[PATCH] arch: um: rust: Use the generated target.json again

5 views
Skip to first unread message

David Gow

unread,
May 29, 2024, 5:33:43 AMMay 29
to Rae Moar, Thomas Gleixner, Ingo Molnar, Miguel Ojeda, H . Peter Anvin, Masahiro Yamada, Jamie Cunliffe, Catalin Marinas, Richard Weinberger, Anton Ivanov, Johannes Berg, David Gow, kuni...@googlegroups.com, linux-...@vger.kernel.org, linu...@lists.infradead.org, rust-fo...@vger.kernel.org, x...@kernel.org, Wedson Almeida Filho, Borislav Petkov, Dave Hansen, Alex Gaynor, Boqun Feng, Gary Guo, Björn Roy Baron, Benno Lossin, Andreas Hindborg, Alice Ryhl
The Rust compiler can take a target config from 'target.json', which is
generated by scripts/generate_rust_target.rs. It used to be that all
Linux architectures used this to generate a target.json, but now
architectures must opt-in to this, or they will default to the Rust
compiler's built-in target definition.

This is mostly okay for (64-bit) x86 and UML, except that it can
generate SSE instructions, which we can't use in the kernel. So
re-instate the custom target.json, which disables SSE (and generally
enables the 'soft-float' feature). This fixes the following compile
error:

error: <unknown>:0:0: in function _RNvMNtCs5QSdWC790r4_4core3f32f7next_up float (float): SSE register return with SSE disabled

Fixes: f82811e22b48 ("rust: Refactor the build target to allow the use of builtin targets")
Signed-off-by: David Gow <davi...@google.com>
---
arch/x86/Makefile.um | 1 +
1 file changed, 1 insertion(+)

diff --git a/arch/x86/Makefile.um b/arch/x86/Makefile.um
index 2106a2bd152b..a46b1397ad01 100644
--- a/arch/x86/Makefile.um
+++ b/arch/x86/Makefile.um
@@ -9,6 +9,7 @@ core-y += arch/x86/crypto/
#
ifeq ($(CONFIG_CC_IS_CLANG),y)
KBUILD_CFLAGS += -mno-sse -mno-mmx -mno-sse2 -mno-3dnow -mno-avx
+KBUILD_RUSTFLAGS += --target=$(objtree)/scripts/target.json
KBUILD_RUSTFLAGS += -Ctarget-feature=-sse,-sse2,-sse3,-ssse3,-sse4.1,-sse4.2,-avx,-avx2
endif

--
2.45.1.288.g0e0cd299f1-goog

Boqun Feng

unread,
May 30, 2024, 1:00:50 PMMay 30
to David Gow, Rae Moar, Thomas Gleixner, Ingo Molnar, Miguel Ojeda, H . Peter Anvin, Masahiro Yamada, Jamie Cunliffe, Catalin Marinas, Richard Weinberger, Anton Ivanov, Johannes Berg, kuni...@googlegroups.com, linux-...@vger.kernel.org, linu...@lists.infradead.org, rust-fo...@vger.kernel.org, x...@kernel.org, Wedson Almeida Filho, Borislav Petkov, Dave Hansen, Alex Gaynor, Gary Guo, Björn Roy Baron, Benno Lossin, Andreas Hindborg, Alice Ryhl
On Wed, May 29, 2024 at 05:33:35PM +0800, David Gow wrote:
> The Rust compiler can take a target config from 'target.json', which is
> generated by scripts/generate_rust_target.rs. It used to be that all
> Linux architectures used this to generate a target.json, but now
> architectures must opt-in to this, or they will default to the Rust
> compiler's built-in target definition.
>
> This is mostly okay for (64-bit) x86 and UML, except that it can
> generate SSE instructions, which we can't use in the kernel. So
> re-instate the custom target.json, which disables SSE (and generally
> enables the 'soft-float' feature). This fixes the following compile
> error:
>
> error: <unknown>:0:0: in function _RNvMNtCs5QSdWC790r4_4core3f32f7next_up float (float): SSE register return with SSE disabled
>
> Fixes: f82811e22b48 ("rust: Refactor the build target to allow the use of builtin targets")
> Signed-off-by: David Gow <davi...@google.com>

Reviewed-by: Boqun Feng <boqun...@gmail.com>

Regards,
Boqun

Miguel Ojeda

unread,
Jun 11, 2024, 6:08:45 PMJun 11
to David Gow, Rae Moar, Thomas Gleixner, Ingo Molnar, Miguel Ojeda, H . Peter Anvin, Masahiro Yamada, Jamie Cunliffe, Catalin Marinas, Richard Weinberger, Anton Ivanov, Johannes Berg, kuni...@googlegroups.com, linux-...@vger.kernel.org, linu...@lists.infradead.org, rust-fo...@vger.kernel.org, x...@kernel.org, Wedson Almeida Filho, Borislav Petkov, Dave Hansen, Alex Gaynor, Boqun Feng, Gary Guo, Björn Roy Baron, Benno Lossin, Andreas Hindborg, Alice Ryhl
On Wed, May 29, 2024 at 11:33 AM David Gow <davi...@google.com> wrote:
>
> The Rust compiler can take a target config from 'target.json', which is
> generated by scripts/generate_rust_target.rs. It used to be that all
> Linux architectures used this to generate a target.json, but now
> architectures must opt-in to this, or they will default to the Rust
> compiler's built-in target definition.
>
> This is mostly okay for (64-bit) x86 and UML, except that it can
> generate SSE instructions, which we can't use in the kernel. So
> re-instate the custom target.json, which disables SSE (and generally
> enables the 'soft-float' feature). This fixes the following compile
> error:
>
> error: <unknown>:0:0: in function _RNvMNtCs5QSdWC790r4_4core3f32f7next_up float (float): SSE register return with SSE disabled
>
> Fixes: f82811e22b48 ("rust: Refactor the build target to allow the use of builtin targets")
> Signed-off-by: David Gow <davi...@google.com>

I guess this should go through UML, but please let me know otherwise
(I don't see it in next).

Tested-by: Miguel Ojeda <oj...@kernel.org>
Reviewed-by: Miguel Ojeda <oj...@kernel.org>

Should this have a

Cc: sta...@vger.kernel.org

too?

Thanks!

Cheers,
Miguel

Richard Weinberger

unread,
Jun 13, 2024, 4:25:49 PMJun 13
to Miguel Ojeda, davidgow, Rae Moar, tglx, mingo, Miguel Ojeda, hpa, masahiroy, Jamie Cunliffe, Catalin Marinas, anton ivanov, Johannes Berg, kunit-dev, linux-kernel, linux-um, Rust for Linux Kernel, x86, Wedson Almeida Filho, bp, dave hansen, Alex Gaynor, Boqun Feng, Gary Guo, Björn Roy Baron, Benno Lossin, Andreas Hindborg, Alice Ryhl
----- Ursprüngliche Mail -----
> Von: "Miguel Ojeda" <miguel.oje...@gmail.com>
>> Fixes: f82811e22b48 ("rust: Refactor the build target to allow the use of
>> builtin targets")
>> Signed-off-by: David Gow <davi...@google.com>
>
> I guess this should go through UML, but please let me know otherwise

Yeah, that's the plan.

> Tested-by: Miguel Ojeda <oj...@kernel.org>
> Reviewed-by: Miguel Ojeda <oj...@kernel.org>
>
> Should this have a
>
> Cc: sta...@vger.kernel.org
>
> too?

Ok!

Thanks,
//richard
Reply all
Reply to author
Forward
0 new messages