Summary:
This patch tries to vectorize loads of consecutive memory accesses, accessed
in non-consecutive or jumbled way. An earlier attempt was made with patch D26905
which was reverted back due to some basic issue with representing the 'use mask' of
jumbled accesses.
This patch fixes the mask representation by recording the 'use mask' in the usertree entry.
Change-Id: I9fe7f5045f065d84c126fa307ef6ebe0787296df
Reviewers: mkuper, loladiro, Ayal, zvi, danielcdh
Reviewed By: Ayal
Subscribers: mgrang, dcaballe, hans, mzolotukhin
Differential Revision: https://reviews.llvm.org/D36130
llvm-svn: 320548
Summary:
This change makes the call site creation more general if any of the
arguments is predicated on a condition in the call site's predecessors.
If we find a callsite, that potentially can be split, we collect the set
of conditions for the call site's predecessors (currently only 2
predecessors are allowed). To do that, we traverse each predecessor's
predecessors as long as it only has single predecessors and record the
condition, if it is relevant to the call site. For each condition, we
also check if the condition is taken or not. In case it is not taken,
we record the inverse predicate.
We use the recorded conditions to create the new call sites and split
the basic block.
This has 2 benefits: (1) it is slightly easier to see what is going on
(IMO) and (2) we can easily extend it to handle more complex control
flow.
Reviewers: davidxl, junbuml
Reviewed By: junbuml
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D40728
llvm-svn: 320547
Summary: This brings CPU overhead on bzip2 down from 5.5x to 2x.
Reviewers: kcc, alekseyshl
Subscribers: kubamracek, hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D41137
llvm-svn: 320538
Summary:
If we have pattern `store (load(bitcast(select (cmp(V1, V2), &V1,
&V2)))), bitcast)`, but the load is used in other instructions, it leads
to looping in InstCombiner. Patch adds additional check that all users
of the load instructions are stores and then replaces all uses of load
instruction by the new one with new type.
Reviewers: RKSimon, spatel, majnemer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41072
llvm-svn: 320525
This algorithm (explained more in the source code) takes into account
global redundancies by building a "pair map" to find common subexprs.
The primary motivation of this is to handle situations like
foo = (a * b) * c
bar = (a * d) * c
where we currently don't identify that "a * c" is redundant.
Accordingly, it prioritizes the emission of a * c so that CSE
can remove the redundant calculation later.
Does not change the actual reassociation algorithm -- only the
order in which the reassociated operand chain is reconstructed.
Gives ~1.5% floating point math instruction count reduction on
a large offline suite of graphics shaders.
llvm-svn: 320515
Summary:
The PGO gen/use passes currently fail with an assert failure if there's a
critical edge whose source is an IndirectBr instruction and that edge
needs to be instrumented.
To avoid this in certain cases, split IndirectBr critical edges in the PGO
gen/use passes. This works for blocks with single indirectbr predecessors,
but not for those with multiple indirectbr predecessors (splitting an
IndirectBr critical edge isn't always possible.)
Reviewers: davidxl, xur
Reviewed By: davidxl
Subscribers: efriedma, llvm-commits, mehdi_amini
Differential Revision: https://reviews.llvm.org/D40699
llvm-svn: 320511
Summary:
If we have pattern `store (load(bitcast(select (cmp(V1, V2), &V1,
&V2)))), bitcast)`, but the load is used in other instructions, it leads
to looping in InstCombiner. Patch adds additional check that all users
of the load instructions are stores and then replaces all uses of load
instruction by the new one with new type.
Reviewers: RKSimon, spatel, majnemer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41072
llvm-svn: 320510
Summary:
If we have pattern `store (load(bitcast(select (cmp(V1, V2), &V1,
&V2)))), bitcast)`, but the load is used in other instructions, it leads
to looping in InstCombiner. Patch adds additional check that all users
of the load instructions are stores and then replaces all uses of load
instruction by the new one with new type.
Reviewers: RKSimon, spatel, majnemer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41072
llvm-svn: 320499
Summary:
If we have pattern `store (load(bitcast(select (cmp(V1, V2), &V1,
&V2)))), bitcast)`, but the load is used in other instructions, it leads
to looping in InstCombiner. Patch adds additional check that all users
of the load instructions are stores and then replaces all uses of load
instruction by the new one with new type.
Reviewers: RKSimon, spatel, majnemer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41072
llvm-svn: 320496
Summary:
If we have pattern `store (load(bitcast(select (cmp(V1, V2), &V1,
&V2)))), bitcast)`, but the load is used in other instructions, it leads
to looping in InstCombiner. Patch adds additional check that all users
of the load instructions are stores and then replaces all uses of load
instruction by the new one with new type.
Reviewers: RKSimon, spatel, majnemer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41072
llvm-svn: 320488
If we have pattern `store (load(bitcast(select (cmp(V1, V2), &V1,
&V2)))), bitcast)`, but the load is used in other instructions, it leads
to looping in InstCombiner. Patch adds additional check that all users
of the load instructions are stores and then replaces all uses of load
instruction by the new one with new type.
Reviewers: RKSimon, spatel, majnemer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41072
llvm-svn: 320483
Summary:
Currently, in InstCombineLoadStoreAlloca, we have simplification
rules for the following cases:
1. load off a null
2. load off a GEP with null base
3. store to a null
This patch adds support for the fourth case which is store into a
GEP with null base. Since this is UB as well (and directly analogous to
the load off a GEP with null base), we can substitute the stored val
with undef in instcombine, so that SimplifyCFG can optimize this code
into unreachable code.
Note: Right now, simplifyCFG hasn't been taught about optimizing
this to unreachable and adding an llvm.trap (this is already done for
the above 3 cases).
Reviewers: majnemer, hfinkel, sanjoy, davide
Reviewed by: sanjoy, davide
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41026
llvm-svn: 320480
VecValuesToIgnore holds values that will not appear in the vectorized loop.
We should therefore ignore their cost when VF > 1.
Differential Revision: https://reviews.llvm.org/D40883
llvm-svn: 320463
Summary:
This solves PR35616.
We don't want the compiler to generate different code when we compile
with/without -g, so we now ignore debug intrinsics when determining if
the optimization can trigger or not.
Reviewers: junbuml
Subscribers: davide, JDevlieghere, llvm-commits
Differential Revision: https://reviews.llvm.org/D41068
llvm-svn: 320460
The tests fail (opt asserts) on Windows.
> Summary:
> If we have pattern `store (load(bitcast(select (cmp(V1, V2), &V1,
> &V2)))), bitcast)`, but the load is used in other instructions, it leads
> to looping in InstCombiner. Patch adds additional check that all users
> of the load instructions are stores and then replaces all uses of load
> instruction by the new one with new type.
>
> Reviewers: RKSimon, spatel, majnemer
>
> Subscribers: llvm-commits
>
> Differential Revision: https://reviews.llvm.org/D41072
llvm-svn: 320421
The function stack poisioner conditionally stores local variables
either in an alloca or in malloc'ated memory, which has the
unfortunate side-effect, that the actual address of the variable is
only materialized when the variable is accessed, which means that
those variables are mostly invisible to the debugger even when
compiling without optimizations.
This patch stores the address of the local stack base into an alloca,
which can be referred to by the debug info and is available throughout
the function. This adds one extra pointer-sized alloca to each stack
frame (but mem2reg can optimize it away again when optimizations are
enabled, yielding roughly the same debug info quality as before in
optimized code).
rdar://problem/30433661
Differential Revision: https://reviews.llvm.org/D41034
llvm-svn: 320415
Summary:
If we have pattern `store (load(bitcast(select (cmp(V1, V2), &V1,
&V2)))), bitcast)`, but the load is used in other instructions, it leads
to looping in InstCombiner. Patch adds additional check that all users
of the load instructions are stores and then replaces all uses of load
instruction by the new one with new type.
Reviewers: RKSimon, spatel, majnemer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41072
llvm-svn: 320407
This patch introduces getShadowOriginPtr(), a method that obtains both the shadow and origin pointers for an address as a Value pair.
The existing callers of getShadowPtr() and getOriginPtr() are updated to use getShadowOriginPtr().
The rationale for this change is to simplify KMSAN instrumentation implementation.
In KMSAN origins tracking is always enabled, and there's no direct mapping between the app memory and the shadow/origin pages.
Both the shadow and the origin pointer for a given address are obtained by calling a single runtime hook from the instrumentation,
therefore it's easier to work with those pointers together.
Reviewed at https://reviews.llvm.org/D40835.
llvm-svn: 320373
Summary:
Reuse the Linux new mapping as it is.
Sponsored by <The NetBSD Foundation>
Reviewers: joerg, eugenis, vitalybuka
Reviewed By: vitalybuka
Subscribers: llvm-commits, #sanitizers
Tags: #sanitizers
Differential Revision: https://reviews.llvm.org/D41022
llvm-svn: 320219
Summary:
This is LLVM instrumentation for the new HWASan tool. It is basically
a stripped down copy of ASan at this point, w/o stack or global
support. Instrumenation adds a global constructor + runtime callbacks
for every load and store.
HWASan comes with its own IR attribute.
A brief design document can be found in
clang/docs/HardwareAssistedAddressSanitizerDesign.rst (submitted earlier).
Reviewers: kcc, pcc, alekseyshl
Subscribers: srhines, mehdi_amini, mgorny, javed.absar, eraman, llvm-commits, hiraditya
Differential Revision: https://reviews.llvm.org/D40932
llvm-svn: 320217
Summary:
If a partially inlined function has debug info, we have to add debug
locations to the call instruction calling the outlined function.
We use the debug location of the first instruction in the outlined
function, as the introduced call transfers control to this statement and
there is no other equivalent line in the source code.
We also use the same debug location for the branch instruction added
to jump from artificial entry block for the outlined function, which just
jumps to the first actual basic block of the outlined function.
Reviewers: davide, aprantl, rriddle, dblaikie, danielcdh, wmi
Reviewed By: aprantl, rriddle, danielcdh
Subscribers: eraman, JDevlieghere, llvm-commits
Differential Revision: https://reviews.llvm.org/D40413
llvm-svn: 320199
Causes unexpected memory issue with New PM this time.
The new PM invalidates BPI but not BFI, leaving the
reference to BPI from BFI invalid.
Abandon this patch. There is a more general solution
which also handles runtime infinite loop (but not statically).
llvm-svn: 320180
Summary:
If we have the code like this:
```
float a, b;
a = std::max(a ,b);
```
it is converted into something like this:
```
%call = call dereferenceable(4) float* @_ZSt3maxIfERKT_S2_S2_(float* nonnull dereferenceable(4) %a.addr, float* nonnull dereferenceable(4) %b.addr)
%1 = bitcast float* %call to i32*
%2 = load i32, i32* %1, align 4
%3 = bitcast float* %a.addr to i32*
store i32 %2, i32* %3, align 4
```
After inlinning this code is converted to the next:
```
%1 = load float, float* %a.addr
%2 = load float, float* %b.addr
%cmp.i = fcmp fast olt float %1, %2
%__b.__a.i = select i1 %cmp.i, float* %a.addr, float* %b.addr
%3 = bitcast float* %__b.__a.i to i32*
%4 = load i32, i32* %3, align 4
%5 = bitcast float* %arrayidx to i32*
store i32 %4, i32* %5, align 4
```
This pattern is not recognized as minmax pattern.
Patch solves this problem by converting sequence
```
store (bitcast, (load bitcast (select ((cmp V1, V2), &V1, &V2))))
```
to a sequence
```
store (,load (select((cmp V1, V2), &V1, &V2)))
```
After this the code is recognized as minmax pattern.
Reviewers: RKSimon, spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D40304
llvm-svn: 320157
In more recent Linux kernels with 47 bit VMAs the layout of virtual memory
for powerpc64 changed causing the address sanitizer to not work properly. This
patch adds support for 47 bit VMA kernels for powerpc64 and fixes up test
cases.
https://reviews.llvm.org/D40907
There is an associated patch for compiler-rt.
Tested on several 4.x and 3.x kernel releases.
llvm-svn: 320109
Summary:
Make enum ModRefInfo an enum class. Changes to ModRefInfo values should
be done using inline wrappers.
This should prevent future bit-wise opearations from being added, which can be more error-prone.
Reviewers: sanjoy, dberlin, hfinkel, george.burgess.iv
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D40933
llvm-svn: 320107
As a new access is generated spanning across multiple fields, we need to
propagate alias info from all the fields to form the most generic alias info.
rdar://35602528
Differential Revision: https://reviews.llvm.org/D40617
llvm-svn: 319979
This patch factors out the main code transformation utilities in the pgo-driven
indirect call promotion pass and places them in Transforms/Utils. The change is
intended to be a non-functional change, letting non-pgo-driven passes share a
common implementation with the existing pgo-driven pass.
The common utilities are used to conditionally promote indirect call sites to
direct call sites. They perform the underlying transformation, and do not
consider profile information. The pgo-specific details (e.g., the computation
of branch weight metadata) have been left in the indirect call promotion pass.
Differential Revision: https://reviews.llvm.org/D40658
llvm-svn: 319963
Summary:
There is no need to replace the original call instruction if no
VarArgs need to be forwarded.
Reviewers: davide, rnk, majnemer, efriedma
Reviewed By: efriedma
Subscribers: eraman, llvm-commits
Differential Revision: https://reviews.llvm.org/D40412
llvm-svn: 319947
This caused PR35519.
> [memcpyopt] Teach memcpyopt to optimize across basic blocks
>
> This teaches memcpyopt to make a non-local memdep query when a local query
> indicates that the dependency is non-local. This notably allows it to
> eliminate many more llvm.memcpy calls in common Rust code, often by 20-30%.
>
> Fixes PR28958.
>
> Differential Revision: https://reviews.llvm.org/D38374
>
> [memcpyopt] Commit file missed in r319482.
>
> This change was meant to be included with r319482 but was accidentally
> omitted.
llvm-svn: 319873
Summary:
The aim is to make ModRefInfo checks and changes more intuitive
and less error prone using inline methods that abstract the bit operations.
Ideally ModRefInfo would become an enum class, but that change will require
a wider set of changes into FunctionModRefBehavior.
Reviewers: sanjoy, george.burgess.iv, dberlin, hfinkel
Subscribers: nlopes, llvm-commits
Differential Revision: https://reviews.llvm.org/D40749
llvm-svn: 319821
This uses ConstantRange::makeGuaranteedNoWrapRegion's newly-added handling for subtraction to allow CVP to remove some subtraction overflow checks.
Differential Revision: https://reviews.llvm.org/D40039
llvm-svn: 319807
Summary:
A true or false result is expected from a comparison, but it seems the possibility of undef was overlooked, which could lead to a failed assert. This is fixed by this patch by bailing out if we encounter undef.
The bug is old and the assert has been there since the end of 2014, so it seems this is unusual enough to forego optimization.
Patch by JesperAntonsson.
Reviewers: spatel, eeckstein, hans
Reviewed By: hans
Subscribers: uabelho, llvm-commits
Differential Revision: https://reviews.llvm.org/D40639
llvm-svn: 319768
Summary:
Move splitIndirectCriticalEdges() from CodeGenPrepare to BasicBlockUtils.h so
that it can be called from other places.
Reviewers: davidxl
Reviewed By: davidxl
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D40750
llvm-svn: 319689
(This reapplies r314253. r314253 was reverted on r314482 because of a
correctness regression on P100, but that regression was identified to be
something else.)
Summary:
Don't bail out on constant divisors for divisions that can be narrowed without
introducing control flow . This gives us a 32 bit multiply instead of an
emulated 64 bit multiply in the generated PTX assembly.
Reviewers: jlebar
Subscribers: jholewinski, mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D38265
llvm-svn: 319677
Summary:
Currently, we only support predication for forward loops with step
of 1. This patch enables loop predication for reverse or
countdownLoops, which satisfy the following conditions:
1. The step of the IV is -1.
2. The loop has a singe latch as B(X) = X <pred>
latchLimit with pred as s> or u>
3. The IV of the guard is the decrement
IV of the latch condition (Guard is: G(X) = X-1 u< guardLimit).
This patch was downstream for a while and is the last series of patches
that's from our LP implementation downstream.
Reviewers: apilipenko, mkazantsev, sanjoy
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D40353
llvm-svn: 319659
Turns out we can have comparisons which are indirect users of the induction variable that we can make invariant. In this case, there is no loop invariant value contributing and we'd fail an assert.
The test case was found by a java fuzzer and reduced. It's a real cornercase. You have to have a static loop which we've already proven only executes once, but haven't broken the backedge on, and an inner phi whose result can be constant folded by SCEV using exit count reasoning but not proven by isKnownPredicate. To my knowledge, only the fuzzer has hit this case.
llvm-svn: 319583
It causes builds to fail with "Instruction does not dominate all uses" (PR35497).
> Patch tries to improve vectorization of the following code:
>
> void add1(int * __restrict dst, const int * __restrict src) {
> *dst++ = *src++;
> *dst++ = *src++ + 1;
> *dst++ = *src++ + 2;
> *dst++ = *src++ + 3;
> }
> Allows to vectorize even if the very first operation is not a binary add, but just a load.
>
> Fixed issues related to previous commit.
>
> Reviewers: spatel, mzolotukhin, mkuper, hfinkel, RKSimon, filcab, ABataev
>
> Reviewed By: ABataev, RKSimon
>
> Subscribers: llvm-commits, RKSimon
>
> Differential Revision: https://reviews.llvm.org/D28907
llvm-svn: 319550
Summary:
A true or false result is expected from a comparison, but it seems the possibility of undef was overlooked, which could lead to a failed assert. This is fixed by this patch by bailing out if we encounter undef.
The bug is old and the assert has been there since the end of 2014, so it seems this is unusual enough to forego optimization.
Patch by: JesperAntonsson
Reviewers: spatel, eeckstein, hans
Reviewed By: hans
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D40639
llvm-svn: 319537
Patch tries to improve vectorization of the following code:
void add1(int * __restrict dst, const int * __restrict src) {
*dst++ = *src++;
*dst++ = *src++ + 1;
*dst++ = *src++ + 2;
*dst++ = *src++ + 3;
}
Allows to vectorize even if the very first operation is not a binary add, but just a load.
Fixed issues related to previous commit.
Reviewers: spatel, mzolotukhin, mkuper, hfinkel, RKSimon, filcab, ABataev
Reviewed By: ABataev, RKSimon
Subscribers: llvm-commits, RKSimon
Differential Revision: https://reviews.llvm.org/D28907
llvm-svn: 319531
These command line options are not intended for public use, and often
don't even make sense in the context of a particular tool anyway. About
90% of them are already hidden, but when people add new options they
forget to hide them, so if you were to make a brand new tool today, link
against one of LLVM's libraries, and run tool -help you would get a
bunch of junk that doesn't make sense for the tool you're writing.
This patch hides these options. The real solution is to not have
libraries defining command line options, but that's a much larger effort
and not something I'm prepared to take on.
Differential Revision: https://reviews.llvm.org/D40674
llvm-svn: 319505
If the thin module has no references to an internal global in the
merged module, we need to make sure to preserve that property if the
global is a member of a comdat group, as otherwise promotion can end
up adding global symbols to the comdat, which is not allowed.
This situation can arise if the external global in the thin module
has dead constant users, which would cause use_empty() to return
false and would cause us to try to promote it. To prevent this from
happening, discard the dead constant users before asking whether a
global is empty.
Differential Revision: https://reviews.llvm.org/D40593
llvm-svn: 319494
This teaches memcpyopt to make a non-local memdep query when a local query
indicates that the dependency is non-local. This notably allows it to
eliminate many more llvm.memcpy calls in common Rust code, often by 20-30%.
Fixes PR28958.
Differential Revision: https://reviews.llvm.org/D38374
llvm-svn: 319482
Currently, SROA splits loads and stores only when they are accessing the whole alloca.
This patch relaxes this limitation to allow splitting a load/store if all other loads and stores to the alloca are disjoint to or fully included in the current load/store. If there is no other load or store that crosses the boundary of the current load/store, the current splitting implementation works as is.
The whole-alloca loads and stores meet this new condition and so they are still splittable.
Here is a simplified motivating example.
struct record {
long long a;
int b;
int c;
};
int func(struct record r) {
for (int i = 0; i < r.c; i++)
r.b++;
return r.b;
}
When updating r.b (or r.c as well), LLVM generates redundant instructions on some platforms (such as x86_64, ppc64); here, r.b and r.c are packed into one 64-bit GPR when the struct is passed as a method argument.
With this patch, the above example is compiled into only few instructions without loop.
Without the patch, unnecessary loop-carried dependency is introduced by SROA and the loop cannot be eliminated by the later optimizers.
Differential Revision: https://reviews.llvm.org/D32998
llvm-svn: 319407
Support for outlining multiple regions of each function is added, as well as some basic heuristics to determine which regions are good to outline. Outline candidates limited to regions that are single-entry & single-exit. We also avoid outlining regions that produce live-exit variables, which may inhibit some forms of code motion (like commoning).
Fallback to the regular partial inlining scheme is retained when either i) no regions are identified for outlining in the function, or ii) the outlined function could not be inlined in any of its callers.
Differential Revision: https://reviews.llvm.org/D38190
llvm-svn: 319398
An alloca may be larger than a variable that is described to be stored
there. Don't create a dbg.value for fragments that are outside of the
variable.
This fixes PR35447.
https://bugs.llvm.org/show_bug.cgi?id=35447
llvm-svn: 319230
This is needed for cases when the memory access is not as big as the width of
the data type. For instance, storing i1 (1 bit) would be done in a byte (8
bits).
Using 'BitSize >> 3' (or '/ 8') would e.g. give the memory access of an i1 a
size of 0, which for instance makes alias analysis return NoAlias even when
it shouldn't.
There are no tests as this was done as a follow-up to the bugfix for the case
where this was discovered (r318824). This handles more similar cases.
Review: Björn Petterson
https://reviews.llvm.org/D40339
llvm-svn: 319173
The core idea is to (re-)introduce some redundancies where their cost is
hidden by the cost of materializing immediates for constant operands of
PHI nodes. When the cost of the redundancies is covered by this,
avoiding materializing the immediate has numerous benefits:
1) Less register pressure
2) Potential for further folding / combining
3) Potential for more efficient instructions due to immediate operand
As a motivating example, consider the remarkably different cost on x86
of a SHL instruction with an immediate operand versus a register
operand.
This pattern turns up surprisingly frequently, but is somewhat rarely
obvious as a significant performance problem.
The pass is entirely target independent, but it does rely on the target
cost model in TTI to decide when to speculate things around the PHI
node. I've included x86-focused tests, but any target that sets up its
immediate cost model should benefit from this pass.
There is probably more that can be done in this space, but the pass
as-is is enough to get some important performance on our internal
benchmarks, and should be generally performance neutral, but help with
more extensive benchmarking is always welcome.
One awkward part is that this pass has to be scheduled after
*everything* that can eliminate these kinds of redundancies. This
includes SimplifyCFG, GVN, etc. I'm open to suggestions about better
places to put this. We could in theory make it part of the codegen pass
pipeline, but there doesn't really seem to be a good reason for that --
it isn't "lowering" in any sense and only relies on pretty standard cost
model based TTI queries, so it seems to fit well with the "optimization"
pipeline model. Still, further thoughts on the pipeline position are
welcome.
I've also only implemented this in the new pass manager. If folks are
very interested, I can try to add it to the old PM as well, but I didn't
really see much point (my use case is already switched over to the new
PM).
I've tested this pretty heavily without issue. A wide range of
benchmarks internally show no change outside the noise, and I don't see
any significant changes in SPEC either. However, the size class
computation in tcmalloc is substantially improved by this, which turns
into a 2% to 4% win on the hottest path through tcmalloc for us, so
there are definitely important cases where this is going to make
a substantial difference.
Differential revision: https://reviews.llvm.org/D37467
llvm-svn: 319164
Summary:
I think we do not need to analyze debug intrinsics here, as they should
not impact codegen. This has 2 benefits: 1) slightly less work to do and
2) avoiding generating optimization remarks for converting calls to
debug intrinsics to tail calls, which are not really helpful for users.
Based on work by Sander de Smalen.
Reviewers: davide, trentxintong, aprantl
Reviewed By: aprantl
Subscribers: llvm-commits, JDevlieghere
Tags: #debug-info
Differential Revision: https://reviews.llvm.org/D40440
llvm-svn: 319158
This is to address a problem similar to those in D37460 for Scalar PRE. We should not
PRE across an instruction that may not pass execution to its successor unless it is safe
to speculatively execute it.
Differential Revision: https://reviews.llvm.org/D38619
llvm-svn: 319147
Revert "[SROA] Propagate !range metadata when moving loads."
Revert "[Mem2Reg] Clang-format unformatted parts of this file. NFCI."
Davide says they broke a bot.
llvm-svn: 319131
This tries to propagate !range metadata to a pre-existing load
when a load is optimized out. This is done instead of adding an
assume because converting loads to and from assumes creates a
lot of IR.
Patch by Ariel Ben-Yehuda.
Differential Revision: https://reviews.llvm.org/D37216
llvm-svn: 319096
enum TailCallKind { TCK_None = 0, TCK_Tail = 1, TCK_MustTail = 2,
TCK_NoTail = 3 };
TCK_NoTail is greater than TCK_Tail so taking the min does not do the
correct thing.
rdar://35639547
llvm-svn: 319075
MSan used to insert the shadow check of the store pointer operand
_after_ the shadow of the value operand has been written.
This happens to work in the userspace, as the whole shadow range is
always mapped. However in the kernel the shadow page may not exist, so
the bug may cause a crash.
This patch moves the address check in front of the shadow access.
llvm-svn: 318901
In a lambda where we expect to have result within bounds, add respective `nsw/nuw` flags to
help SCEV just in case if it fails to figure them out on its own.
Differential Revision: https://reviews.llvm.org/D40168
llvm-svn: 318898
After the dataflow algorithm proves that an argument is constant,
it replaces it value with the integer constant and drops the lattice
value associated to the DEF.
e.g. in the example we have @f() that's called twice:
call @f(undef, ...)
call @f(2, ...)
`undef` MEET 2 = 2 so we replace the argument and all its uses with
the constant 2.
Shortly after, tryToReplaceWithConstantRange() tries to get the lattice
value for the argument we just replaced, causing an assertion.
This function is a little peculiar as it runs when we're doing replacement
and not as part of the solver but still queries the solver.
The fix is that of checking whether we replaced the value already and
get a temporary lattice value for the constant.
Thanks to Zhendong Su for the report!
Fixes PR35357.
llvm-svn: 318817
Summary:
First step in adding MemorySSA as dependency for loop pass manager.
Adding the dependency under a flag.
New pass manager: MSSA pointer in LoopStandardAnalysisResults can be null.
Legacy and new pass manager: Use cl::opt EnableMSSALoopDependency. Disabled by default.
Reviewers: sanjoy, davide, gberry
Subscribers: mehdi_amini, Prazek, llvm-commits
Differential Revision: https://reviews.llvm.org/D40274
llvm-svn: 318772
properlyDominates() shouldn't be used as sort key. It causes different output between stdlibc++ and libc++.
Instead, I introduced RPOT. In most cases, it works for CSE.
llvm-svn: 318743
Summary:
Add the following heuristics for irreducible loop metadata:
- When an irreducible loop header is missing the loop header weight metadata,
give it the minimum weight seen among other headers.
- Annotate indirectbr targets with the loop header weight metadata (as they are
likely to become irreducible loop headers after indirectbr tail duplication.)
These greatly improve the accuracy of the block frequency info of the Python
interpreter loop (eg. from ~3-16x off down to ~40-55% off) and the Python
performance (eg. unpack_sequence from ~50% slower to ~8% faster than GCC) due to
better register allocation under PGO.
Reviewers: davidxl
Reviewed By: davidxl
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D39980
llvm-svn: 318693
Summary:
SROA can fail in rewriting alloca but still rewrite a phi resulting
in dead instruction elimination. The Changed flag was not being set
correctly, resulting in downstream passes using stale analyses.
The included test case will assert during the second BDCE pass as a
result.
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D39921
llvm-svn: 318677
Summary:
This change reverts r318575 and changes FindDynamicShadowStart() to
keep the memory range it found mapped PROT_NONE to make sure it is
not reused. We also skip MemoryRangeIsAvailable() check, because it
is (a) unnecessary, and (b) would fail anyway.
Reviewers: pcc, vitalybuka, kcc
Subscribers: srhines, kubamracek, mgorny, llvm-commits, hiraditya
Differential Revision: https://reviews.llvm.org/D40203
llvm-svn: 318666
This patch adds a new abstraction layer to VPlan and leverages it to model the planned
instructions that manipulate masks (AND, OR, NOT), introduced during predication.
The new VPValue and VPUser classes model how data flows into, through and out
of a VPlan, forming the vertices of a planned Def-Use graph. The new
VPInstruction class is a generic single-instruction Recipe that models a
planned instruction along with its opcode, operands and users. See
VectorizationPlan.rst for more details.
Differential Revision: https://reviews.llvm.org/D38676
llvm-svn: 318645
In rL316552, we ban intersection of unsigned latch range with signed range check and vice
versa, unless the entire range check iteration space is known positive. It was a correct
functional fix that saved us from dealing with ambiguous values, but it also appeared
to be a very restrictive limitation. In particular, in the following case:
loop:
%iv = phi i32 [ 0, %preheader ], [ %iv.next, %latch]
%iv.offset = add i32 %iv, 10
%rc = icmp slt i32 %iv.offset, %len
br i1 %rc, label %latch, label %deopt
latch:
%iv.next = add i32 %iv, 11
%cond = icmp i32 ult %iv.next, 100
br it %cond, label %loop, label %exit
Here, the unsigned iteration range is `[0, 100)`, and the safe range for range
check is `[-10, %len - 10)`. For unsigned iteration spaces, we use unsigned
min/max functions for range intersection. Given this, we wanted to avoid dealing
with `-10` because it is interpreted as a very big unsigned value. Semantically, range
check's safe range goes through unsigned border, so in fact it is two disjoint
ranges in IV's iteration space. Intersection of such ranges is not trivial, so we prohibited
this case saying that we are not allowed to intersect such ranges.
What semantics of this safe range actually means is that we can start from `-10` and go
up increasing the `%iv` by one until we reach `%len - 10` (for simplicity let's assume that
`%len - 10` is a reasonably big positive value).
In particular, this safe iteration space includes `0, 1, 2, ..., %len - 11`. So if we were able to return
safe iteration space `[0, %len - 10)`, we could safely intersect it with IV's iteration space. All
values in this range are non-negative, so using signed/unsigned min/max for them is unambiguous.
In this patch, we alter the algorithm of safe range calculation so that it returnes a subset of the
original safe space which is represented by one continuous range that does not go through wrap.
In order to reach this, we use modified SCEV substraction function. It can be imagined as a function
that substracts by `1` (or `-1`) as long as the further substraction does not cause a wrap in IV iteration
space. This allows us to perform IRCE in many situations when we deal with IV space and range check
of different types (in terms of signed/unsigned).
We apply this approach for both matching and not matching types of IV iteration space and the
range check. One implication of this is that now IRCE became smarter in detection of empty safe
ranges. For example, in this case:
loop:
%iv = phi i32 [ %begin, %preheader ], [ %iv.next, %latch]
%iv.offset = sub i32 %iv, 10
%rc = icmp ult i32 %iv.offset, %len
br i1 %rc, label %latch, label %deopt
latch:
%iv.next = add i32 %iv, 11
%cond = icmp i32 ult %iv.next, 100
br it %cond, label %loop, label %exit
If `%len` was less than 10 but SCEV failed to trivially prove that `%begin - 10 >u %len- 10`,
we could end up executing entire loop in safe preloop while the main loop was still generated,
but never executed. Now, cutting the ranges so that if both `begin - 10` and `%len - 10` overflow,
we have a trivially empty range of `[0, 0)`. This in some cases prevents us from meaningless optimization.
Differential Revision: https://reviews.llvm.org/D39954
llvm-svn: 318639
As the first test shows, we could transform an llvm intrinsic which never sets errno
into a libcall which could set errno (even though it's marked readnone?), so that's
not ideal.
It's possible that we can also transform a libcall which could set errno to an
intrinsic given the fast-math-flags constraint, but that's deferred to determine
exactly which set of FMF are needed.
Differential Revision: https://reviews.llvm.org/D40150
llvm-svn: 318628
Summary:
With this patch I tried to reduce the complexity of the code sightly, by
removing some indirection. Please let me know what you think.
Reviewers: junbuml, mcrosier, davidxl
Reviewed By: junbuml
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D40037
llvm-svn: 318593
We were not doing that for large shadow granularity. Also add more
stack frame layout tests for large shadow granularity.
Differential Revision: https://reviews.llvm.org/D39475
llvm-svn: 318581
Revert the following commits:
r318369 [asan] Fallback to non-ifunc dynamic shadow on android<22.
r318235 [asan] Prevent rematerialization of &__asan_shadow.
r317948 [sanitizer] Remove unnecessary attribute hidden.
r317943 [asan] Use dynamic shadow on 32-bit Android.
MemoryRangeIsAvailable() reads /proc/$PID/maps into an mmap-ed buffer
that may overlap with the address range that we plan to use for the
dynamic shadow mapping. This is causing random startup crashes.
llvm-svn: 318575
Summary: This change fix PR35342 by replacing only the current use with undef in unreachable blocks.
Reviewers: efriedma, mcrosier, igor-laevsky
Reviewed By: efriedma
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D40184
llvm-svn: 318551
making it no longer even remotely simple.
The pass will now be more of a "full loop unswitching" pass rather than
anything substantively simpler than any other approach. I plan to rename
it accordingly once the dust settles.
The key ideas of the new loop unswitcher are carried over for
non-trivial unswitching:
1) Fully unswitch a branch or switch instruction from inside of a loop to
outside of it.
2) Update the CFG and IR. This avoids needing to "remember" the
unswitched branches as well as avoiding excessively cloning and
reliance on complex parts of simplify-cfg to cleanup the cfg.
3) Update the analyses (where we can) rather than just blowing them away
or relying on something else updating them.
Sadly, #3 is somewhat compromised here as the dominator tree updates
were too complex for me to want to reason about. I will need to make
another attempt to do this now that we have a nice dynamic update API
for dominators. However, we do adhere to #3 w.r.t. LoopInfo.
This approach also adds an important principls specific to non-trivial
unswitching: not *all* of the loop will be duplicated when unswitching.
This fact allows us to compute the cost in terms of how much *duplicate*
code is inserted rather than just on raw size. Unswitching conditions
which essentialy partition loops will work regardless of the total loop
size.
Some remaining issues that I will be addressing in subsequent commits:
- Handling unstructured control flow.
- Unswitching 'switch' cases instead of just branches.
- Moving to the dynamic update API for dominators.
Some high-level, interesting limitationsV that folks might want to push
on as follow-ups but that I don't have any immediate plans around:
- We could be much more clever about not cloning things that will be
deleted. In fact, we should be able to delete *nothing* and do
a minimal number of clones.
- There are many more interesting selection criteria for which branch to
unswitch that we might want to look at. One that I'm interested in
particularly are a set of conditions which all exit the loop and which
can be merged into a single unswitched test of them.
Differential revision: https://reviews.llvm.org/D34200
llvm-svn: 318549
The logic of replacing of a couple `RANGE_CHECK_LOWER + RANGE_CHECK_UPPER`
into `RANGE_CHECK_BOTH` in fact duplicates the logic of range intersection which
happens when we calculate safe iteration space. Effectively, the result of intersection of
these ranges doesn't differ from the range of merged range check.
We chose to remove duplicating logic in favor of code simplicity.
Differential Revision: https://reviews.llvm.org/D39589
llvm-svn: 318508
All these headers already depend on CodeGen headers so moving them into
CodeGen fixes the layering (since CodeGen depends on Target, not the
other way around).
llvm-svn: 318490
The requirement is that shadow memory must be aligned to page
boundaries (4k in this case). Use a closed form equation that always
satisfies this requirement.
Differential Revision: https://reviews.llvm.org/D39471
llvm-svn: 318421
// trunc (binop X, C) --> binop (trunc X, C')
// trunc (binop (ext X), Y) --> binop X, (trunc Y)
I'm grouping sub with the other binops because that makes the code simpler
and the transforms are valid:
https://rise4fun.com/Alive/UeF
...so even though we don't expect a sub with constant Op1 or any of the
other opcodes with constant Op0 due to canonicalization rules, we might as
well handle those situations if non-canonical code somehow reaches this
point (it should just make instcombine more efficient in reaching its
end goal).
This should solve the problem that later manifests in the vectorizers in
PR35295:
https://bugs.llvm.org/show_bug.cgi?id=35295
llvm-svn: 318404
Fix a couple places where the minimum alignment/size should be a
function of the shadow granularity:
- alignment of AllGlobals
- the minimum left redzone size on the stack
Added a test to verify that the metadata_array is properly aligned
for shadow scale of 5, to be enabled when we add build support
for testing shadow scale of 5.
Differential Revision: https://reviews.llvm.org/D39470
llvm-svn: 318395
When expanding exit conditions for pre- and postloops, we may end up expanding a
recurrency from the loop to in its loop's preheader. This produces incorrect IR.
This patch ensures that IRCE uses SCEVExpander correctly and only expands code which
is safe to expand in this particular location.
Differentian Revision: https://reviews.llvm.org/D39234
llvm-svn: 318381
Note that one-use and shouldChangeType() are checked ahead of the switch.
Without the narrowing folds, we can produce inferior vector code as shown in PR35299:
https://bugs.llvm.org/show_bug.cgi?id=35299
llvm-svn: 318323
InstCombine salvages debug info for every instruction it erases from its
worklist, but it wasn't doing it during its initial DCE when populating
its worklist. This fixes that.
This should help improve availability of 'this' in optimized debug info
when casts are necessary.
llvm-svn: 318320
Summary:
Added more remarks to SLP pass, in particular "missed" optimization remarks.
Also proposed several tests for new functionality.
Patch by Vladimir Miloserdov!
For reference you may look at: https://reviews.llvm.org/rL302811
Reviewers: anemet, fhahn
Reviewed By: anemet
Subscribers: javed.absar, lattner, petecoup, yakush, llvm-commits
Differential Revision: https://reviews.llvm.org/D38367
llvm-svn: 318307
Summary:
This patch optimizes a binop sandwiched between 2 selects with the same condition. Since we know its only used by the select we can propagate the appropriate input value from the earlier select.
As I'm writing this I realize I may need to avoid doing this for division in case the select was protecting a divide by zero?
Reviewers: spatel, majnemer
Reviewed By: majnemer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D39999
llvm-svn: 318267
It crashes building sqlite; see reply on the llvm-commits thread.
> [SLPVectorizer] Failure to beneficially vectorize 'copyable' elements in integer binary ops.
>
> Patch tries to improve vectorization of the following code:
>
> void add1(int * __restrict dst, const int * __restrict src) {
> *dst++ = *src++;
> *dst++ = *src++ + 1;
> *dst++ = *src++ + 2;
> *dst++ = *src++ + 3;
> }
> Allows to vectorize even if the very first operation is not a binary add, but just a load.
>
> Fixed issues related to previous commit.
>
> Reviewers: spatel, mzolotukhin, mkuper, hfinkel, RKSimon, filcab, ABataev
>
> Reviewed By: ABataev, RKSimon
>
> Subscribers: llvm-commits, RKSimon
>
> Differential Revision: https://reviews.llvm.org/D28907
llvm-svn: 318239
Simplifying a loop latch changes the IR and we need to make sure the pass manager knows to invalidate analysis passes if that happened.
PR35210 discovered a case where we failed to invalidate the post dominator tree after this simplification because we no changes other than simplifying the loop latch.
Fixes PR35210.
Differential Revision: https://reviews.llvm.org/D40035
llvm-svn: 318237
Summary:
In the mode when ASan shadow base is computed as the address of an
external global (__asan_shadow, currently on android/arm32 only),
regalloc prefers to rematerialize this value to save register spills.
Even in -Os. On arm32 it is rather expensive (2 loads + 1 constant
pool entry).
This changes adds an inline asm in the function prologue to suppress
this behavior. It reduces AsanTest binary size by 7%.
Reviewers: pcc, vitalybuka
Subscribers: aemerson, kristof.beyls, hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D40048
llvm-svn: 318235
Summary:
Instcombine (and probably other passes) sometimes want to change the
type of an alloca. To do this, they generally create a new alloca with
the desired type, create a bitcast to make the new pointer type match
the old pointer type, replace all uses with the cast, and then simplify
the casts. We already knew how to salvage dbg.value instructions when
removing casts, but we can extend it to cover dbg.addr and dbg.declare.
Fixes a debug info quality issue uncovered in Chromium in
http://crbug.com/784609
Reviewers: aprantl, vsk
Subscribers: hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D40042
llvm-svn: 318203
Clang implements the -finstrument-functions flag inherited from GCC, which
inserts calls to __cyg_profile_func_{enter,exit} on function entry and exit.
This is useful for getting a trace of how the functions in a program are
executed. Normally, the calls remain even if a function is inlined into another
function, but it is useful to be able to turn this off for users who are
interested in a lower-level trace, i.e. one that reflects what functions are
called post-inlining. (We use this to generate link order files for Chromium.)
LLVM already has a pass for inserting similar instrumentation calls to
mcount(), which it does after inlining. This patch renames and extends that
pass to handle calls both to mcount and the cygprofile functions, before and/or
after inlining as controlled by function attributes.
Differential Revision: https://reviews.llvm.org/D39287
llvm-svn: 318195
Patch tries to improve vectorization of the following code:
void add1(int * __restrict dst, const int * __restrict src) {
*dst++ = *src++;
*dst++ = *src++ + 1;
*dst++ = *src++ + 2;
*dst++ = *src++ + 3;
}
Allows to vectorize even if the very first operation is not a binary add, but just a load.
Fixed issues related to previous commit.
Reviewers: spatel, mzolotukhin, mkuper, hfinkel, RKSimon, filcab, ABataev
Reviewed By: ABataev, RKSimon
Subscribers: llvm-commits, RKSimon
Differential Revision: https://reviews.llvm.org/D28907
llvm-svn: 318193
This patch is part of D38676.
The patch introduces two new Recipes to handle instructions whose vectorization
involves masking. These Recipes take VPlan-level masks in D38676, but still rely
on ILV's existing createEdgeMask(), createBlockInMask() in this patch.
VPBlendRecipe handles intra-loop phi nodes, which are vectorized as a sequence
of SELECTs. Its execute() code is refactored out of ILV::widenPHIInstruction(),
which now handles only loop-header phi nodes.
VPWidenMemoryInstructionRecipe handles load/store which are to be widened
(but are not part of an Interleave Group). In this patch it simply calls
ILV::vectorizeMemoryInstruction on execute().
Differential Revision: https://reviews.llvm.org/D39068
llvm-svn: 318149
Registers it and everything, updates all the references, etc.
Next patch will add support to Clang's `-fexperimental-new-pass-manager`
path to actually enable BoundsChecking correctly.
Differential Revision: https://reviews.llvm.org/D39084
llvm-svn: 318128
a legacy and new PM pass.
This essentially moves the class state to parameters and re-shuffles the
code to make that reasonable. It also does some minor cleanups along the
way and leaves some comments.
Differential Revision: https://reviews.llvm.org/D39081
llvm-svn: 318124
Summary:
If a compare instruction is same or inverse of the compare in the
branch of the loop latch, then return a constant evolution node.
This shall facilitate computations of loop exit counts in cases
where compare appears in the evolution chain of induction variables.
Will fix PR 34538
Reviewers: sanjoy, hfinkel, junryoungju
Reviewed By: sanjoy, junryoungju
Subscribers: javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D38494
llvm-svn: 318050
In more recent Linux kernels (including those with 47 bit VMAs) the layout of
virtual memory for powerpc64 changed causing the memory sanitizer to not
work properly. This patch adjusts a bit mask in the memory sanitizer to work
on the newer kernels while continuing to work on the older ones as well.
This is the non-runtime part of the patch and finishes it. ref: r317802
Tested on several 4.x and 3.x kernel releases.
llvm-svn: 318045
Summary:
This patch extends the partial inliner to support inlining parts of
vararg functions, if the vararg handling is done in the outlined part.
It adds a `ForwardVarArgsTo` argument to InlineFunction. If it is
non-null, all varargs passed to the inlined function will be added to
all calls to `ForwardVarArgsTo`.
The partial inliner takes care to only pass `ForwardVarArgsTo` if the
varargs handing is done in the outlined function. It checks that vastart
is not part of the function to be inlined.
`test/Transforms/CodeExtractor/PartialInlineNoInline.ll` (already part
of the repo) checks we do not do partial inlining if vastart is used in
a basic block that will be inlined.
Reviewers: davide, davidxl, grosser
Reviewed By: davide, davidxl, grosser
Subscribers: gyiu, grosser, eraman, llvm-commits
Differential Revision: https://reviews.llvm.org/D39607
llvm-svn: 318028
Summary:
This patch adds an early out to visitICmpInst if we are looking at a compare as part of an integer absolute value idiom. Similar is already done for min/max.
In the particular case I observed in a benchmark we had an absolute value of a load from an indexed global. We simplified the compare using foldCmpLoadFromIndexedGlobal into a magic bit vector, a shift, and an and. But the load result was still used for the select and the negate part of the absolute valute idiom. So we overcomplicated the code and lost the ability to recognize it as an absolute value.
I've chosen a simpler case for the test here.
Reviewers: spatel, davide, majnemer
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D39766
llvm-svn: 317994
Summary:
The following kernel change has moved ET_DYN base to 0x4000000 on arm32:
https://marc.info/?l=linux-kernel&m=149825162606848&w=2
Switch to dynamic shadow base to avoid such conflicts in the future.
Reserve shadow memory in an ifunc resolver, but don't use it in the instrumentation
until PR35221 is fixed. This will eventually let use save one load per function.
Reviewers: kcc
Subscribers: aemerson, srhines, kubamracek, kristof.beyls, hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D39393
llvm-svn: 317943
Summary:
The specification of the @llvm.memcpy.element.unordered.atomic intrinsic requires
that the pointer arguments have alignments of at least the element size. The existing
IRBuilder interface to create a call to this intrinsic does not allow for providing
the alignment of these pointer args. Having an interface that makes it easy to
construct invalid intrinsic calls doesn't seem sensible, so this patch simply
adds the requirement that one provide the argument alignments when using IRBuilder
to create atomic memcpy calls.
llvm-svn: 317918