Add -Z no-unique-section-names to reduce ELF header bloat.
This change adds a new compiler flag that can help reduce the size of ELF binaries that contain many functions.
By default, when enabling function sections (which is the default for most targets), the LLVM backend will generate different section names for each function. For example, a function `func` would generate a section called `.text.func`. Normally this is fine because the linker will merge all those sections into a single one in the binary. However, starting with [LLVM 12](https://github.com/llvm/llvm-project/commit/ee5d1a04), the backend will also generate unique section names for exception handling, resulting in thousands of `.gcc_except_table.*` sections ending up in the final binary because some linkers like LLD don't currently merge or strip these EH sections (see discussion [here](https://reviews.llvm.org/D83655)). This can bloat the ELF headers and string table significantly in binaries that contain many functions.
The new option is analogous to Clang's `-fno-unique-section-names`, and instructs LLVM to generate the same `.text` and `.gcc_except_table` section for each function, resulting in a smaller final binary.
The motivation to add this new option was because we have a binary that ended up with so many ELF sections (over 65,000) that it broke some existing ELF tools, which couldn't handle so many sections.
Here's our old binary:
```
$ readelf --sections old.elf | head -1
There are 71746 section headers, starting at offset 0x2a246508:
$ readelf --sections old.elf | grep shstrtab
[71742] .shstrtab STRTAB 0000000000000000 2977204c ad44bb 00 0 0 1
```
That's an 11MB+ string table. Here's the new binary using this option:
```
$ readelf --sections new.elf | head -1
There are 43 section headers, starting at offset 0x29143ca8:
$ readelf --sections new.elf | grep shstrtab
[40] .shstrtab STRTAB 0000000000000000 29143acc 0001db 00 0 0 1
```
The whole binary size went down by over 20MB, which is quite significant.
This change adds a new compiler flag that can help reduce the size of
ELF binaries that contain many functions.
By default, when enabling function sections (which is the default for most
targets), the LLVM backend will generate different section names for each
function. For example, a function "func" would generate a section called
".text.func". Normally this is fine because the linker will merge all those
sections into a single one in the binary. However, starting with LLVM 12
(llvm/llvm-project@ee5d1a0), the backend will
also generate unique section names for exception handling, resulting in
thousands of ".gcc_except_table.*" sections ending up in the final binary
because some linkers don't currently merge or strip these EH sections.
This can bloat the ELF headers and string table significantly in
binaries that contain many functions.
The new option is analogous to Clang's -fno-unique-section-names, and
instructs LLVM to generate the same ".text" and ".gcc_except_table"
section for each function, resulting in smaller object files and
potentially a smaller final binary.
This largely involves implementing the options debug-info-for-profiling
and profile-sample-use and forwarding them on to LLVM.
AutoFDO can be used on x86-64 Linux like this:
rustc -O -Cdebug-info-for-profiling main.rs -o main
perf record -b ./main
create_llvm_prof --binary=main --out=code.prof
rustc -O -Cprofile-sample-use=code.prof main.rs -o main2
Now `main2` will have feedback directed optimization applied to it.
The create_llvm_prof tool can be obtained from this github repository:
https://github.com/google/autofdoFixes#64892.
Introduce -Z remap-cwd-prefix switch
This switch remaps any absolute paths rooted under the current
working directory to a new value. This includes remapping the
debug info in `DW_AT_comp_dir` and `DW_AT_decl_file`.
Importantly, this flag does not require passing the current working
directory to the compiler, such that the command line can be
run on any machine (with the same input files) and produce the
same results. This is critical property for debugging compiler
issues that crop up on remote machines.
This is based on adetaylor's dbc4ae7cba
Major Change Proposal: https://github.com/rust-lang/compiler-team/issues/450
Discussed on #38322. Would resolve issue #87325.
This was removed by #85284 in favor of -Zprofiler-runtime=<name>.
However the suggested -Zprofiler-runtime=None doesn't work because
"None" is treated as a crate name.
Add flag to configure `large_assignments` lint
The `large_assignments` lints detects moves over specified limit. The
limit is configured through `move_size_limit = "N"` attribute placed at
the root of a crate. When attribute is absent, the lint is disabled.
Make it possible to enable the lint without making any changes to the
source code, through a new flag `-Zmove-size-limit=N`. For example, to
detect moves exceeding 1023 bytes in a cargo crate, including all
dependencies one could use:
```
$ env RUSTFLAGS=-Zmove-size-limit=1024 cargo build -vv
```
Lint tracking issue #83518.
The `large_assignments` lints detects moves over specified limit. The
limit is configured through `move_size_limit = "N"` attribute placed at
the root of a crate. When attribute is absent, the lint is disabled.
Make it possible to enable the lint without making any changes to the
source code, through a new flag `-Zmove-size-limit=N`. For example, to
detect moves exceeding 1023 bytes in a cargo crate, including all
dependencies one could use:
```
$ env RUSTFLAGS=-Zmove-size-limit=1024 cargo build -vv
```
This creates a CSV with name "closure_profile_XXXXX.csv", where the
variable part is the process id of the compiler.
To profile a cargo project you can run one of the following depending on
if you're compiling a library or a binary:
```
cargo +stage1 rustc --lib -- -Zprofile-closures
cargo +stage1 rustc --bin -- -Zprofile-closures
```
Allow loading of llvm plugins on nightly
Based on a discussion in #82734 / with `@wsmoses.`
Mainly moves [this](0149bc4e7e) behind a -Z flag, so it can only be used on nightly,
as requested by `@nagisa` in https://github.com/rust-lang/rust/issues/82734#issuecomment-835863940
This change allows loading of llvm plugins like Enzyme.
Right now it also requires a shared library LLVM build of rustc for symbol resolution.
```rust
// test.rs
extern { fn __enzyme_autodiff(_: usize, ...) -> f64; }
fn square(x : f64) -> f64 {
return x * x;
}
fn main() {
unsafe {
println!("Hello, world {} {}!", square(3.0), __enzyme_autodiff(square as usize, 3.0));
}
}
```
```
./rustc test.rs -Z llvm-plugins="./LLVMEnzyme-12.so" -C passes="enzyme"
./test
Hello, world 9 6!
```
I will try to figure out how to simplify the usage and get this into stable in a later iteration,
but having this on nightly will already help testing further steps.
Provide option for specifying the profiler runtime
Currently, if `-Zinstrument-coverage` is enabled, the target is linked
against the `library/profiler_builtins` crate (which pulls in LLVM's
compiler-rt runtime).
This option enables backends to specify an alternative runtime crate for
handling injected instrumentation calls.
Previously, we sorted the vec prior to hashing, making the hash
independent of the original (command-line argument) order. However, the
original vec was still always kept in the original order, so we were
relying on the rest of the compiler always working with it in an
'order-independent' way.
This assumption was not being upheld by the `native_libraries` query -
the order of the entires in its result depends on the order of entries
in `Options.libs`. This lead to an 'unstable fingerprint' ICE when the
`-l` arguments were re-ordered.
This PR removes the sorting logic entirely. Re-ordering command-line
arguments (without adding/removing/changing any arguments) seems like a
really niche use case, and correctly optimizing for it would require
additional work. By always hashing arguments in their original order, we
can entirely avoid a cause of 'unstable fingerprint' errors.
Currently, if `-Zinstrument-coverage` is enabled, the target is linked
against the `library/profiler_builtins` crate (which pulls in LLVM's
compiler-rt runtime).
This option enables backends to specify an alternative runtime crate for
handling injected instrumentation calls.
Introduce the beginning of a THIR unsafety checker
This poses the foundations for the THIR unsafety checker, so that it can be implemented incrementally:
- implements a rudimentary `Visitor` for the THIR (which will definitely need some tweaking in the future)
- introduces a new `-Zthir-unsafeck` flag which tells the compiler to use THIR unsafeck instead of MIR unsafeck
- implements detection of unsafe functions
- adds revisions to the UI tests to test THIR unsafeck alongside MIR unsafeck
This uses a very simple query design, where bodies are unsafety-checked on a body per body basis. This however has some big flaws:
- the unsafety-checker builds the THIR itself, which means a lot of work is duplicated with MIR building constructing its own copy of the THIR
- unsafety-checking closures is currently completely wrong: closures should take into account the "safety context" in which they are created, here we are considering that closures are always a safe context
I had intended to fix these problems in follow-up PRs since they are always gated under the `-Zthir-unsafeck` flag (which is explicitely noted to be unsound).
r? `@nikomatsakis`
cc https://github.com/rust-lang/project-thir-unsafeck/issues/3https://github.com/rust-lang/project-thir-unsafeck/issues/7
Fix `--remap-path-prefix` not correctly remapping `rust-src` component paths and unify handling of path mapping with virtualized paths
This PR fixes#73167 ("Binaries end up containing path to the rust-src component despite `--remap-path-prefix`") by preventing real local filesystem paths from reaching compilation output if the path is supposed to be remapped.
`RealFileName::Named` introduced in #72767 is now renamed as `LocalPath`, because this variant wraps a (most likely) valid local filesystem path.
`RealFileName::Devirtualized` is renamed as `Remapped` to be used for remapped path from a real path via `--remap-path-prefix` argument, as well as real path inferred from a virtualized (during compiler bootstrapping) `/rustc/...` path. The `local_path` field is now an `Option<PathBuf>`, as it will be set to `None` before serialisation, so it never reaches any build output. Attempting to serialise a non-`None` `local_path` will cause an assertion faliure.
When a path is remapped, a `RealFileName::Remapped` variant is created. The original path is preserved in `local_path` field and the remapped path is saved in `virtual_name` field. Previously, the `local_path` is directly modified which goes against its purpose of "suitable for reading from the file system on the local host".
`rustc_span::SourceFile`'s fields `unmapped_path` (introduced by #44940) and `name_was_remapped` (introduced by #41508 when `--remap-path-prefix` feature originally added) are removed, as these two pieces of information can be inferred from the `name` field: if it's anything other than a `FileName::Real(_)`, or if it is a `FileName::Real(RealFileName::LocalPath(_))`, then clearly `name_was_remapped` would've been false and `unmapped_path` would've been `None`. If it is a `FileName::Real(RealFileName::Remapped{local_path, virtual_name})`, then `name_was_remapped` would've been true and `unmapped_path` would've been `Some(local_path)`.
cc `@eddyb` who implemented `/rustc/...` path devirtualisation
This commit implements both the native linking modifiers infrastructure
as well as an initial attempt at the individual modifiers from the RFC.
It also introduces a feature flag for the general syntax along with
individual feature flags for each modifier.
This is necessary for options that should invalidate the incremental
hash but *not* affect the crate hash (e.g. --remap-path-prefix).
This doesn't add `for_crate_hash` to the trait directly because it's not
relevant for *types*, only for *options*, which are fields on a larger
struct. Instead, it adds a new `SUBSTRUCT` directive for options, which
does take a `for_crate_hash` parameter.
- Use TRACKED_NO_CRATE_HASH for --remap-path-prefix
- Add test that `remap_path_prefix` is tracked
- Reduce duplication in the test suite to avoid future churn
- Add back `HirIdVec`, with a comment that it will soon be used.
- Add back `*_region` functions, with a comment they may soon be used.
- Remove `-Z borrowck_stats` completely. It didn't do anything.
- Remove `make_nop` completely.
- Add back `current_loc`, which is used by an out-of-tree tool.
- Fix style nits
- Remove `AtomicCell` with `cfg(parallel_compiler)` for consistency.
coverage bug fixes and optimization support
Adjusted LLVM codegen for code compiled with `-Zinstrument-coverage` to
address multiple, somewhat related issues.
Fixed a significant flaw in prior coverage solution: Every counter
generated a new counter variable, but there should have only been one
counter variable per function. This appears to have bloated .profraw
files significantly. (For a small program, it increased the size by
about 40%. I have not tested large programs, but there is anecdotal
evidence that profraw files were way too large. This is a good fix,
regardless, but hopefully it also addresses related issues.
Fixes: #82144
Invalid LLVM coverage data produced when compiled with -C opt-level=1
Existing tests now work up to at least `opt-level=3`. This required a
detailed analysis of the LLVM IR, comparisons with Clang C++ LLVM IR
when compiled with coverage, and a lot of trial and error with codegen
adjustments.
The biggest hurdle was figuring out how to continue to support coverage
results for unused functions and generics. Rust's coverage results have
three advantages over Clang's coverage results:
1. Rust's coverage map does not include any overlapping code regions,
making coverage counting unambiguous.
2. Rust generates coverage results (showing zero counts) for all unused
functions, including generics. (Clang does not generate coverage for
uninstantiated template functions.)
3. Rust's unused functions produce minimal stubbed functions in LLVM IR,
sufficient for including in the coverage results; while Clang must
generate the complete LLVM IR for each unused function, even though
it will never be called.
This PR removes the previous hack of attempting to inject coverage into
some other existing function instance, and generates dedicated instances
for each unused function. This change, and a few other adjustments
(similar to what is required for `-C link-dead-code`, but with lower
impact), makes it possible to support LLVM optimizations.
Fixes: #79651
Coverage report: "Unexecuted instantiation:..." for a generic function
from multiple crates
Fixed by removing the aforementioned hack. Some "Unexecuted
instantiation" notices are unavoidable, as explained in the
`used_crate.rs` test, but `-Zinstrument-coverage` has new options to
back off support for either unused generics, or all unused functions,
which avoids the notice, at the cost of less coverage of unused
functions.
Fixes: #82875
Invalid LLVM coverage data produced with crate brotli_decompressor
Fixed by disabling the LLVM function attribute that forces inlining, if
`-Z instrument-coverage` is enabled. This attribute is applied to
Rust functions with `#[inline(always)], and in some cases, the forced
inlining breaks coverage instrumentation and reports.
FYI: `@wesleywiser`
r? `@tmandry`
Adjusted LLVM codegen for code compiled with `-Zinstrument-coverage` to
address multiple, somewhat related issues.
Fixed a significant flaw in prior coverage solution: Every counter
generated a new counter variable, but there should have only been one
counter variable per function. This appears to have bloated .profraw
files significantly. (For a small program, it increased the size by
about 40%. I have not tested large programs, but there is anecdotal
evidence that profraw files were way too large. This is a good fix,
regardless, but hopefully it also addresses related issues.
Fixes: #82144
Invalid LLVM coverage data produced when compiled with -C opt-level=1
Existing tests now work up to at least `opt-level=3`. This required a
detailed analysis of the LLVM IR, comparisons with Clang C++ LLVM IR
when compiled with coverage, and a lot of trial and error with codegen
adjustments.
The biggest hurdle was figuring out how to continue to support coverage
results for unused functions and generics. Rust's coverage results have
three advantages over Clang's coverage results:
1. Rust's coverage map does not include any overlapping code regions,
making coverage counting unambiguous.
2. Rust generates coverage results (showing zero counts) for all unused
functions, including generics. (Clang does not generate coverage for
uninstantiated template functions.)
3. Rust's unused functions produce minimal stubbed functions in LLVM IR,
sufficient for including in the coverage results; while Clang must
generate the complete LLVM IR for each unused function, even though
it will never be called.
This PR removes the previous hack of attempting to inject coverage into
some other existing function instance, and generates dedicated instances
for each unused function. This change, and a few other adjustments
(similar to what is required for `-C link-dead-code`, but with lower
impact), makes it possible to support LLVM optimizations.
Fixes: #79651
Coverage report: "Unexecuted instantiation:..." for a generic function
from multiple crates
Fixed by removing the aforementioned hack. Some "Unexecuted
instantiation" notices are unavoidable, as explained in the
`used_crate.rs` test, but `-Zinstrument-coverage` has new options to
back off support for either unused generics, or all unused functions,
which avoids the notice, at the cost of less coverage of unused
functions.
Fixes: #82875
Invalid LLVM coverage data produced with crate brotli_decompressor
Fixed by disabling the LLVM function attribute that forces inlining, if
`-Z instrument-coverage` is enabled. This attribute is applied to
Rust functions with `#[inline(always)], and in some cases, the forced
inlining breaks coverage instrumentation and reports.
This removes all of the code we had in place to work-around LLVM's
handling of forward progress. From this removal excluded is a workaround
where we'd insert a `sideeffect` into clearly infinite loops such as
`loop {}`. This code remains conditionally effective when the LLVM
version is earlier than 12.0, which fixed the forward progress related
miscompilations at their root.
cfg(version): treat nightlies as complete
This PR makes cfg(version) treat the nightlies
for version 1.n.0 as 1.n.0, even though that nightly
version might not have all stabilizations and features
of the released 1.n.0. This is done for greater
convenience for people who want to test a newly
stabilized feature on nightly, or in other words,
give newly stabilized features as many eyeballs
as possible.
For users who wish to pin nightlies, this commit adds
a -Z assume-incomplete-release option that they can
enable if they run into any issues due to this change.
Implements the suggestion in https://github.com/rust-lang/rust/issues/64796#issuecomment-640851454
This commit makes cfg(version) treat the nightlies
for version 1.n.0 as 1.n.0, even though that nightly
version might not have all stabilizations and features
of the released 1.n.0. This is done for greater
convenience for people who want to test a newly
stabilized feature on nightly.
For users who wish to pin nightlies, this commit adds
a -Z assume-incomplete-release option that they can
enable if there are any issues due to this change.
This commit adds a new stable codegen option to rustc,
`-Csplit-debuginfo`. The old `-Zrun-dsymutil` flag is deleted and now
subsumed by this stable flag. Additionally `-Zsplit-dwarf` is also
subsumed by this flag but still requires `-Zunstable-options` to
actually activate. The `-Csplit-debuginfo` flag takes one of
three values:
* `off` - This indicates that split-debuginfo from the final artifact is
not desired. This is not supported on Windows and is the default on
Unix platforms except macOS. On macOS this means that `dsymutil` is
not executed.
* `packed` - This means that debuginfo is desired in one location
separate from the main executable. This is the default on Windows
(`*.pdb`) and macOS (`*.dSYM`). On other Unix platforms this subsumes
`-Zsplit-dwarf=single` and produces a `*.dwp` file.
* `unpacked` - This means that debuginfo will be roughly equivalent to
object files, meaning that it's throughout the build directory
rather than in one location (often the fastest for local development).
This is not the default on any platform and is not supported on Windows.
Each target can indicate its own default preference for how debuginfo is
handled. Almost all platforms default to `off` except for Windows and
macOS which default to `packed` for historical reasons.
Some equivalencies for previous unstable flags with the new flags are:
* `-Zrun-dsymutil=yes` -> `-Csplit-debuginfo=packed`
* `-Zrun-dsymutil=no` -> `-Csplit-debuginfo=unpacked`
* `-Zsplit-dwarf=single` -> `-Csplit-debuginfo=packed`
* `-Zsplit-dwarf=split` -> `-Csplit-debuginfo=unpacked`
Note that `-Csplit-debuginfo` still requires `-Zunstable-options` for
non-macOS platforms since split-dwarf support was *just* implemented in
rustc.
There's some more rationale listed on #79361, but the main gist of the
motivation for this commit is that `dsymutil` can take quite a long time
to execute in debug builds and provides little benefit. This means that
incremental compile times appear that much worse on macOS because the
compiler is constantly running `dsymutil` over every single binary it
produces during `cargo build` (even build scripts!). Ideally rustc would
switch to not running `dsymutil` by default, but that's a problem left
to get tackled another day.
Closes#79361
