The original commit was reverted because of a buildbot problem with LazyCallGraph::SCC handling (not related to the OptBisect handling).
Differential Revision: http://reviews.llvm.org/D19172
llvm-svn: 267231
Avoid quadratic complexity in unusually large basic blocks by limiting
the size of the ready lists.
Differential Revision: http://reviews.llvm.org/D19349
llvm-svn: 267189
This patch implements a optimization bisect feature, which will allow optimizations to be selectively disabled at compile time in order to track down test failures that are caused by incorrect optimizations.
The bisection is enabled using a new command line option (-opt-bisect-limit). Individual passes that may be skipped call the OptBisect object (via an LLVMContext) to see if they should be skipped based on the bisect limit. A finer level of control (disabling individual transformations) can be managed through an addition OptBisect method, but this is not yet used.
The skip checking in this implementation is based on (and replaces) the skipOptnoneFunction check. Where that check was being called, a new call has been inserted in its place which checks the bisect limit and the optnone attribute. A new function call has been added for module and SCC passes that behaves in a similar way.
Differential Revision: http://reviews.llvm.org/D19172
llvm-svn: 267022
Removed some unused headers, replaced some headers with forward class declarations.
Found using simple scripts like this one:
clear && ack --cpp -l '#include "llvm/ADT/IndexedMap.h"' | xargs grep -L 'IndexedMap[<]' | xargs grep -n --color=auto 'IndexedMap'
Patch by Eugene Kosov <claprix@yandex.ru>
Differential Revision: http://reviews.llvm.org/D19219
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 266595
Perform store clustering just like load clustering. This change add
StoreClusterMutation in machine-scheduler. To control StoreClusterMutation,
added enableClusterStores() in TargetInstrInfo.h. This is enabled only on
AArch64 for now.
This change also add support for unscaled stores which were not handled in
getMemOpBaseRegImmOfs().
llvm-svn: 266437
There is no problem with the code today, but the fix will avoid a crash
in test/CodeGen/AMDGPU/subreg-coalescer-undef-use.ll once the
DetectDeadLanes pass is added.
llvm-svn: 265351
The truncation was causing the sorting algorithm to behave oddly when comparing
positive and negative offsets. Fortunately, this doesn't currently happen in
practice and was exposed by a WIP. Thus, I can't test this change now, but the
follow on patch will.
llvm-svn: 263255
Take MachineInstr by reference instead of by pointer in SlotIndexes and
the SlotIndex wrappers in LiveIntervals. The MachineInstrs here are
never null, so this cleans up the API a bit. It also incidentally
removes a few implicit conversions from MachineInstrBundleIterator to
MachineInstr* (see PR26753).
At a couple of call sites it was convenient to convert to a range-based
for loop over MachineBasicBlock::instr_begin/instr_end, so I added
MachineBasicBlock::instrs.
llvm-svn: 262115
Note that this is disabled by default and still requires a patch to
handleMove() which is not upstreamed yet.
If the TrackLaneMasks policy/strategy is enabled the MachineScheduler
will build a schedule graph where definitions of independent
subregisters are no longer serialised.
Implementation comments:
- Without lane mask tracking a sub register def also counts as a use
(except for the first one with the read-undef flag set), with lane
mask tracking enabled this is no longer the case.
- Pressure Diffs where previously maintained per definition of a
vreg with the help of the SSA information contained in the
LiveIntervals. With lanemask tracking enabled we cannot do this
anymore and instead change the pressure diffs for all uses of the vreg
as it becomes live/dead. For this changed style to work correctly we
ignore uses of instructions that define the same register again: They
won't affect register pressure.
- With lanemask tracking we remove all read-undef flags from
sub register defs when building the graph and re-add them later when
all vreg lanes have become dead.
Differential Revision: http://reviews.llvm.org/D14969
llvm-svn: 258259
increase
Summary:
This patch adds a function called getRegPressureSetScore() to
TargetRegisterInfo. The MachineScheduler uses this when comparing
instruction that increase the register pressure of different sets
to determine which set is safer to increase.
This hook is useful for GPU targets where the number of registers in the
class is not the best metric for determing which presser set is safer to
increase.
Future work may include adding more parameters to this function, like
for example, the current pressure level of the set or the amount that
the pressure will be increased/decreased.
Reviewers: qcolombet, escha, arsenm, atrick, MatzeB
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D14806
llvm-svn: 255795
ScheduleDAGInstrs doesn't behave differently before or after register
allocation. It was only used in a method of MachineSchedulerBase which
behaved differently in MachineScheduler/PostMachineScheduler. Change
this to let MachineScheduler/PostMachineScheduler just pass in a
parameter to that function.
The order of the LiveIntervals* and bool RemoveKillFlags paramters have
been switched to make out-of-tree code fail instead of unintentionally
passing a value intended for the IsPostRA flag to the (previously
following and default initialized) RemoveKillFlags.
Differential Revision: http://reviews.llvm.org/D14245
llvm-svn: 251883
This was a layering violation in ScheduleDAGInstrs (and
MachineSchedulerBase) they both shouldn't know directly whether they are
used by the PostMachineScheduler or the MachineScheduler.
llvm-svn: 251608
r248010 changed the -debug output to use short ids, but did not
similarly modify the graph printer. Change to be consistent, for ease of
cross-reference.
llvm-svn: 251465
with the new pass manager, and no longer relying on analysis groups.
This builds essentially a ground-up new AA infrastructure stack for
LLVM. The core ideas are the same that are used throughout the new pass
manager: type erased polymorphism and direct composition. The design is
as follows:
- FunctionAAResults is a type-erasing alias analysis results aggregation
interface to walk a single query across a range of results from
different alias analyses. Currently this is function-specific as we
always assume that aliasing queries are *within* a function.
- AAResultBase is a CRTP utility providing stub implementations of
various parts of the alias analysis result concept, notably in several
cases in terms of other more general parts of the interface. This can
be used to implement only a narrow part of the interface rather than
the entire interface. This isn't really ideal, this logic should be
hoisted into FunctionAAResults as currently it will cause
a significant amount of redundant work, but it faithfully models the
behavior of the prior infrastructure.
- All the alias analysis passes are ported to be wrapper passes for the
legacy PM and new-style analysis passes for the new PM with a shared
result object. In some cases (most notably CFL), this is an extremely
naive approach that we should revisit when we can specialize for the
new pass manager.
- BasicAA has been restructured to reflect that it is much more
fundamentally a function analysis because it uses dominator trees and
loop info that need to be constructed for each function.
All of the references to getting alias analysis results have been
updated to use the new aggregation interface. All the preservation and
other pass management code has been updated accordingly.
The way the FunctionAAResultsWrapperPass works is to detect the
available alias analyses when run, and add them to the results object.
This means that we should be able to continue to respect when various
passes are added to the pipeline, for example adding CFL or adding TBAA
passes should just cause their results to be available and to get folded
into this. The exception to this rule is BasicAA which really needs to
be a function pass due to using dominator trees and loop info. As
a consequence, the FunctionAAResultsWrapperPass directly depends on
BasicAA and always includes it in the aggregation.
This has significant implications for preserving analyses. Generally,
most passes shouldn't bother preserving FunctionAAResultsWrapperPass
because rebuilding the results just updates the set of known AA passes.
The exception to this rule are LoopPass instances which need to preserve
all the function analyses that the loop pass manager will end up
needing. This means preserving both BasicAAWrapperPass and the
aggregating FunctionAAResultsWrapperPass.
Now, when preserving an alias analysis, you do so by directly preserving
that analysis. This is only necessary for non-immutable-pass-provided
alias analyses though, and there are only three of interest: BasicAA,
GlobalsAA (formerly GlobalsModRef), and SCEVAA. Usually BasicAA is
preserved when needed because it (like DominatorTree and LoopInfo) is
marked as a CFG-only pass. I've expanded GlobalsAA into the preserved
set everywhere we previously were preserving all of AliasAnalysis, and
I've added SCEVAA in the intersection of that with where we preserve
SCEV itself.
One significant challenge to all of this is that the CGSCC passes were
actually using the alias analysis implementations by taking advantage of
a pretty amazing set of loop holes in the old pass manager's analysis
management code which allowed analysis groups to slide through in many
cases. Moving away from analysis groups makes this problem much more
obvious. To fix it, I've leveraged the flexibility the design of the new
PM components provides to just directly construct the relevant alias
analyses for the relevant functions in the IPO passes that need them.
This is a bit hacky, but should go away with the new pass manager, and
is already in many ways cleaner than the prior state.
Another significant challenge is that various facilities of the old
alias analysis infrastructure just don't fit any more. The most
significant of these is the alias analysis 'counter' pass. That pass
relied on the ability to snoop on AA queries at different points in the
analysis group chain. Instead, I'm planning to build printing
functionality directly into the aggregation layer. I've not included
that in this patch merely to keep it smaller.
Note that all of this needs a nearly complete rewrite of the AA
documentation. I'm planning to do that, but I'd like to make sure the
new design settles, and to flesh out a bit more of what it looks like in
the new pass manager first.
Differential Revision: http://reviews.llvm.org/D12080
llvm-svn: 247167
Before creating a schedule edge to encourage MacroOpFusion check that:
- The predecessor actually writes a register that the branch reads.
- The predecessor has no successors in the ScheduleDAG so we can
schedule it in front of the branch.
This avoids skewing the scheduling heuristic in cases where macroop
fusion cannot happen.
Differential Revision: http://reviews.llvm.org/D10745
llvm-svn: 242723
The patch is generated using this command:
tools/clang/tools/extra/clang-tidy/tool/run-clang-tidy.py -fix \
-checks=-*,llvm-namespace-comment -header-filter='llvm/.*|clang/.*' \
llvm/lib/
Thanks to Eugene Kosov for the original patch!
llvm-svn: 240137
Summary:
TargetInstrInfo::getLdStBaseRegImmOfs to
TargetInstrInfo::getMemOpBaseRegImmOfs and implement for x86. The
implementation only handles a few easy cases now and will be made more
sophisticated in the future.
This is NFCI: the only user of `getLdStBaseRegImmOfs` (now
`getmemOpBaseRegImmOfs`) is `LoadClusterMotion` and `LoadClusterMotion`
is disabled for x86.
Reviewers: reames, ab, MatzeB, atrick
Reviewed By: MatzeB, atrick
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D10199
llvm-svn: 239741
r213101 changed the behaviour of this method to not only affect the
PostMachineScheduler scheduler but also the PostRAScheduler scheduler,
renaming should make this fact clear. Also document that the preferred
way is to specify this in the scheduling model instead of overriding
this method.
Differential Revision: http://reviews.llvm.org/D10427
llvm-svn: 239659
"Fix the MachineScheduler's logic for updating ready times for in-order.
Now the scheduler updates a node's ready time as soon as it is
scheduled, before releasing dependent nodes."
This fix was only made in one variant of the ScheduleDAGMI driver.
Francois de Ferriere reported the issue in the other bit of code where
it was also needed.
I never got around to coming up with a test case, but it's an
obvious fix that shouldn't be delayed any longer.
I'll try to refactor this code a little better.
I did verify performance on a wide variety of targets and saw no
negative impact with this fix.
llvm-svn: 233366
that control, individually, all of the disparate things it was
controlling.
At the same time move a FIXME in the Hexagon port to a new
subtarget function that will enable a user of the machine
scheduler to avoid using the source scheduler for pre-RA-scheduling.
The FIXME would have this removed, but involves either testcase
changes or adding -pre-RA-sched=source to a few testcases.
llvm-svn: 231980
options.
This commit changes the command line arguments (PassInfo::PassArgument) of two
passes, MachineFunctionPrinter and MachineScheduler, to avoid collisions with
command line options that have the same argument strings.
This bug manifests when the PassList construct (defined in opt.cpp) is used
in a tool that links with codegen passes. To reproduce the bug, paste the
following lines into llc.cpp and run llc.
#include "llvm/IR/LegacyPassNameParser.h"
static llvm:🆑:list<const llvm::PassInfo*, bool, llvm::PassNameParser>
PassList(llvm:🆑:desc("Optimizations available:"));
rdar://problem/19212448
llvm-svn: 224186
This reapplies r224118 with a fix for test 'misched-code-difference-with-debug.ll'.
That test was failing on some buildbots because it was x86 specific but it was
missing a target triple.
Added an explicit triple to test misched-code-difference-with-debug.ll.
llvm-svn: 224126
This patch fixes the issue reported as PR21807. There was a minor difference
in the generated code depending on the -g flag.
The cause was that with -g the machine scheduler used a different
scheduling strategy. This decision was based on the number of instructions
in a schedule region and included debug instructions in that count.
This patch fixes the issue in MISched and provides a test.
Patch by Russell Gallop!
llvm-svn: 224118
Fixes a logic error in the MachineScheduler found by Steve Montgomery (and
confirmed by Andy). This has gone unfixed for months because the fix has been
found to introduce some small performance regressions. However, Andy has
recommended that, at this point, we fix this to avoid further dependence on the
incorrect behavior (and then follow-up separately on any regressions), and I
agree.
Fixes PR18883.
llvm-svn: 219512
shorter/easier and have the DAG use that to do the same lookup. This
can be used in the future for TargetMachine based caching lookups from
the MachineFunction easily.
Update the MIPS subtarget switching machinery to update this pointer
at the same time it runs.
llvm-svn: 214838
This reverts commit r212109, which reverted r212088.
However, disable the assert as it's not necessary for correctness. There are
several corner cases that the assert needed to handle better for in-order
scheduling, but none of them are incorrect scheduler behavior. The assert is
mainly there to collect good unit tests like this and ensure that the
target-independent scheduler is working as expected with the various machine
models.
llvm-svn: 212187
This macro is sometimes defined manually but isn't (and doesn't need to be) in
llvm-config.h so shouldn't appear in the headers, likewise NDEBUG.
Instead switch them over to LLVM_DUMP_METHOD on the definitions.
llvm-svn: 212130
Fixe for Bug 20057 - Assertion failied in llvm::SUnit* llvm::SchedBoundary::pickOnlyChoice(): Assertion `i <= (HazardRec->getMaxLookAhead() + MaxObservedStall) && "permanent hazard"'
Thanks to Chad for the test case.
llvm-svn: 211865
string_ostream is a safe and efficient string builder that combines opaque
stack storage with a built-in ostream interface.
small_string_ostream<bytes> additionally permits an explicit stack storage size
other than the default 128 bytes to be provided. Beyond that, storage is
transferred to the heap.
This convenient class can be used in most places an
std::string+raw_string_ostream pair or SmallString<>+raw_svector_ostream pair
would previously have been used, in order to guarantee consistent access
without byte truncation.
The patch also converts much of LLVM to use the new facility. These changes
include several probable bug fixes for truncated output, a programming error
that's no longer possible with the new interface.
llvm-svn: 211749
Now the scheduler updates a node's ready time as soon as it is
scheduled, before releasing dependent nodes. There was a reason I
didn't do this initially but it no longer applies.
A53 is in-order and was running into an issue where nodes where added
to the readyQ too early. That's now fixed.
This also makes it easier for custom scheduling strategies to build
heuristics based on the actual cycles that the node was scheduled at.
The only impact on OOO (sandybridge/cyclone) is that ready times will
be slightly more accurate. I didn't measure any significant regressions.
llvm-svn: 210390
As requested by AArch64 subtargets.
Note that this will have no effect until the
AArch64 target actually enables the pass like this:
substitutePass(&PostRASchedulerID, &PostMachineSchedulerID);
As soon as armv7 switches over, PostMachineScheduler will become the
default postRA scheduler, so this won't be necessary any more.
Targets using the old postRA schedule would then do:
substitutePass(&PostMachineSchedulerID, &PostRASchedulerID);
llvm-svn: 210167
These were not exposed previously because I didn't want out-of-tree
targets to be too dependent on their internals. They can be reused for
a very wide variety of processors with casual scheduling needs without
exposing the classes by instead using hooks defined in
MachineSchedPolicy (we can add more if needed). When targets are more
aggressively tuned or want to provide custom heuristics, they can
define their own MachineSchedStrategy. I tend to think this is better
once you start customizing heuristics because you can copy over only
what you need. I don't think that layering heuristics generally works
well.
However, Arch64 targets now want to reuse the Generic scheduling logic
but also provide extensions. I don't see much harm in exposing the
Generic scheduling classes with a major caveat: these scheduling
strategies may change in the future without validating performance on
less mainstream processors. If you want to be immune from changes,
just define your own MachineSchedStrategy.
llvm-svn: 210166
define below all header includes in the lib/CodeGen/... tree. While the
current modules implementation doesn't check for this kind of ODR
violation yet, it is likely to grow support for it in the future. It
also removes one layer of macro pollution across all the included
headers.
Other sub-trees will follow.
llvm-svn: 206837
This compiles with no changes to clang/lld/lldb with MSVC and includes
overloads to various functions which are used by those projects and llvm
which have OwningPtr's as parameters. This should allow out of tree
projects some time to move. There are also no changes to libs/Target,
which should help out of tree targets have time to move, if necessary.
llvm-svn: 203083
PostGenericScheduler uses either the new machine model or the hazard
checker for top-down scheduling. Most of the infrastructure for PreRA
machine scheduling is reused.
With a some tuning, this should allow MachineScheduler to be default
for all ARM targets, including cortex-A9, using the new machine
model. Likewise, with additional tuning, it should be able to replace
PostRAScheduler for all targets.
The PostMachineScheduler pass does not currently run the
AntiDepBreaker. There is less need for it on targets that are already
running preRA MachineScheduler. I want to prove it's necessary before
committing to the maintenance burden.
The PostMachineScheduler also currently removes kill flags and adds
them all back later. This is a bit ridiculous. I'd prefer passes to
directly use a liveness utility than rely on flags.
A test case that enables this scheduler will be included in a
subsequent checkin that updates the A9 model.
llvm-svn: 198122
These helper classes take care of the book-keeping the drives the
GenericScheduler heuristics. It is likely that developers writing
target-specific schedulers that work similarly to GenericScheduler
will want to use these helpers too. The immediate goal is to develop a
GenericPostScheduler that can run in place of the old PostRAScheduler,
but will use the new machine model.
No functionality change intended.
llvm-svn: 196643
Not only does it trigger -Wparentheses, I think the assert actually
relies on incorrect operator precedence.
Also, the grammar as questionable, but I might not know enough about the
problem at hand.
llvm-svn: 196567
This allows a target to use MI-Sched as an in-order scheduler that
will model strict resource conflicts without defining a processor
itinerary. Instead, the target can now use the new per-operand machine
model and define in-order resources with BufferSize=0. For example,
this would allow restricting the type of operations that can be formed
into a dispatch group. (Normally NumMicroOps is sufficient to enforce
dispatch groups).
If the intent is to model latency in in-order pipeline, as opposed to
resource conflicts, then a resource with BufferSize=1 should be
defined instead.
This feature is only casually tested as there are no in-tree targets
using it yet. However, Hal will be experimenting with POWER7.
llvm-svn: 196517
This patch removes most of the trivial cases of weak vtables by pinning them to
a single object file. The memory leaks in this version have been fixed. Thanks
Alexey for pointing them out.
Differential Revision: http://llvm-reviews.chandlerc.com/D2068
Reviewed by Andy
llvm-svn: 195064
This change is incorrect. If you delete virtual destructor of both a base class
and a subclass, then the following code:
Base *foo = new Child();
delete foo;
will not cause the destructor for members of Child class. As a result, I observe
plently of memory leaks. Notable examples I investigated are:
ObjectBuffer and ObjectBufferStream, AttributeImpl and StringSAttributeImpl.
llvm-svn: 194997
This patch removes most of the trivial cases of weak vtables by pinning them to
a single object file.
Differential Revision: http://llvm-reviews.chandlerc.com/D2068
Reviewed by Andy
llvm-svn: 194865
The global registry is used to allow command line override of the
scheduler selection, but does not work well as the normal selection
API. For example, the same LLVM process should be able to target
multiple targets or subtargets.
llvm-svn: 191071
This was an experimental scheduler a year ago. It's now used by
several subtargets, both in-order and out-of-order, and it
is about to be enabled by default for x86 and armv7. It will be the
new GenericScheduler for subtargets that don't provide their own
SchedulingStrategy.
llvm-svn: 191051
Arnold's idea.
I generally try to avoid stateful heuristics because it can make
debugging harder. However, we need a way to prevent the latency
priority from dominating, and it somewhat makes sense to schedule
aggressively for latency only within an issue group.
Swift in particular likes this, and it doesn't hurt anyone else:
| Benchmarks/MiBench/consumer-lame | 10.39% |
| Benchmarks/Misc/himenobmtxpa | 9.63% |
llvm-svn: 190360
Allow subtargets to customize the generic scheduling strategy.
This is convenient for targets that don't need to add new heuristics
by specializing the strategy.
llvm-svn: 190176
Fast register pressure tracking currently only takes effect during
bottom up scheduling. Forcing this is a bit faster and simpler for
targets that don't have many scheduling constraints and don't need
top-down scheduling.
llvm-svn: 190014
If the instruction window is < NumRegs/2, pressure tracking is not
likely to be effective. The scheduler has to process a very large
number of tiny blocks. We want this to be fast.
llvm-svn: 189991
Register pressure tracking is half the complexity of the
scheduler. It's useful to be able to turn it off for compile time and
performance comparisons.
llvm-svn: 189987
There was one case that we could hit a DebugValue where I didn't think
to check. DebugValues are evil. No checkinable test case, sorry. It's
an obvious fix.
llvm-svn: 189717
Created SUPressureDiffs array to hold the per node PDiff computed during DAG building.
Added a getUpwardPressureDelta API that will soon replace the old
one. Compute PressureDelta here from the precomputed PressureDiffs.
Updating for liveness will come next.
llvm-svn: 189640
Estimate the cyclic critical path within a single block loop. If the
acyclic critical path is longer, then the loop will exhaust OOO
resources after some number of iterations. If lag between the acyclic
critical path and cyclic critical path is longer the the time it takes
to issue those loop iterations, then aggressively schedule for
latency.
llvm-svn: 189120
When registers must be live throughout the scheduling region, increase
the limit for the register class. Once we exceed the original limit,
they will be spilled, and there's no point further reducing pressure.
This isn't a perfect heuristics but avoids a situation where the
scheduler could become trapped by trying to achieve the impossible.
llvm-svn: 187436
Replace the ill-defined MinLatency and ILPWindow properties with
with straightforward buffer sizes:
MCSchedMode::MicroOpBufferSize
MCProcResourceDesc::BufferSize
These can be used to more precisely model instruction execution if desired.
Disabled some misched tests temporarily. They'll be reenabled in a few commits.
llvm-svn: 184032
"Counts" refer to scaled resource counts within a region. CurrMOps is
simply the number of micro-ops to be issue in the current cycle.
llvm-svn: 184031
Heuristics compare the critical path in the scheduled code, called
ExpectedLatency, with the latency of instructions remaining to be
scheduled. There are two ways to look at remaining latency:
(1) Dependent latency includes the latency between unscheduled and
scheduled instructions.
(2) Independent latency is simply the height (bottom-up) or depth
(top-down) of instructions currently in the ready Q.
llvm-svn: 184029
Andrew Kaylor