The test case included in r280979 wasn't checking what it was supposed to be
checking for the predicated store case. Fixing the test revealed that the
multi-use case (when a pointer is used by both vectorized and scalarized memory
accesses) wasn't being handled properly. We can't skip over
non-consecutive-like pointers since they may have looked consecutive-like with
a different memory access.
llvm-svn: 280992
Previously, all consecutive pointers were marked uniform after vectorization.
However, if a consecutive pointer is used by a memory access that is eventually
scalarized, the pointer won't remain uniform after all. An example is
predicated stores. Even though a predicated store may be consecutive, it will
still be scalarized, making it's pointer operand non-uniform.
This patch updates the logic in collectLoopUniforms to consider the cases where
a memory access may be scalarized. If a memory access may be scalarized, its
pointer operand is not marked uniform. The determination of whether a given
memory instruction will be scalarized or not has been moved into a common
function that is used by the vectorizer, cost model, and legality analysis.
Differential Revision: https://reviews.llvm.org/D24271
llvm-svn: 280979
Summary:
When cloning blocks for prologue/epilogue we need to replicate the loop
structure from the original loop. It wasn't a problem for the innermost
loops, but it led to an incorrect loop info when we unrolled a loop with
a child loop - in this case created prologue-loop had a child loop, but
loop info didn't reflect that.
This fixes PR28888.
Reviewers: chandlerc, sanjoy, hfinkel
Subscribers: llvm-commits, silvas
Differential Revision: https://reviews.llvm.org/D24203
llvm-svn: 280901
We can't create metadata-valued PHIs; don't try to do so when sinking.
I created a test case for this using the @llvm.type.test intrinsic, because it
takes a metadata parameter and does not have severe side effects (thus
SimplifyCFG is willing to otherwise sink it).
Previously, running the test case would crash with:
Invalid use of metadata!
%.sink = select i1 %flag, metadata <...>, metadata <0x4e45dc0>
LLVM ERROR: Broken function found, compilation aborted!
llvm-svn: 280866
This is a revert of r280676 which was a revert of r280637;
ie, this is r280637 again. It was speculatively reverted to
help debug buildbot failures.
llvm-svn: 280861
Summary:
LSV replaces multiple adjacent loads with one vectorized load and a
bunch of extractelement instructions. This patch makes the
extractelement instructions' names match those of the original loads,
for (hopefully) improved readability.
Reviewers: asbirlea, tstellarAMD
Subscribers: arsenm, mzolotukhin
Differential Revision: https://reviews.llvm.org/D23748
llvm-svn: 280818
This fixes a similar issue to the one already fixed by r280804
(revieved in D24256). Revision 280804 fixed the problem with unsafe dyn_casts
in the extrq/extrqi combining logic. However, it turns out that even the
insertq/insertqi logic was affected by the same problem.
llvm-svn: 280807
This patch fixes an assertion failure caused by unsafe dynamic casts on the
constant operands of sse4a intrinsic calls to extrq/extrqi
The combine logic that simplifies sse4a extrq/extrqi intrinsic calls currently
checks if the input operands are constants. Internally, that logic relies on
dyn_casts of values returned by calls to method Constant::getAggregateElement.
However, method getAggregateElemet may return nullptr if the constant element
cannot be retrieved. So, all the dyn_casts can potentially fail. This is what
happens for example if a constexpr value is passed in input to an extrq/extrqi
intrinsic call.
This patch fixes the problem by using a dyn_cast_or_null (instead of a simple
dyn_cast) on the result of each call to Constant::getAggregateElement.
Added reproducible test cases to x86-sse4a.ll.
Differential Revision: https://reviews.llvm.org/D24256
llvm-svn: 280804
I should have realised this the first time around, but if we're avoiding sinking stores where the operands come from allocas so they don't create selects, we also have to do the same for loads because SROA will be just as defective looking at loads of selected addresses as stores.
Fixes PR30188 (again).
llvm-svn: 280792
PR30292 showed a case where our PHI checking wasn't correct. We were checking that all values were used by the same PHI before deciding to sink, but we weren't checking that the incoming values for that PHI were what we expected. As a result, we had to bail out after block splitting which caused us to never reach a steady state in SimplifyCFG.
Fixes PR30292.
llvm-svn: 280790
Currently the pass updates branch weights in the IR if the function has
any PGO info (entry frequency is set). However we could still have
regions of the CFG that does not have branch weights collected (e.g. a
cold region). In this case we'd use static estimates. Since static
estimates for branches are determined independently, they are
inconsistent. Updating them can "randomly" inflate block frequencies.
I've run into this in a completely cold loop of h264ref from
SPEC. -Rpass-with-hotness showed the loop to be completely cold during
inlining (before JT) but completely hot during vectorization (after JT).
The new testcase demonstrate the problem. We check array elements
against 1, 2 and 3 in a loop. The check against 3 is the loop-exiting
check. The block names should be self-explanatory.
In this example, jump threading incorrectly updates the weight of the
loop-exiting branch to 0, drastically inflating the frequency of the
loop (in the range of billions).
There is no run-time profile info for edges inside the loop, so branch
probabilities are estimated. These are the resulting branch and block
frequencies for the loop body:
check_1 (16)
(8) / |
eq_1 | (8)
\ |
check_2 (16)
(8) / |
eq_2 | (8)
\ |
check_3 (16)
(1) / |
(loop exit) | (15)
|
(back edge)
First we thread eq_1 -> check_2 to check_3. Frequencies are updated to
remove the frequency of eq_1 from check_2 and then from the false edge
leaving check_2. Changed frequencies are highlighted with * *:
check_1 (16)
(8) / |
eq_1~ | (8)
/ |
/ check_2 (*8*)
/ (8) / |
\ eq_2 | (*0*)
\ \ |
` --- check_3 (16)
(1) / |
(loop exit) | (15)
|
(back edge)
Next we thread eq_1 -> check_3 and eq_2 -> check_3 to check_1 as new
back edges. Frequencies are updated to remove the frequency of eq_1 and
eq_3 from check_3 and then the false edge leaving check_3 (changed
frequencies are highlighted with * *):
check_1 (16)
(8) / |
eq_1~ | (8)
/ |
/ check_2 (*8*)
/ (8) / |
/-- eq_2~ | (*0*)
(back edge) |
check_3 (*0*)
(*0*) / |
(loop exit) | (*0*)
|
(back edge)
As a result, the loop exit edge ends up with 0 frequency which in turn makes
the loop header to have maximum frequency.
There are a few potential problems here:
1. The profile data seems odd. There is a single profile sample of the
loop being entered. On the other hand, there are no weights inside the
loop.
2. Based on static estimation we shouldn't set edges to "extreme"
values, i.e. extremely likely or unlikely.
3. We shouldn't create profile metadata that is calculated from static
estimation. I am not sure what policy is but it seems to make sense to
treat profile metadata as something that is known to originate from
profiling. Estimated probabilities should only be reflected in BPI/BFI.
Any one of these would probably fix the immediate problem. I went for 3
because I think it's a good policy to have and added a FIXME about 2.
Differential Revision: https://reviews.llvm.org/D24118
llvm-svn: 280713
Summary:
Move early uses of spilled variables after CoroBegin.
For example, if a parameter had address taken, we may end up with the code
like:
define @f(i32 %n) {
%n.addr = alloca i32
store %n, %n.addr
...
call @coro.begin
This patch fixes the problem by moving uses of spilled variables after CoroBegin.
Reviewers: majnemer
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D24234
llvm-svn: 280678
This test code previously caused a failure in the module verifier,
because SimplifyCFG created this invalid instruction, which tries to
take the address of inline asm:
%.sink = select i1 %1, i64 ()* asm "mov $0, #1", "=r", i64 ()* asm %"mov $0, #2", "=r"
This has been fixed recently, presumably by James Molloy's patches that
re-wrote and changed parts of SimplifyCFG, so this patch just adds a
regression test for it.
Differential Revision: https://reviews.llvm.org/D24231
llvm-svn: 280660
Summary:
A frontend may designate a particular suspend to be final, by setting the second argument of the coro.suspend intrinsic to true. Such a suspend point has two properties:
* it is possible to check whether a suspended coroutine is at the final suspend point via coro.done intrinsic;
* a resumption of a coroutine stopped at the final suspend point leads to undefined behavior. The only possible action for a coroutine at a final suspend point is destroying it via coro.destroy intrinsic.
This patch adds final suspend handling logic to CoroEarly and CoroSplit passes.
Now, the final suspend point example from docs\Coroutines.rst compiles and produces expected result (see test/Transform/Coroutines/ex5.ll).
Reviewers: majnemer
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D24068
llvm-svn: 280646
memcpy with ld/st.
When InstCombine replaces a memcpy with loads+stores it does not copy over the
llvm.mem.parallel_loop_access from the memcpy instruction. This patch fixes
that.
Differential Revision: https://reviews.llvm.org/D23499
llvm-svn: 280617
Summary:
The inliner may need to determine where a given funclet unwinds to,
and this determination may depend on other funclets throughout the
funclet tree. The code that performs this walk in getUnwindDestToken
memoizes results to avoid redundant computations. In the case that
a funclet's unwind destination is derived from its ancestor, there's
code to walk back down the tree from the ancestor updating the memo
map of its descendants to record the unwind destination. This change
fixes that code to account for the case that some descendant has a
different unwind destination, which can happen if that unwind dest
is a descendant of the EHPad being queried and thus didn't determine
its unwind destination.
Also update test inline-funclets.ll, which is supposed to cover such
scenarios, to include a case that fails an assertion without this fix
but passes with it.
Fixes PR29151.
Reviewers: majnemer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D24117
llvm-svn: 280610
For the store of a wide value merged from a pair of values, especially int-fp pair,
sometimes it is more efficent to split it into separate narrow stores, which can
remove the bitwise instructions or sink them to colder places.
Now the feature is only enabled on x86 target, and only store of int-fp pair is
splitted. It is possible that the application scope gets extended with perf evidence
support in the future.
Differential Revision: https://reviews.llvm.org/D22840
llvm-svn: 280505
The motivating case occurs with SSE/AVX scalar intrinsics, so this is a first step towards
shrinking that to a single shufflevector.
Note that the transform is intentionally limited to shuffles that are equivalent to vector
selects to avoid creating arbitrary shuffle masks that may not lower well.
This should solve PR29126:
https://llvm.org/bugs/show_bug.cgi?id=29126
Differential Revision: https://reviews.llvm.org/D23886
llvm-svn: 280504
For uniform instructions, we're only required to generate a scalar value for
the first vector lane of each unroll iteration. Thus, if we have a reverse
interleaved group, computing the member index off the scalar GEP corresponding
to the last vector lane of its pointer operand technically makes the GEP
non-uniform. We should compute the member index off the first scalar GEP
instead.
I've added the updated member index computation to the existing reverse
interleaved group test.
llvm-svn: 280497
This patch fixes a crash caused by an incorrect folding of an ordered comparison
between a packed floating point vector and a splat vector of NaN.
An ordered comparison between a vector and a constant vector of NaN, should
always be folded into a constant vector where each element is i1 false.
Since revision 266175, SimplifyFCmpInst folds the ordered fcmp into a scalar
'false'. Later on, this would cause an assertion failure, since the value type
of the folded value doesn't match the expected value type of the uses of the
original instruction: "Assertion failed: New->getType() == getType() &&
"replaceAllUses of value with new value of different type!".
This patch fixes the issue and adds a test case to the already existing test
InstSimplify/floating-point-compares.ll.
Differential Revision: https://reviews.llvm.org/D24143
llvm-svn: 280488
We're sinking stores, which is a good thing, but in the process creating selects for the store address operand, which SROA/Mem2Reg can't look through, which caused serious regressions.
The real fix is in SROA, which I'll be looking into.
llvm-svn: 280470
While removing a scalar shackle from an icmp fold, I noticed that I couldn't find any tests to trigger
this code path.
The 'and' shrinking transform should be handled by InstCombiner::foldCastedBitwiseLogic()
or eliminated with InstSimplify. The icmp narrowing is part of InstCombiner::foldICmpWithCastAndCast().
Differential Revision: https://reviews.llvm.org/D24031
llvm-svn: 280370
This was a real restriction in the original version of SinkIfThenCodeToEnd. Now it's been rewritten, the restriction can be lifted.
As part of this, we handle a very common and useful case where one of the incoming branches is actually conditional. Consider:
if (a)
x(1);
else if (b)
x(2);
This produces the following CFG:
[if]
/ \
[x(1)] [if]
| | \
| | \
| [x(2)] |
\ | /
[ end ]
[end] has two unconditional predecessor arcs and one conditional. The conditional refers to the implicit empty 'else' arc. This same pattern can also be caused by an empty default block in a switch.
We can't sink the call to x() down to end because no call to x() happens on the third incoming arc (assume that x() has sideeffects for the sake of argument; if something is safe to speculate we could indeed sink nevertheless but this cannot happen in the general case and causes many extra selects).
We are now able to detect this case and split off the unconditional arcs to a common successor:
[if]
/ \
[x(1)] [if]
| | \
| | \
| [x(2)] |
\ / |
[sink.split] |
\ /
[ end ]
Now we can sink the call to x() into %sink.split. This can cause significant code simplification in many testcases.
llvm-svn: 280364
r279460 rewrote this function to be able to handle more than two incoming edges and took pains to ensure this didn't regress anything.
This time we change the logic for determining if an instruction should be sunk. Previously we used a single pass greedy algorithm - sink instructions until one requires more than one PHI node or we run out of instructions to sink.
This had the problem that sinking instructions that had non-identical but trivially the same operands needed extra logic so we sunk them aggressively. For example:
%a = load i32* %b %d = load i32* %b
%c = gep i32* %a, i32 0 %e = gep i32* %d, i32 1
Sinking %c and %e would naively require two PHI merges as %a != %d. But the loads are obviously equivalent (and maybe can't be hoisted because there is no common predecessor).
This is why we implemented the fairly complex function areValuesTriviallySame(), to look through trivial differences like this. However it's just not clever enough.
Instead, throw areValuesTriviallySame away, use pointer equality to check equivalence of operands and switch to a two-stage algorithm.
In the "scan" stage, we look at every sinkable instruction in isolation from end of block to front. If it's sinkable, we keep track of all operands that required PHI merging.
In the "sink" stage, we iteratively sink the last non-terminator in the source blocks. But when calculating how many PHIs are actually required to be inserted (to work out if we should stop or not) we remove any values that have already been sunk from the set of PHI-merges required, which allows us to be more aggressive.
This turns an algorithm with potentially recursive lookahead (looking through GEPs, casts, loads and any other instruction potentially not CSE'd) to two linear scans.
llvm-svn: 280351
As discussed in https://reviews.llvm.org/D22666, our current mechanism to
support -pg profiling, where we insert calls to mcount(), or some similar
function, is fundamentally broken. We insert these calls in the frontend, which
means they get duplicated when inlining, and so the accumulated execution
counts for the inlined-into functions are wrong.
Because we don't want the presence of these functions to affect optimizaton,
they should be inserted in the backend. Here's a pass which would do just that.
The knowledge of the name of the counting function lives in the frontend, so
we're passing it here as a function attribute. Clang will be updated to use
this mechanism.
Differential Revision: https://reviews.llvm.org/D22825
llvm-svn: 280347
-fprofile-dir=path allows the user to specify where .gcda files should be
emitted when the program is run. In particular, this is the first flag that
causes the .gcno and .o files to have different paths, LLVM is extended to
support this. -fprofile-dir= does not change the file name in the .gcno (and
thus where lcov looks for the source) but it does change the name in the .gcda
(and thus where the runtime library writes the .gcda file). It's different from
a GCOV_PREFIX because a user can observe that the GCOV_PREFIX_STRIP will strip
paths off of -fprofile-dir= but not off of a supplied GCOV_PREFIX.
To implement this we split -coverage-file into -coverage-data-file and
-coverage-notes-file to specify the two different names. The !llvm.gcov
metadata node grows from a 2-element form {string coverage-file, node dbg.cu}
to 3-elements, {string coverage-notes-file, string coverage-data-file, node
dbg.cu}. In the 3-element form, the file name is already "mangled" with
.gcno/.gcda suffixes, while the 2-element form left that to the middle end
pass.
llvm-svn: 280306
Summary:
Use MemorySSA, if requested, to do less conservative memory dependency
checking.
This change doesn't enable the MemorySSA enhanced EarlyCSE in the
default pipelines, so should be NFC.
Reviewers: dberlin, sanjoy, reames, majnemer
Subscribers: mcrosier, llvm-commits
Differential Revision: http://reviews.llvm.org/D19821
llvm-svn: 280279
This is a first step towards supporting deopt value lowering and reporting entirely with the register allocator. I hope to build on this in the near future to support live-on-return semantics, but I have a use case which allows me to test and investigate code quality with just the live-in semantics so I've chosen to start there. For those curious, my use cases is our implementation of the "__llvm_deoptimize" function we bind to @llvm.deoptimize. I'm choosing not to hard code that fact in the patch and instead make it configurable via function attributes.
The basic approach here is modelled on what is done for the "Live In" values on stackmaps and patchpoints. (A secondary goal here is to remove one of the last barriers to merging the pseudo instructions.) We start by adding the operands directly to the STATEPOINT SDNode. Once we've lowered to MI, we extend the remat logic used by the register allocator to fold virtual register uses into StackMap::Indirect entries as needed. This does rely on the fact that the register allocator rematerializes. If it didn't along some code path, we could end up with more vregs than physical registers and fail to allocate.
Today, we *only* fold in the register allocator. This can create some weird effects when combined with arguments passed on the stack because we don't fold them appropriately. I have an idea how to fix that, but it needs this patch in place to work on that effectively. (There's some weird interaction with the scheduler as well, more investigation needed.)
My near term plan is to land this patch off-by-default, experiment in my local tree to identify any correctness issues and then start fixing codegen problems one by one as I find them. Once I have the live-in lowering fully working (both correctness and code quality), I'm hoping to move on to the live-on-return semantics. Note: I don't have any *known* miscompiles with this patch enabled, but I'm pretty sure I'll find at least a couple. Thus, the "experimental" tag and the fact it's off by default.
Differential Revision: https://reviews.llvm.org/D24000
llvm-svn: 280250
Matthew Simpson