Re-commit r357452 (take 3): "SimplifyCFG
SinkCommonCodeFromPredecessors: Also sink function calls without used
results (PR41259)"
Third time's the charm.
This was reverted in r363220 due to being suspected of an internal
benchmark regression and a test failure, none of which turned out to
be caused by this.
As reported in https://bugs.llvm.org/show_bug.cgi?id=43269, this causes
UNREACHABLE errors when compiling if_malo_pci.c for arm and aarch64.
Clarify comments on helpers used by LFTR [NFC]
I'm slowly wrapping my head around this code, and am making comment
improvements where I can.
Pull in r360972 from upstream llvm trunk (by Philip Reames):
[LFTR] Factor out a helper function for readability purpose [NFC]
Pull in r360976 from upstream llvm trunk (by Philip Reames):
[IndVars] Don't reimplement Loop::isLoopInvariant [NFC]
Using dominance vs a set membership check is indistinguishable from a
compile time perspective, and the two queries return equivelent
results. Simplify code by using the existing function.
Pull in r360978 from upstream llvm trunk (by Philip Reames):
[LFTR] Strengthen assertions in genLoopLimit [NFCI]
Pull in r362292 from upstream llvm trunk (by Nikita Popov):
[IndVarSimplify] Fixup nowrap flags during LFTR (PR31181)
Fix for https://bugs.llvm.org/show_bug.cgi?id=31181 and partial fix
for LFTR poison handling issues in general.
When LFTR moves a condition from pre-inc to post-inc, it may now
depend on value that is poison due to nowrap flags. To avoid this, we
clear any nowrap flag that SCEV cannot prove for the post-inc addrec.
Additionally, LFTR may switch to a different IV that is dynamically
dead and as such may be arbitrarily poison. This patch will correct
nowrap flags in some but not all cases where this happens. This is
related to the adoption of IR nowrap flags for the pre-inc addrec.
(See some of the switch_to_different_iv tests, where flags are not
dropped or insufficiently dropped.)
Finally, there are likely similar issues with the handling of GEP
inbounds, but we don't have a test case for this yet.
Differential Revision: https://reviews.llvm.org/D60935
Pull in r362971 from upstream llvm trunk (by Philip Reames):
Prepare for multi-exit LFTR [NFC]
This change does the plumbing to wire an ExitingBB parameter through
the LFTR implementation, and reorganizes the code to work in terms of
a set of individual loop exits. Most of it is fairly obvious, but
there's one key complexity which makes it worthy of consideration.
The actual multi-exit LFTR patch is in D62625 for context.
Specifically, it turns out the existing code uses the backedge taken
count from before a IV is widened. Oddly, we can end up with a
different (more expensive, but semantically equivelent) BE count for
the loop when requerying after widening. For the nestedIV example
from elim-extend, we end up with the following BE counts:
BEFORE: (-2 + (-1 * %innercount) + %limit)
AFTER: (-1 + (sext i32 (-1 + %limit) to i64) + (-1 * (sext i32 %innercount to i64))<nsw>)
This is the only test in tree which seems sensitive to this
difference. The actual result of using the wider BETC on this example
is that we actually produce slightly better code. :)
In review, we decided to accept that test change. This patch is
structured to preserve the old behavior, but a separate change will
immediate follow with the behavior change. (I wanted it separate for
problem attribution purposes.)
Differential Revision: https://reviews.llvm.org/D62880
Pull in r362975 from upstream llvm trunk (by Philip Reames):
[LFTR] Use recomputed BE count
This was discussed as part of D62880. The basic thought is that
computing BE taken count after widening should produce (on average)
an equally good backedge taken count as the one before widening.
Since there's only one test in the suite which is impacted by this
change, and it's essentially equivelent codegen, that seems to be a
reasonable assertion. This change was separated from r362971 so that
if this turns out to be problematic, the triggering piece is obvious
and easily revertable.
For the nestedIV example from elim-extend.ll, we end up with the
following BE counts:
BEFORE: (-2 + (-1 * %innercount) + %limit)
AFTER: (-1 + (sext i32 (-1 + %limit) to i64) + (-1 * (sext i32 %innercount to i64))<nsw>)
Note that before is an i32 type, and the after is an i64. Truncating
the i64 produces the i32.
Pull in r362980 from upstream llvm trunk (by Philip Reames):
Factor out a helper function for readability and reuse in a future
patch [NFC]
Pull in r363613 from upstream llvm trunk (by Philip Reames):
Fix a bug w/inbounds invalidation in LFTR (recommit)
Recommit r363289 with a bug fix for crash identified in pr42279.
Issue was that a loop exit test does not have to be an icmp, leading
to a null dereference crash when new logic was exercised for that
case. Test case previously committed in r363601.
Original commit comment follows:
This contains fixes for two cases where we might invalidate inbounds
and leave it stale in the IR (a miscompile). Case 1 is when switching
to an IV with no dynamically live uses, and case 2 is when doing
pre-to-post conversion on the same pointer type IV.
The basic scheme used is to prove that using the given IV (pre or
post increment forms) would have to already trigger UB on the path to
the test we're modifying. As such, our potential UB triggering use
does not change the semantics of the original program.
As was pointed out in the review thread by Nikita, this is defending
against a separate issue from the hasConcreteDef case. This is about
poison, that's about undef. Unfortunately, the two are different, see
Nikita's comment for a fuller explanation, he explains it well.
(Note: I'm going to address Nikita's last style comment in a separate
commit just to minimize chance of subtle bugs being introduced due to
typos.)
Differential Revision: https://reviews.llvm.org/D62939
Pull in r363875 from upstream llvm trunk (by Philip Reames):
[LFTR] Rename variable to minimize confusion [NFC]
(Recommit of r363293 which was reverted when a dependent patch was.)
As pointed out by Nikita in D62625, BackedgeTakenCount is generally
used to refer to the backedge taken count of the loop. A conditional
backedge taken count - one which only applies if a particular exit is
taken - is called a ExitCount in SCEV code, so be consistent here.
Pull in r363877 from upstream llvm trunk (by Philip Reames):
[LFTR] Stylistic cleanup as suggested in last review comment of
D62939 [NFC]
(Resumbit of r363292 which was reverted along w/an earlier patch)
Pull in r364346 from upstream llvm trunk (by Philip Reames):
[LFTR] Adjust debug output to include extensions (if any)
Pull in r364693 from upstream llvm trunk (by Philip Reames):
[IndVars] Remove a bit of manual constant folding [NFC]
SCEV is more than capable of folding (add x, trunc(0)) to x.
Pull in r364709 from upstream llvm trunk (by Nikita Popov):
[LFTR] Fix post-inc pointer IV with truncated exit count (PR41998)
Fixes https://bugs.llvm.org/show_bug.cgi?id=41998. Usually when we
have a truncated exit count we'll truncate the IV when comparing
against the limit, in which case exit count overflow in post-inc form
doesn't matter. However, for pointer IVs we don't do that, so we have
to be careful about incrementing the IV in the wide type.
I'm fixing this by removing the IVCount variable (which was ExitCount
or ExitCount+1) and replacing it with a UsePostInc flag, and then
moving the actual limit adjustment to the individual cases (which
are: pointer IV where we add to the wide type, integer IV where we
add to the narrow type, and constant integer IV where we add to the
wide type).
Differential Revision: https://reviews.llvm.org/D63686
Together, these should fix a hang when building the textproc/htmldoc
port, due to an incorrect loop optimization.
PR: 237515
MFC after: 1 week
6.0.0 (branches/release_60 r324090).
This introduces retpoline support, with the -mretpoline flag. The
upstream initial commit message (r323155 by Chandler Carruth) contains
quite a bit of explanation. Quoting:
Introduce the "retpoline" x86 mitigation technique for variant #2 of
the speculative execution vulnerabilities disclosed today,
specifically identified by CVE-2017-5715, "Branch Target Injection",
and is one of the two halves to Spectre.
Summary:
First, we need to explain the core of the vulnerability. Note that
this is a very incomplete description, please see the Project Zero
blog post for details:
https://googleprojectzero.blogspot.com/2018/01/reading-privileged-memory-with-side.html
The basis for branch target injection is to direct speculative
execution of the processor to some "gadget" of executable code by
poisoning the prediction of indirect branches with the address of
that gadget. The gadget in turn contains an operation that provides a
side channel for reading data. Most commonly, this will look like a
load of secret data followed by a branch on the loaded value and then
a load of some predictable cache line. The attacker then uses timing
of the processors cache to determine which direction the branch took
*in the speculative execution*, and in turn what one bit of the
loaded value was. Due to the nature of these timing side channels and
the branch predictor on Intel processors, this allows an attacker to
leak data only accessible to a privileged domain (like the kernel)
back into an unprivileged domain.
The goal is simple: avoid generating code which contains an indirect
branch that could have its prediction poisoned by an attacker. In
many cases, the compiler can simply use directed conditional branches
and a small search tree. LLVM already has support for lowering
switches in this way and the first step of this patch is to disable
jump-table lowering of switches and introduce a pass to rewrite
explicit indirectbr sequences into a switch over integers.
However, there is no fully general alternative to indirect calls. We
introduce a new construct we call a "retpoline" to implement indirect
calls in a non-speculatable way. It can be thought of loosely as a
trampoline for indirect calls which uses the RET instruction on x86.
Further, we arrange for a specific call->ret sequence which ensures
the processor predicts the return to go to a controlled, known
location. The retpoline then "smashes" the return address pushed onto
the stack by the call with the desired target of the original
indirect call. The result is a predicted return to the next
instruction after a call (which can be used to trap speculative
execution within an infinite loop) and an actual indirect branch to
an arbitrary address.
On 64-bit x86 ABIs, this is especially easily done in the compiler by
using a guaranteed scratch register to pass the target into this
device. For 32-bit ABIs there isn't a guaranteed scratch register
and so several different retpoline variants are introduced to use a
scratch register if one is available in the calling convention and to
otherwise use direct stack push/pop sequences to pass the target
address.
This "retpoline" mitigation is fully described in the following blog
post: https://support.google.com/faqs/answer/7625886
We also support a target feature that disables emission of the
retpoline thunk by the compiler to allow for custom thunks if users
want them. These are particularly useful in environments like
kernels that routinely do hot-patching on boot and want to hot-patch
their thunk to different code sequences. They can write this custom
thunk and use `-mretpoline-external-thunk` *in addition* to
`-mretpoline`. In this case, on x86-64 thu thunk names must be:
```
__llvm_external_retpoline_r11
```
or on 32-bit:
```
__llvm_external_retpoline_eax
__llvm_external_retpoline_ecx
__llvm_external_retpoline_edx
__llvm_external_retpoline_push
```
And the target of the retpoline is passed in the named register, or in
the case of the `push` suffix on the top of the stack via a `pushl`
instruction.
There is one other important source of indirect branches in x86 ELF
binaries: the PLT. These patches also include support for LLD to
generate PLT entries that perform a retpoline-style indirection.
The only other indirect branches remaining that we are aware of are
from precompiled runtimes (such as crt0.o and similar). The ones we
have found are not really attackable, and so we have not focused on
them here, but eventually these runtimes should also be replicated for
retpoline-ed configurations for completeness.
For kernels or other freestanding or fully static executables, the
compiler switch `-mretpoline` is sufficient to fully mitigate this
particular attack. For dynamic executables, you must compile *all*
libraries with `-mretpoline` and additionally link the dynamic
executable and all shared libraries with LLD and pass `-z
retpolineplt` (or use similar functionality from some other linker).
We strongly recommend also using `-z now` as non-lazy binding allows
the retpoline-mitigated PLT to be substantially smaller.
When manually apply similar transformations to `-mretpoline` to the
Linux kernel we observed very small performance hits to applications
running typic al workloads, and relatively minor hits (approximately
2%) even for extremely syscall-heavy applications. This is largely
due to the small number of indirect branches that occur in
performance sensitive paths of the kernel.
When using these patches on statically linked applications,
especially C++ applications, you should expect to see a much more
dramatic performance hit. For microbenchmarks that are switch,
indirect-, or virtual-call heavy we have seen overheads ranging from
10% to 50%.
However, real-world workloads exhibit substantially lower performance
impact. Notably, techniques such as PGO and ThinLTO dramatically
reduce the impact of hot indirect calls (by speculatively promoting
them to direct calls) and allow optimized search trees to be used to
lower switches. If you need to deploy these techniques in C++
applications, we *strongly* recommend that you ensure all hot call
targets are statically linked (avoiding PLT indirection) and use both
PGO and ThinLTO. Well tuned servers using all of these techniques saw
5% - 10% overhead from the use of retpoline.
We will add detailed documentation covering these components in
subsequent patches, but wanted to make the core functionality
available as soon as possible. Happy for more code review, but we'd
really like to get these patches landed and backported ASAP for
obvious reasons. We're planning to backport this to both 6.0 and 5.0
release streams and get a 5.0 release with just this cherry picked
ASAP for distros and vendors.
This patch is the work of a number of people over the past month:
Eric, Reid, Rui, and myself. I'm mailing it out as a single commit
due to the time sensitive nature of landing this and the need to
backport it. Huge thanks to everyone who helped out here, and
everyone at Intel who helped out in discussions about how to craft
this. Also, credit goes to Paul Turner (at Google, but not an LLVM
contributor) for much of the underlying retpoline design.
Reviewers: echristo, rnk, ruiu, craig.topper, DavidKreitzer
Subscribers: sanjoy, emaste, mcrosier, mgorny, mehdi_amini, hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D41723
MFC after: 3 months
X-MFC-With: r327952
PR: 224669
[LV] Don't call recordVectorLoopValueForInductionCast for
newly-created IV from a trunc.
Summary:
This method is supposed to be called for IVs that have casts in their
use-def chains that are completely ignored after vectorization under
PSE. However, for truncates of such IVs the same InductionDescriptor
is used during creation/widening of both original IV based on PHINode
and new IV based on TruncInst.
This leads to unintended second call to
recordVectorLoopValueForInductionCast with a VectorLoopVal set to the
newly created IV for a trunc and causes an assert due to attempt to
store new information for already existing entry in the map. This is
wrong and should not be done.
Fixes PR35773.
Reviewers: dorit, Ayal, mssimpso
Reviewed By: dorit
Subscribers: RKSimon, dim, dcaballe, hsaito, llvm-commits, hiraditya
Differential Revision: https://reviews.llvm.org/D41913
This should fix "Vector value already set for part" assertions when
building the net/iodine and sysutils/daa2iso ports.
Reported by: jbeich
PR: 224867,224868
[SLP] Fix PR35777: Incorrect handling of aggregate values.
Summary:
Fixes the bug with incorrect handling of InsertValue|InsertElement
instrucions in SLP vectorizer. Currently, we may use incorrect
ExtractElement instructions as the operands of the original
InsertValue|InsertElement instructions.
Reviewers: mkuper, hfinkel, RKSimon, spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41767
This should fix "Invalid InsertValueInst operands!" errors when building
certain parts of editors/libreoffice.
Reported by: jbeich
PR: 225086
the upstream release_50 branch.
As of this version, lib/msun's trig test should also work correctly
again (see bug 220989 for more information).
PR: 220989
MFC after: 2 months
X-MFC-with: r321369
