The tale starts with r212808 which attempted to fix inversion of the low
and high bits when lowering MUL_LOHI. Sadly, that commit did not include
any positive test cases, and just removed some operations from a test
case where the actual logic being changed isn't fully visible from the
test.
What this commit did was two things. First, it reversed the low and high
results in the formation of the MERGE_VALUES node for the multiple
results. This is entirely correct.
Second it changed the shuffles for extracting the low and high
components from the i64 results of the multiplies to extract them
assuming a big-endian-style encoding of the multiply results. This
second change is wrong. There is no big-endian encoding in x86, the
results of the multiplies are normal v2i64s: when cast to v4i32, the low
i32s are at offsets 0 and 2, and the high i32s are at offsets 1 and 3.
However, the first change wasn't enough to actually fix the bug, which
is (I assume) why the second change was also made. There was another bug
in the MERGE_VALUES formation: we weren't using a VTList, and so were
getting a single result node! When grabbing the *second* result from the
node, we got... well.. colud be anything. I think this *appeared* to
invert things, but had to be causing other problems as well.
Fortunately, I fixed the MERGE_VALUES issue in r213931, so we should
have been fine, right? NOOOPE! Because the core bug was never addressed,
the test in vector-idiv failed when I fixed the MERGE_VALUES node.
Because there are essentially no docs for this node, I had to guess at
how to fix it and tried swapping the operands, restoring the order of
the original code before r212808. While this "fixed" the test case (in
that we produced the write instructions) we were still extracting the
wrong elements of the i64s, and thus PR20355 was still broken.
This commit essentially reverts the big-endian-style extraction part of
r212808 and goes back to the original masks which were correct. Now that
the MERGE_VALUES node formation is also correct, everything works. I've
also included a more detailed test from PR20355 to make sure this stays
fixed.
llvm-svn: 214011
The clever way to implement signed multiplication with unsigned *is
already implemented* and tested and working correctly. The bug is
somewhere else. Re-investigating.
This will teach me to not scroll far enough to read the code that did
what I thought needed to be done.
llvm-svn: 214009
signed multiplication is requested. While there is not a difference in
the *low* half of the result, the *high* half (used specifically to
implement the signed division by these constants) certainly is used. The
test case I've nuked was actively asserting wrong code.
There is a delightful solution to doing signed multiplication even when
we don't have it that Richard Smith has crafted, but I'll add the
machinery back and implement that in a follow-up patch. This at least
restores correctness.
llvm-svn: 214007
address of the stack guard was being spilled to the stack.
Previously the address of the stack guard would get spilled to the stack if it
was impossible to keep it in a register. This patch introduces a new target
independent node and pseudo instruction which gets expanded post-RA to a
sequence of instructions that load the stack guard value. Register allocator
can now just remat the value when it can't keep it in a register.
<rdar://problem/12475629>
llvm-svn: 213967
Reverted by Eric Christopher (Thanks!) in r212203 after Bob Wilson
reported LTO issues. Duncan Exon Smith and Aditya Nandakumar helped
provide a reduced reproduction, though the failure wasn't too hard to
guess, and even easier with the example to confirm.
The assertion that the subprogram metadata associated with an
llvm::Function matches the scope data referenced by the DbgLocs on the
instructions in that function is not valid under LTO. In LTO, a C++
inline function might exist in multiple CUs and the subprogram metadata
nodes will refer to the same llvm::Function. In this case, depending on
the order of the CUs, the first intance of the subprogram metadata may
not be the one referenced by the instructions in that function and the
assertion will fail.
A test case (test/DebugInfo/cross-cu-linkonce-distinct.ll) is added, the
assertion removed and a comment added to explain this situation.
This was then reverted again in r213581 as it caused PR20367. The root
cause of this was the early exit in LiveDebugVariables meant that
spurious DBG_VALUE intrinsics that referenced dead variables were not
removed, causing an assertion/crash later on. The fix is to have
LiveDebugVariables strip all DBG_VALUE intrinsics in functions without
debug info as they're not needed anyway. Test case added to cover this
situation (that occurs when a debug-having function is inlined into a
nodebug function) in test/DebugInfo/X86/nodebug_with_debug_loc.ll
Original commit message:
If a function isn't actually in a CU's subprogram list in the debug info
metadata, ignore all the DebugLocs and don't try to build scopes, track
variables, etc.
While this is possibly a minor optimization, it's also a correctness fix
for an incoming patch that will add assertions to LexicalScopes and the
debug info verifier to ensure that all scope chains lead to debug info
for the current function.
Fix up a few test cases that had broken/incomplete debug info that could
violate this constraint.
Add a test case where this occurs by design (inlining a
debug-info-having function in an attribute nodebug function - we want
this to work because /if/ the nodebug function is then inlined into a
debug-info-having function, it should be fine (and will work fine - we
just stitch the scopes up as usual), but should the inlining not happen
we need to not assert fail either).
llvm-svn: 213952
* Add CUs to the named CU node
* Add missing DW_TAG_subprogram nodes
* Add llvm::Functions to the DW_TAG_subprogram nodes
This cleans up the tests so that they don't break under a
soon-to-be-made change that is more strict about such things.
llvm-svn: 213951
which have successfully round-tripped through the combine phase, and use
this to ensure all operands to DAG nodes are visited by the combiner,
even if they are only added during the combine phase.
This is critical to have the combiner reach nodes that are *introduced*
during combining. Previously these would sometimes be visited and
sometimes not be visited based on whether they happened to end up on the
worklist or not. Now we always run them through the combiner.
This fixes quite a few bad codegen test cases lurking in the suite while
also being more principled. Among these, the TLS codegeneration is
particularly exciting for programs that have this in the critical path
like TSan-instrumented binaries (although I think they engineer to use
a different TLS that is faster anyways).
I've tried to check for compile-time regressions here by running llc
over a merged (but not LTO-ed) clang bitcode file and observed at most
a 3% slowdown in llc. Given that this is essentially a worst case (none
of opt or clang are running at this phase) I think this is tolerable.
The actual LTO case should be even less costly, and the cost in normal
compilation should be negligible.
With this combining logic, it is possible to re-legalize as we combine
which is necessary to implement PSHUFB formation on x86 as
a post-legalize DAG combine (my ultimate goal).
Differential Revision: http://reviews.llvm.org/D4638
llvm-svn: 213898
vector operation legalization with support for custom target lowering
and fallback to expand when it fails, and use this to implement sext and
anyext load lowering for x86 in a more principled way.
Previously, the x86 backend relied on a target DAG combine to "combine
away" sextload and extload nodes prior to legalization, or would expand
them during legalization with terrible code. This is particularly
problematic because the DAG combine relies on running over non-canonical
DAG nodes at just the right time to match several common and important
patterns. It used a combine rather than lowering because we didn't have
good lowering support, and to expose some tricks being employed to more
combine phases.
With this change it becomes a proper lowering operation, the backend
marks that it can lower these nodes, and I've added support for handling
the canonical forms that don't have direct legal representations such as
sextload of a v4i8 -> v4i64 on AVX1. With this change, our test cases
for this behavior continue to pass even after the DAG combiner beigns
running more systematically over every node.
There is some noise caused by this in the test suite where we actually
use vector extends instead of subregister extraction. This doesn't
really seem like the right thing to do, but is unlikely to be a critical
regression. We do regress in one case where by lowering to the
target-specific patterns early we were able to combine away extraneous
legal math nodes. However, this regression is completely addressed by
switching to a widening based legalization which is what I'm working
toward anyways, so I've just switched the test to that mode.
Differential Revision: http://reviews.llvm.org/D4654
llvm-svn: 213897
This patch minimizes the number of nops that must be emitted on X86 to satisfy
stackmap shadow constraints.
To minimize the number of nops inserted, the X86AsmPrinter now records the
size of the most recent stackmap's shadow in the StackMapShadowTracker class,
and tracks the number of instruction bytes emitted since the that stackmap
instruction was encountered. Padding is emitted (if it is required at all)
immediately before the next stackmap/patchpoint instruction, or at the end of
the basic block.
This optimization should reduce code-size and improve performance for people
using the llvm stackmap intrinsic on X86.
<rdar://problem/14959522>
llvm-svn: 213892
When we had a vector_shuffle where we had an input from each vector, we
could miscompile it because we were assuming the input from V2 wouldn't
be moved from where it was on the vector.
Added a test case.
llvm-svn: 213826
It isn't reasonable to test storing things using undef pointers --
storing through those is at best "good luck" and really should be
transformed to "unreachable". Random changes in the combiner can
randomly break these tests for no good reason. I'm following up on the
original commit regarding the right long-term strategy here.
llvm-svn: 213810
The transform to constant fold unary operations with an AND across a
vector comparison applies when the constant is not a splat of a scalar
as well.
llvm-svn: 213800
The folding of unary operations through a vector compare and mask operation
is only safe if the unary operation result is of the same size as its input.
For example, it's not safe for [su]itofp from v4i32 to v4f64.
llvm-svn: 213799
Constant fold the lanes of the input constant build_vector individually
so we correctly handle when the vector elements are not all the same
constant value.
PR20394
llvm-svn: 213798
This chang fully reverts r211771.
That revision added a canonicalization rule which has the potential to causes a
combine-cycle in the target-independent canonicalizing DAG combine.
The plan is to move the logic that forms target specific addsub nodes as part of
the lowering of shuffles.
llvm-svn: 213736
instruction sequences with CHECK-NEXT for these test cases.
This notably exposes how absolutely horrible the generated code is for
several of these test cases, and will make any future updates to the
test as our vector instruction selection gets better.
llvm-svn: 213732
insertions.
The old behavior could cause arbitrarily bad memory usage in the DAG
combiner if there was heavy traffic of adding nodes already on the
worklist to it. This commit switches the DAG combine worklist to work
the same way as the instcombine worklist where we null-out removed
entries and only add new entries to the worklist. My measurements of
codegen time shows slight improvement. The memory utilization is
unsurprisingly dominated by other factors (the IR and DAG itself
I suspect).
This change results in subtle, frustrating churn in the particular order
in which DAG combines are applied which causes a number of minor
regressions where we fail to match a pattern previously matched by
accident. AFAICT, all of these should be using AddToWorklist to directly
or should be written in a less brittle way. None of the changes seem
drastically bad, and a few of the changes seem distinctly better.
A major change required to make this work is to significantly harden the
way in which the DAG combiner handle nodes which become dead
(zero-uses). Previously, we relied on the ability to "priority-bump"
them on the combine worklist to achieve recursive deletion of these
nodes and ensure that the frontier of remaining live nodes all were
added to the worklist. Instead, I've introduced a routine to just
implement that precise logic with no indirection. It is a significantly
simpler operation than that of the combiner worklist proper. I suspect
this will also fix some other problems with the combiner.
I think the x86 changes are really minor and uninteresting, but the
avx512 change at least is hiding a "regression" (despite the test case
being just noise, not testing some performance invariant) that might be
looked into. Not sure if any of the others impact specific "important"
code paths, but they didn't look terribly interesting to me, or the
changes were really minor. The consensus in review is to fix any
regressions that show up after the fact here.
Thanks to the other reviewers for checking the output on other
architectures. There is a specific regression on ARM that Tim already
has a fix prepped to commit.
Differential Revision: http://reviews.llvm.org/D4616
llvm-svn: 213727
Canonicalize shuffles according to rules:
* shuffle(A, shuffle(A, B)) -> shuffle(shuffle(A,B), A)
* shuffle(B, shuffle(A, B)) -> shuffle(shuffle(A,B), B)
* shuffle(B, shuffle(A, Undef)) -> shuffle(shuffle(A, Undef), B)
This patch helps identifying more shuffle pairs that could be combined reusing
the already existing rules in the DAGCombiner.
Added new test 'combine-vec-shuffle-5.ll' to verify that the canonicalized
shuffles are now folded into a single shuffle node by the DAGCombiner.
Added more test cases to 'combine-vec-shuffle-4.ll'.
llvm-svn: 213504
Function @test3c should check that the DAGCombiner is able to fold a pair of
shuffles into a new shuffle with a permute mask of <6,7,2,3>. However, one of
the shuffles in @test3c had a wrong permute mask; this prevented the DAGCombiner
from folding the shuffles into the expected result.
Now that the shuffle mask is fixed, the backend correctly folds the two shuffles
in function @test3c into a single movhlps instruction.
llvm-svn: 213451
Actual support for softening f16 operations is still limited, and can be added
when it's needed. But Soften is much closer to being a useful thing to try
than keeping it Legal when no registers can actually hold such values.
Longer term, we probably want something between Soften and Promote semantics
for most targets, it'll be more efficient to promote the 4 basic operations to
f32 than libcall them.
llvm-svn: 213372
Since the result of a SETCC for X86 is 0 or -1 in each lane, we can
move unary operations, in this case [su]int_to_fp through the mask
operation and constant fold the operation away. Generally speaking:
UNARYOP(AND(VECTOR_CMP(x,y), constant))
--> AND(VECTOR_CMP(x,y), constant2)
where constant2 is UNARYOP(constant).
This implements the transform where UNARYOP is [su]int_to_fp.
For example, consider the simple function:
define <4 x float> @foo(<4 x float> %val, <4 x float> %test) nounwind {
%cmp = fcmp oeq <4 x float> %val, %test
%ext = zext <4 x i1> %cmp to <4 x i32>
%result = sitofp <4 x i32> %ext to <4 x float>
ret <4 x float> %result
}
Before this change, the SSE code is generated as:
LCPI0_0:
.long 1 ## 0x1
.long 1 ## 0x1
.long 1 ## 0x1
.long 1 ## 0x1
.section __TEXT,__text,regular,pure_instructions
.globl _foo
.align 4, 0x90
_foo: ## @foo
cmpeqps %xmm1, %xmm0
andps LCPI0_0(%rip), %xmm0
cvtdq2ps %xmm0, %xmm0
retq
After, the code is improved to:
LCPI0_0:
.long 1065353216 ## float 1.000000e+00
.long 1065353216 ## float 1.000000e+00
.long 1065353216 ## float 1.000000e+00
.long 1065353216 ## float 1.000000e+00
.section __TEXT,__text,regular,pure_instructions
.globl _foo
.align 4, 0x90
_foo: ## @foo
cmpeqps %xmm1, %xmm0
andps LCPI0_0(%rip), %xmm0
retq
The cvtdq2ps has been constant folded away and the floating point 1.0f
vector lanes are materialized directly via the ModRM operand of andps.
llvm-svn: 213342
Previously we asserted on this code. Currently compiler-rt doesn't
actually implement any of these new libcalls, but external help is
pretty much the only viable option for LLVM.
I've followed the much more generic "__truncST2" naming, as opposed to
the odd name for f32 -> f16 truncation. This can obviously be changed
later, or overridden by any targets that need to.
llvm-svn: 213252
x86 has no native ability to extend an f16 to f64, but the same result
is obtained if we expand it into two separate extensions: f16 -> f32
-> f64.
Unfortunately the same is not true for truncate, so that still results
in a compilation failure.
llvm-svn: 213251
This makes the two intrinsics @llvm.convert.from.f16 and
@llvm.convert.to.f16 accept types other than simple "float". This is
only strictly needed for the truncate operation, since otherwise
double rounding occurs and there's no way to represent the strict IEEE
conversion. However, for symmetry we allow larger types in the extend
too.
During legalization, we can expand an "fp16_to_double" operation into
two extends for convenience, but abort when the truncate isn't legal. A new
libcall is probably needed here.
Even after this commit, various target tweaks are needed to actually use the
extended intrinsics. I've put these into separate commits for clarity, so there
are no actual tests of f64 conversion here.
llvm-svn: 213248
Before this change, method 'isShuffleMaskLegal' didn't know that shuffles
implementing a 'movhlps' operation were perfectly legal for SSE targets.
This patch adds the missing check for 'isMOVHLPSMask' inside method
'isShuffleMaskLegal' to fix the problem.
The reason why it is important to do this is because the DAGCombiner
conservatively avoids combining a pair of shuffles if the resulting shuffle
node has an illegal mask. Before this patch, shuffles with a MOVHLPS mask were
wrongly considered not to be legal. This was the root cause of some poor-code
generation bugs.
llvm-svn: 213137
This patch adds two new rules to the DAGCombiner:
1. shuffle (shuffle A, Undef, M0), B, M1 -> shuffle A, B, M2
2. shuffle (shuffle A, Undef, M0), A, M1 -> shuffle A, Undef, M2
We only do this if the combined shuffle is legal for the target.
Example:
;;
define <4 x float> @test(<4 x float> %a, <4 x float> %b) {
%1 = shufflevector <4 x float> %a, <4 x float> undef, <4 x i32><i32 6, i32 0, i32 1, i32 7>
%2 = shufflevector <4 x float> %1, <4 x float> %b, <4 x i32><i32 1, i32 2, i32 4, i32 5>
ret <4 x i32> %2
}
;;
(using llc -mcpu=corei7 -march=x86-64)
Before, the x86 backend generated:
pshufd $120, %xmm0, %xmm0
shufps $-108, %xmm0, %xmm1
movaps %xmm1, %xmm0
Now the x86 backend generates:
movsd %xmm1, %xmm0
llvm-svn: 213069
Instead of specifying 32-bit x86, specify 32-bit x86 linux.
This test is testing a very specific behavior which changed with
WinCOFF's constant pools.
llvm-svn: 213041
NAKAMURA Takumi