pm24.git/scripts, branch rust-6.12

kbuild: rust: Enable KASAN support

2024-09-16T16:04:37Z

Rust supports KASAN via LLVM, but prior to this patch, the flags aren't set properly. Suggested-by: Miguel Ojeda Signed-off-by: Matthew Maurer Reviewed-by: Andrey Konovalov Link: https://lore.kernel.org/r/20240820194910.187826-4-mmaurer@google.com [ Applied "SW_TAGS KASAN" nit. - Miguel ] Signed-off-by: Miguel Ojeda

kbuild: rust: Define probing macros for rustc

2024-09-16T16:04:37Z

Creates flag probe macro variants for `rustc`. These are helpful because: 1. The kernel now supports a minimum `rustc` version rather than a single version. 2. `rustc` links against a range of LLVM revisions, occasionally even ones without an official release number. Since the availability of some Rust flags depends on which LLVM it has been linked against, probing is necessary. Signed-off-by: Matthew Maurer Link: https://github.com/Rust-for-Linux/linux/pull/1087 Link: https://lore.kernel.org/r/20240820194910.187826-2-mmaurer@google.com Signed-off-by: Miguel Ojeda

kasan: simplify and clarify Makefile

2024-09-16T16:04:32Z

When KASAN support was being added to the Linux kernel, GCC did not yet support all of the KASAN-related compiler options. Thus, the KASAN Makefile had to probe the compiler for supported options. Nowadays, the Linux kernel GCC version requirement is 5.1+, and thus we don't need the probing of the -fasan-shadow-offset parameter: it exists in all 5.1+ GCCs. Simplify the KASAN Makefile to drop CFLAGS_KASAN_MINIMAL. Also add a few more comments and unify the indentation. Signed-off-by: Andrey Konovalov Acked-by: Marco Elver Link: https://lore.kernel.org/r/20240814161052.10374-1-andrey.konovalov@linux.dev Signed-off-by: Miguel Ojeda

rust: cfi: add support for CFI_CLANG with Rust

2024-09-16T15:29:58Z

Make it possible to use the Control Flow Integrity (CFI) sanitizer when Rust is enabled. Enabling CFI with Rust requires that CFI is configured to normalize integer types so that all integer types of the same size and signedness are compatible under CFI. Rust and C use the same LLVM backend for code generation, so Rust KCFI is compatible with the KCFI used in the kernel for C. In the case of FineIBT, CFI also depends on -Zpatchable-function-entry for rewriting the function prologue, so we set that flag for Rust as well. The flag for FineIBT requires rustc 1.80.0 or later, so include a Kconfig requirement for that. Enabling Rust will select CFI_ICALL_NORMALIZE_INTEGERS because the flag is required to use Rust with CFI. Using select rather than `depends on` avoids the case where Rust is not visible in menuconfig due to CFI_ICALL_NORMALIZE_INTEGERS not being enabled. One disadvantage of select is that RUST must `depends on` all of the things that CFI_ICALL_NORMALIZE_INTEGERS depends on to avoid invalid configurations. Alice has been using KCFI on her phone for several months, so it is reasonably well tested on arm64. Signed-off-by: Matthew Maurer Co-developed-by: Alice Ryhl Signed-off-by: Alice Ryhl Reviewed-by: Sami Tolvanen Tested-by: Gatlin Newhouse Acked-by: Kees Cook Acked-by: Peter Zijlstra (Intel) Link: https://lore.kernel.org/r/20240801-kcfi-v2-2-c93caed3d121@google.com [ Replaced `!FINEIBT` requirement with `!CALL_PADDING` to prevent a build error on older Rust compilers. Fixed typo. - Miguel ] Signed-off-by: Miguel Ojeda

kbuild: rust: add `CONFIG_RUSTC_VERSION`

2024-09-05T20:44:18Z

Now that we support several Rust versions, introduce `CONFIG_RUSTC_VERSION` so that it can be used in Kconfig to enable and disable configuration options based on the `rustc` version. The approach taken resembles `pahole`'s -- see commit 613fe1692377 ("kbuild: Add CONFIG_PAHOLE_VERSION"), i.e. a simple version parsing without trying to identify several kinds of compilers, since so far there is only one (`rustc`). However, unlike `pahole`'s, we also print a zero if executing failed for any reason, rather than checking if the command is found and executable (which still leaves things like a file that exists and is executable, but e.g. is built for another platform [1]). An equivalent approach to the one here was also submitted for `pahole` [2]. Link: https://lore.kernel.org/rust-for-linux/CANiq72=4vX_tJMJLE6e+bg7ZECHkS-AQpm8GBzuK75G1EB7+Nw@mail.gmail.com/ [1] Link: https://lore.kernel.org/linux-kbuild/20240728125527.690726-1-ojeda@kernel.org/ [2] Reviewed-by: Nicolas Schier Tested-by: Alice Ryhl Acked-by: Masahiro Yamada Link: https://lore.kernel.org/r/20240902165535.1101978-2-ojeda@kernel.org Signed-off-by: Miguel Ojeda

rust: support arrays in target JSON

2024-08-23T04:25:59Z

Some configuration options such as the supported sanitizer list are arrays. To support using Rust with sanitizers on x86, we must update the target.json generator to support this case. The Push trait is removed in favor of the From trait because the Push trait doesn't work well in the nested case where you are not really pushing values to a TargetSpec. Signed-off-by: Matthew Maurer Signed-off-by: Alice Ryhl Reviewed-by: Gary Guo Tested-by: Gatlin Newhouse Link: https://lore.kernel.org/r/20240730-target-json-arrays-v1-1-2b376fd0ecf4@google.com Signed-off-by: Miguel Ojeda

objtool/kbuild/rust: enable objtool for Rust

2024-08-18T21:34:37Z

Now that we should be `objtool`-warning free, enable `objtool` for Rust too. Before this patch series, we were already getting warnings under e.g. IBT builds, since those would see Rust code via `vmlinux.o`. Tested-by: Alice Ryhl Tested-by: Benno Lossin Reviewed-by: Gary Guo Link: https://lore.kernel.org/r/20240725183325.122827-7-ojeda@kernel.org [ Solved trivial conflict. - Miguel ] Signed-off-by: Miguel Ojeda

x86/rust: support MITIGATION_SLS

2024-08-18T21:34:37Z

Support `MITIGATION_SLS` by enabling the target features that Clang does. Without this, `objtool` would complain if enabled for Rust, e.g.: rust/core.o: warning: objtool: _R...next_up+0x44: missing int3 after ret These should be eventually enabled via `-Ctarget-feature` when `rustc` starts recognizing them (or via a new dedicated flag) [1]. Link: https://github.com/rust-lang/rust/issues/116851 [1] Reviewed-by: Gary Guo Tested-by: Alice Ryhl Tested-by: Benno Lossin Link: https://lore.kernel.org/r/20240725183325.122827-5-ojeda@kernel.org Signed-off-by: Miguel Ojeda

x86/rust: support MITIGATION_RETPOLINE

2024-08-18T21:34:37Z

Support `MITIGATION_RETPOLINE` by enabling the target features that Clang does. The existing target feature being enabled was a leftover from our old `rust` branch, and it is not enough: the target feature `retpoline-external-thunk` only implies `retpoline-indirect-calls`, but not `retpoline-indirect-branches` (see LLVM's `X86.td`), unlike Clang's flag of the same name `-mretpoline-external-thunk` which does imply both (see Clang's `lib/Driver/ToolChains/Arch/X86.cpp`). Without this, `objtool` would complain if enabled for Rust, e.g.: rust/core.o: warning: objtool: _R...escape_default+0x13: indirect jump found in RETPOLINE build In addition, change the comment to note that LLVM is the one disabling jump tables when retpoline is enabled, thus we do not need to use `-Zno-jump-tables` for Rust here -- see commit c58f2166ab39 ("Introduce the "retpoline" x86 mitigation technique ...") [1]: The goal is simple: avoid generating code which contains an indirect branch that could have its prediction poisoned by an attacker. In many cases, the compiler can simply use directed conditional branches and a small search tree. LLVM already has support for lowering switches in this way and the first step of this patch is to disable jump-table lowering of switches and introduce a pass to rewrite explicit indirectbr sequences into a switch over integers. As well as a live example at [2]. These should be eventually enabled via `-Ctarget-feature` when `rustc` starts recognizing them (or via a new dedicated flag) [3]. Cc: Daniel Borkmann Link: https://github.com/llvm/llvm-project/commit/c58f2166ab3987f37cb0d7815b561bff5a20a69a [1] Link: https://godbolt.org/z/G4YPr58qG [2] Link: https://github.com/rust-lang/rust/issues/116852 [3] Reviewed-by: Gary Guo Tested-by: Alice Ryhl Tested-by: Benno Lossin Link: https://github.com/Rust-for-Linux/linux/issues/945 Link: https://lore.kernel.org/r/20240725183325.122827-3-ojeda@kernel.org Signed-off-by: Miguel Ojeda

Merge tag 'rust-fixes-6.11' of https://github.com/Rust-for-Linux/linux

2024-08-16T18:24:06Z

Pull rust fixes from Miguel Ojeda: - Fix '-Os' Rust 1.80.0+ builds adding more intrinsics (also tweaked in upstream Rust for the upcoming 1.82.0). - Fix support for the latest version of rust-analyzer due to a change on rust-analyzer config file semantics (considered a fix since most developers use the latest version of the tool, which is the only one actually supported by upstream). I am discussing stability of the config file with upstream -- they may be able to start versioning it. - Fix GCC 14 builds due to '-fmin-function-alignment' not skipped for libclang (bindgen). - A couple Kconfig fixes around '{RUSTC,BINDGEN}_VERSION_TEXT' to suppress error messages in a foreign architecture chroot and to use a proper default format. - Clean 'rust-analyzer' target warning due to missing recursive make invocation mark. - Clean Clippy warning due to missing indentation in docs. - Clean LLVM 19 build warning due to removed 3dnow feature upstream. * tag 'rust-fixes-6.11' of https://github.com/Rust-for-Linux/linux: rust: x86: remove `-3dnow{,a}` from target features kbuild: rust-analyzer: mark `rust_is_available.sh` invocation as recursive rust: add intrinsics to fix `-Os` builds kbuild: rust: skip -fmin-function-alignment in bindgen flags rust: Support latest version of `rust-analyzer` rust: macros: indent list item in `module!`'s docs rust: fix the default format for CONFIG_{RUSTC,BINDGEN}_VERSION_TEXT rust: suppress error messages from CONFIG_{RUSTC,BINDGEN}_VERSION_TEXT