https://reviews.llvm.org/D116179 introduced some changes to
`InstrProfData.inc` which broke some downstream builds. This commit
reverts those changes since they only changes two field names.
Reviewed By: phosek
Differential Revision: https://reviews.llvm.org/D117631
Existing code tended to assume that counters had type `uint64_t` and
computed size from the number of counters. Fix this code to directly
compute the counters size in number of bytes where possible. When the
number of counters is needed, use `__llvm_profile_counter_entry_size()`
or `getCounterTypeSize()`. In a later diff these functions will depend
on the profile mode.
Change the meaning of `DataSize` and `CountersSize` to make them more clear.
* `DataSize` (`CountersSize`) - the size of the data (counter) section in bytes.
* `NumData` (`NumCounters`) - the number of data (counter) entries.
Reviewed By: kyulee
Differential Revision: https://reviews.llvm.org/D116179
When using debug info for profile correlation, avoid adding duplicate
functions in the synthetic Data section.
Before this patch, n duplicate function entries in the Data section would
cause counter values to be a factor of n larger. I built instrumented
clang with and without debug info correlation and got these summaries.
```
# With Debug Info Correlate
$ llvm-profdata show default.profdata
Instrumentation level: IR entry_first = 0
Total functions: 182530
Maximum function count: 52034
Maximum internal block count: 5763
# Without
$ llvm-profdata show default.profdata
Instrumentation level: IR entry_first = 0
Total functions: 183212
Maximum function count: 52034
Maximum internal block count: 5766
```
The slight difference in counts seem to be mostly from FileSystem and
Map functions and the difference in the number of instrumented functions
seems to come from missing debug info like destructors without source.
Reviewed By: kyulee
Differential Revision: https://reviews.llvm.org/D116051
Extend `llvm-profdata` to read in a `.proflite` file and also a debug info file to generate a normal `.profdata` profile. This reduces the binary size by 8.4% when building an instrumented Clang binary without value profiling (164 MB vs 179 MB).
This work is part of the "lightweight instrumentation" RFC: https://groups.google.com/g/llvm-dev/c/r03Z6JoN7d4
This was first landed in https://reviews.llvm.org/D114566 but had to be reverted due to build errors.
Reviewed By: kyulee
Differential Revision: https://reviews.llvm.org/D115915
For some reason this did not fail to compile for me locally, but did
fail on one of the builders. Here is the error:
https://lab.llvm.org/buildbot/#/builders/58/builds/14969
```
/home/buildbot/as-builder-4/lld-x86_64-ubuntu-fast/llvm-project/llvm/lib/ProfileData/InstrProfCorrelator.cpp: In static member function ‘static llvm::Expected<std::unique_ptr<llvm::InstrProfCorrelator::Context> > llvm::InstrProfCorrelator::Context::get(std::unique_ptr<llvm::MemoryBuffer>, const llvm::object::ObjectFile&)’:
/home/buildbot/as-builder-4/lld-x86_64-ubuntu-fast/llvm-project/llvm/lib/ProfileData/InstrProfCorrelator.cpp:44:10: error: could not convert ‘C’ from ‘std::unique_ptr<llvm::InstrProfCorrelator::Context>’ to ‘llvm::Expected<std::unique_ptr<llvm::InstrProfCorrelator::Context> >’
return C;
```
Reviewed By: kyulee
Differential Revision: https://reviews.llvm.org/D115911
Extend `llvm-profdata` to read in a `.proflite` file and also a debug info file to generate a normal `.profdata` profile. This reduces the binary size by 8.4% when building an instrumented Clang binary without value profiling (164 MB vs 179 MB).
This work is part of the "lightweight instrumentation" RFC: https://groups.google.com/g/llvm-dev/c/r03Z6JoN7d4
Reviewed By: kyulee
Differential Revision: https://reviews.llvm.org/D114566
Add the llvm flag `-debug-info-correlate` to attach debug info to instrumentation counters so we can correlate raw profile data to their functions. Raw profiles are dumped as `.proflite` files. The next diff enables `llvm-profdata` to consume `.proflite` and debug info files to produce a normal `.profdata` profile.
Part of the "lightweight instrumentation" work: https://groups.google.com/g/llvm-dev/c/r03Z6JoN7d4
The original diff https://reviews.llvm.org/D114565 was reverted because of the `Instrumentation/InstrProfiling/debug-info-correlate.ll` test, which is fixed in this commit.
Reviewed By: kyulee
Differential Revision: https://reviews.llvm.org/D115693
For some reason, the user-defined implicit conversion from StringRef to
std::string is not invoked by std::map::emplace in libstdc++ 5.4.0, even
though it works fine on modern systems.
CSSPGO currently employs a flat profile format for context-sensitive profiles. Such a flat profile allows for precisely manipulating contexts that is either inlined or not inlined. This is a benefit over the nested profile format used by non-CS AutoFDO. A downside of this is the longer build time due to parsing the indexing the full CS contexts.
For a CS flat profile, though only the context profiles relevant to a module are loaded when that module is compiled, the cost to figure out what profiles are relevant is noticeably high when there're many contexts, since the sample reader will need to scan all context strings anyway. On the contrary, a nested function profile has its related inline subcontexts isolated from other unrelated contexts. Therefore when compiling a set of functions, unrelated contexts will never need to be scanned.
In this change we are exploring using nested profile format for CSSPGO. This is expected to work based on an assumption that with a preinliner-computed profile all contexts are precomputed and expected to be inlined by the compiler. Contexts not expected to be inlined will be cut off and returned to corresponding base profiles (for top-level outlined functions). This naturally forms a nested profile where all nested contexts are expected to be inlined. The compiler will less likely optimize on derived contexts that are not precomputed.
A CS-nested profile will look exactly the same with regular nested profile except that each nested profile can come with an attributes. With pseudo probes, a nested profile shown as below can also have a CFG checksum.
```
main:1968679:12
2: 24
3: 28 _Z5funcAi:18
3.1: 28 _Z5funcBi:30
3: _Z5funcAi:1467398
0: 10
1: 10 _Z8funcLeafi:11
3: 24
1: _Z8funcLeafi:1467299
0: 6
1: 6
3: 287884
4: 287864 _Z3fibi:315608
15: 23
!CFGChecksum: 138828622701
!Attributes: 2
!CFGChecksum: 281479271677951
!Attributes: 2
```
Specific work included in this change:
- A recursive profile converter to convert CS flat profile to nested profile.
- Extend function checksum and attribute metadata to be stored in nested way for text profile and extbinary profile.
- Unifiy sample loader inliner path for CS and preinlined nested profile.
- Changes in the sample loader to support probe-based nested profile.
I've seen promising results regarding build time. A nested profile can result in a 20% shorter build time than a CS flat profile while keep an on-par performance. This is with -duplicate-contexts-into-base=1.
Test Plan:
Reviewed By: wenlei
Differential Revision: https://reviews.llvm.org/D115205
This reverts commit 800bf8ed29.
The `Instrumentation/InstrProfiling/debug-info-correlate.ll` test was
failing because I forgot the `llc` commands are architecture specific.
I'll follow up with a fix.
Differential Revision: https://reviews.llvm.org/D115689
Add the llvm flag `-debug-info-correlate` to attach debug info to instrumentation counters so we can correlate raw profile data to their functions. Raw profiles are dumped as `.proflite` files. The next diff enables `llvm-profdata` to consume `.proflite` and debug info files to produce a normal `.profdata` profile.
Part of the "lightweight instrumentation" work: https://groups.google.com/g/llvm-dev/c/r03Z6JoN7d4
Reviewed By: kyulee
Differential Revision: https://reviews.llvm.org/D114565
The first 8b of each raw profile section need to be aligned to 8b since
the first item in each section is a u64 count of the number of items in
the section.
Summary of changes:
* Assert alignment when reading counts.
* Update test to check alignment, relax some size checks to allow padding.
* Update raw binary inputs for llvm-profdata tests.
Differential Revision: https://reviews.llvm.org/D114826
This commit adds initial support to llvm-profdata to read and print
summaries of raw memprof profiles.
Summary of changes:
* Refactor shared defs to MemProfData.inc
* Extend show_main to display memprof profile summaries.
* Add a simple raw memprof profile reader.
* Add a couple of tests to tools/llvm-profdata.
Differential Revision: https://reviews.llvm.org/D114286
This patch fixes some issues introduced in
https://reviews.llvm.org/D108942:
1) Remove the default label to fix the bots that use
-Werror,-Wcovered-switch-default
2) Modify the malformed test to fix the bots that are
built without zlib support
3) Modify some error messages in malformed profiles
This patch fixes:
llvm/lib/ProfileData/InstrProf.cpp:146:3: error: default label in
switch which covers all enumeration values
[-Werror,-Wcovered-switch-default]
If profile data is malformed for any kind of reason, we generate
an error that only reports "malformed instrumentation profile data"
without any further information. This patch extends InstrProfError
class to receive an optional error message argument, so that we can
do better error reporting.
Differential Revision: https://reviews.llvm.org/D108942
When linkage name isn't available in dwarf (ususally the case of C code), looking up callee samples should be based on the dwarf name instead of using an empty string.
Also fixing a test issue where using empty string to look up callee samples accidentally returns the correct samples because it is treated as indirect call.
Reviewed By: wenlei
Differential Revision: https://reviews.llvm.org/D112948
Adding support to the CS preinliner to trim cold base profiles. This makes trimming consistent with the inline decision made by the preinliner. Also disable the existing profile merger when preinliner is on unless explicitly specified.
Reviewed By: wenlei, wlei
Differential Revision: https://reviews.llvm.org/D112489
Some tests with binary IDs would fail with error: no profile can be merged.
This is because raw profiles could have unaligned headers when emitting binary
IDs. This means padding should be emitted after binary IDs are emitted to
ensure everything else is aligned. This patch adds padding after each binary ID
to ensure the next binary ID size is 8-byte aligned. This also adds extra
checks to ensure we aren't reading corrupted data when printing binary IDs.
Differential Revision: https://reviews.llvm.org/D110365
With the context split work, the context-based (an array of strings) sorting performed at profile load time is way more expansive than single-string-based sorting. This is likely due to auxiliary operations done on each array element, such as indirect references, std::min operations, also likely cache misses. In this change I'm presorting profiles during profile generation time to avoid sorting at compile time.
Compared to the previous context-split work, this effectively cuts down compile time by 20% for one of our large services and brings us closer to non-CS build, with still a small gap in build time.
Reviewed By: wenlei, wmi
Differential Revision: https://reviews.llvm.org/D109036
Adding the compiler support of MD5 CS profile based on pervious context split work D107299. A MD5 CS profile is about 40% smaller than the string-based extbinary profile. As a result, the compilation is 15% faster.
There are a few conversion from real names to md5 names that have been made on the sample loader and context tracker side to get it work.
Reviewed By: wenlei, wmi
Differential Revision: https://reviews.llvm.org/D108342
Currently context strings contain a lot of duplicated function names and that significantly increase the profile size. This change split the context into a series of {name, offset, discriminator} tuples so function names used in the context can be replaced by the index into the name table and that significantly reduce the size consumed by context.
A follow-up improvement made in the compiler and profiling tools is to avoid reconstructing full context strings which is time- and memory- consuming. Instead a context vector of `StringRef` is adopted to represent the full context in all scenarios. As a result, the previous prevalent profile map which was implemented as a `StringRef` is now engineered as an unordered map keyed by `SampleContext`. `SampleContext` is reshaped to using an `ArrayRef` to represent a full context for CS profile. For non-CS profile, it falls back to use `StringRef` to represent a contextless function name. Both the `ArrayRef` and `StringRef` objects are underpinned by real array and string objects that are stored in producer buffers. For compiler, they are maintained by the sample reader. For llvm-profgen, they are maintained in `ProfiledBinary` and `ProfileGenerator`. Full context strings can be generated only in those cases of debugging and printing.
When it comes to profile format, nothing has changed to the text format, though internally CS context is implemented as a vector. Extbinary format is only changed for CS profile, with an additional `SecCSNameTable` section which stores all full contexts logically in the form of `vector<int>`, which each element as an offset points to `SecNameTable`. All occurrences of contexts elsewhere are redirected to using the offset of `SecCSNameTable`.
Testing
This is no-diff change in terms of code quality and profile content (for text profile).
For our internal large service (aka ads), the profile generation is cut to half, with a 20x smaller string-based extbinary format generated.
The compile time of ads is dropped by 25%.
Differential Revision: https://reviews.llvm.org/D107299
The change adds a switch to allow sample loader to use global pre-inliner's decision instead. The pre-inliner in llvm-profgen makes inline decision globally based on whole program profile and function byte size as cost proxy.
Since pre-inliner also adjusts/merges context profile based on its inline decision, honoring its inline decision in sample loader would lead to better post-inline profile quality especially for thinlto where cross module profile merging isn't possible without pre-inliner.
Minor fix in profile reader is also included. When pre-inliner is use, we now also turn off the default merging and trimming logic unless it's explicitly asked.
Differential Revision: https://reviews.llvm.org/D108677
We have "-profile-isfs" internal option for text, binary, and
compactbinary format (mostly for debug and test purpose). We
need to set the related flag in FunctionSamples so that ProfileIsFS
is written to the header in extbinary format.
Differential Revision: https://reviews.llvm.org/D108707
The IRPGOFlag symbol (__llvm_profile_raw_version) is dropped when
identified as non-prevailing for either regular or thin LTO during
the mixed-LTO mode compilation. This happens in the module where
IRPGOFlag is marked as non-prevailing. This variable
is emitted in the final object from the prevailing module.
This is still problematic because we currently query this symbol
to coordinate some actions between PGOInstrumentation pass
and InstrProfiling lowering pass, like whether to do value
profiling, whether to do comdat renaming.
This problem is bought up by YolandaCY in
https://reviews.llvm.org/D107034
YolandCY reported unresolved symbol linker errors in
CSPGO instrumentation build for chromium.
This patch let LTO retain IRPGOFlag decl by adding it to
CompilerUsed list and relax the check in isIRPGOFlagSet() when
doing the InstrProfiling lowering.
The test case in the patch is from D107034
<https://reviews.llvm.org/D107034>.
Differential Revision: https://reviews.llvm.org/D108581
Sample profiles are stored in a string map which is basically an unordered map. Printing out profiles by simply walking the string map doesn't enforce an order. I'm sorting the map in the decreasing order of total samples to enable a more stable dump, which is good for comparing two dumps.
Reviewed By: wenlei, wlei
Differential Revision: https://reviews.llvm.org/D108147
This fixes support for merging profiles which broke as a consequence
of e50a38840d. The issue was missing
adjustment in merge logic to account for the binary IDs which are
now included in the raw profile just after header.
In addition, this change also:
* Includes the version in module signature that's used for merging
to avoid accidental attempts to merge incompatible profiles.
* Moves the binary IDs size field after version field in the header
as was suggested in the review.
Differential Revision: https://reviews.llvm.org/D107143
This fixes support for merging profiles which broke as a consequence
of e50a38840d. The issue was missing
adjustment in merge logic to account for the binary IDs which are
now included in the raw profile just after header.
In addition, this change also:
* Includes the version in module signature that's used for merging
to avoid accidental attempts to merge incompatible profiles.
* Moves the binary IDs size field after version field in the header
as was suggested in the review.
Differential Revision: https://reviews.llvm.org/D107143
This fixes support for merging profiles which broke as a consequence
of e50a38840d. The issue was missing
adjustment in merge logic to account for the binary IDs which are
now included in the raw profile just after header.
In addition, this change also:
* Includes the version in module signature that's used for merging
to avoid accidental attempts to merge incompatible profiles.
* Moves the binary IDs size field after version field in the header
as was suggested in the review.
Differential Revision: https://reviews.llvm.org/D107143
Change `CountersPtr` in `__profd_` to a label difference, which is a link-time
constant. On ELF, when linking a shared object, this requires that `__profc_` is
either private or linkonce/linkonce_odr hidden. On COFF, we need D104564 so that
`.quad a-b` (64-bit label difference) can lower to a 32-bit PC-relative relocation.
```
# ELF: R_X86_64_PC64 (PC-relative)
.quad .L__profc_foo-.L__profd_foo
# Mach-O: a pair of 8-byte X86_64_RELOC_UNSIGNED and X86_64_RELOC_SUBTRACTOR
.quad l___profc_foo-l___profd_foo
# COFF: we actually use IMAGE_REL_AMD64_REL32/IMAGE_REL_ARM64_REL32 so
# the high 32-bit value is zero even if .L__profc_foo < .L__profd_foo
# As compensation, we truncate CountersDelta in the header so that
# __llvm_profile_merge_from_buffer and llvm-profdata reader keep working.
.quad .L__profc_foo-.L__profd_foo
```
(Note: link.exe sorts `.lprfc` before `.lprfd` even if the object writer
has `.lprfd` before `.lprfc`, so we cannot work around by reordering
`.lprfc` and `.lprfd`.)
With this change, a stage 2 (`-DLLVM_TARGETS_TO_BUILD=X86 -DLLVM_BUILD_INSTRUMENTED=IR`)
`ld -pie` linked clang is 1.74% smaller due to fewer R_X86_64_RELATIVE relocations.
```
% readelf -r pie | awk '$3~/R.*/{s[$3]++} END {for (k in s) print k, s[k]}'
R_X86_64_JUMP_SLO 331
R_X86_64_TPOFF64 2
R_X86_64_RELATIVE 476059 # was: 607712
R_X86_64_64 2616
R_X86_64_GLOB_DAT 31
```
The absolute function address (used by llvm-profdata to collect indirect call
targets) can be converted to relative as well, but is not done in this patch.
Differential Revision: https://reviews.llvm.org/D104556
Ellis Hoag