This patch adds a new flag named -bottleneck-analysis to print out information
about throughput bottlenecks.
MCA knows how to identify and classify dynamic dispatch stalls. However, it
doesn't know how to analyze and highlight kernel bottlenecks. The goal of this
patch is to teach MCA how to correlate increases in backend pressure to backend
stalls (and therefore, the loss of throughput).
From a Scheduler point of view, backend pressure is a function of the scheduler
buffer usage (i.e. how the number of uOps in the scheduler buffers changes over
time). Backend pressure increases (or decreases) when there is a mismatch
between the number of opcodes dispatched, and the number of opcodes issued in
the same cycle. Since buffer resources are limited, continuous increases in
backend pressure would eventually leads to dispatch stalls. So, there is a
strong correlation between dispatch stalls, and how backpressure changed over
time.
This patch teaches how to identify situations where backend pressure increases
due to:
- unavailable pipeline resources.
- data dependencies.
Data dependencies may delay execution of instructions and therefore increase the
time that uOps have to spend in the scheduler buffers. That often translates to
an increase in backend pressure which may eventually lead to a bottleneck.
Contention on pipeline resources may also delay execution of instructions, and
lead to a temporary increase in backend pressure.
Internally, the Scheduler classifies instructions based on whether register /
memory operands are available or not.
An instruction is marked as "ready to execute" only if data dependencies are
fully resolved.
Every cycle, the Scheduler attempts to execute all instructions that are ready
to execute. If an instruction cannot execute because of unavailable pipeline
resources, then the Scheduler internally updates a BusyResourceUnits mask with
the ID of each unavailable resource.
ExecuteStage is responsible for tracking changes in backend pressure. If backend
pressure increases during a cycle because of contention on pipeline resources,
then ExecuteStage sends a "backend pressure" event to the listeners.
That event would contain information about instructions delayed by resource
pressure, as well as the BusyResourceUnits mask.
Note that ExecuteStage also knows how to identify situations where backpressure
increased because of delays introduced by data dependencies.
The SummaryView observes "backend pressure" events and prints out a "bottleneck
report".
Example of bottleneck report:
```
Cycles with backend pressure increase [ 99.89% ]
Throughput Bottlenecks:
Resource Pressure [ 0.00% ]
Data Dependencies: [ 99.89% ]
- Register Dependencies [ 0.00% ]
- Memory Dependencies [ 99.89% ]
```
A bottleneck report is printed out only if increases in backend pressure
eventually caused backend stalls.
About the time complexity:
Time complexity is linear in the number of instructions in the
Scheduler::PendingSet.
The average slowdown tends to be in the range of ~5-6%.
For memory intensive kernels, the slowdown can be significant if flag
-noalias=false is specified. In the worst case scenario I have observed a
slowdown of ~30% when flag -noalias=false was specified.
We can definitely recover part of that slowdown if we optimize class LSUnit (by
doing extra bookkeeping to speedup queries). For now, this new analysis is
disabled by default, and it can be enabled via flag -bottleneck-analysis. Users
of MCA as a library can enable the generation of pressure events through the
constructor of ExecuteStage.
This patch partially addresses https://bugs.llvm.org/show_bug.cgi?id=37494
Differential Revision: https://reviews.llvm.org/D58728
llvm-svn: 355308
X86TargetLowering::EmitLoweredSelect presently detects sequences of CMOV pseudo
instructions without accounting for debug intrinsics. This leads to different
codegen with and without option -g, if a DBG_VALUE instruction lands in the
middle of several lowered selects.
Work around this by skipping over debug instructions when looking for CMOV
sequences, and sinking those debug insts into the EmitLoweredSelect sunk block.
This might slightly shift where variables appear in the instruction sequence,
but won't re-order assignments.
Differential Revision: https://reviews.llvm.org/D58672
llvm-svn: 355307
The isScaledConstantInRange function takes upper and lower bounds which are
checked after dividing by the scale, so the bounds checks for half, single and
double precision should all be the same. Previously, we had wrong bounds checks
for half precision, so selected an immediate the instructions can't actually
represent.
Differential revision: https://reviews.llvm.org/D58822
llvm-svn: 355305
1) GCC complains that KnownValid is set but not used.
2) In ARMInstructionSelector::selectGlobal() the code is mixing "enumeral
and non-enumeral type in conditional expression". Solve this by casting
to unsigned which is the final type anyway.
Differential Revision: https://reviews.llvm.org/D58834
llvm-svn: 355304
Summary:
Before when we implemented the first EH proposal, 'catch <tag>'
instruction may not catch an exception so there were multiple EH pads an
exception can unwind to. That means a BB could have multiple EH pad
successors.
Now after we switched to the new proposal, every 'catch' instruction
catches an exception, and there is only one catchpad per catchswitch, so
we at most have one EH pad successor, making `ThrowUnwindDest` map in
`WasmEHInfo` unnecessary.
Keeping `ThrowUnwindDest` map in `WasmEHInfo` has its own problems,
because other optimization passes can split a BB that contains possibly
throwing calls (previously invokes), and we have to update the map every
time that happens, which is not easy for common CodeGen passes.
This also correctly updates successor info in LateEHPrepare when we add
a rethrow instruction.
Reviewers: dschuff
Subscribers: sbc100, jgravelle-google, sunfish, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D58486
llvm-svn: 355296
There are no tests for this case, and I'm not sure how it could ever work,
so I'm just removing this option from the matcher. This should fix PR40940:
https://bugs.llvm.org/show_bug.cgi?id=40940
llvm-svn: 355292
InputIsKnownDead check is shared by all operands. Compute it once.
For non-integer instructions, use Visited.insert(I).second to replace a
find() and an insert().
llvm-svn: 355290
Use sysctl() to implement getMainExecutable() on NetBSD, rather than
trying to guess the correct path from argv[0]. This is one
of the fixes to recent clang-check-mac-libcxx-fixed-compilation-db.cpp
test failure on NetBSD.
This has been historically done on both FreeBSD and NetBSD in r303015,
and reverted in r303285 due to buggy implementation on FreeBSD.
However, FWIK the NetBSD implementation does not suffer from the same
bugs and is more reliable than playing with argv[0].
Differential Revision: https://reviews.llvm.org/D56975
llvm-svn: 355283
We were using VPBLENDW for v2i64 and VBLENDPD for v4i64. VPBLENDD has better throughput than VPBLENDW on some CPUs so it makes sense to use it when possible. VBLENDPD will probably become VBLENDD during execution domain fixing, but we might as well use integer in isel while we can.
This should work around some issues with the domain fixing pass prefering PBLENDW when we start with PBLENDW. There may still be some v8i16 cases that could use PBLENDD.
llvm-svn: 355281
This build target is currently unused, but after r355144 the sync script
started complaining about cfi.cpp not being listed, and this makes the
script happy again.
llvm-svn: 355275
Summary:
This prevents crashes in instruction selection when these operations
are used. The tests check that the scalar version of the instruction
is used where applicable, although some expansions do not use the
scalar version.
Reviewers: aheejin
Subscribers: dschuff, sbc100, jgravelle-google, hiraditya, sunfish, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D58859
llvm-svn: 355261
Summary:
This extends the variety of pattern that can generate a SHLD instead of using two shifts.
This fixes a regression that would be introduced by D57367 or D33587
Reviewers: RKSimon, craig.topper
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D57389
llvm-svn: 355260
In some cases, MaxBECount can be less precise than ExactBECount for AND
and OR (the AND case was PR26207). In the OR test case, both ExactBECounts are
undef, but MaxBECount are different, so we hit the assertion below. This
patch uses the same solution the AND case already uses.
Assertion failed:
((isa<SCEVCouldNotCompute>(ExactNotTaken) || !isa<SCEVCouldNotCompute>(MaxNotTaken))
&& "Exact is not allowed to be less precise than Max"), function ExitLimit
This patch also consolidates test cases for both AND and OR in a single
test case.
Fixes https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=13245
Reviewers: sanjoy, efriedma, mkazantsev
Reviewed By: sanjoy
Differential Revision: https://reviews.llvm.org/D58853
llvm-svn: 355259
The value stored in SCEVConstant is of type ConstantInt*, which can
never be UndefValue. So we should never hit that code.
Reviewers: mkazantsev, sanjoy
Reviewed By: sanjoy
Differential Revision: https://reviews.llvm.org/D58851
llvm-svn: 355257
Summary:
Add an SMLoc to CodeInit that records the source line it originated from.
This allows tablegen to point precisely at portions of code when reporting
errors within the CodeInit. For example, in the upcoming GlobalISel
combiner, it can report undefined expansions and point at the instance of
the expansion. This is achieved using something like:
SMLoc::getFromPointer(SMLoc::getPointer() +
(StringRef - CodeInit::getValue()))
The location is lost when producing a CodeInit by string concatenation so
a fallback SMLoc is required (e.g. the Record::getLoc()) but that's pretty
rare for CodeInits.
There's a reasonable case for extending tracking of a couple other Init
objects, for example StringInit's are often parsed and it would be good to
point inside the string when reporting errors about that. However, location
tracking also harms de-duplication. This is fine for CodeInit where there's
only a few hundred of them (~160 for X86) and it may be worth it for
StringInit (~86k up to ~1.9M for roughly 15MB increase for X86).
However the origin tracking would be a _terrible_ idea for IntInit, BitInit,
and UnsetInit. I haven't measured either of those three but BitInit would
most likely be on the order of increasing the current 2 BitInit values up
to billions.
Reviewers: volkan, aditya_nandakumar, bogner, paquette, aemerson
Reviewed By: paquette
Subscribers: javed.absar, kristof.beyls, dexonsmith, llvm-commits, kristina
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D58141
llvm-svn: 355245
Kamil Rytarowski