Currently prof metadata with branch counts is added only for BranchInst and SwitchInst, but not for IndirectBrInst. As a result, BPI/BFI make incorrect inferences for indirect branches, which can be very hot.
This diff adds metadata for IndirectBrInst, in addition to BranchInst and SwitchInst.
Reviewed By: wmi, wenlei
Differential Revision: https://reviews.llvm.org/D99550
Use profiled call edges to augment the top-down order. There are cases that the top-down order computed based on the static call graph doesn't reflect real execution order. For example:
1. Incomplete static call graph due to unknown indirect call targets. Adjusting the order by considering indirect call edges from the profile can enable the inlining of indirect call targets by allowing the caller processed before them.
2. Mutual call edges in an SCC. The static processing order computed for an SCC may not reflect the call contexts in the context-sensitive profile, thus may cause potential inlining to be overlooked. The function order in one SCC is being adjusted to a top-down order based on the profile to favor more inlining.
3. Transitive indirect call edges due to inlining. When a callee function is inlined into into a caller function in LTO prelink, every call edge originated from the callee will be transferred to the caller. If any of the transferred edges is indirect, the original profiled indirect edge, even if considered, would not enforce a top-down order from the caller to the potential indirect call target in LTO postlink since the inlined callee is gone from the static call graph.
4. #3 can happen even for direct call targets, due to functions defined in header files. Header functions, when included into source files, are defined multiple times but only one definition survives due to ODR. Therefore, the LTO prelink inlining done on those dropped definitions can be useless based on a local file scope. More importantly, the inlinee, once fully inlined to a to-be-dropped inliner, will have no profile to consume when its outlined version is compiled. This can lead to a profile-less prelink compilation for the outlined version of the inlinee function which may be called from external modules. while this isn't easy to fix, we rely on the postlink AutoFDO pipeline to optimize the inlinee. Since the survived copy of the inliner (defined in headers) can be inlined in its local scope in prelink, it may not exist in the merged IR in postlink, and we'll need the profiled call edges to enforce a top-down order for the rest of the functions.
Considering those cases, a profiled call graph completely independent of the static call graph is constructed based on profile data, where function objects are not even needed to handle case #3 and case 4.
I'm seeing an average 0.4% perf win out of SPEC2017. For certain benchmark such as Xalanbmk and GCC, the win is bigger, above 2%.
The change is an enhancement to https://reviews.llvm.org/D95988.
Reviewed By: wmi, wenlei
Differential Revision: https://reviews.llvm.org/D99351
Use SmallVector instead of SmallSet to track the context profiles mapped. Doing this
can help avoid non-determinism caused by iterating over unordered containers.
This bug was found with reverse iteration turning on,
--extra-llvm-cmake-variables="-DLLVM_REVERSE_ITERATION=ON".
Failing LLVM test profile-context-tracker-debug.ll .
Reviewed By: MaskRay, wenlei
Differential Revision: https://reviews.llvm.org/D99547
Lookup tables generate non PIC-friendly code, which requires dynamic relocation as described in:
https://bugs.llvm.org/show_bug.cgi?id=45244
This patch adds a new pass that converts lookup tables to relative lookup tables to make them PIC-friendly.
Differential Revision: https://reviews.llvm.org/D94355
This change sets up a framework in llvm-profgen to estimate inline decision and adjust context-sensitive profile based on that. We call it a global pre-inliner in llvm-profgen.
It will serve two purposes:
1) Since context profile for not inlined context will be merged into base profile, if we estimate a context will not be inlined, we can merge the context profile in the output to save profile size.
2) For thinLTO, when a context involving functions from different modules is not inined, we can't merge functions profiles across modules, leading to suboptimal post-inline count quality. By estimating some inline decisions, we would be able to adjust/merge context profiles beforehand as a mitigation.
Compiler inline heuristic uses inline cost which is not available in llvm-profgen. But since inline cost is closely related to size, we could get an estimate through function size from debug info. Because the size we have in llvm-profgen is the final size, it could also be more accurate than the inline cost estimation in the compiler.
This change only has the framework, with a few TODOs left for follow up patches for a complete implementation:
1) We need to retrieve size for funciton//inlinee from debug info for inlining estimation. Currently we use number of samples in a profile as place holder for size estimation.
2) Currently the thresholds are using the values used by sample loader inliner. But they need to be tuned since the size here is fully optimized machine code size, instead of inline cost based on not yet fully optimized IR.
Differential Revision: https://reviews.llvm.org/D99146
Using $ breaks demangling of the symbols. For example,
$ c++filt _Z3foov\$123
_Z3foov$123
This causes problems for developers who would like to see nice stack traces
etc., but also for automatic crash tracking systems which try to organize
crashes based on the stack traces.
Instead, use the period as suffix separator, since Itanium demanglers normally
ignore such suffixes:
$ c++filt _Z3foov.123
foo() [clone .123]
This is already done in some places; try to do it everywhere.
Differential revision: https://reviews.llvm.org/D97484
During context promotion, intermediate nodes that are on a call path but do not come with a profile can be promoted together with their parent nodes. Do not print sample context string for such nodes since they do not have profile.
Reviewed By: wenlei
Differential Revision: https://reviews.llvm.org/D99441
When prioritize call site to consider for inlining in sample loader, use number of samples as a first tier breaker before using name/guid comparison. This would favor smaller functions when hotness is the same (from the same block). We could try to retrieve accurate function size if this turns out to be more important.
Differential Revision: https://reviews.llvm.org/D99370
In DeadArgumentElimination pass, if a function's argument is never used, corresponding caller's parameter can be changed to undef. If the param/arg has attribute noundef or other related attributes, LLVM LangRef(https://llvm.org/docs/LangRef.html#parameter-attributes) says its behavior is undefined. SimplifyCFG(D97244) takes advantage of this behavior and does bad transformation on valid code.
To avoid this undefined behavior when change caller's parameter to undef, this patch removes noundef attribute and other attributes imply noundef on param/arg.
Differential Revision: https://reviews.llvm.org/D98899
Lookup tables generate non PIC-friendly code, which requires dynamic relocation as described in:
https://bugs.llvm.org/show_bug.cgi?id=45244
This patch adds a new pass that converts lookup tables to relative lookup tables to make them PIC-friendly.
Differential Revision: https://reviews.llvm.org/D94355
Lookup tables generate non PIC-friendly code, which requires dynamic relocation as described in:
https://bugs.llvm.org/show_bug.cgi?id=45244
This patch adds a new pass that converts lookup tables to relative lookup tables to make them PIC-friendly.
Differential Revision: https://reviews.llvm.org/D94355
This attribute represents the minimum and maximum values vscale can
take. For now this attribute is not hooked up to anything during
codegen, this will be added in the future when such codegen is
considered stable.
Additionally hook up the -msve-vector-bits=<x> clang option to emit this
attribute.
Differential Revision: https://reviews.llvm.org/D98030
This patch is plumbing to support work towards the goal outlined in the recent llvm-dev post "[llvm-dev] RFC: Decomposing deref(N) into deref(N) + nofree".
The point of this change is purely to simplify iteration on other pieces on way to making the switch. Rebuilding with a change to Value.h is slow and painful, so I want to get the API change landed. Once that's done, I plan to more closely audit each caller, add the inference rules in their own patch, then post a patch with the langref changes and test diffs. The value of the command line flag is that we can exercise the inference logic in standalone patches without needing the whole switch ready to go just yet.
Differential Revision: https://reviews.llvm.org/D98908
value profile annotated after inlining.
In https://reviews.llvm.org/D96806 and https://reviews.llvm.org/D97350, we
use the magic number -1 in the value profile to avoid repeated indirect call
promotion to the same target for an indirect call. Function updateIDTMetaData
is used to mark an target as being promoted in the value profile with the
magic number. updateIDTMetaData is also used to update the value profile
when an indirect call is inlined and new inline instance profile should be
applied. For the second case, currently updateIDTMetaData mixes up the
existing value profile of the indirect call with the new profile, leading
to the problematic senario that a target count is larger than the total count
in the value profile.
The patch fixes the problem. When updateIDTMetaData is used to update the
value profile after inlining, all the values in the existing value profile
will be dropped except the values with the magic number counts.
Differential Revision: https://reviews.llvm.org/D98835
Not doing it here can lead to subtle bugs - the analysis results are
associated by the Function object's address. Nothing stops the memory
allocator from allocating new functions at the same address.
For ThinLTO's prelink compilation, we need to put external inline candidates into an import list attached to function's entry count metadata. This enables ThinLink to treat such cross module callee as hot in summary index, and later helps postlink to import them for profile guided cross module inlining.
For AutoFDO, the import list is retrieved by traversing the nested inlinee functions. For CSSPGO, since profile is flatterned, a few things need to happen for it to work:
- When loading input profile in extended binary format, we need to load all child context profile whose parent is in current module, so context trie for current module includes potential cross module inlinee.
- In order to make the above happen, we need to know whether input profile is CSSPGO profile before start reading function profile, hence a flag for profile summary section is added.
- When searching for cross module inline candidate, we need to walk through the context trie instead of nested inlinee profile (callsite sample of AutoFDO profile).
- Now that we have more accurate counts with CSSPGO, we swtiched to use entry count instead of total count to decided if an external callee is potentially beneficial to inline. This make it consistent with how we determine whether call tagert is potential inline candidate.
Differential Revision: https://reviews.llvm.org/D98590
Since D86233 we have `mustprogress` which, in combination with
`readonly`, implies `willreturn`. The idea is that every side-effect
has to be modeled as a "write". Consequently, `readonly` means there
is no side-effect, and `mustprogress` guarantees that we cannot "loop"
forever without side-effect.
Reviewed By: fhahn
Differential Revision: https://reviews.llvm.org/D94125
Splitting this out as the change is non-trivial: The way this code
handled pointer types doesn't really make sense, as GEPs can only
apply an offset to the outermost pointer, but can't drill down
into interior pointer types (which would require dereferencing
memory).
Instead give special treatment to the first (pointer) index.
I've hardcoded it to zero as that's the only way the function is
used right now, but handling non-zero indexes would be
straightforward.
The original goal here was to have an element type for CreateGEP.
For CGSCC inline, we need to scale down a function's branch weights and entry counts when thee it's inlined at a callsite. This is done through updateCallProfile. Additionally, we also scale the weigths for the inlined clone based on call site count in updateCallerBFI. Neither is needed for inlining during sample profile loader as it's using context profile that is separated from inlinee's own profile. This change skip the inlinee profile scaling for sample loader inlining.
Differential Revision: https://reviews.llvm.org/D98187
This patch makes uses of the context bridges introduced in D83299 to make
AAValueConstantRange call site specific.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D83744
now -funique-internal-linkage-name flag is available, and we want to flip
it on by default since it is beneficial to have separate sample profiles
for different internal symbols with the same name. As a preparation, we
want to avoid regression caused by the flip.
When we flip -funique-internal-linkage-name on, the profile is collected
from binary built without -funique-internal-linkage-name so it has no uniq
suffix, but the IR in the optimized build contains the suffix. This kind of
mismatch may introduce transient regression.
To avoid such mismatch, we introduce a NameTable section flag indicating
whether there is any name in the profile containing uniq suffix. Compiler
will decide whether to keep uniq suffix during name canonicalization
depending on the NameTable section flag. The flag is only available for
extbinary format. For other formats, by default compiler will keep uniq
suffix so they will only experience transient regression when
-funique-internal-linkage-name is just flipped.
Another type of regression is caused by places where we miss to call
getCanonicalFnName. Those places are fixed.
Differential Revision: https://reviews.llvm.org/D96932
For CS profile, the callsite count of previously inlined callees is populated with the entry count of the callees. Therefore when trying to get a weight for calliste probe after inlinining, the callsite count should always be used. The same fix has already been made for non-probe case.
Reviewed By: wenlei
Differential Revision: https://reviews.llvm.org/D98094
We have the `enable-loopinterchange` option in legacy pass manager but not in NPM.
Add `LoopInterchange` pass to the optimization pipeline (at the same position as before)
when `enable-loopinterchange` is turned on.
Reviewed By: aeubanks, fhahn
Differential Revision: https://reviews.llvm.org/D98116
Initial support for using the OpenMPIRBuilder by clang to generate loops using the OpenMPIRBuilder. This initial support is intentionally limited to:
* Only the worksharing-loop directive.
* Recognizes only the nowait clause.
* No loop nests with more than one loop.
* Untested with templates, exceptions.
* Semantic checking left to the existing infrastructure.
This patch introduces a new AST node, OMPCanonicalLoop, which becomes parent of any loop that has to adheres to the restrictions as specified by the OpenMP standard. These restrictions allow OMPCanonicalLoop to provide the following additional information that depends on base language semantics:
* The distance function: How many loop iterations there will be before entering the loop nest.
* The loop variable function: Conversion from a logical iteration number to the loop variable.
These allow the OpenMPIRBuilder to act solely using logical iteration numbers without needing to be concerned with iterator semantics between calling the distance function and determining what the value of the loop variable ought to be. Any OpenMP logical should be done by the OpenMPIRBuilder such that it can be reused MLIR OpenMP dialect and thus by flang.
The distance and loop variable function are implemented using lambdas (or more exactly: CapturedStmt because lambda implementation is more interviewed with the parser). It is up to the OpenMPIRBuilder how they are called which depends on what is done with the loop. By default, these are emitted as outlined functions but we might think about emitting them inline as the OpenMPRuntime does.
For compatibility with the current OpenMP implementation, even though not necessary for the OpenMPIRBuilder, OMPCanonicalLoop can still be nested within OMPLoopDirectives' CapturedStmt. Although OMPCanonicalLoop's are not currently generated when the OpenMPIRBuilder is not enabled, these can just be skipped when not using the OpenMPIRBuilder in case we don't want to make the AST dependent on the EnableOMPBuilder setting.
Loop nests with more than one loop require support by the OpenMPIRBuilder (D93268). A simple implementation of non-rectangular loop nests would add another lambda function that returns whether a loop iteration of the rectangular overapproximation is also within its non-rectangular subset.
Reviewed By: jdenny
Differential Revision: https://reviews.llvm.org/D94973
sample loader pass.
In https://reviews.llvm.org/rG5fb65c02ca5e91e7e1a00e0efdb8edc899f3e4b9,
to prevent repeated indirect call promotion for the same indirect call
and the same target, we used zero-count value profile to indicate an
indirect call has been promoted for a certain target. We removed
PromotedInsns cache in the same patch. However, there was a problem in
that patch described below, and that problem led me to add PromotedInsns
back as a mitigation in
https://reviews.llvm.org/rG4ffad1fb489f691825d6c7d78e1626de142f26cf.
When we get value profile from metadata by calling getValueProfDataFromInst,
we need to specify the maximum possible number of values we expect to read.
We uses MaxNumPromotions in the last patch so the maximum number of value
information extracted from metadata is MaxNumPromotions. If we have many
values including zero-count values when we write the metadata, some of them
will be dropped when we read them because we only read MaxNumPromotions
values. It will allow repeated indirect call promotion again. We need to
make sure if there are values indicating promoted targets, those values need
to be saved in metadata with higher priority than other values.
The patch fixed that problem. We change to use -1 to represent the count
of a promoted target instead of 0 so it is easier to sort the values.
When we prepare to update the metadata in updateIDTMetaData, we will sort
the values in the descending count order and extract only MaxNumPromotions
values to write into metadata. Since -1 is the max uint64_t number, if we
have equal to or less than MaxNumPromotions of -1 count values, they will
all be kept in metadata. If we have more than MaxNumPromotions of -1 count
values, we will only save MaxNumPromotions such values maximally. In such
case, we have logic in place in doesHistoryAllowICP to guarantee no more
promotion in sample loader pass will happen for the indirect call, because
it has been promoted enough.
With this change, now we can remove PromotedInsns without problem.
Differential Revision: https://reviews.llvm.org/D97350
This change fixes a couple places where the pseudo probe intrinsic blocks optimizations because they are not naturally removable. To unblock those optimizations, the blocking pseudo probes are moved out of the original blocks and tagged dangling, instead of allowing pseudo probes to be literally removed. The reason is that when the original block is removed, we won't be able to sample it. Instead of assigning it a zero weight, moving all its pseudo probes into another block and marking them dangling should allow the counts inference a chance to assign them a more reasonable weight. We have not seen counts quality degradation from our experiments.
The optimizations being unblocked are:
1. Removing conditional probes for if-converted branches. Conditional probes are tagged dangling when their homing branch arms are folded so that they will not be over-counted.
2. Unblocking jump threading from removing empty blocks. Pseudo probe prevents jump threading from removing logically empty blocks that only has one unconditional jump instructions.
3. Unblocking SimplifyCFG and MIR tail duplicate to thread empty blocks and blocks with redundant branch checks.
Since dangling probes are logically deleted, they should not consume any samples in LTO postLink. This can be achieved by setting their distribution factors to zero when dangled.
Reviewed By: wmi
Differential Revision: https://reviews.llvm.org/D97481
We don't need a bool and an enum to express the three options we
currently have. This makes the interface nicer and much easier to
use optional dependencies. Also avoids mistakes where the bool is
false and enum ignored.
This seems to be more of a Clang thing rather than a generic LLVM thing,
so this moves it out of LLVM pipelines and as Clang extension hooks into
LLVM pipelines.
Move the post-inline EEInstrumentation out of the backend pipeline and
into a late pass, similar to other sanitizer passes. It doesn't fit
into the codegen pipeline.
Also fix up EntryExitInstrumentation not running at -O0 under the new
PM. PR49143
Reviewed By: hans
Differential Revision: https://reviews.llvm.org/D97608
In SanitizerCoverage, the metadata sections (`__sancov_guards`,
`__sancov_cntrs`, `__sancov_bools`) are referenced by functions. After
inlining, such a `__sancov_*` section can be referenced by more than one
functions, but its sh_link still refers to the original function's section.
(Note: a SHF_LINK_ORDER section referenced by a section other than its linked-to
section violates the invariant.)
If the original function's section is discarded (e.g. LTO internalization +
`ld.lld --gc-sections`), ld.lld may report a `sh_link points to discarded section` error.
This above reasoning means that `!associated` is not appropriate to be called by
an inlinable function. Non-interposable functions are inline candidates, so we
have to drop `!associated`. A `__sancov_pcs` is not referenced by other sections
but is expected to parallel a metadata section, so we have to make sure the two
sections are retained or discarded at the same time. A section group does the
trick. (Note: we have a module ctor, so `getUniqueModuleId` guarantees to
return a non-empty string, and `GetOrCreateFunctionComdat` guarantees to return
non-null.)
For interposable functions, we could keep using `!associated`, but
LTO can change the linkage to `internal` and allow such functions to be inlinable,
so we have to drop `!associated`, too. To not interfere with section
group resolution, we need to use the `noduplicates` variant (section group flag 0).
(This allows us to get rid of the ModuleID parameter.)
In -fno-pie and -fpie code (mostly dso_local), instrumented interposable
functions have WeakAny/LinkOnceAny linkages, which are rare. So the
section group header overload should be low.
This patch does not change the object file output for COFF (where `!associated` is ignored).
Reviewed By: morehouse, rnk, vitalybuka
Differential Revision: https://reviews.llvm.org/D97430
This patch makes SampleProfileLoaderBaseImpl a template class so it
can be used in CodeGen transformation.
Noticeable changes:
* use one template parameter and use IRTraits to get other used
types an type specific functions.
* remove the temporary "inline" keywords in previous refactor
patch.
* change the template function findEquivalencesFor to a regular
function. This function has a single caller with type of
PostDominatorTree. It's simpler to use the type directly
because MachinePostDominatorTree is not a derived type of
template DominatorTreeBase.
Differential Revision: https://reviews.llvm.org/D96981
And delete the SmallPtrSetImpl overload.
While here, decrease inline element counts from 8 to 4. See D97128 for the choice.
Reviewed By: tejohnson
Differential Revision: https://reviews.llvm.org/D97257
While here, decrease inline element counts from 8 to 4. See D97128 for the choice.
Depends on D97128 (which added a new SmallVecImpl overload for collectUsedGlobalVariables).
Reviewed By: tejohnson
Differential Revision: https://reviews.llvm.org/D97139
spupyrev