Summary: Port Dead Loop Deletion Pass to the new pass manager.
Reviewers: silvas, davide
Subscribers: llvm-commits, sanjoy, mcrosier
Differential Revision: https://reviews.llvm.org/D21483
llvm-svn: 275453
This patch prevents increases in the number of instructions, pre-instcombine,
due to induction variable scalarization. An increase in instructions can lead
to an increase in the compile-time required to simplify the induction
variables. We now maintain a new map for scalarized induction variables to
prevent us from converting between the scalar and vector forms.
This patch should resolve compile-time regressions seen after r274627.
llvm-svn: 275419
enables the code size optimisation to fold a rem and div into a single
aeabi_uidivmod call. This was not happening before because sdiv was converted
but srem not, and instructions with different signedness are not combined.
Differential Revision: http://reviews.llvm.org/D22214
llvm-svn: 275403
This pass hoists duplicated computations in the program. The primary goal of
gvn-hoist is to reduce the size of functions before inline heuristics to reduce
the total cost of function inlining.
Pass written by Sebastian Pop, Aditya Kumar, Xiaoyu Hu, and Brian Rzycki.
Important algorithmic contributions by Daniel Berlin under the form of reviews.
Differential Revision: http://reviews.llvm.org/D19338
llvm-svn: 275401
constant hoisting. It not only takes into account the number of uses and the
cost of expressions in which constants appear, but now also the resulting
integer range of the offsets. Thus, the algorithm maximizes the number of uses
within an integer range that will enable more efficient code generation. On
ARM, for example, this will enable code size optimisations because less
negative offsets will be created. Negative offsets/immediates are not supported
by Thumb1 thus preventing more compact instruction encoding.
Differential Revision: http://reviews.llvm.org/D21183
llvm-svn: 275382
We were able to fold masked loads with an all-ones mask to a normal
load. However, we couldn't turn a masked load with a mask with mixed
ones and undefs into a normal load.
llvm-svn: 275380
This now should also work with the interprocedural variant of the pass.
Slightly easier now that the yak is shaved.
Differential Revision: http://reviews.llvm.org/D22329
llvm-svn: 275363
Summary:
In Scalarizer::gather we see if we already have a scattered form of Op,
and in that case use the new form.
In the particular case of PR28108, the found ValueVector SV has size 2,
where the first Value is nullptr, and the second is indeed a proper Value.
The nullptr then caused an assert to blow when we tried to do
cast<Instruction>(SV[I]).
With this patch we check SV[I] before doing the cast, and if it's nullptr
we just skip over it.
I don't know the Scalarizer well enough to know if this is the best fix
or if something should be done else where to prevent the nullptr from
being in the ValueVector at all, but at least this avoids the crash
and looking at the test case output it looks reasonable.
Reviewers: hfinkel, frasercrmck, wala, mehdi_amini
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D21518
llvm-svn: 275359
Treat loads which clip before the start of a global initializer the same
way we treat clipping beyond the end of the initializer: use zeros.
llvm-svn: 275345
Summary:
LSV used to abort vectorizing a chain for interleaved load/store accesses that alias.
Allow a valid prefix of the chain to be vectorized, mark just the prefix and retry vectorizing the remaining chain.
Reviewers: llvm-commits, jlebar, arsenm
Subscribers: mzolotukhin
Differential Revision: http://reviews.llvm.org/D22119
llvm-svn: 275317
In D21740, we discussed trying to make this a more general matcher. However, I didn't see a clean
way to handle the regular m_Not cases and these non-splat vector patterns, so I've opted for the
direct approach here. If there are other potential uses of areInverseVectorBitmasks(), we could
move that helper function to a higher level.
There is an open question as to which is of these forms should be considered the canonical IR:
%sel = select <4 x i1> <i1 true, i1 false, i1 false, i1 true>, <4 x i32> %a, <4 x i32> %b
%shuf = shufflevector <4 x i32> %a, <4 x i32> %b, <4 x i32> <i32 0, i32 5, i32 6, i32 3>
Differential Revision: http://reviews.llvm.org/D22114
llvm-svn: 275289
The LCSSA pass itself will not generate several redundant PHI nodes in a single
exit block. However, such redundant PHI nodes don't violate LCSSA form, and may
be introduced by passes that preserve LCSSA, and/or preserved by the LCSSA pass
itself. So, assuming a single PHI node per exit block is not safe.
llvm-svn: 275217
Summary:
It's useful to have some visibility about which call sites are devirtualized,
especially for debug purposes. Another use case is a regression test on the
application side (like, Chromium).
Reviewers: pcc
Differential Revision: http://reviews.llvm.org/D22252
llvm-svn: 275145
Summary:
Aiming to correct the ordering of loads/stores. This patch changes the
insert point for loads to the position of the first load.
It updates the ordering method for loads to insert before, rather than after.
Before this patch the following sequence:
"load a[1], store a[1], store a[0], load a[2]"
Would incorrectly vectorize to "store a[0,1], load a[1,2]".
The correctness check was assuming the insertion point for loads is at
the position of the first load, when in practice it was at the last
load. An alternative fix would have been to invert the correctness check.
The current fix changes insert position but also requires reordering of
instructions before the vectorized load.
Updated testcases to reflect the changes.
Reviewers: tstellarAMD, llvm-commits, jlebar, arsenm
Subscribers: mzolotukhin
Differential Revision: http://reviews.llvm.org/D22071
llvm-svn: 275117
Make some AVX and AVX512 cast costs more precise.
Based on part of a patch by Elena Demikhovsky (D15604).
Differential Revision: http://reviews.llvm.org/D22064
llvm-svn: 275106
There's a little bit of churn in this patch because the initialization
mechanism is now shared between the old and the new PM. Other than
that, it's just a pretty mechanical translation.
llvm-svn: 275082
Summary:
For sample-based PGO, using BFI to calculate callsite count is sometime not accurate. This is because with sampling based approach, if a callsite resides in a hot loop deeply nested in a bunch of cold branches, the callsite's BFI frequency would be inaccurately calculated due to lack of samples in the cold branch.
E.g.
if (A1 && A2 && A3 && ..... && A10) {
for (i=0; i < 100000000; i++) {
callsite();
}
}
Assume that A1 to A100 are all 100% taken, and callsite has 1000 samples and thus is considerred hot. Because the loop's trip count is huge, it's normal that all branches outside the loop has no sample at all. As a result, we can only use static branch probability to derive the the frequency of the loop header. Assuming that static heuristic thinks each branch is 50% taken, then the count calculated from BFI will be 1/(2^10) of the actual value.
In order to get more accurate callsite count, we directly annotate the weight on the call instruction, and directly use it when checking callsite hotness.
Note that this mechanism can also be shared by instrumentation based callsite hotness analysis. The side benefit is that it breaks the dependency from Inliner to BFI as call count is embedded in the IR.
Reviewers: davidxl, eraman, dnovillo
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D22118
llvm-svn: 275073
Summary: Handle the case when there is only one incoming/outgoing edge for a visited basic block: use the block weight to adjust edge weight even when the edge has been visited before. This can help reduce inaccuracies introduced by incorrect basic block profile, as shown in the updated unittest.
Reviewers: davidxl, dnovillo
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D22180
llvm-svn: 275072
For functions which are known to return a specific argument, pointer-comparison
folding can look through the function calls as part of its analysis.
Differential Revision: http://reviews.llvm.org/D9387
llvm-svn: 275039
If a function is known to return one of its arguments, we can use that in order
to compute known bits of the return value.
Differential Revision: http://reviews.llvm.org/D9397
llvm-svn: 275036
A function can have one argument with the 'returned' attribute, indicating that
the associated argument is always the return value of the function. Add
FuncAttrs inference logic.
Differential Revision: http://reviews.llvm.org/D22202
llvm-svn: 275027
ConstantInt::getSExtValue may fail on >64-bit integers. Add checks to call
getSExtValue only on narrow integers.
As a minor aside, simplify slsr-gep.ll to remove unnecessary load instructions.
llvm-svn: 274982
While here move simplifyLoop() function to the new header, as
suggested by Chandler in the review.
Differential Revision: http://reviews.llvm.org/D21404
llvm-svn: 274959
Summary:
This way the metadata will be only generated when asserts enabled,
or when -enable-import-metadata specified
FIXED missing colon on requires.
Reviewers: tejohnson, eraman, mehdi_amini
Subscribers: mehdi_amini, llvm-commits
Differential Revision: http://reviews.llvm.org/D22167
llvm-svn: 274947
Summary:
This way the metadata will be only generated when asserts enabled,
or when -enable-import-metadata specified
Reviewers: tejohnson, eraman, mehdi_amini
Subscribers: mehdi_amini, llvm-commits
Differential Revision: http://reviews.llvm.org/D22167
llvm-svn: 274938
This isn't a sure thing (are 2 extra bitcasts less expensive than a logic op?),
but we'll try to err on the conservative side by going with the case that has
less IR instructions.
Note: This question came up in http://reviews.llvm.org/D22114 , but this part is
independent of that patch proposal, so I'm making this small change ahead of that
one.
See also:
http://reviews.llvm.org/rL274926
llvm-svn: 274932
Summary: As we will move to use uniformed hotness check in inliner, we do not need inline hints in SampleProfile pass any more.
Reviewers: dnovillo, davidxl
Subscribers: eraman, llvm-commits
Differential Revision: http://reviews.llvm.org/D19287
llvm-svn: 274918
This should be slightly more efficient and could avoid spurious overdefined
markings, as Eli pointed out.
Differential Revision: http://reviews.llvm.org/D22122
llvm-svn: 274905
We can fold truncs whose operand feeds from a load, if the trunc value
is available through a prior load/store.
This change is from: http://reviews.llvm.org/D21246, which folded the
trunc but missed the bitcast or ptrtoint/inttoptr required in the RAUW
call, when the load type didnt match the prior load/store type.
Differential Revision: http://reviews.llvm.org/D21791
llvm-svn: 274853
As discussed on D22106, improve the testing for constant folding sse scalar conversion intrinsics to ensure we are correctly handling special/out of range cases
llvm-svn: 274846
We can remove dead stores in the presence of fence instructions. Fence
does not change an otherwise thread local store to visible.
reviewers: reames, dexonsmith, jfb
Differential Revision: http://reviews.llvm.org/D22001
llvm-svn: 274795
By replacing dyn_cast of ConstantInt with m_Zero/m_One/m_AllOnes, we
allow these transforms for splat vectors.
Differential Revision: http://reviews.llvm.org/D21899
llvm-svn: 274696
Added metadata to be able to make statistics on how many functions
that have been imported have been removed. Also module name might
be helpfull when debugging.
Reviewers: tejohnson, eraman
Subscribers: mehdi_amini, llvm-commits
Differential Revision: http://reviews.llvm.org/D21943
llvm-svn: 274668
The dse_with_dbg_value.ll test committed with r273141 is removed because this
we no longer performs any type of back tracking, which is what was causing the
codegen differences with and without debug information.
Differential Revision: http://reviews.llvm.org/D21613
llvm-svn: 274660
We were still crashing in the "no change" case because LVI was not
getting invalidated.
See the thread "Should analyses be able to hold AssertingVH to IR?
(related to PR28400)" for more discussion.
llvm-svn: 274656
The cost model should not assume vector casts get completely scalarized, since
on targets that have vector support, the common case is a partial split up to
the legal vector size. So, when a vector cast gets split, the resulting casts
end up legal and cheap.
Instead of pessimistically assuming scalarization, base TTI can use the costs
the concrete TTI provides for the split vector, plus a fudge factor to account
for the cost of the split itself. This fudge factor is currently 1 by default,
except on AMDGPU where inserts and extracts are considered free.
Differential Revision: http://reviews.llvm.org/D21251
llvm-svn: 274642
We currently always vectorize induction variables. However, if an induction
variable is only used for counting loop iterations or computing addresses with
getelementptr instructions, we don't need to do this. Vectorizing these trivial
induction variables can create vector code that is difficult to simplify later
on. This is especially true when the unroll factor is greater than one, and we
create vector arithmetic when computing step vectors. With this patch, we check
if an induction variable is only used for counting iterations or computing
addresses, and if so, scalarize the arithmetic when computing step vectors
instead. This allows for greater simplification.
This patch addresses the suboptimal pointer arithmetic sequence seen in
PR27881.
Reference: https://llvm.org/bugs/show_bug.cgi?id=27881
Differential Revision: http://reviews.llvm.org/D21620
llvm-svn: 274627
This actually uncovered a surprisingly large chain of ultimately unused
TLI args.
From what I can gather, this argument is a remnant of when
isKnownNonNull would look at the TLI directly.
The current approach seems to be that InferFunctionAttrs runs early in
the pipeline and uses TLI to annotate the TLI-dependent non-null
information as return attributes.
This also removes the dependence of functionattrs on TLI altogether.
llvm-svn: 274455
Summary:
GetBoundryInstruction returns the last instruction as the instruction which follows or end(). Otherwise the last instruction in the boundry set is not being tested by isVectorizable().
Partially solve reordering of instructions. More extensive solution to follow.
Reviewers: tstellarAMD, llvm-commits, jlebar
Subscribers: escha, arsenm, mzolotukhin
Differential Revision: http://reviews.llvm.org/D21934
llvm-svn: 274389
integer.
Fixes issues on some architectures where we use arithmetic ops to build
vectors, which can cause bad things to happen for loads/stores of mixed
types.
Patch by Fiona Glaser
llvm-svn: 274307
This pass hoists duplicated computations in the program. The primary goal of
gvn-hoist is to reduce the size of functions before inline heuristics to reduce
the total cost of function inlining.
Pass written by Sebastian Pop, Aditya Kumar, Xiaoyu Hu, and Brian Rzycki.
Important algorithmic contributions by Daniel Berlin under the form of reviews.
Differential Revision: http://reviews.llvm.org/D19338
llvm-svn: 274305
Except the seed uniform instructions (conditional branch and consecutive ptr
instructions), dependencies to be added into uniform set should only be used
by existing uniform instructions or intructions outside of current loop.
Differential Revision: http://reviews.llvm.org/D21755
llvm-svn: 274262
Summary:
Found cases where DSE incorrectly add partially-overwritten intervals.
Please see the test case for details.
Reviewers: mcrosier, eeckstein, hfinkel
Subscribers: mcrosier, llvm-commits
Differential Revision: http://reviews.llvm.org/D21859
llvm-svn: 274237
This is a fix for PR27842.
An IR-level implementation of stack coloring tailored to work with
SafeStack. It is a bit weaker than the MI implementation in that it
does not the "lifetime start at first access" logic. This can be
improved in the future.
This patch also replaces the naive implementation of stack frame
layout with a greedy algorithm that can split existing stack slots
and even fit small objects inside the alignment padding of other
objects.
llvm-svn: 274162
Summary: This adds tests for covering all cases that FoldAndOfFCmps and FoldOrOfFCmps handle.
Reviewers: spatel
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D21844
llvm-svn: 274144
This is breaking an optimizaton remark test in clang. I've identified a couple fixes for that, but want to understand it better before I commit to anything.
llvm-svn: 274102
If a operation for a recurrence is an addition with no signed wrap and both input sign bits are 0, then the result sign bit must also be 0. Similar for the negative case.
I found this deficiency while playing around with a loop in the x86 backend that contained a signed division that could be optimized into an unsigned division if we could prove both inputs were positive. One of them being the loop induction variable. With this patch we can perform the conversion for this case. One of the test cases here is a contrived variation of the loop I was looking at.
Differential revision: http://reviews.llvm.org/D21493
llvm-svn: 274098
Revert "[InstCombine] Combine A->B->A BitCast"
as this appears to cause PR27996 and as discussed in http://reviews.llvm.org/D20847
This reverts commits r270135 and r263734.
llvm-svn: 274094
Summary:
This fixes bug: https://llvm.org/bugs/show_bug.cgi?id=28282
Currently the cost model of constant hoisting checks the bit width of the data type of the constants.
However, the actual immediate value is small enough and not need to be hoisted.
This patch checks for the actual bit width needed for the constant.
Reviewers: t.p.northover, rengolin
Subscribers: aemerson, rengolin, llvm-commits
Differential Revision: http://reviews.llvm.org/D21668
llvm-svn: 274073
This is a resubmittion of 263158 change after fixing the existing problem with intrinsics mangling (see LTO and intrinsics mangling llvm-dev thread for details).
This patch fixes the problem which occurs when loop-vectorize tries to use @llvm.masked.load/store intrinsic for a non-default addrspace pointer. It fails with "Calling a function with a bad signature!" assertion in CallInst constructor because it tries to pass a non-default addrspace pointer to the pointer argument which has default addrspace.
The fix is to add pointer type as another overloaded type to @llvm.masked.load/store intrinsics.
Reviewed By: reames
Differential Revision: http://reviews.llvm.org/D17270
llvm-svn: 274043
If the load is conditional we can't hoist its 0-iteration instance to
the preheader because that would make it unconditional. Thus we would
access a memory location that the original loop did not access.
llvm-svn: 273991
This is a resubmittion of 263158 change after fixing the existing problem with intrinsics mangling (see LTO and intrinsics mangling llvm-dev thread for details).
This patch fixes the problem which occurs when loop-vectorize tries to use @llvm.masked.load/store intrinsic for a non-default addrspace pointer. It fails with "Calling a function with a bad signature!" assertion in CallInst constructor because it tries to pass a non-default addrspace pointer to the pointer argument which has default addrspace.
The fix is to add pointer type as another overloaded type to @llvm.masked.load/store intrinsics.
Reviewed By: reames
Differential Revision: http://reviews.llvm.org/D17270
llvm-svn: 273892
It did not handle correctly cases without GEP.
The following loop wasn't vectorized:
for (int i=0; i<len; i++)
*to++ = *from++;
I use getPtrStride() to find Stride for memory access and return 0 is the Stride is not 1 or -1.
Re-commit rL273257 - revision: http://reviews.llvm.org/D20789
llvm-svn: 273864
Summary:
This is a straightforward extension of what LoopUnswitch does to
branches to guards. That is, we unswitch
```
for (;;) {
...
guard(loop_invariant_cond);
...
}
```
into
```
if (loop_invariant_cond) {
for (;;) {
...
// There is no need to emit guard(true)
...
}
} else {
for (;;) {
...
guard(false);
// SimplifyCFG will clean this up by adding an
// unreachable after the guard(false)
...
}
}
```
Reviewers: majnemer
Subscribers: mcrosier, llvm-commits, mzolotukhin
Differential Revision: http://reviews.llvm.org/D21725
llvm-svn: 273801
SimplifyCFG had logic to insert calls to llvm.trap for two very
particular IR patterns: stores and invokes of undef/null.
While InstCombine canonicalizes certain undefined behavior IR patterns
to stores of undef, phase ordering means that this cannot be relied upon
in general.
There are much better tools than llvm.trap: UBSan and ASan.
N.B. I could be argued into reverting this change if a clear argument as
to why it is important that we synthesize llvm.trap for stores, I'd be
hard pressed to see why it'd be useful for invokes...
llvm-svn: 273778
There are two separate issues:
- LLVM doesn't consider infinite loops to be side effects: we happily
hoist/sink above/below loops whose bounds are unknown.
- The absence of the noreturn attribute is insufficient for us to know
if a function will definitely return. Relying on noreturn in the
middle-end for any property is an accident waiting to happen.
llvm-svn: 273762
This intrinsic safely loads a function pointer from a virtual table pointer
using type metadata. This intrinsic is used to implement control flow integrity
in conjunction with virtual call optimization. The virtual call optimization
pass will optimize away llvm.type.checked.load intrinsics associated with
devirtualized calls, thereby removing the type check in cases where it is
not needed to enforce the control flow integrity constraint.
This patch also introduces the capability to copy type metadata between
global variables, and teaches the virtual call optimization pass to do so.
Differential Revision: http://reviews.llvm.org/D21121
llvm-svn: 273756
r273711 was reverted by r273743. The inliner needs to know about any
call sites in the inlined function. These were obscured if we replaced
a call to undef with an undef but kept the call around.
This fixes PR28298.
llvm-svn: 273753
Summary: Set ProfileSummary in SampleProfilerLoader.
Reviewers: davidxl, eraman
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D21702
llvm-svn: 273745
The bitset metadata currently used in LLVM has a few problems:
1. It has the wrong name. The name "bitset" refers to an implementation
detail of one use of the metadata (i.e. its original use case, CFI).
This makes it harder to understand, as the name makes no sense in the
context of virtual call optimization.
2. It is represented using a global named metadata node, rather than
being directly associated with a global. This makes it harder to
manipulate the metadata when rebuilding global variables, summarise it
as part of ThinLTO and drop unused metadata when associated globals are
dropped. For this reason, CFI does not currently work correctly when
both CFI and vcall opt are enabled, as vcall opt needs to rebuild vtable
globals, and fails to associate metadata with the rebuilt globals. As I
understand it, the same problem could also affect ASan, which rebuilds
globals with a red zone.
This patch solves both of those problems in the following way:
1. Rename the metadata to "type metadata". This new name reflects how
the metadata is currently being used (i.e. to represent type information
for CFI and vtable opt). The new name is reflected in the name for the
associated intrinsic (llvm.type.test) and pass (LowerTypeTests).
2. Attach metadata directly to the globals that it pertains to, rather
than using the "llvm.bitsets" global metadata node as we are doing now.
This is done using the newly introduced capability to attach
metadata to global variables (r271348 and r271358).
See also: http://lists.llvm.org/pipermail/llvm-dev/2016-June/100462.html
Differential Revision: http://reviews.llvm.org/D21053
llvm-svn: 273729
By putting all the possible commutations together, we simplify the code.
Note that this is NFCI, but I'm adding tests that actually exercise each
commutation pattern because we don't have this anywhere else.
llvm-svn: 273702
The interleaved access analysis currently assumes that the inserted run-time
pointer aliasing checks ensure the absence of dependences that would prevent
its instruction reordering. However, this is not the case.
Issues can arise from how code generation is performed for interleaved groups.
For a load group, all loads in the group are essentially moved to the location
of the first load in program order, and for a store group, all stores in the
group are moved to the location of the last store. For groups having members
involved in a dependence relation with any other instruction in the loop, this
reordering can violate the dependence.
This patch teaches the interleaved access analysis how to avoid breaking such
dependences, and should fix PR27626.
An assumption of the original analysis was that the accesses had been collected
in "program order". The analysis was then simplified by visiting the accesses
bottom-up. However, this ordering was never guaranteed for anything other than
single basic block loops. Thus, this patch also enforces the desired ordering.
Reference: https://llvm.org/bugs/show_bug.cgi?id=27626
Differential Revision: http://reviews.llvm.org/D19984
llvm-svn: 273687
reduce the number of comparisons.
Specifically, InstCombine can turn:
(i == 5334 || i == 5335)
into:
((i | 1) == 5335)
SimplifyCFG was already able to detect the pattern:
(i == 5334 || i == 5335)
to:
((i & -2) == 5334)
This patch supersedes D21315 and resolves PR27555
(https://llvm.org/bugs/show_bug.cgi?id=27555).
Thanks to David and Chandler for the suggestions!
Author: Thomas Jablin (tjablin)
Reviewers: majnemer chandlerc halfdan cycheng
http://reviews.llvm.org/D21397
llvm-svn: 273639
Summary:
This instcombine rule folds away trunc operations that have value available from a prior load or store.
This kind of code can be generated as a result of GVN widening the load or from source code as well.
Reviewers: reames, majnemer, sanjoy
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D21246
llvm-svn: 273608
When simplifying a load we need to make sure that the type of the
simplified value matches the type of the instruction we're processing.
In theory, we can handle casts here as we deal with constant data, but
since it's not implemented at the moment, we at least need to bail out.
This fixes PR28262.
llvm-svn: 273562
DeadStoreElimination can currently remove a small store rendered unnecessary by
a later larger one, but could not remove a larger store rendered unnecessary by
a series of later smaller ones. This adds that capability.
It works by keeping a map, which is used as an effective interval map, for each
store later overwritten only partially, and filling in that interval map as
more such stores are discovered. No additional walking or aliasing queries are
used. In the map forms an interval covering the the entire earlier store, then
it is dead and can be removed. The map is used as an interval map by storing a
mapping between the ending offset and the beginning offset of each interval.
I discovered this problem when investigating a performance issue with code like
this on PowerPC:
#include <complex>
using namespace std;
complex<float> bar(complex<float> C);
complex<float> foo(complex<float> C) {
return bar(C)*C;
}
which produces this:
define void @_Z4testSt7complexIfE(%"struct.std::complex"* noalias nocapture sret %agg.result, i64 %c.coerce) {
entry:
%ref.tmp = alloca i64, align 8
%tmpcast = bitcast i64* %ref.tmp to %"struct.std::complex"*
%c.sroa.0.0.extract.shift = lshr i64 %c.coerce, 32
%c.sroa.0.0.extract.trunc = trunc i64 %c.sroa.0.0.extract.shift to i32
%0 = bitcast i32 %c.sroa.0.0.extract.trunc to float
%c.sroa.2.0.extract.trunc = trunc i64 %c.coerce to i32
%1 = bitcast i32 %c.sroa.2.0.extract.trunc to float
call void @_Z3barSt7complexIfE(%"struct.std::complex"* nonnull sret %tmpcast, i64 %c.coerce)
%2 = bitcast %"struct.std::complex"* %agg.result to i64*
%3 = load i64, i64* %ref.tmp, align 8
store i64 %3, i64* %2, align 4 ; <--- ***** THIS SHOULD NOT BE HERE ****
%_M_value.realp.i.i = getelementptr inbounds %"struct.std::complex", %"struct.std::complex"* %agg.result, i64 0, i32 0, i32 0
%4 = lshr i64 %3, 32
%5 = trunc i64 %4 to i32
%6 = bitcast i32 %5 to float
%_M_value.imagp.i.i = getelementptr inbounds %"struct.std::complex", %"struct.std::complex"* %agg.result, i64 0, i32 0, i32 1
%7 = trunc i64 %3 to i32
%8 = bitcast i32 %7 to float
%mul_ad.i.i = fmul fast float %6, %1
%mul_bc.i.i = fmul fast float %8, %0
%mul_i.i.i = fadd fast float %mul_ad.i.i, %mul_bc.i.i
%mul_ac.i.i = fmul fast float %6, %0
%mul_bd.i.i = fmul fast float %8, %1
%mul_r.i.i = fsub fast float %mul_ac.i.i, %mul_bd.i.i
store float %mul_r.i.i, float* %_M_value.realp.i.i, align 4
store float %mul_i.i.i, float* %_M_value.imagp.i.i, align 4
ret void
}
the problem here is not just that the i64 store is unnecessary, but also that
it blocks further backend optimizations of the other uses of that i64 value in
the backend.
In the future, we might want to add a special case for handling smaller
accesses (e.g. using a bit vector) if the map mechanism turns out to be
noticeably inefficient. A sorted vector is also a possible replacement for the
map for small numbers of tracked intervals.
Differential Revision: http://reviews.llvm.org/D18586
llvm-svn: 273559
This is similar to the computeKnownBits improvement in rL268479.
There's probably more we can do for vector logic instructions, but
this should let us see non-splat constant masking ops that can
become vector selects instead of and/andn/or sequences.
Differential Revision: http://reviews.llvm.org/D21610
llvm-svn: 273459
We no longer have corresponding code in autoupgrade and the vast majority of the tests were fixed long time ago. Fix the remaining few. One of the verifier test cases is marked as XFAIL because it was passing only because the signature was incorrect.
llvm-svn: 273428
It did not handle correctly cases without GEP.
The following loop wasn't vectorized:
for (int i=0; i<len; i++)
*to++ = *from++;
I use getPtrStride() to find Stride for memory access and return 0 is the Stride is not 1 or -1.
Differential revision: http://reviews.llvm.org/D20789
llvm-svn: 273257
The BSWAP of vector types is quite efficiently implemented using vector shuffles on SSE/AVX targets, we should reflect the typical cost of this to encourage vectorization.
Differential Revision: http://reviews.llvm.org/D21521
llvm-svn: 273217
By moving this transform to InstSimplify from InstCombine, we sidestep the problem/question
raised by PR27869:
https://llvm.org/bugs/show_bug.cgi?id=27869
...where InstCombine turns an icmp+zext into a shift causing us to miss the fold.
Credit to David Majnemer for a draft patch of the changes to InstructionSimplify.cpp.
Differential Revision: http://reviews.llvm.org/D21512
llvm-svn: 273200
Summary: Inliner needs ACT when calling InlineFunction. Instead of nullptr, we need to pass it in from SampleProfileLoader
Reviewers: davidxl
Subscribers: eraman, vsk, danielcdh, llvm-commits
Differential Revision: http://reviews.llvm.org/D21205
llvm-svn: 273199
This reverts commit r273160, reapplying r273132.
RecursivelyDeleteTriviallyDeadInstructions cannot be called on a
parentless Instruction.
llvm-svn: 273162
This reverts commit r273132.
Breaks multiple test under /llvm/test:Transforms (e.g.
llvm/test:Transforms/LoopIdiom/basic.ll.test) under asan.
llvm-svn: 273160
After a store has been eliminated, when making sure that the
instruction iterator points to a valid instruction, dbg intrinsics are
now ignored as a new instruction.
Patch by Henric Karlsson.
Reviewed by Daniel Berlin.
Differential Revision: http://reviews.llvm.org/D21076
llvm-svn: 273141
Removing dead instructions requires remembering which operands have
already been removed. RecursivelyDeleteTriviallyDeadInstructions has
this logic, don't partially reimplement it in LoopIdiomRecognize.
This fixes PR28196.
llvm-svn: 273132
Change the underlying offset and comparisons to use int64_t instead of
uint64_t.
Patch by River Riddle!
Differential Revision: http://reviews.llvm.org/D21499
llvm-svn: 273105
The way we elide max expressions when computing trip counts is incorrect
-- it breaks cases like this:
```
static int wrapping_add(int a, int b) {
return (int)((unsigned)a + (unsigned)b);
}
void test() {
volatile int end_buf = 2147483548; // INT_MIN - 100
int end = end_buf;
unsigned counter = 0;
for (int start = wrapping_add(end, 200); start < end; start++)
counter++;
print(counter);
}
```
Note: the `NoWrap` variable that was being tested has little to do with
the values flowing into the max expression; it is a property of the
induction variable.
test/Transforms/LoopUnroll/nsw-tripcount.ll was added to solely test
functionality I'm reverting in this change, so I've deleted the test
fully.
llvm-svn: 273079
This is a functional change for LLE and LDist. The other clients (LV,
LVerLICM) already had this explicitly enabled.
The temporary boolean parameter to LAA is removed that allowed turning
off speculation of symbolic strides. This makes LAA's caching interface
LAA::getInfo only take the loop as the parameter. This makes the
interface more friendly to the new Pass Manager.
The flag -enable-mem-access-versioning is moved from LV to a LAA which
now allows turning off speculation globally.
llvm-svn: 273064
This is indeed a much cleaner approach (thanks to Daniel Berlin
for pointing out), and also David/Sean for review.
Differential Revision: http://reviews.llvm.org/D21454
llvm-svn: 273032
Many CPUs only have the ability to do a 4-byte cmpxchg (or ll/sc), not 1
or 2-byte. For those, you need to mask and shift the 1 or 2 byte values
appropriately to use the 4-byte instruction.
This change adds support for cmpxchg-based instruction sets (only SPARC,
in LLVM). The support can be extended for LL/SC-based PPC and MIPS in
the future, supplanting the ISel expansions those architectures
currently use.
Tests added for the IR transform and SPARCv9.
Differential Revision: http://reviews.llvm.org/D21029
llvm-svn: 273025
The motivating example for this transform is similar to D20774 where bitcasts interfere
with a single cmp/select sequence, but in this case we have 2 uses of each bitcast to
produce min and max ops:
define void @minmax_bc_store(<4 x float> %a, <4 x float> %b, <4 x float>* %ptr1, <4 x float>* %ptr2) {
%cmp = fcmp olt <4 x float> %a, %b
%bc1 = bitcast <4 x float> %a to <4 x i32>
%bc2 = bitcast <4 x float> %b to <4 x i32>
%sel1 = select <4 x i1> %cmp, <4 x i32> %bc1, <4 x i32> %bc2
%sel2 = select <4 x i1> %cmp, <4 x i32> %bc2, <4 x i32> %bc1
%bc3 = bitcast <4 x float>* %ptr1 to <4 x i32>*
store <4 x i32> %sel1, <4 x i32>* %bc3
%bc4 = bitcast <4 x float>* %ptr2 to <4 x i32>*
store <4 x i32> %sel2, <4 x i32>* %bc4
ret void
}
With this patch, we move the selects up to use the input args which allows getting rid of
all of the bitcasts:
define void @minmax_bc_store(<4 x float> %a, <4 x float> %b, <4 x float>* %ptr1, <4 x float>* %ptr2) {
%cmp = fcmp olt <4 x float> %a, %b
%sel1.v = select <4 x i1> %cmp, <4 x float> %a, <4 x float> %b
%sel2.v = select <4 x i1> %cmp, <4 x float> %b, <4 x float> %a
store <4 x float> %sel1.v, <4 x float>* %ptr1, align 16
store <4 x float> %sel2.v, <4 x float>* %ptr2, align 16
ret void
}
The asm for x86 SSE then improves from:
movaps %xmm0, %xmm2
cmpltps %xmm1, %xmm2
movaps %xmm2, %xmm3
andnps %xmm1, %xmm3
movaps %xmm2, %xmm4
andnps %xmm0, %xmm4
andps %xmm2, %xmm0
orps %xmm3, %xmm0
andps %xmm1, %xmm2
orps %xmm4, %xmm2
movaps %xmm0, (%rdi)
movaps %xmm2, (%rsi)
To:
movaps %xmm0, %xmm2
minps %xmm1, %xmm2
maxps %xmm0, %xmm1
movaps %xmm2, (%rdi)
movaps %xmm1, (%rsi)
The TODO comments show that we're limiting this transform only to vectors and only to bitcasts
because we need to improve other transforms or risk creating worse codegen.
Differential Revision: http://reviews.llvm.org/D21190
llvm-svn: 273011
This is a fix for PR27844.
When replacing uses of unsafe allocas, emit the new location
immediately after each use. Without this, the pointer stays live from
the function entry to the last use, while it's usually cheaper to
recalculate.
llvm-svn: 272969
When moving unsafe allocas to the unsafe stack, dbg.declare intrinsics are
updated to refer to the new location.
This change does the same to dbg.value intrinsics.
llvm-svn: 272968
Redundant invariant loads can be CSE'ed with very little extra effort
over what early-cse already tracks, so it looks reasonable to make
early-cse handle this case.
llvm-svn: 272954
This test checks that the string 'bar' (no quotes) doesn't exist in the
output after running opt. But opt embeds the absolute path to the
filename, and on my machine, the filename contains the string 'jlebar',
causing the test to fail.
This patch changes the test to look for the string '"bar"' instead.
llvm-svn: 272941
Daniel Berlin expressed some real concerns about the port and proposed
and alternative approach. I'll revert this for now while working on a
new patch, which I hope to put up for review shortly. Sorry for the churn.
llvm-svn: 272925
This is currently only performed in the Vectorizer. I will change this
as symbolic stride collection is moved to LAA.
This test will track when the actual functional change occurs.
llvm-svn: 272918
We should update results of the BranchProbabilityInfo after removing block in JumpThreading. Otherwise
we will get dangling pointer inside BranchProbabilityInfo cache.
Differential Revision: http://reviews.llvm.org/D20957
llvm-svn: 272891
Added checks to make sure the Scalarizer::transferMetadata() don't
remove valid debug locations from instructions. This is important as
the verifier pass require that e.g. inlinable callsites have a valid
debug location.
https://llvm.org/bugs/show_bug.cgi?id=27938
Patch by Karl-Johan Karlsson
Reviewers: dblaikie
Differential Revision: http://reviews.llvm.org/D20807
llvm-svn: 272884
(i == 5334 || i == 5335)
to:
((i & -2) == 5334)
This transformation has some incorrect side conditions. Specifically, the
transformation is only applied when the right-hand side constant (5334 in
the example) is a power of two not equal and not equal to the negated mask.
These side conditions were added in r258904 to fix PR26323. The correct side
condition is that: ((Constant & Mask) == Constant)[(5334 & -2) == 5334].
It's a little bit hard to see why these transformations are correct and what
the side conditions ought to be. Here is a CVC3 program to verify them for
64-bit values:
ONE : BITVECTOR(64) = BVZEROEXTEND(0bin1, 63);
x : BITVECTOR(64);
y : BITVECTOR(64);
z : BITVECTOR(64);
mask : BITVECTOR(64) = BVSHL(ONE, z);
QUERY( (y & ~mask = y) =>
((x & ~mask = y) <=> (x = y OR x = (y | mask)))
);
Please note that each pattern must be a dual implication (<--> or iff). One
directional implication can create spurious matches. If the implication is
only one-way, an unsatisfiable condition on the left side can imply a
satisfiable condition on the right side. Dual implication ensures that
satisfiable conditions are transformed to other satisfiable conditions and
unsatisfiable conditions are transformed to other unsatisfiable conditions.
Here is a concrete example of a unsatisfiable condition on the left
implying a satisfiable condition on the right:
mask = (1 << z)
(x & ~mask) == y --> (x == y || x == (y | mask))
Substituting y = 3, z = 0 yields:
(x & -2) == 3 --> (x == 3 || x == 2)
The version of this code before r258904 had no side-conditions and
incorrectly justified itself in comments through one-directional
implication.
Thanks to Chandler for the suggestion!
Author: Thomas Jablin (tjablin)
Reviewers: chandlerc majnemer hfinkel cycheng
http://reviews.llvm.org/D21417
llvm-svn: 272873
We would fail to validate the type of the tan function which would cause
downstream users of isValidProtoForLibFunc to assert.
This fixes PR28143.
llvm-svn: 272802
This uses the "runImpl" approach to share code with the old PM.
Porting to the new PM meant abandoning the anonymous namespace enclosing
most of SLPVectorizer.cpp which is a bit of a bummer (but not a big deal
compared to having to pull the pass class into a header which the new PM
requires since it calls the constructor directly).
llvm-svn: 272766
r272715 broke libcxx because it did not correctly handle cases where the
last iteration of one IV is the second-to-last iteration of another.
Original commit message:
Vectorizing loops with "escaping" IVs has been disabled since r190790, due to
PR17179. This re-enables it, with support for external use of both
"post-increment" (last iteration) and "pre-increment" (second-to-last iteration)
IVs.
llvm-svn: 272742
We do not support splitting cleanuppad or catchswitches. This is
problematic for passes which assume that a loop is in loop simplify
form (the loop would have a dedicated exit block instead of sharing it).
While it isn't great that we don't support this for cleanups, we still
cannot make loop-simplify form an assertable precondition because
indirectbr will also disable these sorts of CFG cleanups.
This fixes PR28132.
llvm-svn: 272739
Nearly all the changes to this pass have been done while maintaining and
updating other parts of LLVM. LLVM has had another pass, SROA, which
has superseded ScalarReplAggregates for quite some time.
Differential Revision: http://reviews.llvm.org/D21316
llvm-svn: 272737
Vectorizing loops with "escaping" IVs has been disabled since r190790, due to
PR17179. This re-enables it, with support for external use of both
"post-increment" (last iteration) and "pre-increment" (second-to-last iteration)
IVs.
Differential Revision: http://reviews.llvm.org/D21048
llvm-svn: 272715
If a local_unnamed_addr attribute is attached to a global, the address
is known to be insignificant within the module. It is distinct from the
existing unnamed_addr attribute in that it only describes a local property
of the module rather than a global property of the symbol.
This attribute is intended to be used by the code generator and LTO to allow
the linker to decide whether the global needs to be in the symbol table. It is
possible to exclude a global from the symbol table if three things are true:
- This attribute is present on every instance of the global (which means that
the normal rule that the global must have a unique address can be broken without
being observable by the program by performing comparisons against the global's
address)
- The global has linkonce_odr linkage (which means that each linkage unit must have
its own copy of the global if it requires one, and the copy in each linkage unit
must be the same)
- It is a constant or a function (which means that the program cannot observe that
the unique-address rule has been broken by writing to the global)
Although this attribute could in principle be computed from the module
contents, LTO clients (i.e. linkers) will normally need to be able to compute
this property as part of symbol resolution, and it would be inefficient to
materialize every module just to compute it.
See:
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20160509/356401.htmlhttp://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20160516/356738.html
for earlier discussion.
Part of the fix for PR27553.
Differential Revision: http://reviews.llvm.org/D20348
llvm-svn: 272709
This change teaches llvm::isGuaranteedToTransferExecutionToSuccessor
that calls to @llvm.assume always terminate. Most other relevant
intrinsics should be covered by the "CS.onlyReadsMemory() ||
CS.onlyAccessesArgMemory()" bit but we were missing @llvm.assumes
because we state that it clobbers memory.
Added an LICM test case, but this change is not specific to LICM.
llvm-svn: 272703
We only used to add the edge from the cloned loop to PHIs that
corresponded to values defined by the loop. We need to do this for all
PHIs obviously since we need a PHI operand for each incoming edge.
This includes things like PHIs with a constant value or with values
defined before the original loop (see the testcases).
After the patch the PHIs are added to the exit block in two passes.
In the first pass we ensure there is a single-operand (LCSSA) PHI for
each value defined by the loop.
In the second pass we loop through each (single-operand) PHI and add the
value for the edge from the cloned loop. If the value is defined in the
loop we'll use the cloned instruction from the cloned loop.
Fixes PR28037
llvm-svn: 272649
The need for all these Lookup* functions is just because of calls to
getAnalysis inside methods (i.e. not at the top level) of the
runOnFunction method. They should be straightforward to clean up when
the old PM is gone.
llvm-svn: 272615
This reverts commit r272603 and adds a fix.
Big thanks to Davide for pointing me at r216244 which gives some insight
into how to fix this VS2013 issue. VS2013 can't synthesize a move
constructor. So the fix here is to add one explicitly to the
JumpThreadingPass class.
llvm-svn: 272607
This follows the approach in r263208 (for GVN) pretty closely:
- move the bulk of the body of the function to the new PM class.
- expose a runImpl method on the new-PM class that takes the IRUnitT and
pointers/references to any analyses and use that to implement the
old-PM class.
- use a private namespace in the header for stuff that used to be file
scope
llvm-svn: 272597
Below are my super rough notes when porting. They can probably serve as
a basic guide for porting other passes to the new PM. As I port more
passes I'll expand and generalize this and make a proper
docs/HowToPortToNewPassManager.rst document. There is also missing
documentation for general concepts and API's in the new PM which will
require some documentation.
Once there is proper documentation in place we can put up a list of
passes that have to be ported and game-ify/crowdsource the rest of the
porting (at least of the middle end; the backend is still unclear).
I will however be taking personal responsibility for ensuring that the
LLD/ELF LTO pipeline is ported in a timely fashion. The remaining passes
to be ported are (do something like
`git grep "<the string in the bullet point below>"` to find the pass):
General Scalar:
[ ] Simplify the CFG
[ ] Jump Threading
[ ] MemCpy Optimization
[ ] Promote Memory to Register
[ ] MergedLoadStoreMotion
[ ] Lazy Value Information Analysis
General IPO:
[ ] Dead Argument Elimination
[ ] Deduce function attributes in RPO
Loop stuff / vectorization stuff:
[ ] Alignment from assumptions
[ ] Canonicalize natural loops
[ ] Delete dead loops
[ ] Loop Access Analysis
[ ] Loop Invariant Code Motion
[ ] Loop Vectorization
[ ] SLP Vectorizer
[ ] Unroll loops
Devirtualization / CFI:
[ ] Cross-DSO CFI
[ ] Whole program devirtualization
[ ] Lower bitset metadata
CGSCC passes:
[ ] Function Integration/Inlining
[ ] Remove unused exception handling info
[ ] Promote 'by reference' arguments to scalars
Please let me know if you are interested in working on any of the passes
in the above list (e.g. reply to the post-commit thread for this patch).
I'll probably be tackling "General Scalar" and "General IPO" first FWIW.
Steps as I port "Deduce function attributes in RPO"
---------------------------------------------------
(note: if you are doing any work based on these notes, please leave a
note in the post-commit review thread for this commit with any
improvements / suggestions / incompleteness you ran into!)
Note: "Deduce function attributes in RPO" is a module pass.
1. Do preparatory refactoring.
Do preparatory factoring. In this case all I had to do was to pull out a static helper (r272503).
(TODO: give more advice here e.g. if pass holds state or something)
2. Rename the old pass class.
llvm/lib/Transforms/IPO/FunctionAttrs.cpp
Rename class ReversePostOrderFunctionAttrs -> ReversePostOrderFunctionAttrsLegacyPass
in preparation for adding a class ReversePostOrderFunctionAttrs as the pass in the new PM.
(edit: actually wait what? The new class name will be
ReversePostOrderFunctionAttrsPass, so it doesn't conflict. So this step is
sort of useless churn).
llvm/include/llvm/InitializePasses.h
llvm/lib/LTO/LTOCodeGenerator.cpp
llvm/lib/Transforms/IPO/IPO.cpp
llvm/lib/Transforms/IPO/FunctionAttrs.cpp
Rename initializeReversePostOrderFunctionAttrsPass -> initializeReversePostOrderFunctionAttrsLegacyPassPass
(note that the "PassPass" thing falls out of `s/ReversePostOrderFunctionAttrs/ReversePostOrderFunctionAttrsLegacyPass/`)
Note that the INITIALIZE_PASS macro is what creates this identifier name, so renaming the class requires this renaming too.
Note that createReversePostOrderFunctionAttrsPass does not need to be
renamed since its name is not generated from the class name.
3. Add the new PM pass class.
In the new PM all passes need to have their
declaration in a header somewhere, so you will often need to add a header.
In this case
llvm/include/llvm/Transforms/IPO/FunctionAttrs.h is already there because
PostOrderFunctionAttrsPass was already ported.
The file-level comment from the .cpp file can be used as the file-level
comment for the new header. You may want to tweak the wording slightly
from "this file implements" to "this file provides" or similar.
Add declaration for the new PM pass in this header:
class ReversePostOrderFunctionAttrsPass
: public PassInfoMixin<ReversePostOrderFunctionAttrsPass> {
public:
PreservedAnalyses run(Module &M, AnalysisManager<Module> &AM);
};
Its name should end with `Pass` for consistency (note that this doesn't
collide with the names of most old PM passes). E.g. call it
`<name of the old PM pass>Pass`.
Also, move the doxygen comment from the old PM pass to the declaration of
this class in the header.
Also, include the declaration for the new PM class
`llvm/Transforms/IPO/FunctionAttrs.h` at the top of the file (in this case,
it was already done when the other pass in this file was ported).
Now define the `run` method for the new class.
The main things here are:
a) Use AM.getResult<...>(M) to get results instead of `getAnalysis<...>()`
b) If the old PM pass would have returned "false" (i.e. `Changed ==
false`), then you should return PreservedAnalyses::all();
c) In the old PM getAnalysisUsage method, observe the calls
`AU.addPreserved<...>();`.
In the case `Changed == true`, for each preserved analysis you should do
call `PA.preserve<...>()` on a PreservedAnalyses object and return it.
E.g.:
PreservedAnalyses PA;
PA.preserve<CallGraphAnalysis>();
return PA;
Note that calls to skipModule/skipFunction are not supported in the new PM
currently, so optnone and optimization bisect support do not work. You can
just drop those calls for now.
4. Add the pass to the new PM pass registry to make it available in opt.
In llvm/lib/Passes/PassBuilder.cpp add a #include for your header.
`#include "llvm/Transforms/IPO/FunctionAttrs.h"`
In this case there is already an include (from when
PostOrderFunctionAttrsPass was ported).
Add your pass to llvm/lib/Passes/PassRegistry.def
In this case, I added
`MODULE_PASS("rpo-functionattrs", ReversePostOrderFunctionAttrsPass())`
The string is from the `INITIALIZE_PASS*` macros used in the old pass
manager.
Then choose a test that uses the pass and use the new PM `-passes=...` to
run it.
E.g. in this case there is a test that does:
; RUN: opt < %s -basicaa -functionattrs -rpo-functionattrs -S | FileCheck %s
I have added the line:
; RUN: opt < %s -aa-pipeline=basic-aa -passes='require<targetlibinfo>,cgscc(function-attrs),rpo-functionattrs' -S | FileCheck %s
The `-aa-pipeline=basic-aa` and
`require<targetlibinfo>,cgscc(function-attrs)` are what is needed to run
functionattrs in the new PM (note that in the new PM "functionattrs"
becomes "function-attrs" for some reason). This is just pulled from
`readattrs.ll` which contains the change from when functionattrs was ported
to the new PM.
Adding rpo-functionattrs causes the pass that was just ported to run.
llvm-svn: 272505
It isn't legal to hoist a load past a call which might not return;
even if it doesn't throw, it could, for example, call exit().
Fixes http://llvm.org/PR27953.
llvm-svn: 272495
Summary:
Make isGuaranteedToExecute use the
isGuaranteedToTransferExecutionToSuccessor helper, and make that helper
a bit more accurate.
There's a potential performance impact here from assuming that arbitrary
calls might not return. This probably has little impact on loads and
stores to a pointer because most things alias analysis can reason about
are dereferenceable anyway. The other impacts, like less aggressive
hoisting of sdiv by a variable and less aggressive hoisting around
volatile memory operations, are unlikely to matter for real code.
This also impacts SCEV, which uses the same helper. It's a minor
improvement there because we can tell that, for example, memcpy always
returns normally. Strictly speaking, it's also introducing
a bug, but it's not any worse than everywhere else we assume readonly
functions terminate.
Fixes http://llvm.org/PR27857.
Reviewers: hfinkel, reames, chandlerc, sanjoy
Subscribers: broune, llvm-commits
Differential Revision: http://reviews.llvm.org/D21167
llvm-svn: 272489
This reapplies commit r272385 with a fix. The build was failing when compiled
with gcc, but not with clang. With the fix, we now get the data layout from the
current TTI implementation, which will hopefully solve the issue.
llvm-svn: 272395
This patch refines the default cost for interleaved load groups having gaps. If
a load group has gaps, the legalized instructions corresponding to the unused
elements will be dead. Thus, we don't need to account for them in the cost
model. Instead, we only need to account for the fraction of legalized loads
that will actually be used.
Differential Revision: http://reviews.llvm.org/D20873
llvm-svn: 272385
Instead of directly using MaxFunctionCount and function entry count to determine callee hotness, use the isHotFunction/isColdFunction methods provided by ProfileSummaryInfo.
Differential revision: http://reviews.llvm.org/D21045
llvm-svn: 272321
Previously, we materialized secondary vector IVs from the primary scalar IV,
by offseting the primary to match the correct start value, and then broadcasting
it - inside the loop body. Instead, we can use a real vector IV, like we do for
the primary.
This enables using vector IVs for secondary integer IVs whose type matches the
type of the primary.
Differential Revision: http://reviews.llvm.org/D20932
llvm-svn: 272283
We can safely rely on a NoWrap add recurrence causing UB down the road
only if we know the loop does not have a exit expressed in a way that is
opaque to ScalarEvolution (e.g. by a function call that conditionally
calls exit(0)).
I believe with this change PR28012 is fixed.
Note: I had to change some llvm-lit tests in LoopReroll, since it looks
like they were depending on this incorrect behavior.
llvm-svn: 272237
Summary:
This fixes PR26682. Also add LCSSA as a preserved pass to LoopSimplify,
that looks correct to me and allows to write a test for the issue.
Reviewers: chandlerc, bogner, sanjoy
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D21112
llvm-svn: 272224
Summary:
This fixes PR27617.
Bug description: The SLPVectorizer asserts on encountering GEPs with different index types, such as i8 and i64.
The patch includes a simple relaxation of the assert to allow constants being of different types, along with a regression test that will provoke the unrelaxed assert.
Reviewers: nadav, mzolotukhin
Subscribers: JesperAntonsson, llvm-commits, mzolotukhin
Differential Revision: http://reviews.llvm.org/D20685
Patch by Jesper Antonsson!
llvm-svn: 272206
with user specified count has been applied.
Summary:
Previously SetLoopAlreadyUnrolled() set the disable pragma only if
there was some loop metadata.
Now it set the pragma in all cases. This helps to prevent multiple
unroll when -unroll-count=N is given.
Reviewers: mzolotukhin
Differential Revision: http://reviews.llvm.org/D20765
From: Evgeny Stupachenko <evstupac@gmail.com>
llvm-svn: 272195
Unlike native shifts, the AVX2 per-element shift instructions VPSRAV/VPSRLV/VPSLLV handle out of range shift values (logical shifts set the result to zero, arithmetic shifts splat the sign bit).
If the shift amount is constant we can sometimes convert these instructions to native shifts:
1 - if all shift amounts are in range then the conversion is trivial.
2 - out of range arithmetic shifts can be clamped to the (bitwidth - 1) (a legal shift amount) before conversion.
3 - logical shifts just return zero if all elements have out of range shift amounts.
In addition, UNDEF shift amounts are handled - either as an UNDEF shift amount in a native shift or as an UNDEF in the logical 'all out of range' zero constant special case for logical shifts.
Differential Revision: http://reviews.llvm.org/D19675
llvm-svn: 271996
This patch adds support for folding undef/zero/constant inputs to MOVMSK instructions.
The SSE/AVX versions can be fully folded, but the MMX version can only handle undef inputs.
Differential Revision: http://reviews.llvm.org/D20998
llvm-svn: 271990
scalarizePHI only looked for phis that have exactly two uses - the "latch"
use, and an extract. Unfortunately, we can not assume all equivalent extracts
are CSE'd, since InstCombine itself may create an extract which is a duplicate
of an existing one. This extends it to handle several distinct extracts from
the same index.
This should fix at least some of the performance regressions from PR27988.
Differential Revision: http://reviews.llvm.org/D20983
llvm-svn: 271961
In some cases, when simplifying with SCEV, we might consider pointer values as
just usual integer values. Thus, we might get a different type from what we
had originally in the map of simplified values, and hence we need to check
types before operating on the values.
This fixes PR28015.
llvm-svn: 271931
Summary:
Fix LSRInstance::HoistInsertPosition() to check the original insert
position block first for a canonical insertion point that is dominated
by all inputs. This leads to SCEV being able to reuse more instructions
since it currently tracks the instructions it creates for reuse by
keeping a table of <Value, insert point> pairs.
Originally reviewed in http://reviews.llvm.org/D18001
Reviewers: atrick
Subscribers: llvm-commits, mzolotukhin, mcrosier
Differential Revision: http://reviews.llvm.org/D18480
llvm-svn: 271929
In r271810 ( http://reviews.llvm.org/rL271810 ), I loosened the check
above this to work for any Constant rather than ConstantInt. AFAICT,
that part makes sense if we can determine that the shrunken/extended
constant remained equal. But it doesn't make sense for this later
transform where we assume that the constant DID change.
This could assert for a ConstantExpr:
https://llvm.org/bugs/show_bug.cgi?id=28011
And it could be wrong for a vector as shown in the added regression test.
llvm-svn: 271908
Summary:
This hasn't been caught before because it requires noalias or similarly
strong alias analysis to actually reproduce.
Fixes http://llvm.org/PR27952 .
Reviewers: hfinkel, sanjoy
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D20944
llvm-svn: 271858
Since FoldOpIntoPhi speculates the binary operation to potentially each
of the predecessors of the PHI node (pulling it out of arbitrary control
dependence in the process), we can FoldOpIntoPhi only if we know the
operation doesn't have UB.
This also brings up an interesting profitability question -- the way it
is written today, commonIRemTransforms will hoist out work from
dynamically dead code into code that will execute at runtime. Perhaps
that isn't the best canonicalization?
Fixes PR27968.
llvm-svn: 271857
Summary:
There are some rough corners, since the new pass manager doesn't have
(as far as I can tell) LoopSimplify and LCSSA, so I've updated the
tests to run them separately in the old pass manager in the lit tests.
We also don't have an equivalent for AU.setPreservesCFG() in the new
pass manager, so I've left a FIXME.
Reviewers: bogner, chandlerc, davide
Subscribers: sanjoy, mcrosier, llvm-commits
Differential Revision: http://reviews.llvm.org/D20783
llvm-svn: 271846
A basic block could contain:
%cp = cleanuppad []
cleanupret from %cp unwind to caller
This basic block is empty and is thus a candidate for removal. However,
there can be other uses of %cp outside of this basic block. This is
only possible in unreachable blocks.
Make our transform more correct by checking that the pad has a single
user before removing the BB.
This fixes PR28005.
llvm-svn: 271816
Add the MMX implementation to the SimplifyDemandedUseBits SSE/AVX MOVMSK support added in D19614
Requires a minor tweak as llvm.x86.mmx.pmovmskb takes a x86_mmx argument - so we have to be explicit about the implied v8i8 vector type.
llvm-svn: 271789
There was concern that creating bitcasts for the simpler potential select pattern:
define <2 x i64> @vecBitcastOp1(<4 x i1> %cmp, <2 x i64> %a) {
%a2 = add <2 x i64> %a, %a
%sext = sext <4 x i1> %cmp to <4 x i32>
%bc = bitcast <4 x i32> %sext to <2 x i64>
%and = and <2 x i64> %a2, %bc
ret <2 x i64> %and
}
might lead to worse code for some targets, so this patch is matching the larger
patterns seen in the test cases.
The motivating example for this patch is this IR produced via SSE intrinsics in C:
define <2 x i64> @gibson(<2 x i64> %a, <2 x i64> %b) {
%t0 = bitcast <2 x i64> %a to <4 x i32>
%t1 = bitcast <2 x i64> %b to <4 x i32>
%cmp = icmp sgt <4 x i32> %t0, %t1
%sext = sext <4 x i1> %cmp to <4 x i32>
%t2 = bitcast <4 x i32> %sext to <2 x i64>
%and = and <2 x i64> %t2, %a
%neg = xor <4 x i32> %sext, <i32 -1, i32 -1, i32 -1, i32 -1>
%neg2 = bitcast <4 x i32> %neg to <2 x i64>
%and2 = and <2 x i64> %neg2, %b
%or = or <2 x i64> %and, %and2
ret <2 x i64> %or
}
For an AVX target, this is currently:
vpcmpgtd %xmm1, %xmm0, %xmm2
vpand %xmm0, %xmm2, %xmm0
vpandn %xmm1, %xmm2, %xmm1
vpor %xmm1, %xmm0, %xmm0
retq
With this patch, it becomes:
vpmaxsd %xmm1, %xmm0, %xmm0
Differential Revision: http://reviews.llvm.org/D20774
llvm-svn: 271676
The original tests were intended to show a missing transform that would
be solved by D20774:
http://reviews.llvm.org/D20774
But it's not clear that the transform for the simpler tests is a win for
all targets. Make the tests show a larger pattern that should be a win
regardless of the cost of bitcast instructions.
llvm-svn: 271603
Inline virtual functions has linkeonceodr linkage (emitted in comdat on
supporting targets). If the vtable for the class is not emitted in the
defining module, function won't be address taken thus its address is not
recorded. At the mercy of the linker, if the per-func prf_data from this
module (in comdat) is picked at link time, we will lose mapping from
function address to its hash val. This leads to missing icall promotion.
The second test case (currently disabled) in compiler_rt (r271528):
instrprof-icall-prom.test demostrates the bug. The first profile-use
subtest is fine due to linker order difference.
With this change, no missing icall targets is found in instrumented clang's
raw profile.
llvm-svn: 271532
Add support for the new pass manager to MemorySSA pass.
Change MemorySSA to be computed eagerly upon construction.
Change MemorySSAWalker to be owned by the MemorySSA object that creates
it.
Reviewers: dberlin, george.burgess.iv
Subscribers: mcrosier, llvm-commits
Differential Revision: http://reviews.llvm.org/D19664
llvm-svn: 271432
Patch by Taewook Oh
Summary: Patch for Bug 27478. Make BasicAliasAnalysis claims NoAlias if two GEPs index different fields of the same structure.
Reviewers: hfinkel, dberlin
Subscribers: dberlin, mcrosier, llvm-commits
Differential Revision: http://reviews.llvm.org/D20665
llvm-svn: 271415
Previously, whenever we needed a vector IV, we would create it on the fly,
by splatting the scalar IV and adding a step vector. Instead, we can create a
real vector IV. This tends to save a couple of instructions per iteration.
This only changes the behavior for the most basic case - integer primary
IVs with a constant step.
Differential Revision: http://reviews.llvm.org/D20315
llvm-svn: 271410
This patch fixes bug https://llvm.org/bugs/show_bug.cgi?id=27897.
When query memory access cost, current SLP always passes in alignment value of 1 (unaligned), so it gets a very high cost of scalar memory access, and wrongly vectorize memory loads in the test case.
It can be fixed by simply giving correct alignment.
llvm-svn: 271333
The assumption, made in insert() that weak functions are always inserted after strong functions,
is only true in the first round of adding functions.
In subsequent rounds this is no longer guaranteed , because we might remove a strong function from the tree (because it's modified) and add it later,
where an equivalent weak function already exists in the tree.
This change removes the assert in insert() and explicitly enforces a weak->strong order.
This also removes the need of two separate loops in runOnModule().
llvm-svn: 271299
Summary:
If we can prove that an op.with.overflow intrinsic does not overflow, we
can get rid of the intrinsic, and replace it with non-wrapping
arithmetic.
This was first checked in at r265913 but reverted in r265950 because it
exposed some issues around how SCEV handled post-inc add recurrences.
Those issues have now been fixed.
Reviewers: atrick, regehr
Subscribers: sanjoy, mcrosier, llvm-commits
Differential Revision: http://reviews.llvm.org/D18685
llvm-svn: 271153
Fixes PR27315.
The post-inc version of an add recurrence needs to "follow the same
rules" as a normal add or subtract expression. Otherwise we miscompile
programs like
```
int main() {
int a = 0;
unsigned a_u = 0;
volatile long last_value;
do {
a_u += 3;
last_value = (long) ((int) a_u);
if (will_add_overflow(a, 3)) {
// Leave, and don't actually do the increment, so no UB.
printf("last_value = %ld\n", last_value);
exit(0);
}
a += 3;
} while (a != 46);
return 0;
}
```
This patch changes SCEV to put no-wrap flags on post-inc add recurrences
only when the poison from a potential overflow will go ahead to cause
undefined behavior.
To avoid regressing performance too much, I've assumed infinite loops
without side effects is undefined behavior to prove poison<->UB
equivalence in more cases. This isn't ideal, but is not new to LLVM as
a whole, and far better than the situation I'm trying to fix.
llvm-svn: 271151
This patch removes the llvm intrinsics VPMOVSX and (V)PMOVZX sign/zero extension intrinsics and auto-upgrades to SEXT/ZEXT calls instead. We already did this for SSE41 PMOVSX sometime ago so much of that implementation can be reused.
Reapplied now that the the companion patch (D20684) removes/auto-upgrade the clang intrinsics has been committed.
Differential Revision: http://reviews.llvm.org/D20686
llvm-svn: 271131
It was triggering an msan bot.
Revert "[IRPGO] Set the function entry count metadata."
This reverts commit r271090.
Revert "[IRPGO] Centralize the function attribute inliner hint logic. NFC."
This reverts commit r271089.
llvm-svn: 271091
Summary:
Unroll factor (Count) calculations moved to a new function.
Early exits on pragma and "-unroll-count" defined factor added.
New type of unrolling "Force" introduced (previously used implicitly).
New unroll preference "AllowRemainder" introduced and set "true" by default.
(should be set to false for architectures that suffers from it).
Reviewers: hfinkel, mzolotukhin, zzheng
Differential Revision: http://reviews.llvm.org/D19553
From: Evgeny Stupachenko <evstupac@gmail.com>
llvm-svn: 271071
When we traced through a phi node looking for floating-point reductions, we
forgot whether we'd ever seen an instruction without fast-math flags (that
would block vectorization). This propagates it through to the end.
llvm-svn: 271015
Currently we consider that each constant has itself as a base value. I.e "base(const) = const".
This introduces couple of problems when we are trying to avoid reporting constants in statepoint live sets:
1. When querying "base( phi(const1, const2) )" we will get "phi(const1, const2)" as a base pointer. Since
it's not a constant we will record it in a stack map. However on practice we don't want this to happen
(constant are never relocated).
2. base( phi(const, gc ptr) ) = phi( const, base(gc ptr) ). This particular case imposes challenge on our
runtime - we don't expect to see constant base pointers other than null. This problems can be avoided
by treating all constant as if they were derived from null pointer base. I.e in a first case we will
not include constant pointer in a stack map at all. In a second case we will get "phi(null, base(gc ptr))"
as a base pointer which is a lot more convenient.
Differential Revision: http://reviews.llvm.org/D20584
llvm-svn: 270993
This patch removes the llvm intrinsics VPMOVSX and (V)PMOVZX sign/zero extension intrinsics and auto-upgrades to SEXT/ZEXT calls instead. We already did this for SSE41 PMOVSX sometime ago so much of that implementation can be reused.
A companion patch (D20684) removes/auto-upgrade the clang intrinsics.
Differential Revision: http://reviews.llvm.org/D20686
llvm-svn: 270973
objc_storeStrong can be formed from a sequence such as
%0 = tail call i8* @objc_retain(i8* %p) nounwind
%tmp = load i8*, i8** @x, align 8
store i8* %0, i8** @x, align 8
tail call void @objc_release(i8* %tmp) nounwind
The code was already looking through bitcasts for most of the values
involved, but had missed one case where the pointer operand for the
store was a bitcast. Ultimately the pointer for the load and store
have to be the same value, after stripping casts.
llvm-svn: 270955
Condition might be simplified to a Constant, but it doesn't have to be
ConstantInt, so we should dyn_cast, instead of cast.
This fixes PR27886.
llvm-svn: 270924
An exception could prevent a store from occurring but MemCpyOpt's
callslot optimization would fire anyway, causing the store to occur.
This fixes PR27849.
llvm-svn: 270892
The memory location that corresponds to a volatile operation is very
special. They are observed by the machine in ways which we cannot
reason about.
Differential Revision: http://reviews.llvm.org/D20555
llvm-svn: 270879
It is unsafe to hoist a load before a function call which may throw, the
throw might prevent a pointer dereference.
Likewise, it is unsafe to sink a store after a call which may throw.
The caller might be able to observe the difference.
This fixes PR27858.
llvm-svn: 270828
Summary:
If an index for a vector or array type is out-of-range GEP constant
folding tries to factor it into preceding dimensions. The code however
does not consider addressing of structure field padding which should not
qualify as out-of-range index.
As demonstrated by the testcase, this can occur if the indexing
performed on a vector type and the preceding index is an array type.
SROA generates GEPs for example involving padding bytes as it slices an
alloca.
My fix disables this folding if the element type is a vector type. I
believe that this is the only way we can end up with padding. (We have
no access to DataLayout so I am not sure if there is actual robust way
of actually checking the presence of padding.)
Reviewers: majnemer
Subscribers: llvm-commits, Gerolf
Differential Revision: http://reviews.llvm.org/D20663
llvm-svn: 270826
It turns out that too many passes are relying on alias analysis results
for control dependencies. Until we fix that by introducing a more accurate
modelling of control dependencies, special case assume in MemorySSA instead.
Also introduce tests to ensure we don't regress the FunctionAttrs or LICM
passes.
Differential Revision: http://reviews.llvm.org/D20658
llvm-svn: 270823
After this change, we do the expected thing for cases like
```
Check0Passed = /* range check IRCE can optimize */
Check1Passed = /* range check IRCE can optimize */
if (!(Check0Passed && Check1Passed))
throw_Exception();
```
llvm-svn: 270804
Getting accurate locations for loops is important, because those locations are
used by the frontend to generate optimization remarks. Currently, optimization
remarks for loops often appear on the wrong line, often the first line of the
loop body instead of the loop itself. This is confusing because that line might
itself be another loop, or might be somewhere else completely if the body was
inlined function call. This happens because of the way we find the loop's
starting location. First, we look for a preheader, and if we find one, and its
terminator has a debug location, then we use that. Otherwise, we look for a
location on an instruction in the loop header.
The fallback heuristic is not bad, but will almost always find the beginning of
the body, and not the loop statement itself. The preheader location search
often fails because there's often not a preheader, and even when there is a
preheader, depending on how it was formed, it sometimes carries the location of
some preceeding code.
I don't see any good theoretical way to fix this problem. On the other hand,
this seems like a straightforward solution: Put the debug location in the
loop's llvm.loop metadata. A companion Clang patch will cause Clang to insert
llvm.loop metadata with appropriate locations when generating debugging
information. With these changes, our loop remarks have much more accurate
locations.
Differential Revision: http://reviews.llvm.org/D19738
llvm-svn: 270771
There was a typo in r267758. It caused invalid accesses when
given something like "void @free(...)", as NumParams == 0, and
we then try to look at the 0th parameter.
Turns out, most of these were untested; add both attribute
and missing-prototype checks for all libc libfuncs.
Differential Revision: http://reviews.llvm.org/D20543
llvm-svn: 270750
This is probably correct for all uses except cross-module IR linking,
where we need to move the comdat from the source module to the
destination module.
Fixes PR27870.
Reviewers: majnemer
Differential Revision: http://reviews.llvm.org/D20631
llvm-svn: 270743
By making pointer extraction from a vector more expensive in the cost model,
we avoid the vectorization of a loop that is very likely to be memory-bound:
https://llvm.org/bugs/show_bug.cgi?id=27826
There are still bugs related to this, so we may need a more general solution
to avoid vectorizing obviously memory-bound loops when we don't have HW gather
support.
Differential Revision: http://reviews.llvm.org/D20601
llvm-svn: 270729
A volatile load has side effects beyond what callers expect readonly to
signify. For example, it is not safe to reorder two function calls
which each perform a volatile load to the same memory location.
llvm-svn: 270671
Similar in spirit to D20497 :
If all elements of a constant vector are known non-zero, then we can say that the
whole vector is known non-zero.
It seems like we could extend this to FP scalar/vector too, but isKnownNonZero()
says it only works for integers and pointers for now.
Differential Revision: http://reviews.llvm.org/D20544
llvm-svn: 270562
This changes IRCE to optimize uses, and not branches. This change is
NFCI since the uses we do inspect are in practice only ever going to be
the condition use in conditional branches; but this flexibility will
later allow us to analyze more complex expressions than just a direct
branch on a range check.
llvm-svn: 270500
When an aggregate contains an opaque type its size cannot be
determined. This triggers an "Invalid GetElementPtrInst indices for type" assert
in function checkGEPType. The fix suppresses the conversion in this case.
http://reviews.llvm.org/D20319
llvm-svn: 270479
Summary:
This patch turns on LoopUnrollAnalyzer by default. To mitigate compile
time regressions, I chose very conservative thresholds for now. Later we
can make them more aggressive, but it might require being smarter in
which loops we're optimizing. E.g. currently the biggest issue is that
with more agressive thresholds we unroll many cold loops, which
increases compile time for no performance benefit (performance of those
loops is improved, but it doesn't matter since they are cold).
Test results for compile time(using 4 samples to reduce noise):
```
MultiSource/Benchmarks/VersaBench/ecbdes/ecbdes 5.19%
SingleSource/Benchmarks/Polybench/medley/reg_detect/reg_detect 4.19%
MultiSource/Benchmarks/FreeBench/fourinarow/fourinarow 3.39%
MultiSource/Applications/JM/lencod/lencod 1.47%
MultiSource/Benchmarks/Fhourstones-3_1/fhourstones3_1 -6.06%
```
I didn't see any performance changes in the testsuite, but it improves
some internal tests.
Reviewers: hfinkel, chandlerc
Subscribers: llvm-commits, mzolotukhin
Differential Revision: http://reviews.llvm.org/D20482
llvm-svn: 270478
We could try harder to handle non-splat vector constants too,
but that seems much rarer to me.
Note that the div test isn't resolved because there's a check
for isIntegerTy() guarding that transform.
Differential Revision: http://reviews.llvm.org/D20497
llvm-svn: 270369
A cleanuppad is not cheap, they turn into many instructions and result
in additional spills and fills. It is not worth keeping a cleanuppad
around if all it does is hold a lifetime.end instruction.
N.B. We first try to merge the cleanuppad with another cleanuppad to
avoid dropping the lifetime and debug info markers.
llvm-svn: 270314
I had used `std::remove_if` under the assumption that it moves the
predicate matching elements to the end, but actaully the elements
remaining towards the end (after the iterator returned by
`std::remove_if`) are indeterminate. Fix the bug (and make the code
more straightforward) by using a temporary SmallVector, and add a test
case demonstrating the issue.
llvm-svn: 270306
Check that the incoming blocks of phi nodes are identical, and block
function merging if they are not.
rdar://problem/26255167
Differential Revision: http://reviews.llvm.org/D20462
llvm-svn: 270250
Before r257832, the threshold used by SimpleInliner was explicitly specified or generated from opt levels and passed to the base class Inliner's constructor. There, it was first overridden by explicitly specified -inline-threshold. The refactoring in r257832 did not preserve this behavior for all opt levels. This change brings back the original behavior.
Differential Revision: http://reviews.llvm.org/D20452
llvm-svn: 270153
Sequences of range checks expressed using guards, like
guard((I - 2) u< L)
guard((I - 1) u< L)
guard((I + 0) u< L)
guard((I + 1) u< L)
guard((I + 2) u< L)
can sometimes be combined into a smaller sequence:
guard((I - 2) u< L AND (I + 2) u< L)
if we can prove that (I - 2) u< L AND (I + 2) u< L implies all of checks
expressed in the previous sequence.
This change teaches GuardWidening to do this kind of merging when
feasible.
llvm-svn: 270151
This patch fixes https://llvm.org/bugs/show_bug.cgi?id=27703.
If there is a sequence of one or more load instructions, each loaded value is used as address of later load instruction, bitcast is necessary to change the value type, don't optimize it.
llvm-svn: 270135
Summary:
Implement guard widening in LLVM. Description from GuardWidening.cpp:
The semantics of the `@llvm.experimental.guard` intrinsic lets LLVM
transform it so that it fails more often that it did before the
transform. This optimization is called "widening" and can be used hoist
and common runtime checks in situations like these:
```
%cmp0 = 7 u< Length
call @llvm.experimental.guard(i1 %cmp0) [ "deopt"(...) ]
call @unknown_side_effects()
%cmp1 = 9 u< Length
call @llvm.experimental.guard(i1 %cmp1) [ "deopt"(...) ]
...
```
to
```
%cmp0 = 9 u< Length
call @llvm.experimental.guard(i1 %cmp0) [ "deopt"(...) ]
call @unknown_side_effects()
...
```
If `%cmp0` is false, `@llvm.experimental.guard` will "deoptimize" back
to a generic implementation of the same function, which will have the
correct semantics from that point onward. It is always _legal_ to
deoptimize (so replacing `%cmp0` with false is "correct"), though it may
not always be profitable to do so.
NB! This pass is a work in progress. It hasn't been tuned to be
"production ready" yet. It is known to have quadriatic running time and
will not scale to large numbers of guards
Reviewers: reames, atrick, bogner, apilipenko, nlewycky
Subscribers: mcrosier, llvm-commits
Differential Revision: http://reviews.llvm.org/D20143
llvm-svn: 269997
Summary: Set default branch weight to 1:1 if one of the branch has profile missing when simplifying CFG.
Reviewers: spatel, davidxl
Subscribers: danielcdh, llvm-commits
Differential Revision: http://reviews.llvm.org/D20307
llvm-svn: 269995
This is assertion is no longer necessary since we never record
constants in the live set anyway. (They are never recorded in
the initial live set, and constant bases are removed near line 2119)
Differential Revision: http://reviews.llvm.org/D20293
llvm-svn: 269764
Fix a bug introduced with rL269426 :
[InstCombine] canonicalize* LE/GE vector integer comparisons to LT/GT (PR26701, PR26819)
We were assuming that a ConstantDataVector / ConstantVector / ConstantAggregateZero operand of
an ICMP was composed of ConstantInt elements, but it might have ConstantExpr or UndefValue
elements. Handle those appropriately.
Also, refactor this function to join the scalar and vector paths and eliminate the switches.
Differential Revision: http://reviews.llvm.org/D20289
llvm-svn: 269728
This patch renames the option enabling the store-to-load forwarding conflict
detection optimization. This change was requested in the review of D20241.
llvm-svn: 269668
The selection of the vectorization factor currently doesn't consider
interleaved accesses. The vectorization factor is based on the maximum safe
dependence distance computed by LAA. However, for loops with interleaved
groups, we should instead base the vectorization factor on the maximum safe
dependence distance divided by the maximum interleave factor of all the
interleaved groups. Interleaved accesses not in a group will be scalarized.
Differential Revision: http://reviews.llvm.org/D20241
llvm-svn: 269659
Vector GEP with mixed (vector and scalar) indices failed on the InstSimplify Pass when all indices are constants.
Differential revision http://reviews.llvm.org/D20149
llvm-svn: 269590
Summary: This change fix the bug in isProfitableToUseMemset() where MaxIntSize shoule be in byte, not bit.
Reviewers: arsenm, joker.eph, mcrosier
Subscribers: mcrosier, llvm-commits
Differential Revision: http://reviews.llvm.org/D20176
llvm-svn: 269433
*We don't currently handle the edge case constants (min/max values), so it's not a complete
canonicalization.
To fully solve the motivating bugs, we need to enhance this to recognize a zero vector
too because that's a ConstantAggregateZero which is a ConstantData, not a ConstantVector
or a ConstantDataVector.
Differential Revision: http://reviews.llvm.org/D17859
llvm-svn: 269426
Summary:
...loop after the last iteration.
This is really hard to do correctly. The core problem is that we need to
model liveness through the induction PHIs from iteration to iteration in
order to get the correct results, and we need to correctly de-duplicate
the common subgraphs of instructions feeding some subset of the
induction PHIs. All of this can be driven either from a side effect at
some iteration or from the loop values used after the loop finishes.
This patch implements this by storing the forward-propagating analysis
of each instruction in a cache to recall whether it was free and whether
it has become live and thus counted toward the total unroll cost. Then,
at each sink for a value in the loop, we recursively walk back through
every value that feeds the sink, including looping back through the
iterations as needed, until we have marked the entire input graph as
live. Because we cache this, we never visit instructions more than twice
-- once when we analyze them and put them into the cache, and once when
we count their cost towards the unrolled loop. Also, because the cache
is only two bits and because we are dealing with relatively small
iteration counts, we can store all of this very densely in memory to
avoid this from becoming an excessively slow analysis.
The code here is still pretty gross. I would appreciate suggestions
about better ways to factor or split this up, I've stared too long at
the algorithmic side to really have a good sense of what the design
should probably look at.
Also, it might seem like we should do all of this bottom-up, but I think
that is a red herring. Specifically, the simplification power is *much*
greater working top-down. We can forward propagate very effectively,
even across strange and interesting recurrances around the backedge.
Because we use data to propagate, this doesn't cause a state space
explosion. Doing this level of constant folding, etc, would be very
expensive to do bottom-up because it wouldn't be until the last moment
that you could collapse everything. The current solution is essentially
a top-down simplification with a bottom-up cost accounting which seems
to get the best of both worlds. It makes the simplification incremental
and powerful while leaving everything dead until we *know* it is needed.
Finally, a core property of this approach is its *monotonicity*. At all
times, the current UnrolledCost is a conservatively low estimate. This
ensures that we will never early-exit from the analysis due to exceeding
a threshold when if we had continued, the cost would have gone back
below the threshold. These kinds of bugs can cause incredibly hard to
track down random changes to behavior.
We could use a techinque similar (but much simpler) within the inliner
as well to avoid considering speculated code in the inline cost.
Reviewers: chandlerc
Subscribers: sanjoy, mzolotukhin, llvm-commits
Differential Revision: http://reviews.llvm.org/D11758
llvm-svn: 269388
Summary:
Currently we consider such instructions as simplified, which is incorrect,
because if their user isn't simplified, we can't actually simplify them too.
This biases our estimates of profitability: for instance the analyzer expects
much more gains from unrolling memcpy loops than there actually are.
Reviewers: hfinkel, chandlerc
Subscribers: mzolotukhin, llvm-commits
Differential Revision: http://reviews.llvm.org/D17365
llvm-svn: 269387
Shifts beyond the bitwidth are undef but SCCP resolved them to zero.
Instead, DTRT and resolve them to undef.
This reimplements the transform which caused PR27712.
llvm-svn: 269269
This new verifier rule lets us unambigously pick a calling convention
when creating a new declaration for
`@llvm.experimental.deoptimize.<ty>`. It is also congruent with our
lowering strategy -- since all calls to `@llvm.experimental.deoptimize`
are lowered to calls to `__llvm_deoptimize`, it is reasonable to enforce
a unique calling convention.
Some of the tests that were breaking this verifier rule have had to be
split up into different .ll files.
The inliner was violating this rule as well, and has been fixed to avoid
producing invalid IR.
llvm-svn: 269261
Summary: In sample profile, some branches may have profile missing due to profile inaccuracy. We want existing branch probability still valid after propagation.
Reviewers: hfinkel, davidxl, spatel
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D19948
llvm-svn: 269137
Sort of the BB-local equivalent to idiom-recognizer: if we have a basic-block
that really implements a memcpy operation, backends can benefit from seeing
this.
llvm-svn: 269125
Before r268509, Clang would disable the loop unroll pass when optimizing
for size. That commit enabled it to be able to support unroll pragmas
in -Os builds. However, this regressed binary size in one of Chromium's
DLLs with ~100 KB.
This restores the original behaviour of no unrolling at -Os, but doing it
in LLVM instead of Clang makes more sense, and also allows the pragmas to
keep working.
Differential revision: http://reviews.llvm.org/D20115
llvm-svn: 269124
This patch extend loopreroll to allow the instruction chain
of loop control only IV has sext.
Differential Revision: http://reviews.llvm.org/D19820
llvm-svn: 269121
Do simplifications common to all shift instructions based on the amount shifted:
1. If the shift amount is known larger than the bitwidth, the result is undefined.
2. If the valid bits of the shift amount are all known to be 0, it's a shift by zero, so the shift operand is the result.
Note that we could generalize the shift-by-zero transform into a shift-by-constant if all of the valid bits in the shift
amount are known, but that would have to be done in InstCombine rather than here because it would mean we need to create
a new shift instruction.
Differential Revision: http://reviews.llvm.org/D19874
llvm-svn: 269114
The LoopPassManager needs to calculate the loops analysis in order to
iterate over the loops at all. Requiring it is redundant and just adds
noise to the RUN lines here.
llvm-svn: 269097
An oddity of the .ll syntax is that the "@var = " in
@var = global i32 42
is optional. Writing just
global i32 42
is equivalent to
@0 = global i32 42
This means that there is a pretty big First set at the top level. The
current implementation maintains it manually. I was trying to refactor
it, but then started wondering why keep it a all. I personally find the
above syntax confusing. It looks like something is missing.
This patch removes the feature and simplifies the parser.
llvm-svn: 269096
This patch extend loopreroll to allow the instruction chain
of loop control only IV has sext.
Differential Revision: http://reviews.llvm.org/D19820
llvm-svn: 269093
Loop rotation clones instruction from the old header into the preheader. If
there were uses of values produced by these instructions that were outside
the loop, we have to insert PHI nodes to merge the two values. If the values
are used by DbgIntrinsics they will be used as a MetadataAsValue of a
ValueAsMetadata of the original values, and iterating all of the uses of the
original value will not update the DbgIntrinsics. The new code checks if the
values are used by DbgIntrinsics and if so, updates them using essentially
the same logic as the original code.
The attached testcase demonstrates the issue. Without the fix, the
DbgIntrinic outside the loop uses values computed inside the loop, even
though these values do not dominate the DbgIntrinsic.
Author: Thomas Jablin (tjablin)
Reviewers: dblaikie aprantl kbarton hfinkel cycheng
http://reviews.llvm.org/D19564
llvm-svn: 269034
When a va_start or va_copy is immediately followed by a va_end (ignoring
debug information or other start/end in between), then it is safe to
remove the pair. As this code shares some commonalities with the lifetime
markers, this has been factored to helper functions.
This InstCombine pattern kicks-in 3 times when running the LLVM test
suite.
llvm-svn: 269033
This reverts commits r268969, r268979 and r268984. They had target specific test
in generic directories without the correct specifiers and made it hard for us to
come up with a good solution by rapidly committing untested changes.
This test needs to be in a target specific directory or have the correct REQUIRED
identifier.
llvm-svn: 269027
Allow vectorization when the step is a loop-invariant variable.
This is the loop example that is getting vectorized after the patch:
int int_inc;
int bar(int init, int *restrict A, int N) {
int x = init;
for (int i=0;i<N;i++){
A[i] = x;
x += int_inc;
}
return x;
}
"x" is an induction variable with *loop-invariant* step.
But it is not a primary induction. Primary induction variable with non-constant step is not handled yet.
Differential Revision: http://reviews.llvm.org/D19258
llvm-svn: 269023
When we encounter unsafe memory dependencies, loop distribution could
help.
Even though, the diagnostics is in LAA, it's only currently emitted in
the vectorizer.
llvm-svn: 268987
IR instrumentation generates a COMDAT symbol __llvm_profile_raw_version to
overwrite the same symbol in profile run-time to distinguish IR profiles from
Clang generated profiles. In MACHO, LinkOnceODR linkage is used due to the
lack of COMDAT support.
But LinkOnceODR linkage might have .weak_def_can_be_hidden assembly directive,
while the weak variable in run-time has a .weak_definition directive. Linker
will not merge these two symbols even they have the same name. The end result
is IR profiles are not properly flagged in MACHO.
This patch changes the linkage for __llvm_profile_raw_version in each module to
LinkOnceAny so that it has same .weak_definition directive as in the run-time.
Differential Revision: http://reviews.llvm.org/D20078
llvm-svn: 268969
When deciding if a vector calculation can be done in a smaller bitwidth, use sign bit information from ValueTracking to add more information and allow more truncations.
llvm-svn: 268921
This moves the code that handles stripping debug info intrinsic from
StripDebugInfo(Module) to StripDebugInfo(Function). The latter is
already walking every instructions so it makes sense to do it at the
same time.
This makes also stripDebugInfo(Function) as an API more useful: it
is really dropping every debug info in the Function.
Finally the existing code is trigerring an assertion when the Module
is not fully materialized.
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 268847
Original Commit Message
Extend load/store type canonicalization to handle unordered operations
Extend the type canonicalization logic to work for unordered atomic loads and stores. Note that while this change itself is fairly simple and low risk, there's a reasonable chance this will expose problems in the backends by suddenly generating IR they wouldn't have seen before. Anything of this nature will be an existing bug in the backend (you could write an atomic float load), but this will definitely change the frequency with which such cases are encountered. If you see problems, feel free to revert this change, but please make sure you collect a test case.
Note that the concern about lowering is now much less likely. PR27490 proved that we already *were* mucking with the types of ordered atomics and volatiles. As a result, this change doesn't introduce as much new behavior as originally thought.
llvm-svn: 268809
Again, fairly simple. Only change is ensuring that we actually copy the property of the load correctly. The aliasing legality constraints were already handled by the FRE patches. There's nothing special about unorder atomics from the perspective of the PRE algorithm itself.
llvm-svn: 268804
You'll note there are essentially no code changes here. Cross block FRE heavily reuses code from the block local FRE. All of the tricky parts were done as part of the previous patch and the refactoring that removed the original code duplication.
llvm-svn: 268775
This patch is the first in a small series teaching GVN to optimize unordered loads aggressively. This change just handles block local FRE because that's the simplest thing which lets me test MDA, and the AvailableValue pieces. Somewhat suprisingly, MDA appears fine and only a couple of small changes are needed in GVN.
Once this is in, I'll tackle non-local FRE and PRE. The former looks like a natural extension of this, the later will require a couple of minor changes.
Differential Revision: http://reviews.llvm.org/D19440
llvm-svn: 268770
Retrying r268550/r268751 which were reverted at r268577/r268765 due a memory sanitizer failure.
I have not been able to reproduce that failure, but I've taken another guess at fixing
the problem in this version of the patch and will watch for another failure.
Original commit message:
Unlike earlier similar fixes, we need to recalculate the branch weights
in this case.
Differential Revision: http://reviews.llvm.org/D19674
llvm-svn: 268767
Retrying r268550 which was reverted at r268577 due a memory sanitizer failure.
I have not been able to reproduce that failure, but I've taken a guess at fixing
the problem in this version of the patch and will watch for another failure.
Original commit message:
Unlike earlier similar fixes, we need to recalculate the branch weights
in this case.
Differential Revision: http://reviews.llvm.org/D19674
llvm-svn: 268751
Summary: We need to clean up CFG before assigning discriminator to minimize the impact of optimization on debug info.
Reviewers: davidxl, dblaikie, dnovillo
Subscribers: dnovillo, danielcdh, llvm-commits
Differential Revision: http://reviews.llvm.org/D19926
llvm-svn: 268675
Summary:
Some PHIs can have expressions that are not AddRecExprs due to the presence
of sext/zext instructions. In order to prevent the Loop Vectorizer from
bailing out when encountering these PHIs, we now coerce the SCEV
expressions to AddRecExprs using SCEV predicates (when possible).
We only do this when the alternative would be to not vectorize.
Reviewers: mzolotukhin, anemet
Subscribers: mssimpso, sanjoy, mzolotukhin, llvm-commits
Differential Revision: http://reviews.llvm.org/D17153
llvm-svn: 268633
MemorySanitizer: use-of-uninitialized-value
0x4910e47 in count /mnt/b/sanitizer-buildbot2/sanitizer-x86_64-linux-bootstrap/build/llvm/include/llvm/Support/MathExtras.h:159:12
0x4910e47 in countLeadingZeros<unsigned long> /mnt/b/sanitizer-buildbot2/sanitizer-x86_64-linux-bootstrap/build/llvm/include/llvm/Support/MathExtras.h:183
0x4910e47 in FitWeights /mnt/b/sanitizer-buildbot2/sanitizer-x86_64-linux-bootstrap/build/llvm/lib/Transforms/Utils/SimplifyCFG.cpp:855
0x4910e47 in SimplifyCondBranchToCondBranch /mnt/b/sanitizer-buildbot2/sanitizer-x86_64-linux-bootstrap/build/llvm/lib/Transforms/Utils/SimplifyCFG.cpp:2895
This reverts commit 609f4dd4bf3bc735c8c047a4d4b0a8e9e4d202e2.
llvm-svn: 268577
This reapplies commit r268521, that was reverted in r268530 due to a test failure in select-implied.ll
Modified the test case to reflect the new change.
llvm-svn: 268557
Unlike earlier similar fixes, we need to recalculate the branch weights
in this case.
Differential Revision: http://reviews.llvm.org/D19674
llvm-svn: 268550
ConstantFold has logic to take icmp (bitcast x to y), null and strip the
bitcast. This makes sense in general, but not if x has floating-point type. In
this case, we'd need a fcmp, not an icmp, and the code will assert. We normally
don't see this situation because we constant fold fp -> int bitcasts, however,
we'll see it for bitcasts of ppc_fp128 -> i128. This is because that bitcast is
Endian-dependent, and as a result, we don't simplify it in ConstantFold (we
could, but no one has yet added the necessary logic). Regardless, ConstantFold
should not depend on that canonicalization for correctness.
llvm-svn: 268534
The unroll pass was disabled by clang in /Os. Those new test cases shows that the pass will behave correctly even if it is not fully disabled. This patch is related in some way to the clang commit (http://reviews.llvm.org/D19827), which re-enables the pass in /Os.
Differential Revision: http://reviews.llvm.org/D19870
llvm-svn: 268524
This patch fixes PR27615.
@llvm.dbg.value instructions no longer count towards the maximum number of
instructions to look back at in the instruction list when searching for a
store instruction. This should make the output consistent between debug and
non-debug build.
Patch by Henric Karlsson <henric.karlsson@ericsson.com>!
Differential Revision: http://reviews.llvm.org/D19912
llvm-svn: 268512
Goal of this change is to guarantee stable ordering of the statepoint arguments and other
newly inserted values such as gc.relocates. Previously we had explicit sorting in a couple
of places. However for unnamed values ordering was partial and overall we didn't have any
strong invariant regarding it. This change switches all data structures to use SetVector's
and MapVector's which provide possibility for deterministic iteration over them.
Explicit sorting is now redundant and was removed.
Differential Revision: http://reviews.llvm.org/D19669
llvm-svn: 268502
We assumed that ConstantVectors would be rather uninteresting from the
perspective of analysis. However, this is not the case due to a quirk
of how LLVM handles vectors of i1. Vectors of i1 are not
ConstantDataVectors like vectors of i8, i16, i32 or i64 because i1's
SizeInBits differs from it's StoreSizeInBytes. This leads to it being
categorized as a ConstantVector instead of a ConstantDataVector.
Instead, treat ConstantVector more uniformly.
This fixes PR27591.
llvm-svn: 268479
We forgot to consider the target of ifuncs when considering if a
function was alive or dead.
N.B. Also update a few auxiliary tools like bugpoint and
verify-uselistorder.
This fixes PR27593.
llvm-svn: 268468
pointing to the same addr space. This can prevent SROA from creating a bitcast
between pointers with different addr spaces.
Differential Revision: http://reviews.llvm.org/D19697
llvm-svn: 268424
SCEV caches whether SCEV expressions are loop invariant, variant or
computable. LICM breaks this cache, almost by definition; so clear the
SCEV disposition cache if LICM changed anything.
llvm-svn: 268408
`Loop::makeLoopInvariant` can hoist instructions out of loops, so loop
dispositions for the loop it operated on may need to be cleared. We can
be smarter here (especially around how `forgetLoopDispositions` is
implemented), but let's be correct first.
Fixes PR27570.
llvm-svn: 268406
Summary
When a non-escaping pointer is compared to a global value, the
comparison can be folded even if the corresponding malloc/allocation
call cannot be elided.
We need to make sure the global value is not null, since comparisons to
null cannot be folded.
In future, we should also handle cases when the the comparison
instruction dominates the pointer escape.
Reviewers: sanjoy
Subscribers s.egerton, llvm-commits
Differential Revision: http://reviews.llvm.org/D19549
llvm-svn: 268390
A few benchmarks with lots of accesses to global variables in the hot
loops regressed a lot since r266399, which added the
SpeculativeExecution pass to the default pipeline. The problem is that
this pass doesn't mark Globals Alias Analysis as preserved. Globals
Alias Analysis is computed in a module pass, whereas
SpeculativeExecution is a function pass, and a lot of passes dependent
on the Globals Alias Analysis to optimize these benchmarks are also
function passes. As such, the Globals Alias Analysis information cannot
be recomputed between SpeculativeExecution and the following function
passes needing that information.
SpeculativeExecution doesn't invalidate Globals Alias Analysis, so mark
it as such to fix those performance regressions.
Differential Revision: http://reviews.llvm.org/D19806
llvm-svn: 268370
We were overly cautious in our analysis of loops which have invokes
which unwind to EH pads. The loop unroll transform is safe because it
only clones blocks in the loop body, it does not try to split critical
edges involving EH pads. Instead, move the necessary safety check to
LoopUnswitch.
N.B. The safety check for loop unswitch is covered by an existing test
which fails without it.
llvm-svn: 268357
There is not point in importing a "weak" or a "linkonce" function
since we won't be able to inline it anyway.
We already had a targeted check for WeakAny, this is using the
same check on GlobalValue as the inline, i.e.
isMayBeOverriddenLinkage()
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 268341
There is not point in importing a "weak" or a "linkonce" function
since we won't be able to inline it anyway.
We already had a targeted check for WeakAny, this is using the
same check on GlobalValue as the inline, i.e.
isMayBeOverriddenLinkage()
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 268315
Make it possible that TryToSimplifyUncondBranchFromEmptyBlock merges empty
basic block including lifetime intrinsics as well as phi nodes and
unconditional branch into its successor or predecessor(s).
If successor of empty block has single predecessor, all contents including
lifetime intrinsics are sinked into the successor. Otherwise, they are
hoisted into its predecessor(s) and then merged into the predecessor(s).
Patch by Josh Yoon <josh.yoon@samsung.com>!
Differential Revision: http://reviews.llvm.org/D19257
llvm-svn: 268254
If a guard call being lowered by LowerGuardIntrinsics has the
`!make.implicit` metadata attached, then reattach the metadata to the
branch in the resulting expanded form of the intrinsic. This allows us
to implement null checks as guards and still get the benefit of implicit
null checks.
llvm-svn: 268148
support multiple induction variables
This patch enable loop reroll for the following case:
for(int i=0; i<N; i += 2) {
S += *a++;
S += *a++;
};
Differential Revision: http://reviews.llvm.org/D16550
llvm-svn: 268147
This moves some logic added to EarlyCSE in rL268120 into
`llvm::isInstructionTriviallyDead`. Adds a test case for DCE to
demonstrate that passes other than EarlyCSE can now pick up on the new
information.
llvm-svn: 268126
Summary:
This change teaches EarlyCSE some basic properties of guard intrinsics:
- Guard intrinsics read all memory, but don't write to any memory
- After a guard has executed, the condition it was guarding on can be
assumed to be true
- Guard intrinsics on a constant `true` are no-ops
Reviewers: reames, hfinkel
Subscribers: mcrosier, llvm-commits
Differential Revision: http://reviews.llvm.org/D19578
llvm-svn: 268120
matchSelectPattern attempts to see through casts which mask min/max
patterns from being more obvious. Under certain circumstances, it would
misidentify a sequence of instructions as a min/max because it assumed
that folding casts would preserve the result. This is not the case for
floating point <-> integer casts.
This fixes PR27575.
llvm-svn: 268086
This was being treated the same as private, which has an immediate
offset. For unknown, it probably means it's for a computation not
actually being used for accessing memory, so it should not have a
nontrivial addressing mode.
llvm-svn: 268002
We need to keep loop hints from the original loop on the new vector loop.
Failure to do this meant that, for example:
void foo(int *b) {
#pragma clang loop unroll(disable)
for (int i = 0; i < 16; ++i)
b[i] = 1;
}
this loop would be unrolled. Why? Because we'd vectorize it, thus dropping the
hints that unrolling should be disabled, and then we'd unroll it.
llvm-svn: 267970
I closely followed the precedents set by the vectorizer:
* With -Rpass-missed, the loop is reported with further details pointing
to -Rpass--analysis.
* -Rpass-analysis reports the details why distribution has failed.
* Regardless of -Rpass*, when distribution fails for a loop where
distribution was forced with the pragma, a warning is produced according
to -Wpass-failed. In this case the analysis info is also printed even
without -Rpass-analysis.
llvm-svn: 267952
When inlining a call site with llvm.mem.parallel_loop_access metadata, this
metadata needs to be propagated to all cloned memory-accessing instructions.
Otherwise, inlining parts of the loop body will invalidate the annotation.
With this functionality, we now vectorize the following as expected:
void Body(int *res, int *c, int *d, int *p, int i) {
res[i] = (p[i] == 0) ? res[i] : res[i] + d[i];
}
void Test(int *res, int *c, int *d, int *p, int n) {
int i;
#pragma clang loop vectorize(assume_safety)
for (i = 0; i < 1600; i++) {
Body(res, c, d, p, i);
}
}
llvm-svn: 267949
The MOVMSK instructions copies a vector elements' sign bits to the low bits of a scalar register and zeros the high bits.
This patch adds MOVMSK support to SimplifyDemandedUseBits so that its aware that the upper bits are known to be zero. It also removes the call to MOVMSK if none of the lower bits are actually required and just returns zero.
Differential Revision: http://reviews.llvm.org/D19614
llvm-svn: 267873
This patch implements the transformation that promotes indirect calls to
conditional direct calls when the indirect-call value profile meta-data is
available.
Differential Revision: http://reviews.llvm.org/D17864
llvm-svn: 267815
The sink cast machinery is supposed to sink casts as close to their user
as possible. However, an EH pad is the first instruction in it's basic
block. Don't sink if the user is an EH pad.
This fixes PR27536.
llvm-svn: 267767
"inferattrs" will deduce the attribute, but it will be too late for
many optimizations. Set it ourselves when creating the call.
Differential Revision: http://reviews.llvm.org/D17598
llvm-svn: 267762
We previously disallowed interleaved load groups that may cause us to
speculatively access memory out-of-bounds (r261331). We did this by ensuring
each load group had an access corresponding to the first and last member.
Instead of bailing out for these interleaved groups, this patch enables us to
peel off the last vector iteration, ensuring that we execute at least one
iteration of the scalar remainder loop. This solution was proposed in the
review of the previous patch.
Differential Revision: http://reviews.llvm.org/D19487
llvm-svn: 267751
This change adds a new hook for estimating the cost of vector extracts followed
by zero- and sign-extensions. The motivating example for this change is the
SMOV and UMOV instructions on AArch64. These instructions move data from vector
to general purpose registers while performing the corresponding extension
(sign-extend for SMOV and zero-extend for UMOV) at the same time. For these
operations, TargetTransformInfo can assume the extensions are free and only
report the cost of the vector extract. The SLP vectorizer has been updated to
make use of the new hook.
Differential Revision: http://reviews.llvm.org/D18523
llvm-svn: 267725
Summary:
D19403 adds a new pragma for loop distribution. This change adds
support for the corresponding metadata that the pragma is translated to
by the FE.
As part of this I had to rethink the flag -enable-loop-distribute. My
goal was to be backward compatible with the existing behavior:
A1. pass is off by default from the optimization pipeline
unless -enable-loop-distribute is specified
A2. pass is on when invoked directly from opt (e.g. for unit-testing)
The new pragma/metadata overrides these defaults so the new behavior is:
B1. A1 + enable distribution for individual loop with the pragma/metadata
B2. A2 + disable distribution for individual loop with the pragma/metadata
The default value whether the pass is on or off comes from the initiator
of the pass. From the PassManagerBuilder the default is off, from opt
it's on.
I moved -enable-loop-distribute under the pass. If the flag is
specified it overrides the default from above.
Then the pragma/metadata can further modifies this per loop.
As a side-effect, we can now also use -enable-loop-distribute=0 from opt
to emulate the default from the optimization pipeline. So to be precise
this is the new behavior:
C1. pass is off by default from the optimization pipeline
unless -enable-loop-distribute or the pragma/metadata enables it
C2. pass is on when invoked directly from opt
unless -enable-loop-distribute=0 or the pragma/metadata disables it
Reviewers: hfinkel
Subscribers: joker.eph, mzolotukhin, llvm-commits
Differential Revision: http://reviews.llvm.org/D19431
llvm-svn: 267672
Summary:
It is incorrect to compare TripCount (which is BECount + 1)
with extraiters (or Count) to check if we should enter unrolled
loop or not, because TripCount can potentially overflow
(when BECount is max unsigned integer).
While comparing BECount with (Count - 1) is overflow safe and
therefore correct.
Reviewer: hfinkel
Differential Revision: http://reviews.llvm.org/D19256
From: Evgeny Stupachenko <evstupac@gmail.com>
llvm-svn: 267662
When encountering a non-local pointer, LVI would eagerly scan the block for dereferences of the given object to prove the pointer to be non null. That's all well and good, but *then* we'd go recurse through our input blocks. As a result, we could end up scanning each and every block we traverse, even if the final definition was obviously non null or we found a constant value somewhere up the chain. The previous code papered over this by using the isKnownNonNull routine from value tracking. This made the duplication less painful in the common case.
Instead, we know do the block scan only *after* we've gotten the recursive results back. This lets us stop scanning individual blocks as soon as we've determined it to be non-null in any predecessor block and use our usual merge rules to propagate that information cheaply through successor blocks. For a pointer which can be found non-null, this does strictly less work and sometimes substaintially so.
Note that the case where we *can't* prove something non-null is still the really expensive case. We end up scanning each and every block looking for a dereference and never end up finding one.
llvm-svn: 267642
As pointed out by John Regehr over in http://reviews.llvm.org/D19485, LVI was being incredibly stupid about applying its transfer rules. Rather than gathering local facts from the expression itself, it was simply giving up entirely if one of the inputs was overdefined. This greatly impacts the precision of the overall analysis and makes it far more fragile as well.
This patch builds on 267609 which did the same thing for unary casts.
llvm-svn: 267620
Essentially, I was using the wrong size function. For types which were sized, but not primitive, I wasn't getting a useful size for the operand and failed an assert. I fixed this, and also added a guard that the input is a sized type. Test case is for the original mistake. I'm not sure how to actually exercise the sized type check.
llvm-svn: 267618
We need the default ratio to be sufficiently large that it triggers transforms
based on block frequency info (BFI) and plays well with the recently introduced
BranchProbability used by CGP.
Differential Revision: http://reviews.llvm.org/D19435
llvm-svn: 267615
As pointed out by John Regehr over in http://reviews.llvm.org/D19485, LVI was being incredibly stupid about applying its transfer rules. Rather than gathering local facts from the expression itself, it was simply giving up entirely if one of the inputs was overdefined. This greatly impacts the precision of the overall analysis and makes it far more fragile as well.
This patch implements only the unary operation case. Once this is in, I'll implement the same for the binary operations.
Differential Revision: http://reviews.llvm.org/D19492
llvm-svn: 267609
The destination buffer that sprintf uses is restrict qualified, we do
not need to worry about derived pointers referenced via format
specifiers.
This reverts commit r267580.
llvm-svn: 267605
sprintf doesn't read or copy the terminating null byte from it's string
operands. sprintf will append it's own after processing all of the
format specifiers.
This fixes PR27526.
llvm-svn: 267580
Summary:
Instead of using maximum IR weight as the basic block weight, this patch uses the voting algorithm to find the most likely weight for the basic block. This can effectively avoid the cases when some IRs are annotated incorrectly due to code motion of the profiled binary.
This patch also updates propagate.ll unittest to include discriminator in the input file so that it is testing something meaningful.
Reviewers: davidxl, dnovillo
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D19301
llvm-svn: 267519
When SimplifyCFG merges identical instructions from both sides of a diamond, it
can preserve !llvm.mem.parallel_loop_access (as it does with most of the other
metadata). There's no real data or control dependency change in this case.
llvm-svn: 267515
I really thought we were doing this already, but we were not. Given this input:
void Test(int *res, int *c, int *d, int *p) {
for (int i = 0; i < 16; i++)
res[i] = (p[i] == 0) ? res[i] : res[i] + d[i];
}
we did not vectorize the loop. Even with "assume_safety" the check that we
don't if-convert conditionally-executed loads (to protect against
data-dependent deferenceability) was not elided.
One subtlety: As implemented, it will still prefer to use a masked-load
instrinsic (given target support) over the speculated load. The choice here
seems architecture specific; the best option depends on how expensive the
masked load is compared to a regular load. Ideally, using the masked load still
reduces unnecessary memory traffic, and so should be preferred. If we'd rather
do it the other way, flipping the order of the checks is easy.
The LangRef is updated to make explicit that llvm.mem.parallel_loop_access also
implies that if conversion is okay.
Differential Revision: http://reviews.llvm.org/D19512
llvm-svn: 267514
This patch is what was the "instcombine" portion of D14185, with an additional
test added (see julia_pseudovec in test/Transforms/InstCombine/insert-val-extract-elem.ll).
The patch causes instcombine to replace sequences of extractelement-insertvalue-store
that act essentially like a bitcast followed by a store.
Differential review: http://reviews.llvm.org/D14260
llvm-svn: 267482
The current logic assumes that any constant global will never be SRA'd. I presume this is because normally constant globals can be pushed into their uses and deleted. However, that sometimes can't happen (which is where you really want SRA, so the elements that can be eliminated, are!).
There seems to be no reason why we can't SRA constants too, so let's do it.
llvm-svn: 267393
As discussed on D19318, if we only demand the first element of a DIVSS/DIVSD intrinsic, then reduce to a FDIV call. This matches the existing FADD/FSUB/FMUL patterns.
llvm-svn: 267359
Split from D17490. This patch improves support for determining the demanded vector elements through SSE scalar intrinsics:
1 - demanded vector element support for unary and some extra binary scalar intrinsics (RCP/RSQRT/SQRT/FRCZ and ADD/CMP/DIV/ROUND).
2 - addss/addsd get simplified to a fadd call if we aren't interested in the pass through elements
3 - if we don't need the lowest element of a scalar operation then just use the first argument (the pass through elements) directly
We can add support for propagating demanded elements through any equivalent packed SSE intrinsics in a future patch (these wouldn't use the pass through patterns).
Differential Revision: http://reviews.llvm.org/D19318
llvm-svn: 267357
This patch improves support for determining the demanded vector elements through SSE scalar intrinsics:
1 - recognise that we only need the lowest element of the second input for binary scalar operations (and all the elements of the first input)
2 - recognise that the roundss/roundsd intrinsics use the lowest element of the second input and the remaining elements from the first input
Differential Revision: http://reviews.llvm.org/D17490
llvm-svn: 267356
Eliminate DITypeIdentifierMap and make DITypeRef a thin wrapper around
DIType*. It is no longer legal to refer to a DICompositeType by its
'identifier:', and DIBuilder no longer retains all types with an
'identifier:' automatically.
Aside from the bitcode upgrade, this is mainly removing logic to resolve
an MDString-based reference to an actualy DIType. The commits leading
up to this have made the implicit type map in DICompileUnit's
'retainedTypes:' field superfluous.
This does not remove DITypeRef, DIScopeRef, DINodeRef, and
DITypeRefArray, or stop using them in DI-related metadata. Although as
of this commit they aren't serving a useful purpose, there are patchces
under review to reuse them for CodeView support.
The tests in LLVM were updated with deref-typerefs.sh, which is attached
to the thread "[RFC] Lazy-loading of debug info metadata":
http://lists.llvm.org/pipermail/llvm-dev/2016-April/098318.html
llvm-svn: 267296
This intrinsic takes two arguments, ``%ptr`` and ``%offset``. It loads
a 32-bit value from the address ``%ptr + %offset``, adds ``%ptr`` to that
value and returns it. The constant folder specifically recognizes the form of
this intrinsic and the constant initializers it may load from; if a loaded
constant initializer is known to have the form ``i32 trunc(x - %ptr)``,
the intrinsic call is folded to ``x``.
LLVM provides that the calculation of such a constant initializer will
not overflow at link time under the medium code model if ``x`` is an
``unnamed_addr`` function. However, it does not provide this guarantee for
a constant initializer folded into a function body. This intrinsic can be
used to avoid the possibility of overflows when loading from such a constant.
Differential Revision: http://reviews.llvm.org/D18367
llvm-svn: 267223
Summary:
We can fold compares to false when two distinct allocations within a
function are compared for equality.
Patch by Anna Thomas!
Reviewers: majnemer, reames, sanjoy
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D19390
llvm-svn: 267214
Extend the type canonicalization logic to work for unordered atomic loads and stores. Note that while this change itself is fairly simple and low risk, there's a reasonable chance this will expose problems in the backends by suddenly generating IR they wouldn't have seen before. Anything of this nature will be an existing bug in the backend (you could write an atomic float load), but this will definitely change the frequency with which such cases are encountered. If you see problems, feel free to revert this change, but please make sure you collect a test case.
llvm-svn: 267210
Summary: This change will shorten memset if the beginning of memset is overwritten by later stores.
Reviewers: hfinkel, eeckstein, dberlin, mcrosier
Subscribers: mgrang, mcrosier, llvm-commits
Differential Revision: http://reviews.llvm.org/D18906
llvm-svn: 267197
In the next change, I am generalizing the function
findStringMetadataForLoop and I want to make sure I don't break this.
Looks like there was no coverage for this so far.
llvm-svn: 267182
Summary: eq imply [u|s]ge and [u|s]le are true.
Remove redundant logic by implementing isImpliedFalseByMatchingCmp(Pred1, Pred2)
as isImpliedTrueByMatchingCmp(Pred1, getInversePredicate(Pred2)).
llvm-svn: 267177
Summary: [u|s]gt and [u|s]lt imply [u|s]ge and [u|s]le are true, respectively.
I've simplified the existing tests and added additional tests to cover the new
cases mentioned above. I've also added tests for all the cases where the
first compare doesn't imply anything about the second compare.
llvm-svn: 267171
A followup commit will replace these tests with simplified and more inclusive
tests. The diff is unreadable if this were to be done in a single commit.
llvm-svn: 267170
We take the intersection of overflow flags while CSE'ing.
This permits us to consider two instructions with different overflow
behavior to be replaceable.
llvm-svn: 267153
Summary:
When optimizing PHIs which have inputs floating point binary
operators, we preserve all IR flags except the fast math
flags.
This change removes the logic which tracked some of the IR flags
(no wrap, exact) and replaces it by doing an and on the IR flags of
all inputs to the PHI - which will also handle the fast math
flags.
Reviewers: majnemer
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D19370
llvm-svn: 267139
EarlyCSE had inconsistent behavior with regards to flag'd instructions:
- In some cases, it would pessimize if the available instruction had
different flags by not performing CSE.
- In other cases, it would miscompile if it replaced an instruction
which had no flags with an instruction which has flags.
Fix this by being more consistent with our flag handling by utilizing
andIRFlags.
llvm-svn: 267111
Summary:
If we know that the pointer allocated within a function does not escape,
we can fold away comparisons that are done with global pointers
Patch by Anna Thomas!
Reviewers: reames, majnemer, sanjoy
Subscribers: mgrang, mcrosier, majnemer, llvm-commits
Differential Revision: http://reviews.llvm.org/D19276
llvm-svn: 267035
This builds on 266999 which made FindAvailableValue do the right thing. Tests included show the newly enabled transforms and those which disabled either due to conservatism or correctness requirements.
llvm-svn: 267006
This change adds a couple of test cases to make sure FindAvailableLoadedValue does the right thing. At the moment, the code added is dead, but separating it makes follow on changes far more obvious.
llvm-svn: 266999
Summary:
`llvm.guard(false)` always bails out of the current compilation unit, so
we can prune any control flow following it.
Reviewers: hfinkel, pcc, reames
Subscribers: majnemer, reames, mcrosier, llvm-commits
Differential Revision: http://reviews.llvm.org/D19245
llvm-svn: 266955
Summary:
The function importer already decided what symbols need to be pulled
in. Also these magically added ones will not be in the export list
for the source module, which can confuse the internalizer for
instance.
Reviewers: tejohnson, rafael
Subscribers: joker.eph, llvm-commits
Differential Revision: http://reviews.llvm.org/D19096
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 266948
No matter what value you OR in to A, the result of (or A, B) is going to be UGE A. When A and B are positive, it's SGE too. If A is negative, OR'ing a value into it can't make it positive, but can increase its value closer to -1, therefore (or A, B) is SGE A. Working through all possible combinations produces this truth table:
```
A is
+, -, +/-
F F F + B is
T F ? -
? F ? +/-
```
The related optimizations are flipping the 'slt' for 'sge' which always NOTs the result (if the result is known), and swapping the LHS and RHS while swapping the comparison predicate.
There are more idioms left to implement (aren't there always!) but I've stopped here because any more would risk becoming unreasonable for reviewers.
llvm-svn: 266939
Summary:
This patch prevents importing from (and therefore exporting from) any
module with a "llvm.used" local value. Local values need to be promoted
and renamed when importing, and their presense on the llvm.used variable
indicates that there are opaque uses that won't see the rename. One such
example is a use in inline assembly.
See also the discussion at:
http://lists.llvm.org/pipermail/llvm-dev/2016-April/098047.html
As part of this, move collectUsedGlobalVariables out of Transforms/Utils
and into IR/Module so that it can be used more widely. There are several
other places in LLVM that used copies of this code that can be cleaned
up as a follow on NFC patch.
Reviewers: joker.eph
Subscribers: pcc, llvm-commits, joker.eph
Differential Revision: http://reviews.llvm.org/D18986
llvm-svn: 266877
Both AArch64 and ARM support llvm.<arch>.thread.pointer intrinsics that
just return the thread pointer. I have a pending patch that does the same
for SystemZ (D19054), and there are many more targets that could benefit
from one.
This patch merges the ARM and AArch64 intrinsics into a single target
independent one that will also be used by subsequent targets.
Differential Revision: http://reviews.llvm.org/D19098
llvm-svn: 266818
This patch improves SimplifyCFG to catch cases like:
if (a < b) {
if (a > b) <- known to be false
unreachable;
}
Phabricator Revision: http://reviews.llvm.org/D18905
llvm-svn: 266767
The fast register-allocator cannot cope with inter-block dependencies without
spilling. This is fine for ldrex/strex loops coming from atomicrmw instructions
where any value produced within a block is dead by the end, but not for
cmpxchg. So we lower a cmpxchg at -O0 via a pseudo-inst that gets expanded
after regalloc.
Fortunately this is at -O0 so we don't have to care about performance. This
simplifies the various axes of expansion considerably: we assume a strong
seq_cst operation and ensure ordering via the always-present DMB instructions
rather than v8 acquire/release instructions.
Should fix the 32-bit part of PR25526.
llvm-svn: 266679
This reverts commit r266477.
This commit introduces cyclic dependency. This commit has "Analysis" depend on "ProfileData",
while "ProfileData" depends on "Object", which depends on "BitCode", which
depends on "Analysis".
llvm-svn: 266619
I accidentally replaced `mayBeOverridden` with `!isInterposable`.
Remove the negation and add a test case that would've caught this.
Many thanks to Håkan Hjort for spotting this!
llvm-svn: 266551
To be able to work accurately on the reference graph when taking
decision about internalizing, promoting, renaming, etc. We need
to have the alias information explicit.
Differential Revision: http://reviews.llvm.org/D18836
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 266517
Allow explicit section for indirectly called functions in cfi-icall.
Jumptables for functions in the same type class must be contiguous, so they
always go to the default text section.
Fixes PR25079.
llvm-svn: 266486
Adds an interface to get ProfileSummary for a module and makes InlineCost use ProfileSummary to get max function count.
Differential Revision: http://reviews.llvm.org/D18622
llvm-svn: 266477
Divisions by a constant can be converted into multiplies which are usually
cheaper, but this isn't possible if the constant gets separated (particularly
in loops). Fix this by telling ConstantHoisting that the immediate in a DIV is
cheap.
I considered making the check generic, but neither AArch64 (strangely) nor x86
showed any benefit on the tests I had.
llvm-svn: 266464
InstCombine wants to optimize compares of calls to fabs with zero.
However, we didn't have the necessary legality checking to verify that
the function call had the same behavior as fabs.
llvm-svn: 266452
Currently each Function points to a DISubprogram and DISubprogram has a
scope field. For member functions the scope is a DICompositeType. DIScopes
point to the DICompileUnit to facilitate type uniquing.
Distinct DISubprograms (with isDefinition: true) are not part of the type
hierarchy and cannot be uniqued. This change removes the subprograms
list from DICompileUnit and instead adds a pointer to the owning compile
unit to distinct DISubprograms. This would make it easy for ThinLTO to
strip unneeded DISubprograms and their transitively referenced debug info.
Motivation
----------
Materializing DISubprograms is currently the most expensive operation when
doing a ThinLTO build of clang.
We want the DISubprogram to be stored in a separate Bitcode block (or the
same block as the function body) so we can avoid having to expensively
deserialize all DISubprograms together with the global metadata. If a
function has been inlined into another subprogram we need to store a
reference the block containing the inlined subprogram.
Attached to https://llvm.org/bugs/show_bug.cgi?id=27284 is a python script
that updates LLVM IR testcases to the new format.
http://reviews.llvm.org/D19034
<rdar://problem/25256815>
llvm-svn: 266446
This is almost identical to:
http://reviews.llvm.org/rL264527
This doesn't solve PR27344; it just allows the profile weights to survive.
To solve the bug, we need to use the profile weights in the backend.
llvm-svn: 266442
Summary:
This lets us add this pass to the IR pass manager unconditionally; it
will simply not do anything on targets without branch divergence.
Reviewers: tra
Subscribers: llvm-commits, jingyue, rnk, chandlerc
Differential Revision: http://reviews.llvm.org/D18625
llvm-svn: 266398
If the size of an AST entry changes, we also need to make sure we perform
necessary alias set merges, as the new size may overlap pointers in other sets.
We happen to run into this with memset, because memset allows an entry for a
i8* pointer to have a decidedly non-i8 size.
This fixes PR27262.
Differential Revision: http://reviews.llvm.org/D18939
llvm-svn: 266381
Some SIMD implementations are not IEEE-754 compliant, for example ARM's NEON.
This patch teaches the loop vectorizer to only allow transformations of loops
that either contain no floating-point operations or have enough allowance
flags supporting lack of precision (ex. -ffast-math, Darwin).
For that, the target description now has a method which tells us if the
vectorizer is allowed to handle FP math without falling into unsafe
representations, plus a check on every FP instruction in the candidate loop
to check for the safety flags.
This commit makes LLVM behave like GCC with respect to ARM NEON support, but
it stops short of fixing the underlying problem: sub-normals. Neither GCC
nor LLVM have a flag for allowing sub-normal operations. Before this patch,
GCC only allows it using unsafe-math flags and LLVM allows it by default with
no way to turn it off (short of not using NEON at all).
As a first step, we push this change to make it safe and in sync with GCC.
The second step is to discuss a new sub-normal's flag on both communitues
and come up with a common solution. The third step is to improve the FastMath
flags in LLVM to encode sub-normals and use those flags to restrict NEON FP.
Fixes PR16275.
llvm-svn: 266363
https://llvm.org/bugs/show_bug.cgi?id=27105
We can check if all bits outside of a constant mask are set with a
single constant.
As noted in the bug report, although this form should be considered the
canonical IR, backends may want to transform this into an 'andn' / 'andc'
comparison against zero because that could be a single machine instruction.
Differential Revision: http://reviews.llvm.org/D18842
llvm-svn: 266362
At some point, ARM stopped getting any benefit from ConstantHoisting because
the pass called a different variant of getIntImmCost. Reimplementing the
correct variant revealed some problems, however:
+ ConstantHoisting was modifying switch statements. This is simply invalid,
the cases must remain integer constants no matter the notional cost.
+ ConstantHoisting was mangling alloca instructions in the entry block. These
should be handled by FrameLowering, so constants actually have a cost of 0.
Worse, the resulting bitcasts meant they became dynamic allocas.
rdar://25707382
llvm-svn: 266260
LLVM optimization passes may reduce a profiled target expression
to a constant. Removing runtime calls at such instrumentation points
would help speedup the runtime of the instrumented program.
llvm-svn: 266229
Summary:
To be able to work accurately on the reference graph when taking decision
about internalizing, promoting, renaming, etc. We need to have the alias
information explicit.
Reviewers: tejohnson
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D18836
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 266214
This patch fixes calculating of builtin_object_size if it depends on a
condition. Before this patch compiler did not know how to calculate the
object size when it finds a condition that cannot be eliminated.
This patch enables calculating of builtin_object_size even in case when
condition cannot be eliminated by choosing minimum or maximum value as a
result from condition. Choosing minimum or maximum value from condition
is based on the second argument of __builtin_object_size function.
Patch by Strahinja Petrovic.
Differential Revision: http://reviews.llvm.org/D18438
llvm-svn: 266193
Remove an ad-hoc transform in InstCombine and replace it with more
general machinery (ValueTracking, InstructionSimplify and VectorUtils).
This fixes PR27332.
llvm-svn: 266175
This bug was introduced with:
http://reviews.llvm.org/rL262269
AVX masked loads are specified to set vector lanes to zero when the high bit of the mask
element for that lane is zero:
"If the mask is 0, the corresponding data element is set to zero in the load form of these
instructions, and unmodified in the store form." --Intel manual
Differential Revision: http://reviews.llvm.org/D19017
llvm-svn: 266148
Summary:
For correct handling of alias to nameless
function, we need to be able to refer them through a GUID in the summary.
Here we name them using a hash of the non-private global names in the module.
Reviewers: tejohnson
Subscribers: joker.eph, llvm-commits
Differential Revision: http://reviews.llvm.org/D18883
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 266132
Summary:
Let keep llvm-as "dumb": it converts textual IR to bitcode. This
commit removes the dependency from llvm-as to libLLVMAnalysis.
We'll add back summary in llvm-as if we get to a textual
representation for it at some point. In the meantime, opt seems
like a better place for that.
Reviewers: tejohnson
Subscribers: joker.eph, llvm-commits
Differential Revision: http://reviews.llvm.org/D19032
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 266131
(Recommit of r266002, with r266011, r266016, and not accidentally
including an extra unused/uninitialized element in LibcallRoutineNames)
AtomicExpandPass can now lower atomic load, atomic store, atomicrmw, and
cmpxchg instructions to __atomic_* library calls, when the target
doesn't support atomics of a given size.
This is the first step towards moving all atomic lowering from clang
into llvm. When all is done, the behavior of __sync_* builtins,
__atomic_* builtins, and C11 atomics will be unified.
Previously LLVM would pass everything through to the ISelLowering
code. There, unsupported atomic instructions would turn into __sync_*
library calls. Because of that behavior, Clang currently avoids emitting
llvm IR atomic instructions when this would happen, and emits __atomic_*
library functions itself, in the frontend.
This change makes LLVM able to emit __atomic_* libcalls, and thus will
eventually allow clang to depend on LLVM to do the right thing.
It is advantageous to do the new lowering to atomic libcalls in
AtomicExpandPass, before ISel time, because it's important that all
atomic operations for a given size either lower to __atomic_*
libcalls (which may use locks), or native instructions which won't. No
mixing and matching.
At the moment, this code is enabled only for SPARC, as a
demonstration. The next commit will expand support to all of the other
targets.
Differential Revision: http://reviews.llvm.org/D18200
llvm-svn: 266115
This is a resubmittion of 263158 change.
This patch fixes the problem which occurs when loop-vectorize tries to use @llvm.masked.load/store intrinsic for a non-default addrspace pointer. It fails with "Calling a function with a bad signature!" assertion in CallInst constructor because it tries to pass a non-default addrspace pointer to the pointer argument which has default addrspace.
The fix is to add pointer type as another overloaded type to @llvm.masked.load/store intrinsics.
Reviewed By: reames
Differential Revision: http://reviews.llvm.org/D17270
llvm-svn: 266086
They broke the msan bot.
Original message:
Add __atomic_* lowering to AtomicExpandPass.
AtomicExpandPass can now lower atomic load, atomic store, atomicrmw,and
cmpxchg instructions to __atomic_* library calls, when the target
doesn't support atomics of a given size.
This is the first step towards moving all atomic lowering from clang
into llvm. When all is done, the behavior of __sync_* builtins,
__atomic_* builtins, and C11 atomics will be unified.
Previously LLVM would pass everything through to the ISelLowering
code. There, unsupported atomic instructions would turn into __sync_*
library calls. Because of that behavior, Clang currently avoids emitting
llvm IR atomic instructions when this would happen, and emits __atomic_*
library functions itself, in the frontend.
This change makes LLVM able to emit __atomic_* libcalls, and thus will
eventually allow clang to depend on LLVM to do the right thing.
It is advantageous to do the new lowering to atomic libcalls in
AtomicExpandPass, before ISel time, because it's important that all
atomic operations for a given size either lower to __atomic_*
libcalls (which may use locks), or native instructions which won't. No
mixing and matching.
At the moment, this code is enabled only for SPARC, as a
demonstration. The next commit will expand support to all of the other
targets.
Differential Revision: http://reviews.llvm.org/D18200
llvm-svn: 266062
`allocsize` is a function attribute that allows users to request that
LLVM treat arbitrary functions as allocation functions.
This patch makes LLVM accept the `allocsize` attribute, and makes
`@llvm.objectsize` recognize said attribute.
The review for this was split into two patches for ease of reviewing:
D18974 and D14933. As promised on the revisions, I'm landing both
patches as a single commit.
Differential Revision: http://reviews.llvm.org/D14933
llvm-svn: 266032
r237193 fix handling of alloca size / align in MergeFunctions, but only tested one and didn't follow FunctionComparator::cmpOperations's usual comparison pattern. It also didn't update Instruction.cpp:haveSameSpecialState which I'll do separately.
llvm-svn: 266022
This is more robust to changes in the link ordering.
Differential Revision: http://reviews.llvm.org/D18946
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 266018
Add StackProtector to SafeStack. This adds limited protection against
data corruption in the caller frame. Current implementation treats
all stack protector levels as -fstack-protector-all.
llvm-svn: 266004
AtomicExpandPass can now lower atomic load, atomic store, atomicrmw, and
cmpxchg instructions to __atomic_* library calls, when the target
doesn't support atomics of a given size.
This is the first step towards moving all atomic lowering from clang
into llvm. When all is done, the behavior of __sync_* builtins,
__atomic_* builtins, and C11 atomics will be unified.
Previously LLVM would pass everything through to the ISelLowering
code. There, unsupported atomic instructions would turn into __sync_*
library calls. Because of that behavior, Clang currently avoids emitting
llvm IR atomic instructions when this would happen, and emits __atomic_*
library functions itself, in the frontend.
This change makes LLVM able to emit __atomic_* libcalls, and thus will
eventually allow clang to depend on LLVM to do the right thing.
It is advantageous to do the new lowering to atomic libcalls in
AtomicExpandPass, before ISel time, because it's important that all
atomic operations for a given size either lower to __atomic_*
libcalls (which may use locks), or native instructions which won't. No
mixing and matching.
At the moment, this code is enabled only for SPARC, as a
demonstration. The next commit will expand support to all of the other
targets.
Differential Revision: http://reviews.llvm.org/D18200
llvm-svn: 266002
This patch ensures that when we detect first-order recurrences, we reject a phi
node if its previous value is also a phi node. During vectorization the initial
and previous values of the recurrence are shuffled together to create the value
for the current iteration. However, phi nodes are not widened like other
instructions. This fixes PR27246.
Differential Revision: http://reviews.llvm.org/D18971
llvm-svn: 265983
This is the straightforward fix for PR26760:
https://llvm.org/bugs/show_bug.cgi?id=26760
But we still need to make some changes to generalize this helper function
and then send the lshr case into here.
llvm-svn: 265960
Summary:
This is the first step in also serializing the index out to LLVM
assembly.
The per-module summary written to bitcode is moved out of the bitcode
writer and to a new analysis pass (ModuleSummaryIndexWrapperPass).
The pass itself uses a new builder class to compute index, and the
builder class is used directly in places where we don't have a pass
manager (e.g. llvm-as).
Because we are computing summaries outside of the bitcode writer, we no
longer can use value ids created by the bitcode writer's
ValueEnumerator. This required changing the reference graph edge type
to use a new ValueInfo class holding a union between a GUID (combined
index) and Value* (permodule index). The Value* are converted to the
appropriate value ID during bitcode writing.
Also, this enables removal of the BitWriter library's dependence on the
Analysis library that was previously required for the summary computation.
Reviewers: joker.eph
Subscribers: joker.eph, llvm-commits
Differential Revision: http://reviews.llvm.org/D18763
llvm-svn: 265941
Summary:
If we can prove that an op.with.overflow intrinsic does not overflow, we
can get rid of the intrinsic, and replace it with non-wrapping
arithmetic.
Reviewers: atrick, regehr
Subscribers: sanjoy, mcrosier, llvm-commits
Differential Revision: http://reviews.llvm.org/D18685
llvm-svn: 265913
Vectorization cost of uniform load wasn't correctly calculated.
As a result, a simple loop that loads a uniform value wasn't vectorized.
Differential Revision: http://reviews.llvm.org/D18940
llvm-svn: 265901
Summary:
The llvm cos intrinsic currently does not propagate undef's. This change
transforms cos(undef) to null value or 0.
There are 2 test cases added as well.
Patch by Anna Thomas!
Reviewers: sanjoy
Subscribers: majnemer, llvm-commits
Differential Revision: http://reviews.llvm.org/D18863
llvm-svn: 265825
We had a select of a cast of a select but attempted to replace the outer
select with the inner select dispite their incompatible types.
Patch by Anton Korobeynikov!
This fixes PR27236.
llvm-svn: 265805
InstCombine cannot effectively remove redundant assumptions without them
registered in the assumption cache. The vectorizer can create identical
assumptions but doesn't register them with the cache, resulting in
slower compile times because InstCombine tries to reason about a lot
more assumptions.
Fix this by registering the cloned assumptions.
llvm-svn: 265800
This re-commits r265535 which was reverted in r265541 because it
broke the windows bots. The problem was that we had a PointerIntPair
which took a pointer to a struct allocated with new. The problem
was that new doesn't provide sufficient alignment guarantees.
This pattern was already present before r265535 and it just happened
to work. To fix this, we now separate the PointerToIntPair from the
ExitNotTakenInfo struct into a pointer and a bool.
Original commit message:
Summary:
When the backedge taken codition is computed from an icmp, SCEV can
deduce the backedge taken count only if one of the sides of the icmp
is an AddRecExpr. However, due to sign/zero extensions, we sometimes
end up with something that is not an AddRecExpr.
However, we can use SCEV predicates to produce a 'guarded' expression.
This change adds a method to SCEV to get this expression, and the
SCEV predicate associated with it.
In HowManyGreaterThans and HowManyLessThans we will now add a SCEV
predicate associated with the guarded backedge taken count when the
analyzed SCEV expression is not an AddRecExpr. Note that we only do
this as an alternative to returning a 'CouldNotCompute'.
We use new feature in Loop Access Analysis and LoopVectorize to analyze
and transform more loops.
Reviewers: anemet, mzolotukhin, hfinkel, sanjoy
Subscribers: flyingforyou, mcrosier, atrick, mssimpso, sanjoy, mzolotukhin, llvm-commits
Differential Revision: http://reviews.llvm.org/D17201
llvm-svn: 265786
This reverts commit r265765, reapplying r265759 after changing a call from
LocalAsMetadata::get to ValueAsMetadata::get (and adding a unit test). When a
local value is mapped to a constant (like "i32 %a" => "i32 7"), the new debug
intrinsic operand may no longer be pointing at a local.
http://lab.llvm.org:8080/green/job/clang-stage1-configure-RA_build/19020/
The previous coommit message follows:
--
This is a partial re-commit -- maybe more of a re-implementation -- of
r265631 (reverted in r265637).
This makes RF_IgnoreMissingLocals behave (almost) consistently between
the Value and the Metadata hierarchy. In particular:
- MapValue returns nullptr or "metadata !{}" for missing locals in
MetadataAsValue/LocalAsMetadata bridging paris, depending on
the RF_IgnoreMissingLocals flag.
- MapValue doesn't memoize LocalAsMetadata-related results.
- MapMetadata no longer deals with LocalAsMetadata or
RF_IgnoreMissingLocals at all. (This wasn't in r265631 at all, but
I realized during testing it would make the patch simpler with no
loss of generality.)
r265631 went too far, making both functions universally ignore
RF_IgnoreMissingLocals. This broke building (e.g.) compiler-rt.
Reassociate (and possibly other passes) don't currently maintain
dominates-use invariants for metadata operands, resulting in IR like
this:
define void @foo(i32 %arg) {
call void @llvm.some.intrinsic(metadata i32 %x)
%x = add i32 1, i32 %arg
}
If the inliner chooses to inline @foo into another function, then
RemapInstruction will call `MapValue(metadata i32 %x)` and assert that
the return is not nullptr.
I've filed PR27273 to add a Verifier check and fix the underlying
problem in the optimization passes.
As a workaround, return `!{}` instead of nullptr for unmapped
LocalAsMetadata when RF_IgnoreMissingLocals is unset. Otherwise, match
the behaviour of r265631.
Original commit message:
ValueMapper: Make LocalAsMetadata match function-local Values
Start treating LocalAsMetadata similarly to function-local members of
the Value hierarchy in MapValue and MapMetadata.
- Don't memoize them.
- Return nullptr if they are missing.
This also cleans up ConstantAsMetadata to stop listening to the
RF_IgnoreMissingLocals flag.
llvm-svn: 265768
Summary:
Fixes PR26774.
If you're aware of the issue, feel free to skip the "Motivation"
section and jump directly to "This patch".
Motivation:
I define "refinement" as discarding behaviors from a program that the
optimizer has license to discard. So transforming:
```
void f(unsigned x) {
unsigned t = 5 / x;
(void)t;
}
```
to
```
void f(unsigned x) { }
```
is refinement, since the behavior went from "if x == 0 then undefined
else nothing" to "nothing" (the optimizer has license to discard
undefined behavior).
Refinement is a fundamental aspect of many mid-level optimizations done
by LLVM. For instance, transforming `x == (x + 1)` to `false` also
involves refinement since the expression's value went from "if x is
`undef` then { `true` or `false` } else { `false` }" to "`false`" (by
definition, the optimizer has license to fold `undef` to any non-`undef`
value).
Unfortunately, refinement implies that the optimizer cannot assume
that the implementation of a function it can see has all of the
behavior an unoptimized or a differently optimized version of the same
function can have. This is a problem for functions with comdat
linkage, where a function can be replaced by an unoptimized or a
differently optimized version of the same source level function.
For instance, FunctionAttrs cannot assume a comdat function is
actually `readnone` even if it does not have any loads or stores in
it; since there may have been loads and stores in the "original
function" that were refined out in the currently visible variant, and
at the link step the linker may in fact choose an implementation with
a load or a store. As an example, consider a function that does two
atomic loads from the same memory location, and writes to memory only
if the two values are not equal. The optimizer is allowed to refine
this function by first CSE'ing the two loads, and the folding the
comparision to always report that the two values are equal. Such a
refined variant will look like it is `readonly`. However, the
unoptimized version of the function can still write to memory (since
the two loads //can// result in different values), and selecting the
unoptimized version at link time will retroactively invalidate
transforms we may have done under the assumption that the function
does not write to memory.
Note: this is not just a problem with atomics or with linking
differently optimized object files. See PR26774 for more realistic
examples that involved neither.
This patch:
This change introduces a new set of linkage types, predicated as
`GlobalValue::mayBeDerefined` that returns true if the linkage type
allows a function to be replaced by a differently optimized variant at
link time. It then changes a set of IPO passes to bail out if they see
such a function.
Reviewers: chandlerc, hfinkel, dexonsmith, joker.eph, rnk
Subscribers: mcrosier, llvm-commits
Differential Revision: http://reviews.llvm.org/D18634
llvm-svn: 265762
This is a partial re-commit -- maybe more of a re-implementation -- of
r265631 (reverted in r265637).
This makes RF_IgnoreMissingLocals behave (almost) consistently between
the Value and the Metadata hierarchy. In particular:
- MapValue returns nullptr or "metadata !{}" for missing locals in
MetadataAsValue/LocalAsMetadata bridging paris, depending on
the RF_IgnoreMissingLocals flag.
- MapValue doesn't memoize LocalAsMetadata-related results.
- MapMetadata no longer deals with LocalAsMetadata or
RF_IgnoreMissingLocals at all. (This wasn't in r265631 at all, but
I realized during testing it would make the patch simpler with no
loss of generality.)
r265631 went too far, making both functions universally ignore
RF_IgnoreMissingLocals. This broke building (e.g.) compiler-rt.
Reassociate (and possibly other passes) don't currently maintain
dominates-use invariants for metadata operands, resulting in IR like
this:
define void @foo(i32 %arg) {
call void @llvm.some.intrinsic(metadata i32 %x)
%x = add i32 1, i32 %arg
}
If the inliner chooses to inline @foo into another function, then
RemapInstruction will call `MapValue(metadata i32 %x)` and assert that
the return is not nullptr.
I've filed PR27273 to add a Verifier check and fix the underlying
problem in the optimization passes.
As a workaround, return `!{}` instead of nullptr for unmapped
LocalAsMetadata when RF_IgnoreMissingLocals is unset. Otherwise, match
the behaviour of r265631.
Original commit message:
ValueMapper: Make LocalAsMetadata match function-local Values
Start treating LocalAsMetadata similarly to function-local members of
the Value hierarchy in MapValue and MapMetadata.
- Don't memoize them.
- Return nullptr if they are missing.
This also cleans up ConstantAsMetadata to stop listening to the
RF_IgnoreMissingLocals flag.
llvm-svn: 265759
When GVN wants to re-interpret an already available value in a smaller
type, it needs to right-shift the value on big-endian systems to ensure
the correct bytes are accessed. The shift value is the difference of
the sizes of the two types.
This is correct as long as both types occupy multiples of full bytes.
However, when one of them is a sub-byte type like i1, this no longer
holds true: we still need to shift, but only to access the correct
*byte*. Accessing bits within the byte requires no shift in either
endianness; e.g. an i1 resides in the least-significant bit of its
containing byte on both big- and little-endian systems.
Therefore, the appropriate shift value to be used is the difference of
the *storage* sizes of the two types. This is already handled correctly
in one place where such a shift takes place (GetStoreValueForLoad), but
is incorrect in two other places: GetLoadValueForLoad and
CoerceAvailableValueToLoadType.
This patch changes both places to use the storage size as well.
Differential Revision: http://reviews.llvm.org/D18662
llvm-svn: 265684
Updating dominators for exit-blocks of the unrolled loops is not enough,
as shown in PR27157. The proper way is to update dominators for all
dominance-children of original loop blocks.
llvm-svn: 265605
Summary:
When the backedge taken codition is computed from an icmp, SCEV can
deduce the backedge taken count only if one of the sides of the icmp
is an AddRecExpr. However, due to sign/zero extensions, we sometimes
end up with something that is not an AddRecExpr.
However, we can use SCEV predicates to produce a 'guarded' expression.
This change adds a method to SCEV to get this expression, and the
SCEV predicate associated with it.
In HowManyGreaterThans and HowManyLessThans we will now add a SCEV
predicate associated with the guarded backedge taken count when the
analyzed SCEV expression is not an AddRecExpr. Note that we only do
this as an alternative to returning a 'CouldNotCompute'.
We use new feature in Loop Access Analysis and LoopVectorize to analyze
and transform more loops.
Reviewers: anemet, mzolotukhin, hfinkel, sanjoy
Subscribers: flyingforyou, mcrosier, atrick, mssimpso, sanjoy, mzolotukhin, llvm-commits
Differential Revision: http://reviews.llvm.org/D17201
llvm-svn: 265535
To quote the langref "Unlike sqrt in libm, however, llvm.sqrt has
undefined behavior for negative numbers other than -0.0 (which allows
for better optimization, because there is no need to worry about errno
being set). llvm.sqrt(-0.0) is defined to return -0.0 like IEEE sqrt."
This means that it's unsafe to replace sqrt with llvm.sqrt unless the
call is annotated with nnan.
Thanks to Hal Finkel for pointing this out!
llvm-svn: 265521
r265273 added Mapper::mapBlockAddress, which delays mapping a
blockaddress value until the function has a body. The condition was
backwards, and should be checking Function::empty instead of
GlobalValue::isDeclaration.
llvm-svn: 265508
Don't emit a gc.result for a statepoint lowered from
@llvm.experimental.deoptimize since the call into __llvm_deoptimize is
effectively noreturn. Instead follow the corresponding gc.statepoint
with an "unreachable".
llvm-svn: 265485
utils/update_test_checks.py was improved with:
http://reviews.llvm.org/rL265414
to CHECK-NEXT the first line of the IR function. This ensures that nothing bad
has happened before that.
llvm-svn: 265417
Presently, CodeGenPrepare deletes all nearly empty (only phi and branch)
basic blocks. This pass can delete loop preheaders which frequently creates
critical edges. A preheader can be a convenient place to spill registers to
the stack. If the entrance to a loop body is a critical edge, then spills
may occur in the loop body rather than immediately before it. This patch
protects loop preheaders from deletion in CodeGenPrepare even if they are
nearly empty.
Since the patch alters the CFG, it affects a large number of test cases.
In most cases, the changes are merely cosmetic (basic blocks have different
names or instruction orders change slightly). I am somewhat concerned about
the test/CodeGen/Mips/brdelayslot.ll test case. If the loop preheader is not
deleted, then the MIPS backend does not take advantage of a branch delay
slot. Consequently, I would like some close review by a MIPS expert.
The patch also partially subsumes D16893 from George Burgess IV. George
correctly notes that CodeGenPrepare does not actually preserve the dominator
tree. I think the dominator tree was usually not valid when CodeGenPrepare
ran, but I am using LoopInfo to mark preheaders, so the dominator tree is
now always valid before CodeGenPrepare.
Author: Tom Jablin (tjablin)
Reviewers: hfinkel george.burgess.iv vkalintiris dsanders kbarton cycheng
http://reviews.llvm.org/D16984
llvm-svn: 265397
Direct callees' that are cast to other function prototypes,
show up in the Call/Invoke instructions as ConstantExpr's.
Currently llvm::CallSite's getCalledFunction() fails
to return the callees in such expressions as direct calls.
Value profiling should avoid instrumenting such cases. Mostly NFC.
llvm-svn: 265330
Summary:
Useful for debugging since we lose this correlation after the permodule
summary/VST is read and until we later materialize source modules in the
function importer.
Reviewers: joker.eph
Subscribers: llvm-commits, joker.eph
Differential Revision: http://reviews.llvm.org/D18555
llvm-svn: 265327
Sinking comparisons in CGP can undo the job of hoisting them done
earlier by LICM, and soft-FP makes this an expensive mistake.
A common pattern that produces floating point comparisons uniform
over a loop is an explicit check for division by zero. If the divisor
is hoisted out of the loop, the comparison can also be, but hoisting
the function that unwinds is never legal, since it may cause side
effects in the loop body prior to the unwinding to not be executed.
Differential Revision: http://reviews.llvm.org/D18744
llvm-svn: 265264
Floating point intrinsics in LLVM are generally not speculatively
executed, since most of them are defined to behave the same as libm
functions, which set errno.
However, the only error that can happen when executing ceil, floor,
nearbyint, rint and round libm functions per POSIX.1-2001 is -ERANGE,
and that requires the maximum value of the exponent to be smaller
than the number of mantissa bits, which is not the case with any of
the floating point types supported by LLVM.
The trunc and copysign functions never set errno per per POSIX.1-2001.
Differential Revision: http://reviews.llvm.org/D18643
llvm-svn: 265262
A catchswitch cannot be preceded by another instruction in the same
basic block (other than a PHI node).
Instead, insert the extract element right after the materialization of
the vectorized value. This isn't optimal but is a reasonable compromise
given the constraints of WinEH.
This fixes PR27163.
llvm-svn: 265157
They're not necessary (since the stack pointer is trivially restored on
return), and the way LLVM inserts the stackrestore calls breaks the
IR (we get a stackrestore between the deoptimize call and the return).
llvm-svn: 265101
They're not necessary (since the lifetime of the alloca is trivially
over due to the return), and the way LLVM inserts the lifetime.end
markers breaks the IR (we get a lifetime end marker between the
deoptimize call and the return).
llvm-svn: 265100
This mostly cosmetic patch moves the DebugEmissionKind enum from DIBuilder
into DICompileUnit. DIBuilder is not the right place for this enum to live
in — a metadata consumer should not have to include DIBuilder.h.
I also added a Verifier check that checks that the emission kind of a
DICompileUnit is actually legal.
http://reviews.llvm.org/D18612
<rdar://problem/25427165>
llvm-svn: 265077
This patch simply mirrors the attributes we give to @llvm.nvvm.reflect
to the __nvvm_reflect libdevice call. This shaves about 30% of the code
in libdevice away because of CSE opportunities. It's also helps us
figure out that libdevice implementations of transcendental functions
don't have side-effects.
llvm-svn: 265060
Summary:
As discussed on llvm-dev[1].
This change adds the basic boilerplate code around having this intrinsic
in LLVM:
- Changes in Intrinsics.td, and the IR Verifier
- A lowering pass to lower @llvm.experimental.guard to normal
control flow
- Inliner support
[1]: http://lists.llvm.org/pipermail/llvm-dev/2016-February/095523.html
Reviewers: reames, atrick, chandlerc, rnk, JosephTremoulet, echristo
Subscribers: mcrosier, llvm-commits
Differential Revision: http://reviews.llvm.org/D18527
llvm-svn: 264976
Widening a PHI requires us to insert a trunc.
The logical place for this trunc is in the same BB as the PHI.
This is not possible if the BB is terminated by a catchswitch.
This fixes PR27133.
llvm-svn: 264926
The TailDup transform was removed in r138841 in 2011, along with most
of the tests for it. This test, however, was missed. Probably because
it had already been XFAIL'd for 3 years at that point (since r52243!)
and continued to fail when the opt flag for -tailduplicate stopped
being valid.
llvm-svn: 264916
This change prevents the loop vectorizer from vectorizing when all of the vector
types it generates will be scalarized. I've run into this problem on the PPC's QPX
vector ISA, which only holds floating-point vector types. The loop vectorizer
will, however, happily vectorize loops with purely integer computation. Here's
an example:
LV: The Smallest and Widest types: 32 / 32 bits.
LV: The Widest register is: 256 bits.
LV: Found an estimated cost of 0 for VF 1 For instruction: %indvars.iv25 = phi i64 [ 0, %entry ], [ %indvars.iv.next26, %for.body ]
LV: Found an estimated cost of 0 for VF 1 For instruction: %arrayidx = getelementptr inbounds [1600 x i32], [1600 x i32]* %a, i64 0, i64 %indvars.iv25
LV: Found an estimated cost of 0 for VF 1 For instruction: %2 = trunc i64 %indvars.iv25 to i32
LV: Found an estimated cost of 1 for VF 1 For instruction: store i32 %2, i32* %arrayidx, align 4
LV: Found an estimated cost of 1 for VF 1 For instruction: %indvars.iv.next26 = add nuw nsw i64 %indvars.iv25, 1
LV: Found an estimated cost of 1 for VF 1 For instruction: %exitcond27 = icmp eq i64 %indvars.iv.next26, 1600
LV: Found an estimated cost of 0 for VF 1 For instruction: br i1 %exitcond27, label %for.cond.cleanup, label %for.body
LV: Scalar loop costs: 3.
LV: Found an estimated cost of 0 for VF 2 For instruction: %indvars.iv25 = phi i64 [ 0, %entry ], [ %indvars.iv.next26, %for.body ]
LV: Found an estimated cost of 0 for VF 2 For instruction: %arrayidx = getelementptr inbounds [1600 x i32], [1600 x i32]* %a, i64 0, i64 %indvars.iv25
LV: Found an estimated cost of 0 for VF 2 For instruction: %2 = trunc i64 %indvars.iv25 to i32
LV: Found an estimated cost of 2 for VF 2 For instruction: store i32 %2, i32* %arrayidx, align 4
LV: Found an estimated cost of 1 for VF 2 For instruction: %indvars.iv.next26 = add nuw nsw i64 %indvars.iv25, 1
LV: Found an estimated cost of 1 for VF 2 For instruction: %exitcond27 = icmp eq i64 %indvars.iv.next26, 1600
LV: Found an estimated cost of 0 for VF 2 For instruction: br i1 %exitcond27, label %for.cond.cleanup, label %for.body
LV: Vector loop of width 2 costs: 2.
LV: Found an estimated cost of 0 for VF 4 For instruction: %indvars.iv25 = phi i64 [ 0, %entry ], [ %indvars.iv.next26, %for.body ]
LV: Found an estimated cost of 0 for VF 4 For instruction: %arrayidx = getelementptr inbounds [1600 x i32], [1600 x i32]* %a, i64 0, i64 %indvars.iv25
LV: Found an estimated cost of 0 for VF 4 For instruction: %2 = trunc i64 %indvars.iv25 to i32
LV: Found an estimated cost of 4 for VF 4 For instruction: store i32 %2, i32* %arrayidx, align 4
LV: Found an estimated cost of 1 for VF 4 For instruction: %indvars.iv.next26 = add nuw nsw i64 %indvars.iv25, 1
LV: Found an estimated cost of 1 for VF 4 For instruction: %exitcond27 = icmp eq i64 %indvars.iv.next26, 1600
LV: Found an estimated cost of 0 for VF 4 For instruction: br i1 %exitcond27, label %for.cond.cleanup, label %for.body
LV: Vector loop of width 4 costs: 1.
...
LV: Selecting VF: 8.
LV: The target has 32 registers
LV(REG): Calculating max register usage:
LV(REG): At #0 Interval # 0
LV(REG): At #1 Interval # 1
LV(REG): At #2 Interval # 2
LV(REG): At #4 Interval # 1
LV(REG): At #5 Interval # 1
LV(REG): VF = 8
The problem is that the cost model here is not wrong, exactly. Since all of
these operations are scalarized, their cost (aside from the uniform ones) are
indeed VF*(scalar cost), just as the model suggests. In fact, the larger the VF
picked, the lower the relative overhead from the loop itself (and the
induction-variable update and check), and so in a sense, picking the largest VF
here is the right thing to do.
The problem is that vectorizing like this, where all of the vectors will be
scalarized in the backend, isn't really vectorizing, but rather interleaving.
By itself, this would be okay, but then the vectorizer itself also interleaves,
and that's where the problem manifests itself. There's aren't actually enough
scalar registers to support the normal interleave factor multiplied by a factor
of VF (8 in this example). In other words, the problem with this is that our
register-pressure heuristic does not account for scalarization.
While we might want to improve our register-pressure heuristic, I don't think
this is the right motivating case for that work. Here we have a more-basic
problem: The job of the vectorizer is to vectorize things (interleaving aside),
and if the IR it generates won't generate any actual vector code, then
something is wrong. Thus, if every type looks like it will be scalarized (i.e.
will be split into VF or more parts), then don't consider that VF.
This is not a problem specific to PPC/QPX, however. The problem comes up under
SSE on x86 too, and as such, this change fixes PR26837 too. I've added Sanjay's
reduced test case from PR26837 to this commit.
Differential Revision: http://reviews.llvm.org/D18537
llvm-svn: 264904
We already try not to truncate PHIs in computeMinimalBitwidths. LoopVectorize can't handle it and we really don't need to, because both induction and reduction PHIs are truncated by other means.
However, we weren't bailing out in all the places we should have, and we ended up by returning a PHI to be truncated, which has caused PR27018.
This fixes PR17018.
llvm-svn: 264852
Prior to this patch, the MemorySSA caching visitor would cache all
calls that it visited. When paired with phi optimization, this can be
problematic. Consider:
define void @foo() {
; 1 = MemoryDef(liveOnEntry)
call void @clobberFunction()
br i1 undef, label %if.end, label %if.then
if.then:
; MemoryUse(??)
call void @readOnlyFunction()
; 2 = MemoryDef(1)
call void @clobberFunction()
br label %if.end
if.end:
; 3 = MemoryPhi(...)
; MemoryUse(?)
call void @readOnlyFunction()
ret void
}
When optimizing MemoryUse(?), we visit defs 1 and 2, so we note to
cache them later. We ultimately end up not being able to optimize
passed the Phi, so we set MemoryUse(?) to point to the Phi. We then
cache the clobbering call for def 1 to be the Phi.
This commit changes this behavior so that we wipe out any calls
added to VisistedCalls while visiting the defs of a phi we couldn't
optimize.
Aside: With this patch, we now can bootstrap clang/LLVM without a
single MemorySSA verifier failure. Woohoo. :)
llvm-svn: 264820
This patch teaches the caching MemorySSA walker a few things:
1. Not to walk Phis we've walked before. It seems that we tried to do
this before, but it didn't work so well in cases like:
define void @foo() {
%1 = alloca i8
%2 = alloca i8
br label %begin
begin:
; 3 = MemoryPhi({%0,liveOnEntry},{%end,2})
; 1 = MemoryDef(3)
store i8 0, i8* %2
br label %end
end:
; MemoryUse(?)
load i8, i8* %1
; 2 = MemoryDef(1)
store i8 0, i8* %2
br label %begin
}
Because we wouldn't put Phis in Q.Visited until we tried to visit them.
So, when trying to optimize MemoryUse(?):
- We would visit 3 above
- ...Which would make us put {%0,liveOnEntry} in Q.Visited
- ...Which would make us visit {%0,liveOnEntry}
- ...Which would make us put {%end,2} in Q.Visited
- ...Which would make us visit {%end,2}
- ...Which would make us visit 3
- ...Which would realize we've already visited everything in 3
- ...Which would make us conservatively return 3.
In the added test-case, (@looped_visitedonlyonce) this behavior would
cause us to give incorrect results. Specifically, we'd visit 4 twice
in the same query, but on the second visit, we'd skip while.cond because
it had been visited, visit if.then/if.then2, and cache "1" as the
clobbering def on the way back.
2. If we try to walk the defs of a {Phi,MemLoc} and see it has been
visited before, just hand back the Phi we're trying to optimize.
I promise this isn't as terrible as it seems. :)
We now insert {Phi,MemLoc} pairs just before walking the Phi's upward
defs. So, we check the cache for the {Phi,MemLoc} pair before checking
if we've already walked the Phi.
The {Phi,MemLoc} pair is (almost?) always guaranteed to have a cache
entry if we've already fully walked it, because we cache as we go.
So, if the {Phi,MemLoc} pair isn't in cache, either:
(a) we must be in the process of visiting it (in which case, we can't
give a better answer in a cache-as-we-go DFS walker)
(b) we visited it, but didn't cache it on the way back (...which seems
to require `ModifyingAccess` to not dominate `StartingAccess`,
so I'm 99% sure that would be an error. If it's not an error, I
haven't been able to get it to happen locally, so I suspect it's
rare.)
- - - - -
As a consequence of this change, we no longer skip upward defs of phis,
so we can kill the `VisitedOnlyOne` check. This gives us better accuracy
than we had before, at the cost of potentially doing a bit more work
when we have a loop.
llvm-svn: 264814
This is effectively NFC, minus the renaming of the options
(-cyclone-prefetch-distance -> -prefetch-distance).
The change was requested by Tim in D17943.
llvm-svn: 264806
During ADCE, track which debug info scopes still have live references
from the code, and delete debug info intrinsics for the dead ones.
These intrinsics describe the locations of variables (in registers or
stack slots). If there's no code left corresponding to a variable's
scope, then there's no way to reference the variable in the debugger and
it doesn't matter what its value is.
I add a DEBUG printout when the described location in an SSA register,
in case it helps some trying to track down why locations get lost.
However, we still delete these; the scope itself isn't attached to any
real code, so the ship has already sailed.
llvm-svn: 264800
Add function soft attribute to the generation of Jump Tables in CodeGen
as initial step towards clang support of gcc's no-jump-table support
Reviewers: hans, echristo
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D18321
llvm-svn: 264756
When eliminating or merging almost empty basic blocks, the existence of non-trivial PHI nodes
is currently used to recognize potential loops of which the block is the header and keep the block.
However, the current algorithm fails if the loops' exit condition is evaluated only with volatile
values hence no PHI nodes in the header. Especially when such a loop is an outer loop of a nested
loop, the loop is collapsed into a single loop which prevent later optimizations from being
applied (e.g., transforming nested loops into simplified forms and loop vectorization).
The patch augments the existing PHI node-based check by adding a pre-test if the BB actually
belongs to a set of loop headers and not eliminating it if yes.
llvm-svn: 264697
A DICompileUnit that is not listed in llvm.dbg.cu will cause assertion
failures and/or crashes in the backend. The Verifier should reject this.
rdar://problem/25369499
llvm-svn: 264657
When eliminating or merging almost empty basic blocks, the existence of non-trivial PHI nodes
is currently used to recognize potential loops of which the block is the header and keep the block.
However, the current algorithm fails if the loops' exit condition is evaluated only with volatile
values hence no PHI nodes in the header. Especially when such a loop is an outer loop of a nested
loop, the loop is collapsed into a single loop which prevent later optimizations from being
applied (e.g., transforming nested loops into simplified forms and loop vectorization).
The patch augments the existing PHI node-based check by adding a pre-test if the BB actually
belongs to a set of loop headers and not eliminating it if yes.
llvm-svn: 264596
Summary:
Add a statistic to count the number of imported functions. Also, add a
new -print-imports option to emit a trace of imported functions, that
works even for an NDEBUG build.
Note that emitOptimizationRemark does not work for the above printing as
it expects a Function object and DebugLoc, neither of which we have
with summary-based importing.
This is part 2 of D18487, the first part was committed separately as
r264536.
Reviewers: joker.eph
Subscribers: llvm-commits, joker.eph
Differential Revision: http://reviews.llvm.org/D18487
llvm-svn: 264537
Reject the following IR as malformed (assuming that %entry, %next are
not in a loop):
next:
%y = phi i32 [ 0, %entry ]
%x = phi i32 [ %y, %entry ]
Such PHI nodes came up in PR26718. While there was no consensus on
whether or not this is valid IR, most opinions on that bug and in a
discussion on the llvm-dev mailing list tended towards a
"strict interpretation" (term by Joseph Tremoulet) of PHI node uses.
Also, the language reference explicitly states that "the use of each
incoming value is deemed to occur on the edge from the corresponding
predecessor block to the current block" and
`DominatorTree::dominates(Instruction*, Use&)` uses this definition as
well.
For the code mentioned in PR15384, clang does not compile to such PHIs
(anymore?). The test case still hangs when replacing `%tmp6` with `%tmp`
in revisions before r176366 (where PR15384 has been fixed). The
occurrence of %tmp6 therefore was probably unintentional. Its value is
not used except in other PHIs.
Reviewers: majnemer, reames, JosephTremoulet, bkramer, grosser, jdoerfert, kparzysz, sanjoy
Differential Revision: http://reviews.llvm.org/D18443
llvm-svn: 264528
Summary:
Now that the summary contains the full reference/call graph, we can
replace the existing function importer that loads and inspect the IR
to iteratively walk the call graph by a traversal based purely on the
summary information. Decouple the actual importing decision from any
IR manipulation.
Reviewers: tejohnson
Subscribers: llvm-commits, joker.eph
Differential Revision: http://reviews.llvm.org/D18343
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 264503
A release fence acts as a publication barrier for stores within the current thread to become visible to other threads which might observe the release fence. It does not require the current thread to observe stores performed on other threads. As a result, we can allow store-load and load-load forwarding across a release fence.
We choose to be much more conservative about stores. In theory, nothing prevents us from shifting a store from after a release fence to before it, and then eliminating the preceeding (previously fenced) store. Doing this without actually moving the second store is likely also legal, but we chose to be conservative at this time.
The LangRef indicates only atomic loads and stores are effected by fences. This patch chooses to be far more conservative then that.
This is the GVN companion to http://reviews.llvm.org/D11434 which applied the same logic in EarlyCSE and has been baking in tree for a while now.
Differential Revision: http://reviews.llvm.org/D11436
llvm-svn: 264472
I didn't notice any significant changes in the actual checks here;
all of these tests already used FileCheck, so a script can batch
update them in one shot.
This commit is just to show the value of automating this process:
We have uniform formatting as opposed to a mish-mash of check
structure that changes based on individual prefs and the current
fashion. This makes it simpler to update when we find a bug or
make an enhancement.
llvm-svn: 264457
This changes RS4GC to lower calls to ``@llvm.experimental.deoptimize``
to gc.statepoints wrapping ``__llvm_deoptimize``, and changes
``callsGCLeafFunction`` to recognize ``@llvm.experimental.deoptimize``
as a non GC leaf function.
I've had to hard code the ``"__llvm_deoptimize"`` name in
RewriteStatepointsForGC; since ``TargetLibraryInfo`` is available only
during codegen. This isn't without precedent in the codebase, so I'm
not overtly concerned.
llvm-svn: 264456
We try to hoist the insertion point as high as possible to encourage
sharing. However, we must be careful not to hoist into a catchswitch as
it is both an EHPad and a terminator.
llvm-svn: 264344
isDependenceDistanceOfOne asserts that the store and the load access
through the same type. This function is also used by
removeDependencesFromMultipleStores so we need to make sure we filter
out mismatching types before reaching this point.
Now we do this when the initial candidates are gathered.
This is a refinement of the fix made in r262267.
Fixes PR27048.
llvm-svn: 264313
There are a few bugs in the walker that this patch addresses.
Primarily:
- Caching can break when we have multiple BBs without phis
- We weren't optimizing some phis properly
- Because of how the DFS iterator works, there were times where we
wouldn't cache any results of our DFS
I left the test cases with FIXMEs in, because I'm not sure how much
effort it will take to get those to work (read: We'll probably
ultimately have to end up redoing the walker, or we'll have to come up
with some creative caching tricks), and more test coverage = better.
Differential Revision: http://reviews.llvm.org/D18065
llvm-svn: 264180
CGP modifies the domtree in some cases, so saying that it preserves the
domtree is a lie. We'll be able to selectively preserve it with the new
pass manager.
Differential Revision: http://reviews.llvm.org/D16893
llvm-svn: 264099
When you have multiple LCSSA (single-operand) PHIs that are converted
into two-operand PHIs due to versioning, only assert that the PHI
currently being converted has a single operand. I.e. we don't want to
check PHIs that were converted earlier in the loop.
Fixes PR27023.
Thanks to Karl-Johan Karlsson for the minimized testcase!
llvm-svn: 264081
It's a bug fix.
For rerolled loops SE trip count remains unchanged. It leads to incorrect work of the next passes.
My patch just resets SE info for rerolled loop forcing SE to re-evaluate it next time it requested.
I also added a verifier call in the exisitng test to be sure no invalid SE data remain. Without my fix this test would fail with -verify-scev.
Differential Revision: http://reviews.llvm.org/D18316
llvm-svn: 264051
If we have a BB with only MemoryDefs, live-in calculations will ignore
it. This means we get results like this:
define void @foo(i8* %p) {
; 1 = MemoryDef(liveOnEntry)
store i8 0, i8* %p
br i1 undef, label %if.then, label %if.end
if.then:
; 2 = MemoryDef(1)
store i8 1, i8* %p
br label %if.end
if.end:
; 3 = MemoryDef(1)
store i8 2, i8* %p
ret void
}
...When there should be a MemoryPhi in the `if.end` BB.
This patch fixes that behavior.
llvm-svn: 263991
Summary:
replaceCongruentIVs can break LCSSA when trying to replace IV increments
since it tries to replace all uses of a phi node with another phi node
while both of the phi nodes are not necessarily in the processed loop.
This will cause an assert in IndVars.
To fix this, we add a check to make sure that the replacement maintains
LCSSA.
Reviewers: sanjoy
Subscribers: mzolotukhin, llvm-commits
Differential Revision: http://reviews.llvm.org/D18266
llvm-svn: 263941
The sinpi/cospi can be replaced with sincospi to remove unnecessary
computations. However, we need to make sure that the calls are within
the same function!
This fixes PR26993.
llvm-svn: 263875
Summary:
ThinLTO is relying on linkInModule to import selected function.
However a lot of "magic" was hidden in linkInModule and the IRMover,
who would rename and promote global variables on the fly.
This is moving to an approach where the steps are decoupled and the
client is reponsible to specify the list of globals to import.
As a consequence some test are changed because they were relying on
the previous behavior which was importing the definition of *every*
single global without control on the client side.
Now the burden is on the client to decide if a global has to be imported
or not.
Reviewers: tejohnson
Subscribers: joker.eph, llvm-commits
Differential Revision: http://reviews.llvm.org/D18122
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 263863
While not strictly necessary, since we don't support large integer
types, this avoids bugs due to silent truncation from uint64_t to a
32-bit unsigned (e.g. DL.isLegalInteger(DL.getTypeSizeInBits(Ty) )
This fixes PR26972.
Differential Revision: http://reviews.llvm.org/D18258
llvm-svn: 263850
The loop on IVOperand's incoming values assumes IVOperand to be an
induction variable on the loop over which `S Pred X` is invariant;
otherwise loop invariant incoming values to IVOperand are not guaranteed
to dominate the comparision.
This fixes PR26973.
llvm-svn: 263827
Summary:
It can hurt performance to prefetch ahead too much. Be conservative for
now and don't prefetch ahead more than 3 iterations on Cyclone.
Reviewers: hfinkel
Subscribers: llvm-commits, mzolotukhin
Differential Revision: http://reviews.llvm.org/D17949
llvm-svn: 263772
Summary:
And use this TTI for Cyclone. As it was explained in the original RFC
(http://thread.gmane.org/gmane.comp.compilers.llvm.devel/92758), the HW
prefetcher work up to 2KB strides.
I am also adding tests for this and the previous change (D17943):
* Cyclone prefetching accesses with a large stride
* Cyclone not prefetching accesses with a small stride
* Generic Aarch64 subtarget not prefetching either
Reviewers: hfinkel
Subscribers: aemerson, rengolin, llvm-commits, mzolotukhin
Differential Revision: http://reviews.llvm.org/D17945
llvm-svn: 263771
Summary:
Use the new LoopVersioning facility (D16712) to add noalias metadata in
the vector loop if we versioned with memchecks. This can enable some
optimization opportunities further down the pipeline (see the included
test or the benchmark improvement quoted in D16712).
The test also covers the bug I had in the initial version in D16712.
The vectorizer did not previously use LoopVersioning. The reason is
that the vectorizer performs its transformations in single shot. It
creates an empty single-block vector loop that it then populates with
the widened, if-converted instructions. Thus creating an intermediate
versioned scalar loop seems wasteful.
So this patch (rather than bringing in LoopVersioning fully) adds a
special interface to LoopVersioning to allow the vectorizer to add
no-alias annotation while still performing its own versioning.
As the vectorizer propagates metadata from the instructions in the
original loop to the vector instructions we also check the pointer in
the original instruction and see if LoopVersioning can add no-alias
metadata based on the issued memchecks.
Reviewers: hfinkel, nadav, mzolotukhin
Subscribers: mzolotukhin, llvm-commits
Differential Revision: http://reviews.llvm.org/D17191
llvm-svn: 263744
Summary:
If we decide to version a loop to benefit a transformation, it makes
sense to record the now non-aliasing accesses in the newly versioned
loop. This allows non-aliasing information to be used by subsequent
passes.
One example is 456.hmmer in SPECint2006 where after loop distribution,
we vectorize one of the newly distributed loops. To vectorize we
version this loop to fully disambiguate may-aliasing accesses. If we
add the noalias markers, we can use the same information in a later DSE
pass to eliminate some dead stores which amounts to ~25% of the
instructions of this hot memory-pipeline-bound loop. The overall
performance improves by 18% on our ARM64.
The scoped noalias annotation is added in LoopVersioning. The patch
then enables this for loop distribution. A follow-on patch will enable
it for the vectorizer. Eventually this should be run by default when
versioning the loop but first I'd like to get some feedback whether my
understanding and application of scoped noalias metadata is correct.
Essentially my approach was to have a separate alias domain for each
versioning of the loop. For example, if we first version in loop
distribution and then in vectorization of the distributed loops, we have
a different set of memchecks for each versioning. By keeping the scopes
in different domains they can conveniently be defined independently
since different alias domains don't affect each other.
As written, I also have a separate domain for each loop. This is not
necessary and we could save some metadata here by using the same domain
across the different loops. I don't think it's a big deal either way.
Probably the best is to review the tests first to see if I mapped this
problem correctly to scoped noalias markers. I have plenty of comments
in the tests.
Note that the interface is prepared for the vectorizer which needs the
annotateInstWithNoAlias API. The vectorizer does not use LoopVersioning
so we need a way to pass in the versioned instructions. This is also
why the maps have to become part of the object state.
Also currently, we only have an AA-aware DSE after the vectorizer if we
also run the LTO pipeline. Depending how widely this triggers we may
want to schedule a DSE toward the end of the regular pass pipeline.
Reviewers: hfinkel, nadav, ashutosh.nema
Subscribers: mssimpso, aemerson, llvm-commits, mcrosier
Differential Revision: http://reviews.llvm.org/D16712
llvm-svn: 263743
This is similar to D18133 where we allowed profile weights on select instructions.
This extends that change to also allow the 'unpredictable' attribute of branches to apply to selects.
A test to check that 'unpredictable' metadata is preserved when cloning instructions was checked in at:
http://reviews.llvm.org/rL263648
Differential Revision: http://reviews.llvm.org/D18220
llvm-svn: 263716
I'm testing out a script that auto-generates the check lines.
It's 98% copied from utils/update_llc_test_checks.py.
If others think this is useful, please let me know.
llvm-svn: 263668
I'm testing out a script that auto-generates the check lines.
It's 98% copied from utils/update_llc_test_checks.py.
If others think this is useful, please let me know.
llvm-svn: 263667
Summary:
Fix LSRInstance::HoistInsertPosition() to check the original insert
position block first for a canonical insertion point that is dominated
by all inputs. This leads to SCEV being able to reuse more instructions
since it currently tracks the instructions it creates for reuse by
keeping a table of <Value, insert point> pairs.
Reviewers: atrick
Subscribers: mcrosier, mzolotukhin, llvm-commits
Differential Revision: http://reviews.llvm.org/D18001
llvm-svn: 263644
This was a latent bug that got exposed by the change to add LoopSimplify
as a dependence to LoopLoadElimination. Since LoopInfo was corrupted
after LV, LoopSimplify mis-compiled nbench in the test-suite (more
details in the PR).
The problem was that when we create the blocks for predicated stores we
didn't add those to any loops.
The original testcase for store predication provides coverage for this
assuming we verify LI on the way out of LV.
Fixes PR26952.
llvm-svn: 263565
(Resubmitting after fixing missing file issue)
With the changes in r263275, there are now more than just functions in
the summary. Completed the renaming of data structures (started in
r263275) to reflect the wider scope. In particular, changed the
FunctionIndex* data structures to ModuleIndex*, and renamed related
variables and comments. Also renamed the files to reflect the changes.
A companion clang patch will immediately succeed this patch to reflect
this renaming.
llvm-svn: 263513
Summary:
Specifically, when we perform runtime loop unrolling of a loop that
contains a convergent op, we can only unroll k times, where k divides
the loop trip multiple.
Without this change, we'll happily unroll e.g. the following loop
for (int i = 0; i < N; ++i) {
if (i == 0) convergent_op();
foo();
}
into
int i = 0;
if (N % 2 == 1) {
convergent_op();
foo();
++i;
}
for (; i < N - 1; i += 2) {
if (i == 0) convergent_op();
foo();
foo();
}.
This is unsafe, because we've just added a control-flow dependency to
the convergent op in the prelude.
In general, runtime unrolling loops that contain convergent ops is safe
only if we don't have emit a prelude, which occurs when the unroll count
divides the trip multiple.
Reviewers: resistor
Subscribers: llvm-commits, mzolotukhin
Differential Revision: http://reviews.llvm.org/D17526
llvm-svn: 263509
Summary: This now try to reorder instructions in order to help create the optimizable pattern.
Reviewers: craig.topper, spatel, dexonsmith, Prazek, chandlerc, joker.eph, majnemer
Differential Revision: http://reviews.llvm.org/D16523
llvm-svn: 263503
With the changes in r263275, there are now more than just functions in
the summary. Completed the renaming of data structures (started in
r263275) to reflect the wider scope. In particular, changed the
FunctionIndex* data structures to ModuleIndex*, and renamed related
variables and comments. Also renamed the files to reflect the changes.
A companion clang patch will immediately succeed this patch to reflect
this renaming.
llvm-svn: 263490
As noted in:
https://llvm.org/bugs/show_bug.cgi?id=26636
This doesn't accomplish anything on its own. It's the first step towards preserving
and using branch weights with selects.
The next step would be to make sure we're propagating the info in all of the other
places where we create selects (SimplifyCFG, InstCombine, etc). I don't think there's
an easy fix to make this happen; we have to look at each transform individually to
determine how to correctly propagate the weights.
Along with that step, we need to then use the weights when making subsequent transform
decisions such as discussed in http://reviews.llvm.org/D16836.
The inliner test is independent but closely related. It verifies that metadata is
preserved when both branches and selects are cloned.
Differential Revision: http://reviews.llvm.org/D18133
llvm-svn: 263482
Summary:
Previously we had a notion of convergent functions but not of convergent
calls. This is insufficient to correctly analyze calls where the target
is unknown, e.g. indirect calls.
Now a call is convergent if it targets a known-convergent function, or
if it's explicitly marked as convergent. As usual, we can remove
convergent where we can prove that no convergent operations are
performed in the call.
Originally landed as r261544, then reverted in r261544 for (incidental)
build breakage. Re-landed here with no changes.
Reviewers: chandlerc, jingyue
Subscribers: llvm-commits, tra, jhen, hfinkel
Differential Revision: http://reviews.llvm.org/D17739
llvm-svn: 263481
The motivating example is this
for (j = n; j > 1; j = i) {
i = j / 2;
}
The signed division is safely to be changed to an unsigned division (j is known
to be larger than 1 from the loop guard) and later turned into a single shift
without considering the sign bit.
llvm-svn: 263406
This follows up on the related AVX instruction transforms, but this
one is too strange to do anything more with. Intel's behavioral
description of this instruction in its Software Developer's Manual
is tragi-comic.
llvm-svn: 263340
Summary: As we now have unit-tests for UnrollAnalyzer, we can convert some existing tests to this format. It should make the tests more robust.
Reviewers: chandlerc, sanjoy
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D17904
llvm-svn: 263318
Summary:
This intrinsic, together with deoptimization operand bundles, allow
frontends to express transfer of control and frame-local state from
one (typically more specialized, hence faster) version of a function
into another (typically more generic, hence slower) version.
In languages with a fully integrated managed runtime this intrinsic can
be used to implement "uncommon trap" like functionality. In unmanaged
languages like C and C++, this intrinsic can be used to represent the
slow paths of specialized functions.
Note: this change does not address how `@llvm.experimental_deoptimize`
is lowered. That will be done in a later change.
Reviewers: chandlerc, rnk, atrick, reames
Subscribers: llvm-commits, kmod, mjacob, maksfb, mcrosier, JosephTremoulet
Differential Revision: http://reviews.llvm.org/D17732
llvm-svn: 263281
Value profile instrumentation treats inline asm calls like they are
indirect calls. This causes problems when the 'Callee' is passed to a
ptrtoint cast -- the verifier rightly claims that this is bogus and
crashes opt.
llvm-svn: 263278
tests to run GVN in both modes.
This is mostly the boring refactoring just like SROA and other complex
transformation passes. There is some trickiness in that GVN's
ValueNumber class requires hand holding to get to compile cleanly. I'm
open to suggestions about a better pattern there, but I tried several
before settling on this. I was trying to balance my desire to sink as
much implementation detail into the source file as possible without
introducing overly many layers of abstraction.
Much like with SROA, the design of this system is made somewhat more
cumbersome by the need to support both pass managers without duplicating
the significant state and logic of the pass. The same compromise is
struck here.
I've also left a FIXME in a doxygen comment as the GVN pass seems to
have pretty woeful documentation within it. I'd like to submit this with
the FIXME and let those more deeply familiar backfill the information
here now that we have a nice place in an interface to put that kind of
documentaiton.
Differential Revision: http://reviews.llvm.org/D18019
llvm-svn: 263208
llvm::getDISubprogram walks the instructions in a function, looking for one in the scope of the current function, so that it can find the !dbg entry for the subprogram itself.
Now that !dbg is attached to functions, this should not be necessary. This patch changes all uses to just query the subprogram directly on the function.
Ideally this should be NFC, but in reality its possible that a function:
has no !dbg (in which case there's likely a bug somewhere in an opt pass), or
that none of the instructions had a scope referencing the function, so we used to not find the !dbg on the function but now we will
Reviewed by Duncan Exon Smith.
Differential Revision: http://reviews.llvm.org/D18074
llvm-svn: 263184
The code assumed that we always had a preheader without making the pass
dependent on LoopSimplify.
Thanks to Mattias Eriksson V for reporting this.
llvm-svn: 263173
This patch fixes the problem which occurs when loop-vectorize tries to use @llvm.masked.load/store intrinsic for a non-default addrspace pointer. It fails with "Calling a function with a bad signature!" assertion in CallInst constructor because it tries to pass a non-default addrspace pointer to the pointer argument which has default addrspace.
The fix is to add pointer type as another overloaded type to @llvm.masked.load/store intrinsics.
Reviewed By: reames
Differential Revision: http://reviews.llvm.org/D17270
llvm-svn: 263158
of, and I misdiagnosed for months and months.
Andrea has had a patch for this forever, but I just couldn't see how
it was fixing the root cause of the problem. It didn't make sense to me,
even though the patch was perfectly good and the analysis of the actual
failure event was *fantastic*.
Well, I came back to it today because the patch has sat for *far* too
long and needs attention and decided I wouldn't let it go until I really
understood what was going on. After quite some time in the debugger,
I finally realized that in fact I had just missed an important case with
my previous attempt to fix PR22093 in r225149. Not only do we need to
handle loads that won't be split, but stores-of-loads that we won't
split. We *do* actually have enough logic in the presplitting to form
new slices for split stores.... *unless* we decided not to split them!
I'm so sorry that it took me this long to come to the realization that
this is the issue. It seems so obvious in hind sight (of course).
Anyways, the fix becomes *much* smaller and more focused. The fact that
we're left doing integer smashing is related to the FIXME in my original
commit: fundamentally, we're not aggressive about pre-splitting for
loads and stores to the same alloca. If we want to get aggressive about
this, it'll need both what Andrea had put into the proposed fix, but
also a *lot* more logic to essentially iteratively pre-split the alloca
until we can't do any more. As I said in that commit log, its really
unclear that this is the right call. Instead, the integer blending and
letting targets lower this to narrower stores seems slightly better. But
we definitely shouldn't really go down that path just to fix this bug.
Again, tons of thanks are owed to Andrea and others at Sony for working
on this bug. I really should have seen what was going on here and
re-directed them sooner. =////
llvm-svn: 263121
This patch teaches CGP to duplicate addressing mode computations into cold paths (detected via explicit cold attribute on calls) if required to let addressing mode be safely sunk into the basic block containing each load and store.
In general, duplicating code into cold blocks may result in code growth, but should not effect performance. In this case, it's better to duplicate some code than to put extra pressure on the register allocator by making it keep the address through the entirely of the fast path.
This patch only handles addressing computations, but in principal, we could implement a more general cold cold scheduling heuristic which tries to reduce register pressure in the fast path by duplicating code into the cold path. Getting the profitability of the general case right seemed likely to be challenging, so I stuck to the existing case (addressing computation) we already had.
Differential Revision: http://reviews.llvm.org/D17652
llvm-svn: 263074
This patch teaches LICM's implementation of store promotion to exploit the fact that the memory location being accessed might be provable thread local. The fact it's thread local weakens the requirements for where we can insert stores since no other thread can observe the write. This allows us perform store promotion even in cases where the store is not guaranteed to execute in the loop.
Two key assumption worth drawing out is that this assumes a) no-capture is strong enough to imply no-escape, and b) standard allocation functions like malloc, calloc, and operator new return values which can be assumed not to have previously escaped.
In future work, it would be nice to generalize this so that it works without directly seeing the allocation site. I believe that the nocapture return attribute should be suitable for this purpose, but haven't investigated carefully. It's also likely that we could support unescaped allocas with similar reasoning, but since SROA and Mem2Reg should destroy those, they're less interesting than they first might seem.
Differential Revision: http://reviews.llvm.org/D16783
llvm-svn: 263072
The irony of this patch is that one CPU that is affected is AMD Jaguar, and Jaguar
has a completely double-pumped AVX implementation. But getting the cost model to
reflect that is a much bigger problem. The small goal here is simply to improve on
the lie that !AVX2 == SandyBridge.
Differential Revision: http://reviews.llvm.org/D18000
llvm-svn: 263069
When checking whether an smin is positive, we can move the comparison to one of the inputs if the other is known positive. If the known positive one is the min, then the other can't be negative. If the other is the min, then we compute the min.
Differential Revision: http://reviews.llvm.org/D17873
llvm-svn: 263059
I somehow missed this. The case in GCC (global_alloc) was similar to
the new testcase except it had an array of structs rather than a two
dimensional array.
Fixes RP26885.
llvm-svn: 263058
As part of r251146 InstCombine was extended to call computeKnownBits on
every value in the function to determine whether it happens to be
constant. This increases typical compiletime by 1-3% (5% in irgen+opt
time) in my measurements. On the other hand this case did not trigger
once in the whole llvm-testsuite.
This patch introduces the notion of ExpensiveCombines which are only
enabled for OptLevel > 2. I removed the check in InstructionSimplify as
that is called from various places where the OptLevel is not known but
given the rarity of the situation I think a check in InstCombine is
enough.
Differential Revision: http://reviews.llvm.org/D16835
llvm-svn: 263047
Original commit message:
calculate builtin_object_size if argument is a removable pointer
This patch fixes calculating correct value for builtin_object_size function
when pointer is used only in builtin_object_size function call and never
after that.
Patch by Strahinja Petrovic.
Differential Revision: http://reviews.llvm.org/D17337
Reland the original change with a small modification (first do a null check
and then do the cast) to satisfy ubsan.
llvm-svn: 263011
The diff is relatively large since I took a chance to rearrange the code I had to touch in a more obvious way, but the key bit is merely using the !range metadata when we can't analyze the instruction further. The previous !range metadata code was essentially just dead since no binary operator or cast will have !range metadata (per Verifier) and it was otherwise dropped on the floor.
llvm-svn: 262751
This experiment was originally about trying to use facts implied dominating conditions to infer more precise known bits. While the compile time was found to be acceptable on several large code bases, we never found sufficiently profitable examples to justify turning on the code by default. Given this, it's time to abandon the experiment.
Several folks have commented that they've found this useful for experimentation, but nothing has come of those experiments. Given how easy the patch is to apply, there's no reason to leave the code in tree.
For anyone interested in further investigation in this area, I recommend finding the summary email I sent on one of the original review threads. In particular, I now believe the use-list based approach is strictly worse than the dom-tree-walking approach.
llvm-svn: 262646
Given that we're not actually reducing the instruction count in the included
regression tests, I think we would call this a canonicalization step.
The motivation comes from the example in PR26702:
https://llvm.org/bugs/show_bug.cgi?id=26702
If we hoist the bitwise logic ahead of the bitcast, the previously unoptimizable
example of:
define <4 x i32> @is_negative(<4 x i32> %x) {
%lobit = ashr <4 x i32> %x, <i32 31, i32 31, i32 31, i32 31>
%not = xor <4 x i32> %lobit, <i32 -1, i32 -1, i32 -1, i32 -1>
%bc = bitcast <4 x i32> %not to <2 x i64>
%notnot = xor <2 x i64> %bc, <i64 -1, i64 -1>
%bc2 = bitcast <2 x i64> %notnot to <4 x i32>
ret <4 x i32> %bc2
}
Simplifies to the expected:
define <4 x i32> @is_negative(<4 x i32> %x) {
%lobit = ashr <4 x i32> %x, <i32 31, i32 31, i32 31, i32 31>
ret <4 x i32> %lobit
}
Differential Revision: http://reviews.llvm.org/D17583
llvm-svn: 262645
This patch provides the following infrastructure for PGO enhancements in inliner:
Enable the use of block level profile information in inliner
Incremental update of block frequency information during inlining
Update the function entry counts of callees when they get inlined into callers.
Differential Revision: http://reviews.llvm.org/D16381
llvm-svn: 262636
The vectorization of first-order recurrences (r261346) caused PR26734. When
detecting these recurrences, we need to ensure that the previous value is
actually defined inside the loop. This patch includes the fix and test case.
llvm-svn: 262624
Summary: This is the last step toward supporting aggregate memory access in instcombine. This explodes stores of arrays into a serie of stores for each element, allowing them to be optimized.
Reviewers: joker.eph, reames, hfinkel, majnemer, mgrang
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D17828
llvm-svn: 262530
Summary: This is another step toward improving fca support. This unpack load of array in a series of load to array's elements.
Reviewers: chandlerc, joker.eph, majnemer, reames, hfinkel
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D15890
llvm-svn: 262521
This is a test that Akira Hatanaka wrote to test GlobalOpt's handling of
aliases with GEP operands. David Majnemer independently made the same
change to GlobalOpt in r212079. Akira's test is a useful addition, so I'm
pulling it over from the llvm repo for Swift on GitHub.
llvm-svn: 262510
As noted in the code comment, I don't think we can do the same transform that we do for
*scalar* integers comparisons to *vector* integers comparisons because it might pessimize
the general case.
Exhibit A for an incomplete integer comparison ISA remains x86 SSE/AVX: it only has EQ and GT
for integer vectors.
But we should now recognize all the variants of this construct and produce the optimal code
for the cases shown in:
https://llvm.org/bugs/show_bug.cgi?id=26701
llvm-svn: 262424
Summary: SampleProfile pass needs to be performed after InstructionCombiningPass, which helps eliminate un-inlinable function calls.
Reviewers: davidxl, dnovillo
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D17742
llvm-svn: 262419
This patch fixes calculating correct value for builtin_object_size function
when pointer is used only in builtin_object_size function call and never
after that.
Patch by Strahinja Petrovic.
Differential Revision: http://reviews.llvm.org/D17337
llvm-svn: 262337
The intended effect of this patch in conjunction with:
http://reviews.llvm.org/rL259392http://reviews.llvm.org/rL260145
is that customers using the AVX intrinsics in C will benefit from combines when
the load mask is constant:
__m128 mload_zeros(float *f) {
return _mm_maskload_ps(f, _mm_set1_epi32(0));
}
__m128 mload_fakeones(float *f) {
return _mm_maskload_ps(f, _mm_set1_epi32(1));
}
__m128 mload_ones(float *f) {
return _mm_maskload_ps(f, _mm_set1_epi32(0x80000000));
}
__m128 mload_oneset(float *f) {
return _mm_maskload_ps(f, _mm_set_epi32(0x80000000, 0, 0, 0));
}
...so none of the above will actually generate a masked load for optimized code.
This is the masked load counterpart to:
http://reviews.llvm.org/rL262064
llvm-svn: 262269
merged into a loop that was subsequently unrolled (or otherwise nuked).
In this case it can't merge in the ASTs for any remaining nested loops,
it needs to re-add their instructions dircetly.
The fix is very isolated, but I've pulled the code for merging blocks
into the AST into a single place in the process. The only behavior
change is in the case which would have crashed before.
This fixes a crash reported by Mikael Holmen on the list after r261316
restored much of the loop pass pipelining and allowed us to actually do
this kind of nested transformation sequenc. I've taken that test case
and further reduced it into the somewhat twisty maze of loops in the
included test case. This does in fact trigger the bug even in this
reduced form.
llvm-svn: 262108
Most of this is fairly straight forward. Add handling for min/max via existing matcher utility and ConstantRange routines. Add handling for clamp by exploiting condition constraints on inputs.
Note that I'm only handling two constant ranges at this point. It would be reasonable to consider treating overdefined as a full range if the instruction is typed as an integer, but that should be a separate change.
Differential Revision: http://reviews.llvm.org/D17184
llvm-svn: 262085
The intended effect of this patch in conjunction with:
http://reviews.llvm.org/rL259392http://reviews.llvm.org/rL260145
is that customers using the AVX intrinsics in C will benefit from combines when
the store mask is constant:
void mstore_zero_mask(float *f, __m128 v) {
_mm_maskstore_ps(f, _mm_set1_epi32(0), v);
}
void mstore_fake_ones_mask(float *f, __m128 v) {
_mm_maskstore_ps(f, _mm_set1_epi32(1), v);
}
void mstore_ones_mask(float *f, __m128 v) {
_mm_maskstore_ps(f, _mm_set1_epi32(0x80000000), v);
}
void mstore_one_set_elt_mask(float *f, __m128 v) {
_mm_maskstore_ps(f, _mm_set_epi32(0x80000000, 0, 0, 0), v);
}
...so none of the above will actually generate a masked store for optimized code.
Differential Revision: http://reviews.llvm.org/D17485
llvm-svn: 262064
The constant folding for sdiv and udiv has a big discrepancy between the
comments and the code, which looks like a typo. Currently, we're folding
X / undef pretty inconsistently:
0 / undef -> undef
C / undef -> 0
undef / undef -> 0
Whereas the comments state we do X / undef -> undef. The logic that
returns zero is actually commented as doing undef / X -> 0, despite that
the LHS isn't undef in many of the cases that hit it.
llvm-svn: 261813
Summary:
Both the hardware and LLVM have changed since 2012.
Now, load-based heuristic don't show big differences any more on OoO cores.
There is no notable regressons and improvements on spec2000/2006. (Cortex-A57, Core i5).
Reviewers: spatel, zansari
Differential Revision: http://reviews.llvm.org/D16836
llvm-svn: 261809
This is part of the payoff for the refactoring in:
http://reviews.llvm.org/rL261649http://reviews.llvm.org/rL261707
In addition to removing a pile of duplicated code, the xor case was
missing the optimization for vector types because it checked
"SrcTy->isIntegerTy()" rather than "SrcTy->isIntOrIntVectorTy()"
like 'and' and 'or' were already doing.
This solves part of:
https://llvm.org/bugs/show_bug.cgi?id=26702
llvm-svn: 261750
DeleteDeadBlock was called indiscriminately, leading to cleanuprets with
undef cleanuppad references.
Instead, try to drain the BB of most of it's instructions if it is
unreachable. We can then remove the BB if it solely consists of a
terminator (and maybe some phis).
llvm-svn: 261731
It is problematic if the inlinee has a cleanupret which unwinds to
caller and we inline it into a call site which doesn't unwind.
If the funclet unwinds anywhere other than to the caller,
then we will give the funclet two unwind destinations.
This will result in a verifier failure.
Seeing as how the caller wasn't an invoke (which would locally unwind)
and that the funclet cannot unwind to caller, we must conclude that an
'unwind to caller' cleanupret is dynamically unreachable.
This fixes PR26698.
Differential Revision: http://reviews.llvm.org/D17536
llvm-svn: 261656
Summary:
Since this is an IR pass it's nice to be able to write tests without
llc. This is the counterpart of the llc test under
CodeGen/PowerPC/loop-data-prefetch.ll.
Reviewers: hfinkel
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D17464
llvm-svn: 261578
The issue was that we only required LCSSA rebuilding if the immediate
parent-loop had values used outside of it. The fix is to enaable the
same logic for all outer loops, not only immediate parent.
llvm-svn: 261575
This flag was part of a migration to a new means of handling vectors-of-points which was described in the llvm-dev thread "FYI: Relocating vector of pointers". The old code path has been off by default for a while without complaints, so time to cleanup.
llvm-svn: 261569
DMB instructions can be expensive, so it's best to avoid them if possible. In
atomicrmw operations there will always be an attempted store so a release
barrier is always needed, but in the cmpxchg case we can delay the DMB until we
know we'll definitely try to perform a store (and so need release semantics).
In the strong cmpxchg case this isn't quite free: we must duplicate the LDREX
instructions to skip the barrier on subsequent iterations. The basic outline
becomes:
ldrex rOld, [rAddr]
cmp rOld, rDesired
bne Ldone
dmb
Lloop:
strex rRes, rNew, [rAddr]
cbz rRes Ldone
ldrex rOld, [rAddr]
cmp rOld, rDesired
beq Lloop
Ldone:
So we'll skip this version for strong operations in "minsize" functions.
llvm-svn: 261568
This change reverts "246133 [RewriteStatepointsForGC] Reduce the number of new instructions for base pointers" and a follow on bugfix 12575.
As pointed out in pr25846, this code suffers from a memory corruption bug. Since I'm (empirically) not going to get back to this any time soon, simply reverting the problematic change is the right answer.
llvm-svn: 261565
Summary:
Previously we had a notion of convergent functions but not of convergent
calls. This is insufficient to correctly analyze calls where the target
is unknown, e.g. indirect calls.
Now a call is convergent if it targets a known-convergent function, or
if it's explicitly marked as convergent. As usual, we can remove
convergent where we can prove that no convergent operations are
performed in the call.
Reviewers: chandlerc, jingyue
Subscribers: hfinkel, jhen, tra, llvm-commits
Differential Revision: http://reviews.llvm.org/D17317
llvm-svn: 261544
Cleanuppads may be merged together if one is the only predecessor of the
other in which case a simple transform can be performed: replace the
a cleanupret with a branch and remove an unnecessary cleanuppad.
Differential Revision: http://reviews.llvm.org/D17459
llvm-svn: 261390
This patch enables the vectorization of first-order recurrences. A first-order
recurrence is a non-reduction recurrence relation in which the value of the
recurrence in the current loop iteration equals a value defined in the previous
iteration. The load PRE of the GVN pass often creates these recurrences by
hoisting loads from within loops.
In this patch, we add a new recurrence kind for first-order phi nodes and
attempt to vectorize them if possible. Vectorization is performed by shuffling
the values for the current and previous iterations. The vectorization cost
estimate is updated to account for the added shuffle instruction.
Contributed-by: Matthew Simpson and Chad Rosier <mcrosier@codeaurora.org>
Differential Revision: http://reviews.llvm.org/D16197
llvm-svn: 261346
Summary:
If we don't have the first and last access of an interleaved load group,
the first and last wide load in the loop can do an out of bounds
access. Even though we discard results from speculative loads,
this can cause problems, since it can technically generate page faults
(or worse).
We now discard interleaved load groups that don't have the first and
load in the group.
Reviewers: hfinkel, rengolin
Subscribers: rengolin, llvm-commits, mzolotukhin, anemet
Differential Revision: http://reviews.llvm.org/D17332
llvm-svn: 261331
routine.
We were getting this wrong in small ways and generally being very
inconsistent about it across loop passes. Instead, let's have a common
place where we do this. One minor downside is that this will require
some analyses like SCEV in more places than they are strictly needed.
However, this seems benign as these analyses are complete no-ops, and
without this consistency we can in many cases end up with the legacy
pass manager scheduling deciding to split up a loop pass pipeline in
order to run the function analysis half-way through. It is very, very
annoying to fix these without just being very pedantic across the board.
The only loop passes I've not updated here are ones that use
AU.setPreservesAll() such as IVUsers (an analysis) and the pass printer.
They seemed less relevant.
With this patch, almost all of the problems in PR24804 around loop pass
pipelines are fixed. The one remaining issue is that we run simplify-cfg
and instcombine in the middle of the loop pass pipeline. We've recently
added some loop variants of these passes that would seem substantially
cleaner to use, but this at least gets us much closer to the previous
state. Notably, the seven loop pass managers is down to three.
I've not updated the loop passes using LoopAccessAnalysis because that
analysis hasn't been fully wired into LoopSimplify/LCSSA, and it isn't
clear that those transforms want to support those forms anyways. They
all run late anyways, so this is harmless. Similarly, LSR is left alone
because it already carefully manages its forms and doesn't need to get
fused into a single loop pass manager with a bunch of other loop passes.
LoopReroll didn't use loop simplified form previously, and I've updated
the test case to match the trivially different output.
Finally, I've also factored all the pass initialization for the passes
that use this technique as well, so that should be done regularly and
reliably.
Thanks to James for the help reviewing and thinking about this stuff,
and Ben for help thinking about it as well!
Differential Revision: http://reviews.llvm.org/D17435
llvm-svn: 261316
Sanjoy Das