We have optimized the implementation of introducing the "matching and
inference" technique into Propeller. In this new implementation, we have
made every effort to avoid introducing new compilation parameters while
ensuring compatibility with Propeller's current usage. Instead of
creating a new profile format, we reused the existing one employed by
Propeller. This new implementation is fully compatible with Propeller's
current usage patterns and reduces the amount of code changes. For
detailed information, please refer to the following RFC:
https://discourse.llvm.org/t/rfc-adding-matching-and-inference-functionality-to-propeller/86238.
We plan to submit the relevant changes in several pull requests (PRs).
The current one is the first PR, which adds the basic block hash to the
SHT_LLVM_BB_ADDR_MAP section.
co-authors: lifengxiang1025 <lifengxiang@kuaishou.com>; zcfh
<wuminghui03@kuaishou.com>
Co-authored-by: lifengxiang1025 <lifengxiang@kuaishou.com>
Co-authored-by: zcfh <wuminghui03@kuaishou.com>
Co-authored-by: Rahman Lavaee <rahmanl@google.com>
yaml2obj should determine the program header offset (and other
properties) based on the intended values rather than the final
`sh_offset` of the section header.
`setProgramHeaderLayout` uses section offsets for determining
`p_offset`. Move section header overriding after
`setProgramHeaderLayout` to prevent `ShOffset` from affecting program
header `p_offset`.
This change adjusts the timing of when the section header is overridden
to ensure that the program headers are set correctly.
More details
[here](https://github.com/llvm/llvm-project/pull/126537#issuecomment-2700421989).
---------
Signed-off-by: Ruoyu Qiu <cabbaken@outlook.com>
Signed-off-by: Ruoyu Qiu <qiuruoyu@xiaomi.com>
Co-authored-by: Ruoyu Qiu <qiuruoyu@xiaomi.com>
This PR simply moves the callsite anchors from the beginning of
callsites to their end.
Emitting the end of callsites is more sensible as it allows breaking the
basic block into subblocks which end with control transfer instructions.
Recently, we have been looking at some optimizations targeting
individual calls. In particular, we plan to extend the address mapping
technique to map to individual callsites. For example, in this piece of
code for a basic blocks:
```
<BB>:
1200: lea 0x1(%rcx), %rdx
1204: callq foo
1209: cmpq 0x10, %rdx
120d: ja L1
```
We want to emit 0x9 as the call site offset for `callq foo` (the offset
from the block entry to right after the call), so that we know if a
sampled address is before the call or after.
This PR implements the decode/encode/emit capability. The Codegen change
will be implemented in a later PR.
All notes in the note section must be correctly aligned, including the
first. The tool should refuse to generate notes if the section offset is
incorrect in this respect.
This relands #118157 with a fix for the use of an uninitialized
variable and additional tests.
The System V ABI
(https://www.sco.com/developers/gabi/latest/ch5.pheader.html#note_section)
states that the note entries and their descriptor fields must be aligned
to 4 or 8 bytes for 32-bit or 64-bit objects respectively. In practice,
64-bit systems can use both alignments, with the actual format being
determined by the alignment of the segment. For example, the Linux
gABI extension (https://github.com/hjl-tools/linux-abi/wiki/linux-abi-draft.pdf)
contains a special note on this, see 2.1.7 "Alignment of Note Sections".
This patch adjusts the format of the generated notes to the specified
section alignment. Since `llvm-readobj` was fixed in a similar way in
https://reviews.llvm.org/D150022, "[Object] Fix handling of Elf_Nhdr
with sh_addralign=8", the generated notes can now be parsed
successfully by the tool.
The [System V ABI](https://www.sco.com/developers/gabi/latest/ch5.pheader.html#note_section)
states that the note entries and their descriptor fields must be aligned
to 4 or 8 bytes for 32-bit or 64-bit objects respectively. In practice,
64-bit systems can use both alignments, with the actual format being
determined by the alignment of the segment. For example, the
[Linux gABI extension](https://github.com/hjl-tools/linux-abi/wiki/linux-abi-draft.pdf)
contains a special note on this, see 2.1.7 "Alignment of Note Sections".
This patch adjusts the format of the generated notes to the specified
section alignment. Since `llvm-readobj` was fixed in a similar way in
[D150022](https://reviews.llvm.org/D150022), "[Object] Fix handling of
Elf_Nhdr with sh_addralign=8", the generated notes can now be parsed
successfully by the tool.
Sometimes we want to use a `PgoAnalysisMap` feature that doesn't require
the BB entries info, e.g. only the `FuncEntryCount`, but the BB entries
is emitted by default, so I'm adding an option to skip the info for this
case to save the binary size(can save ~90% size of the section). For
implementation, it extends a new field(`OmitBBEntries`) in
`BBAddrMap::Features` for this and it's controlled by a switch
`--basic-block-address-map-skip-bb-entries`.
Note that this naturally supports backwards compatibility as the field
is zero for the old version, matches the decoding in the new version
llvm.
CREL is a compact relocation format for the ELF object file format.
This patch adds integrated assembler support (using the RELA form)
available with `llvm-mc -filetype=obj -crel a.s -o a.o`.
A dependent patch will add `clang -c -Wa,--crel,--allow-experimental-crel`.
Also add llvm-readobj support (for both REL and RELA forms) to
facilitate testing the assembler. Additionally, yaml2obj gains support
for the RELA form to aid testing with llvm-readobj.
We temporarily assign the section type code 0x40000020 from the generic
range to `SHT_CREL`. We avoided using `SHT_LLVM_` or `SHT_GNU_` to
avoid code churn and maintain broader applicability for interested psABIs.
Similarly, `DT_CREL` is temporarily 0x40000026.
LLVM will change the code and break compatibility. This is not an issue
if all relocatable files using CREL are regenerated (aka no prebuilt
relocatable files).
Link: https://discourse.llvm.org/t/rfc-crel-a-compact-relocation-format-for-elf/77600
Pull Request: https://github.com/llvm/llvm-project/pull/91280
`TargetEndianness` is long and unwieldy. "Target" in the name is confusing. Rename it to "Endianness".
I cannot find noticeable out-of-tree users of `TargetEndianness`, but
keep `TargetEndianness` to make this patch safer. `TargetEndianness`
will be removed by a subsequent change.
Today `-split-machine-functions` and `-fbasic-block-sections={all,list}`
cannot be combined with `-basic-block-sections=labels` (the labels
option will be ignored).
The inconsistency comes from the way basic block address map -- the
underlying mechanism for basic block labels -- encodes basic block
addresses
(https://lists.llvm.org/pipermail/llvm-dev/2020-July/143512.html).
Specifically, basic block offsets are computed relative to the function
begin symbol. This relies on functions being contiguous which is not the
case for MFS and basic block section binaries. This means Propeller
cannot use binary profiles collected from these binaries, which limits
the applicability of Propeller for iterative optimization.
To make the `SHT_LLVM_BB_ADDR_MAP` feature work with basic block section
binaries, we propose modifying the encoding of this section as follows.
First let us review the current encoding which emits the address of each
function and its number of basic blocks, followed by basic block entries
for each basic block.
| | |
|--|--|
| Address of the function | Function Address |
| Number of basic blocks in this function | NumBlocks |
| BB entry 1
| BB entry 2
| ...
| BB entry #NumBlocks
To make this work for basic block sections, we treat each basic block
section similar to a function, except that basic block sections of the
same function must be encapsulated in the same structure so we can map
all of them to their single function.
We modify the encoding to first emit the number of basic block sections
(BB ranges) in the function. Then we emit the address map of each basic
block section section as before: the base address of the section, its
number of blocks, and BB entries for its basic block. The first section
in the BB address map is always the function entry section.
| | |
|--|--|
| Number of sections for this function | NumBBRanges |
| Section 1 begin address | BaseAddress[1] |
| Number of basic blocks in section 1 | NumBlocks[1] |
| BB entries for Section 1
|..................|
| Section #NumBBRanges begin address | BaseAddress[NumBBRanges] |
| Number of basic blocks in section #NumBBRanges |
NumBlocks[NumBBRanges] |
| BB entries for Section #NumBBRanges
The encoding of basic block entries remains as before with the minor
change that each basic block offset is now computed relative to the
begin symbol of its containing BB section.
This patch adds a new boolean codegen option `-basic-block-address-map`.
Correspondingly, the front-end flag `-fbasic-block-address-map` and LLD
flag `--lto-basic-block-address-map` are introduced.
Analogously, we add a new TargetOption field `BBAddrMap`. This means BB
address maps are either generated for all functions in the compiling
unit, or for none (depending on `TargetOptions::BBAddrMap`).
This patch keeps the functionality of the old
`-fbasic-block-sections=labels` option but does not remove it. A
subsequent patch will remove the obsolete option.
We refactor the `BasicBlockSections` pass by separating the BB address
map and BB sections handing to their own functions (named
`handleBBAddrMap` and `handleBBSections`). `handleBBSections` renumbers
basic blocks and places them in their assigned sections.
`handleBBAddrMap` is invoked after `handleBBSections` (if requested) and
only renumbers the blocks.
- New tests added:
- Two tests basic-block-address-map-with-basic-block-sections.ll and
basic-block-address-map-with-mfs.ll to exercise the combination of
`-basic-block-address-map` with `-basic-block-sections=list` and
'-split-machine-functions`.
- A driver sanity test for the `-fbasic-block-address-map` option
(basic-block-address-map.c).
- An LLD test for testing the `--lto-basic-block-address-map` option.
This reuses the LLVM IR from `lld/test/ELF/lto/basic-block-sections.ll`.
- Renamed and modified the two existing codegen tests for basic block
address map (`basic-block-sections-labels-functions-sections.ll` and
`basic-block-sections-labels.ll`)
- Removed `SHT_LLVM_BB_ADDR_MAP_V0` tests. Full deprecation of
`SHT_LLVM_BB_ADDR_MAP_V0` and `SHT_LLVM_BB_ADDR_MAP` version less than 2
will happen in a separate PR in a few months.
Reviewed in PR (#71750). A part of [RFC - PGO Accuracy Metrics: Emitting and Evaluating Branch
and Block
Analysis](https://discourse.llvm.org/t/rfc-pgo-accuracy-metrics-emitting-and-evaluating-branch-and-block-analysis/73902).
This PR adds the PGOAnalysisMap data structure and implements encoding and
decoding through Object and ObjectYAML along with associated tests. When
emitted into the bb-addr-map section, each function is followed by the associated
pgo-analysis-map for that function. The emitting of each analysis in the map is
controlled by a bit in the bb-addr-map feature byte. All existing bb-addr-map
code can ignore the pgo-analysis-map if the caller does not request the data.
This patch replaces uses of StringRef::{starts,ends}with with
StringRef::{starts,ends}_with for consistency with
std::{string,string_view}::{starts,ends}_with in C++20.
I'm planning to deprecate and eventually remove
StringRef::{starts,ends}with.
Now that llvm::support::endianness has been renamed to
llvm::endianness, we can use the shorter form. This patch replaces
support::endianness with llvm::endianness.
Let Propeller use specialized IDs for basic blocks, instead of MBB number.
This allows optimizations not just prior to asm-printer, but throughout the entire codegen.
This patch only implements the functionality under the new `LLVM_BB_ADDR_MAP` version, but the old version is still being used. A later patch will change the used version.
####Background
Today Propeller uses machine basic block (MBB) numbers, which already exist, to map native assembly to machine IR. This is done as follows.
- Basic block addresses are captured and dumped into the `LLVM_BB_ADDR_MAP` section just before the AsmPrinter pass which writes out object files. This ensures that we have a mapping that is close to assembly.
- Profiling mapping works by taking a virtual address of an instruction and looking up the `LLVM_BB_ADDR_MAP` section to find the MBB number it corresponds to.
- While this works well today, we need to do better when we scale Propeller to target other Machine IR optimizations like spill code optimization. Register allocation happens earlier in the Machine IR pipeline and we need an annotation mechanism that is valid at that point.
- The current scheme will not work in this scenario because the MBB number of a particular basic block is not fixed and changes over the course of codegen (via renumbering, adding, and removing the basic blocks).
- In other words, the volatile MBB numbers do not provide a one-to-one correspondence throughout the lifetime of Machine IR. Profile annotation using MBB numbers is restricted to a fixed point; only valid at the exact point where it was dumped.
- Further, the object file can only be dumped before AsmPrinter and cannot be dumped at an arbitrary point in the Machine IR pass pipeline. Hence, MBB numbers are not suitable and we need something else.
####Solution
We propose using fixed unique incremental MBB IDs for basic blocks instead of volatile MBB numbers. These IDs are assigned upon the creation of machine basic blocks. We modify `MachineFunction::CreateMachineBasicBlock` to assign the fixed ID to every newly created basic block. It assigns `MachineFunction::NextMBBID` to the MBB ID and then increments it, which ensures having unique IDs.
To ensure correct profile attribution, multiple equivalent compilations must generate the same Propeller IDs. This is guaranteed as long as the MachineFunction passes run in the same order. Since the `NextBBID` variable is scoped to `MachineFunction`, interleaving of codegen for different functions won't cause any inconsistencies.
The new encoding is generated under the new version number 2 and we keep backward-compatibility with older versions.
####Impact on Size of the `LLVM_BB_ADDR_MAP` Section
Emitting the Propeller ID results in a 23% increase in the size of the `LLVM_BB_ADDR_MAP` section for the clang binary.
Reviewed By: tmsriram
Differential Revision: https://reviews.llvm.org/D100808
Use deduction guides instead of helper functions.
The only non-automatic changes have been:
1. ArrayRef(some_uint8_pointer, 0) needs to be changed into ArrayRef(some_uint8_pointer, (size_t)0) to avoid an ambiguous call with ArrayRef((uint8_t*), (uint8_t*))
2. CVSymbol sym(makeArrayRef(symStorage)); needed to be rewritten as CVSymbol sym{ArrayRef(symStorage)}; otherwise the compiler is confused and thinks we have a (bad) function prototype. There was a few similar situation across the codebase.
3. ADL doesn't seem to work the same for deduction-guides and functions, so at some point the llvm namespace must be explicitly stated.
4. The "reference mode" of makeArrayRef(ArrayRef<T> &) that acts as no-op is not supported (a constructor cannot achieve that).
Per reviewers' comment, some useless makeArrayRef have been removed in the process.
This is a follow-up to https://reviews.llvm.org/D140896 that introduced
the deduction guides.
Differential Revision: https://reviews.llvm.org/D140955
Let Propeller use specialized IDs for basic blocks, instead of MBB number.
This allows optimizations not just prior to asm-printer, but throughout the entire codegen.
This patch only implements the functionality under the new `LLVM_BB_ADDR_MAP` version, but the old version is still being used. A later patch will change the used version.
####Background
Today Propeller uses machine basic block (MBB) numbers, which already exist, to map native assembly to machine IR. This is done as follows.
- Basic block addresses are captured and dumped into the `LLVM_BB_ADDR_MAP` section just before the AsmPrinter pass which writes out object files. This ensures that we have a mapping that is close to assembly.
- Profiling mapping works by taking a virtual address of an instruction and looking up the `LLVM_BB_ADDR_MAP` section to find the MBB number it corresponds to.
- While this works well today, we need to do better when we scale Propeller to target other Machine IR optimizations like spill code optimization. Register allocation happens earlier in the Machine IR pipeline and we need an annotation mechanism that is valid at that point.
- The current scheme will not work in this scenario because the MBB number of a particular basic block is not fixed and changes over the course of codegen (via renumbering, adding, and removing the basic blocks).
- In other words, the volatile MBB numbers do not provide a one-to-one correspondence throughout the lifetime of Machine IR. Profile annotation using MBB numbers is restricted to a fixed point; only valid at the exact point where it was dumped.
- Further, the object file can only be dumped before AsmPrinter and cannot be dumped at an arbitrary point in the Machine IR pass pipeline. Hence, MBB numbers are not suitable and we need something else.
####Solution
We propose using fixed unique incremental MBB IDs for basic blocks instead of volatile MBB numbers. These IDs are assigned upon the creation of machine basic blocks. We modify `MachineFunction::CreateMachineBasicBlock` to assign the fixed ID to every newly created basic block. It assigns `MachineFunction::NextMBBID` to the MBB ID and then increments it, which ensures having unique IDs.
To ensure correct profile attribution, multiple equivalent compilations must generate the same Propeller IDs. This is guaranteed as long as the MachineFunction passes run in the same order. Since the `NextBBID` variable is scoped to `MachineFunction`, interleaving of codegen for different functions won't cause any inconsistencies.
The new encoding is generated under the new version number 2 and we keep backward-compatibility with older versions.
####Impact on Size of the `LLVM_BB_ADDR_MAP` Section
Emitting the Propeller ID results in a 23% increase in the size of the `LLVM_BB_ADDR_MAP` section for the clang binary.
Reviewed By: tmsriram
Differential Revision: https://reviews.llvm.org/D100808
This patch mechanically replaces None with std::nullopt where the
compiler would warn if None were deprecated. The intent is to reduce
the amount of manual work required in migrating from Optional to
std::optional.
This is part of an effort to migrate from llvm::Optional to
std::optional:
https://discourse.llvm.org/t/deprecating-llvm-optional-x-hasvalue-getvalue-getvalueor/63716
This is a resurrection of D106421 with the change that it keeps backward-compatibility. This means decoding the previous version of `LLVM_BB_ADDR_MAP` will work. This is required as the profile mapping tool is not released with LLVM (AutoFDO). As suggested by @jhenderson we rename the original section type value to `SHT_LLVM_BB_ADDR_MAP_V0` and assign a new value to the `SHT_LLVM_BB_ADDR_MAP` section type. The new encoding adds a version byte to each function entry to specify the encoding version for that function. This patch also adds a feature byte to be used with more flexibility in the future. An use-case example for the feature field is encoding multi-section functions more concisely using a different format.
Conceptually, the new encoding emits basic block offsets and sizes as label differences between each two consecutive basic block begin and end label. When decoding, offsets must be aggregated along with basic block sizes to calculate the final offsets of basic blocks relative to the function address.
This encoding uses smaller values compared to the existing one (offsets relative to function symbol).
Smaller values tend to occupy fewer bytes in ULEB128 encoding. As a result, we get about 17% total reduction in the size of the bb-address-map section (from about 11MB to 9MB for the clang PGO binary).
The extra two bytes (version and feature fields) incur a small 3% size overhead to the `LLVM_BB_ADDR_MAP` section size.
Reviewed By: jhenderson
Differential Revision: https://reviews.llvm.org/D121346
Currently when .llvm.call-graph-profile is created by llvm it explicitly encodes the symbol indices. This section is basically a black box for post processing tools. For example, if we run strip -s on the object files the symbol table changes, but indices in that section do not. In non-visible behavior indices point to wrong symbols. The visible behavior indices point outside of Symbol table: "invalid symbol index".
This patch changes the format by using R_*_NONE relocations to indicate the from/to symbols. The Frequency (Weight) will still be in the .llvm.call-graph-profile, but symbol information will be in relocation section. In LLD information from both sections is used to reconstruct call graph profile. Relocations themselves will never be applied.
With this approach post processing tools that handle relocations correctly work for this section also. Tools can add/remove symbols and as long as they handle relocation sections with this approach information stays correct.
Doing a quick experiment with clang-13.
The size went up from 107KB to 322KB, aggregate of all the input sections. Size of clang-13 binary is ~118MB. For users of -fprofile-use/-fprofile-sample-use the size of object files will go up slightly, it will not impact final binary size.
Reviewed By: jhenderson, MaskRay
Differential Revision: https://reviews.llvm.org/D104080
This patch adds support for a new field in the FileHeader, which states
the name to use for the section header string table. This also allows
combining the string table with another string table in the object, e.g.
the symbol name string table. The field is optional. By default,
.shstrtab will continue to be used.
This partially fixes https://bugs.llvm.org/show_bug.cgi?id=50506.
Reviewed by: Higuoxing
Differential Revision: https://reviews.llvm.org/D104035
There was an off-by-one error caused by an index (which included an
index for the null section header) being used to check against the size
of a list of sections (which didn't include the null section header).
This is a partial fix for https://bugs.llvm.org/show_bug.cgi?id=50506.
Reviewed by: MaskRay
Differential Revision: https://reviews.llvm.org/D104098
As discussed in D95511, this allows us to encode invalid BBAddrMap
sections to be used in more rigorous testing.
Reviewed By: jhenderson
Differential Revision: https://reviews.llvm.org/D96831
As discussed in D95511, this allows us to encode invalid BBAddrMap
sections to be used in more rigorous testing.
Reviewed By: jhenderson
Differential Revision: https://reviews.llvm.org/D96831
Currently we don't allow the following definition:
```
Sections:
- Type: SectionHeaderTable
- Name: .foo
Type: SHT_PROGBITS
```
We report an error: "SectionHeaderTable can't be empty. Use 'NoHeaders' key to drop the section header table".
It was implemented in this way earlier, when `SectionHeaderTable`
was a dedicated key outside of the `Sections` list. And we did not
allow to select where the table is written.
Currently it makes sense to allow it, because a user might
want to place the default section header table at an arbitrary position,
e.g. before other sections. In this case it is not convenient and error prone
to require specifying all sections:
```
Sections:
- Type: SectionHeaderTable
Sections:
- Name: .foo
- Name: .strtab
- Name: .shstrtab
- Name: .foo
Type: SHT_PROGBITS
```
This patch allows empty SectionHeaderTable definitions.
Differential revision: https://reviews.llvm.org/D95341
We already set the `sh_entsize` field in a single place
for all non-implicit sections.
This patch reorders the logic slightly and with it
we finally have the only one place where the `sh_entsize` is set.
obj2yaml will not dump the `EntSize` key for `SHT_DYNSYM/SHT_SYMTAB` sections anymore,
when the value of `sh_entsize` is equal to `sizeof(Elf_Sym)`
Note that this also seems revealed an issue in llvm-objcopy:
Previously yaml2obj set the `sh_entsize` for the `.symtab` section to 0x18,
now we it sets it for `SHT_SYMTAB` sections, i.e. by type.
But the `llvm-objcopy/ELF/only-keep-debug.test` has a `.symtab` section of type `SHT_STRTAB`,
and now yaml2obj sets the `sh_entsize` to 0 for it.
I had to update the corresponding check lines for `ES`, but the behavior of
`llvm-objcopy` should be fixed instead I think.
I've added a TODO and a comment.
Differential revision: https://reviews.llvm.org/D95364
This refactors the logic that sets the `sh_link` field.
With this patch we set it in a single place for all sections.
Differential revision: https://reviews.llvm.org/D95354
This was discussed in D93678 thread.
Currently we have one special chunk - Fill.
This patch re implements the "SectionHeaderTable" key to become a special chunk too.
With that we are able to place the section header table at any location,
just like we place sections.
Differential revision: https://reviews.llvm.org/D95140
This makes the following improvements.
For `SHT_GNU_versym`:
* yaml2obj: set `sh_link` to index of `.dynsym` section automatically.
For `SHT_GNU_verdef`:
* yaml2obj: set `sh_link` to index of `.dynstr` section automatically.
* yaml2obj: set `sh_info` field automatically.
* obj2yaml: don't dump the `Info` field when its value matches the number of version definitions.
For `SHT_GNU_verneed`:
* yaml2obj: set `sh_link` to index of `.dynstr` section automatically.
* yaml2obj: set `sh_info` field automatically.
* obj2yaml: don't dump the `Info` field when its value matches the number of version dependencies.
Also, simplifies few test cases.
Differential revision: https://reviews.llvm.org/D94956
This reuses the code from yaml2obj (moves it to ELFYAML.h).
With it we can set the `sh_entsize` in a single place in `obj2yaml`.
Note that it also fixes a bug of `yaml2obj`: we do not
set the `sh_entsize` field for the `SHT_ARM_EXIDX` section properly.
Differential revision: https://reviews.llvm.org/D93858
We have the following issues related to group sections:
1) yaml2obj is unable to set the custom `sh_entsize` value, because the `EntSize`
key is currently ignored.
2) obj2yaml is unable to dump the group section which `sh_entsize != 4`.
3) obj2yaml always dumps the "EntSize" for group sections, though
usually we are trying to omit dumping default values when dumping keys.
I.e. we should not print the "EntSize" key when `sh_entsize` == 4.
This patch fixes (1),(3) and adds the test case to document the behavior of (2).
Differential revision: https://reviews.llvm.org/D93854
