* [PATCH AUTOSEL 6.6 08/20] kbuild: rust: skip -fmin-function-alignment in bindgen flags
[not found] <20241012112715.1763241-1-sashal@kernel.org>
@ 2024-10-12 11:26 ` Sasha Levin
2024-10-12 11:26 ` [PATCH AUTOSEL 6.6 09/20] rust: add intrinsics to fix `-Os` builds Sasha Levin
1 sibling, 0 replies; 4+ messages in thread
From: Sasha Levin @ 2024-10-12 11:26 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Zehui Xu, Alice Ryhl, Neal Gompa, Gary Guo, Miguel Ojeda,
Sasha Levin, alex.gaynor, rust-for-linux
From: Zehui Xu <zehuixu@whu.edu.cn>
[ Upstream commit 869b5016e94eced02f2cf99bf53c69b49adcee32 ]
GCC 14 recently added -fmin-function-alignment option and the
root Makefile uses it to replace -falign-functions when available.
However, this flag can cause issues when passed to the Rust
Makefile and affect the bindgen process. Bindgen relies on
libclang to parse C code, and currently does not support the
-fmin-function-alignment flag, leading to compilation failures
when GCC 14 is used.
This patch addresses the issue by adding -fmin-function-alignment
to the bindgen_skip_c_flags in rust/Makefile. This prevents the
flag from causing compilation issues.
[ Matthew and Gary confirm function alignment should not change
the ABI in a way that bindgen would care about, thus we did
not need the extra logic for bindgen from v2. - Miguel ]
Link: https://lore.kernel.org/linux-kbuild/20240222133500.16991-1-petr.pavlu@suse.com/
Signed-off-by: Zehui Xu <zehuixu@whu.edu.cn>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Reviewed-by: Neal Gompa <neal@gompa.dev>
Reviewed-by: Gary Guo <gary@garyguo.net>
Link: https://lore.kernel.org/r/20240731134346.10630-1-zehuixu@whu.edu.cn
[ Reworded title. - Miguel ]
Signed-off-by: Miguel Ojeda <ojeda@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
rust/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/rust/Makefile b/rust/Makefile
index e5619f25b55ca..060ba8cfa9149 100644
--- a/rust/Makefile
+++ b/rust/Makefile
@@ -288,7 +288,7 @@ bindgen_skip_c_flags := -mno-fp-ret-in-387 -mpreferred-stack-boundary=% \
-fno-reorder-blocks -fno-allow-store-data-races -fasan-shadow-offset=% \
-fzero-call-used-regs=% -fno-stack-clash-protection \
-fno-inline-functions-called-once -fsanitize=bounds-strict \
- -fstrict-flex-arrays=% \
+ -fstrict-flex-arrays=% -fmin-function-alignment=% \
--param=% --param asan-%
# Derived from `scripts/Makefile.clang`.
--
2.43.0
^ permalink raw reply related [flat|nested] 4+ messages in thread* [PATCH AUTOSEL 6.6 09/20] rust: add intrinsics to fix `-Os` builds
[not found] <20241012112715.1763241-1-sashal@kernel.org>
2024-10-12 11:26 ` [PATCH AUTOSEL 6.6 08/20] kbuild: rust: skip -fmin-function-alignment in bindgen flags Sasha Levin
@ 2024-10-12 11:26 ` Sasha Levin
1 sibling, 0 replies; 4+ messages in thread
From: Sasha Levin @ 2024-10-12 11:26 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Miguel Ojeda, Gary Guo, Alice Ryhl, Trevor Gross, Boqun Feng,
Sasha Levin, alex.gaynor, nathan, rust-for-linux, llvm
From: Miguel Ojeda <ojeda@kernel.org>
[ Upstream commit 02dfd63afe65f7bacad543ba2b10f77083ae7929 ]
Alice reported [1] that an arm64 build failed with:
ld.lld: error: undefined symbol: __extendsfdf2
>>> referenced by core.a6f5fc5794e7b7b3-cgu.0
>>> rust/core.o:(<f32>::midpoint) in archive vmlinux.a
>>> referenced by core.a6f5fc5794e7b7b3-cgu.0
>>> rust/core.o:(<f32>::midpoint) in archive vmlinux.a
ld.lld: error: undefined symbol: __truncdfsf2
>>> referenced by core.a6f5fc5794e7b7b3-cgu.0
>>> rust/core.o:(<f32>::midpoint) in archive vmlinux.a
Rust 1.80.0 or later together with `CONFIG_CC_OPTIMIZE_FOR_SIZE=y`
is what triggers it.
In addition, x86_64 builds also fail the same way.
Similarly, compiling with Rust 1.82.0 (currently in nightly) makes
another one appear, possibly due to the LLVM 19 upgrade there:
ld.lld: error: undefined symbol: __eqdf2
>>> referenced by core.20495ea57a9f069d-cgu.0
>>> rust/core.o:(<f64>::next_up) in archive vmlinux.a
>>> referenced by core.20495ea57a9f069d-cgu.0
>>> rust/core.o:(<f64>::next_down) in archive vmlinux.a
Gary adds [1]:
> Usually the fix on rustc side is to mark those functions as `#[inline]`
>
> All of {midpoint,next_up,next_down} are indeed unstable functions not
> marked as inline...
Fix all those by adding those intrinsics to our usual workaround.
[ Trevor quickly submitted a fix to upstream Rust [2] that has already
been merged, to be released in Rust 1.82.0 (2024-10-17). - Miguel ]
Cc: Gary Guo <gary@garyguo.net>
Reported-by: Alice Ryhl <aliceryhl@google.com>
Closes: https://rust-for-linux.zulipchat.com/#narrow/stream/x/topic/x/near/455637364 [1]
Reviewed-by: Trevor Gross <tmgross@umich.edu>
Tested-by: Alice Ryhl <aliceryhl@google.com>
Tested-by: Boqun Feng <boqun.feng@gmail.com>
Reviewed-by: Gary Guo <gary@garyguo.net>
Link: https://github.com/rust-lang/rust/pull/128749 [2]
Link: https://lore.kernel.org/r/20240806150619.192882-1-ojeda@kernel.org
[ Shortened Zulip link. - Miguel ]
Signed-off-by: Miguel Ojeda <ojeda@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
rust/Makefile | 4 ++--
rust/compiler_builtins.rs | 3 +++
2 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/rust/Makefile b/rust/Makefile
index 060ba8cfa9149..9e0939b8fa78b 100644
--- a/rust/Makefile
+++ b/rust/Makefile
@@ -411,8 +411,8 @@ rust-analyzer:
$(if $(KBUILD_EXTMOD),$(extmod_prefix),$(objtree))/rust-project.json
redirect-intrinsics = \
- __addsf3 __eqsf2 __gesf2 __lesf2 __ltsf2 __mulsf3 __nesf2 __unordsf2 \
- __adddf3 __ledf2 __ltdf2 __muldf3 __unorddf2 \
+ __addsf3 __eqsf2 __extendsfdf2 __gesf2 __lesf2 __ltsf2 __mulsf3 __nesf2 __truncdfsf2 __unordsf2 \
+ __adddf3 __eqdf2 __ledf2 __ltdf2 __muldf3 __unorddf2 \
__muloti4 __multi3 \
__udivmodti4 __udivti3 __umodti3
diff --git a/rust/compiler_builtins.rs b/rust/compiler_builtins.rs
index bba2922c6ef77..f14b8d7caf899 100644
--- a/rust/compiler_builtins.rs
+++ b/rust/compiler_builtins.rs
@@ -40,16 +40,19 @@ pub extern "C" fn $ident() {
define_panicking_intrinsics!("`f32` should not be used", {
__addsf3,
__eqsf2,
+ __extendsfdf2,
__gesf2,
__lesf2,
__ltsf2,
__mulsf3,
__nesf2,
+ __truncdfsf2,
__unordsf2,
});
define_panicking_intrinsics!("`f64` should not be used", {
__adddf3,
+ __eqdf2,
__ledf2,
__ltdf2,
__muldf3,
--
2.43.0
^ permalink raw reply related [flat|nested] 4+ messages in thread