Open
Conversation
Clippy subtree update r? @Manishearth 1 week late, due to vacation and some technical issues.
…update_cg_gcc_2026-02-14
It was just a dummy implementation to workarround the fact that thin local lto is the default in rustc. By adding a thin_lto_supported thin local lto can be automatically disabled for cg_gcc, removing the need for this dummy implementation. This makes improvements to the LTO handling on the cg_ssa side a lot easier.
This allows us to get rid of box_new entirely
This allows removing the Metadata associated type from BackendTypes.
In Cranelift the regular Type enum can't represent function signatures. Function pointers are represented as plain pointer sized integer.
Previously this was not correctly implemented. Each funclet may need its own terminate block, so this changes the `terminate_block` into a `terminate_blocks` `IndexVec` which can have a terminate_block for each funclet. We key on the first basic block of the funclet -- in particular, this is the start block for the old case of the top level terminate function. Rather than using a catchswitch/catchpad pair, I used a cleanuppad. The reason for the pair is to avoid catching foreign exceptions on MSVC. On wasm, it seems that the catchswitch/catchpad pair is optimized back into a single cleanuppad and a catch_all instruction is emitted which will catch foreign exceptions. Because the new logic is only used on wasm, it seemed better to take the simpler approach seeing as they do the same thing.
In Cranelift a Value can't hold arbitrarily sized values.
Fix: On wasm targets, call `panic_in_cleanup` if panic occurs in cleanup Previously this was not correctly implemented. Each funclet may need its own terminate block, so this changes the `terminate_block` into a `terminate_blocks` `IndexVec` which can have a terminate_block for each funclet. We key on the first basic block of the funclet -- in particular, this is the start block for the old case of the top level terminate function. I also fixed the `terminate` handler to not be invoked when a foreign exception is raised, mimicking the behavior from msvc. On wasm, in order to avoid generating a `catch_all` we need to call `llvm.wasm.get.exception` and `llvm.wasm.get.ehselector`.
Couple of cg_ssa refactorings These should help a bit with using cg_ssa in cg_clif at some point in the future.
This is already CodegenResults without CrateInfo. The driver can calculate the CrateInfo and pass it by-ref to the backend. Using CompiledModules makes it a bit easier to move some other things out of the backend as will be necessary for moving LTO to the link phase.
Make `size`/`align` always correct rather than conditionally on the `safe` field. This makes it less error prone and easier to work with for `MaybeDangling` / potential future pointer kinds like `Aligned<_>`.
Instead of defaulting to `None` it now defaults to `Align::ONE` i.e. no alignment restriction. Codegen test changes are due to us now skipping `align 1` annotations (they are useless; not skipping them makes all the raw pointers gain an `align 1` annotation which doesn't seem any good)
…Jung Implement `MaybeDangling` compiler support Tracking issue: rust-lang/rust#118166 cc @RalfJung
Renaming to remove any ambiguity as to what "vector" refers to in this context
It's defined in `rustc_span::source_map` which doesn't make any sense because it has nothing to do with source maps. This commit moves it to the crate root, a more sensible spot for something this basic.
…uwer Rollup of 7 pull requests Successful merges: - rust-lang/rust#153560 (Introduce granular tidy_ctx's check in extra_checks) - rust-lang/rust#153666 (Add a regression test for rust-lang/rust#153599) - rust-lang/rust#153493 (Remove `FromCycleError` trait) - rust-lang/rust#153549 (tests/ui/binop: add annotations for reference rules) - rust-lang/rust#153641 (Move `Spanned`.) - rust-lang/rust#153663 (Remove `TyCtxt::node_lint` method and `rustc_middle::lint_level` function) - rust-lang/rust#153664 (Add test for rust-lang/rust#109804)
Contributor
Author
|
@bjorn3: We still get the same error. |
Member
|
It now correctly uses fat LTO for cg_gcc afaict. Something seems to have broken fat LTO. There are two diff --git a/src/back/write.rs b/src/back/write.rs
index 24ea2b66..540472ec 100644
--- a/src/back/write.rs
+++ b/src/back/write.rs
@@ -148,7 +148,6 @@ pub(crate) fn codegen(
let path = obj_out.to_str().expect("path to str");
if fat_lto {
- let lto_path = format!("{}.lto", path);
// cSpell:disable
// FIXME(antoyo): The LTO frontend generates the following warning:
// ../build_sysroot/sysroot_src/library/core/src/num/dec2flt/lemire.rs:150:15: warning: type of ‘_ZN4core3num7dec2flt5table17POWER_OF_FIVE_12817ha449a68fb31379e4E’ does not match original declaration [-Wlto-type-mismatch]
@@ -161,24 +160,7 @@ pub(crate) fn codegen(
context.add_driver_option("-Wno-lto-type-mismatch");
// NOTE: this doesn't actually generate an executable. With the above
// flags, it combines the .o files together in another .o.
- context.compile_to_file(OutputKind::Executable, <o_path);
-
- let context = Context::default();
- if cgcx.target_arch == "x86" || cgcx.target_arch == "x86_64" {
- // NOTE: it seems we need to use add_driver_option instead of
- // add_command_line_option here because we use the LTO frontend via gcc.
- context.add_driver_option("-masm=intel");
- }
-
- // NOTE: these two options are needed to invoke LTO to produce an object file.
- // We need to initiate a second compilation because the arguments "-x lto"
- // needs to be at the very beginning.
- context.add_driver_option("-x");
- context.add_driver_option("lto");
- add_pic_option(&context, module.module_llvm.relocation_model);
- context.add_driver_option(lto_path);
-
- context.compile_to_file(OutputKind::ObjectFile, path);
+ context.compile_to_file(OutputKind::Executable, &path);
} else {
// NOTE: this doesn't actually generate an executable. With the above
// flags, it combines the .o files together in another .o.I did have to also apply the following patch (with paths adjusted): https://github.com/rust-lang/rustc_codegen_cranelift/blob/main/patches/0030-sysroot_tests-Add-missing-feature-gate.patch Still need to upstream that one. With those two changes I eventually get an ICE that the |
antoyo
force-pushed
the
sync_from_rust_2026_03_13
branch
from
7d49787 to
cacf74a
Compare
Contributor
Author
This seems to break actual Fat LTO as the CI shows.
This code calls the LTO frontend (via Any idea how to fix Thin LTO while keeping this code? |
Member
For me fat LTO was actually broken without that change. The hello-world test failed with lto=fat. I don't recall having changed anything in the fat lto implementation on the rustc side that could break cg_gcc. |
Contributor
Author
It does work for me. Maybe you forgot the
Fat LTO is not broken by the changes on the rustc side: it does work for me. |
Member
|
I missed that I just noticed that the CI failure is on a job that doesn't support LTO at all. So the testwould never have worked with fat LTO. The only reason it worked with thin LTO previously is because that was effectively disabling LTO. I think the proper fix is to just disable LTO for this test with diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
index 840c0940..2b4a5db5 100644
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@@ -101,7 +101,7 @@ jobs:
- name: Run y.sh cargo build
run: |
- ./y.sh cargo build --manifest-path tests/hello-world/Cargo.toml
+ CARGO_PROFILE_DEV_LTO=no ./y.sh cargo build --manifest-path tests/hello-world/Cargo.toml
- name: Clean
run: | |
Member
|
#864 fixes CI. You may want to squash away the revert commit. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
12 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.