0410-b5e6-96231b3b80d8
One unusual feature of the z architecture is that the result of a
previous load can be reused indefinitely for subsequent loads, even if
a cache-coherent store to that location is performed by another CPU.
A special serializing instruction must be used if you want to force
a load to be reattempted.
Since volatile loads are not supposed to be omitted in this way,
we should insert a serializing instruction before each such load.
The same goes for atomic loads.
The patch implements this at the IR->DAG boundary, in a similar way
to atomic fences. It is a no-op for targets other than SystemZ.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@196906 91177308-0d34-0410-b5e6-96231b3b80d8
One unusual feature of the z architecture is that the result of a
previous load can be reused indefinitely for subsequent loads, even if
a cache-coherent store to that location is performed by another CPU.
A special serializing instruction must be used if you want to force
a load to be reattempted.
Since volatile loads are not supposed to be omitted in this way,
we should insert a serializing instruction before each such load.
The same goes for atomic loads.
The patch implements this at the IR->DAG boundary, in a similar way
to atomic fences. It is a no-op for targets other than SystemZ.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@196905 91177308-0d34-0410-b5e6-96231b3b80d8
For stack frames requiring realignment, three pointers may be needed:
- ebp to address incoming arguments
- esi (could be any callee-saved register) to address locals
- esp to address outgoing arguments
We would use esi unconditionally without verifying that it did not
conflict with inline assembly.
This change doesn't do the verification, it simply emits a fatal error
on functions that use stack realignment, dynamic SP adjustments, and
inline assembly.
Because stack realignment is common on Windows, we also no longer assume
that MS inline assembly clobbers esp. Instead, we analyze the inline
instructions for implicit definitions and check if esp is there. If so,
we require the use of a base pointer and consider it in the condition
above.
Mostly fixes PR16830, but we could try harder to find a non-conflicting
base pointer.
Reviewers: sunfish
Differential Revision: http://llvm-reviews.chandlerc.com/D1317
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@196876 91177308-0d34-0410-b5e6-96231b3b80d8
