Now that `RUSTC_VERSION_TEXT` exists, use it to rebuild `core` when the
version text changes (which in turn will trigger a rebuild of all the
kernel Rust code).

This also applies to proc macros (which only work with the `rustc` that
compiled them), via the already existing dependency on `core.o`. That
is cleaned up in the next commit.

However, this does not cover host programs written in Rust, which is
the same case in the C side.

This is accomplished by referencing directly the generated file, instead
of using the `fixdep` header trick, since we cannot change the Rust
standard library sources. This is not too much of a burden, since it
only needs to be done for `core`.

Signed-off-by: Miguel Ojeda <ojeda@kernel.org>
---
Masahiro: I used `$(objtree)` here since we still use it in the rest of
this `Makefile`, but please let me know if you prefer otherwise. Thanks!

 rust/Makefile | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/rust/Makefile b/rust/Makefile
index 6c0644b6090c..966743a9ee25 100644
--- a/rust/Makefile
+++ b/rust/Makefile
@@ -370,7 +370,8 @@ $(obj)/core.o: private skip_clippy = 1
 $(obj)/core.o: private skip_flags = -Wunreachable_pub
 $(obj)/core.o: private rustc_objcopy = $(foreach sym,$(redirect-intrinsics),--redefine-sym $(sym)=__rust$(sym))
 $(obj)/core.o: private rustc_target_flags = $(core-cfgs)
-$(obj)/core.o: $(RUST_LIB_SRC)/core/src/lib.rs FORCE
+$(obj)/core.o: $(RUST_LIB_SRC)/core/src/lib.rs \
+    $(wildcard $(objtree)/include/config/RUSTC_VERSION_TEXT) FORCE
 	+$(call if_changed_dep,rustc_library)
 ifneq ($(or $(CONFIG_X86_64),$(CONFIG_X86_32)),)
 $(obj)/core.o: scripts/target.json
--
2.46.0

On Fri, Aug 09, 2024 at 12:11:36AM +0200, Miguel Ojeda wrote:
> Now that `RUSTC_VERSION_TEXT` exists, use it to rebuild `core` when the
> version text changes (which in turn will trigger a rebuild of all the
> kernel Rust code).
> 
> This also applies to proc macros (which only work with the `rustc` that
> compiled them), via the already existing dependency on `core.o`. That
> is cleaned up in the next commit.
> 
> However, this does not cover host programs written in Rust, which is
> the same case in the C side.
> 
> This is accomplished by referencing directly the generated file, instead
> of using the `fixdep` header trick, since we cannot change the Rust
> standard library sources. This is not too much of a burden, since it
> only needs to be done for `core`.
> 
> Signed-off-by: Miguel Ojeda <ojeda@kernel.org>
> ---
> Masahiro: I used `$(objtree)` here since we still use it in the rest of
> this `Makefile`, but please let me know if you prefer otherwise. Thanks!
> 
>  rust/Makefile | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/rust/Makefile b/rust/Makefile
> index 6c0644b6090c..966743a9ee25 100644
> --- a/rust/Makefile
> +++ b/rust/Makefile
> @@ -370,7 +370,8 @@ $(obj)/core.o: private skip_clippy = 1
>  $(obj)/core.o: private skip_flags = -Wunreachable_pub
>  $(obj)/core.o: private rustc_objcopy = $(foreach sym,$(redirect-intrinsics),--redefine-sym $(sym)=__rust$(sym))
>  $(obj)/core.o: private rustc_target_flags = $(core-cfgs)
> -$(obj)/core.o: $(RUST_LIB_SRC)/core/src/lib.rs FORCE
> +$(obj)/core.o: $(RUST_LIB_SRC)/core/src/lib.rs \
> +    $(wildcard $(objtree)/include/config/RUSTC_VERSION_TEXT) FORCE
>  	+$(call if_changed_dep,rustc_library)
>  ifneq ($(or $(CONFIG_X86_64),$(CONFIG_X86_32)),)
>  $(obj)/core.o: scripts/target.json
> --
> 2.46.0

Looks good to me, but I'd prefer something like the suggestion from
Masahiro:
https://lore.kernel.org/linux-kbuild/CAK7LNAQBG0nDupXSgAAk-6nOqeqGVkr3H1RjYaqRJ1OxmLm6xA@mail.gmail.com/

On Sat, Aug 17, 2024 at 3:04 PM Nicolas Schier <nicolas@fjasle.eu> wrote:
>
> Looks good to me, but I'd prefer something like the suggestion from
> Masahiro:
> https://lore.kernel.org/linux-kbuild/CAK7LNAQBG0nDupXSgAAk-6nOqeqGVkr3H1RjYaqRJ1OxmLm6xA@mail.gmail.com/

Do you mean for `core` (here), or for `macros` (in the next commit)?
(if you mean for `core`, please see my reply to Masahiro there).

Thanks for taking a look!

Cheers,
Miguel

On Sat, Aug 17, 2024 at 03:36:30PM +0200, Miguel Ojeda wrote:
> On Sat, Aug 17, 2024 at 3:04 PM Nicolas Schier <nicolas@fjasle.eu> wrote:
> >
> > Looks good to me, but I'd prefer something like the suggestion from
> > Masahiro:
> > https://lore.kernel.org/linux-kbuild/CAK7LNAQBG0nDupXSgAAk-6nOqeqGVkr3H1RjYaqRJ1OxmLm6xA@mail.gmail.com/
> 
> Do you mean for `core` (here), or for `macros` (in the next commit)?
> (if you mean for `core`, please see my reply to Masahiro there).

oops, I missed that 'core' is delivered external.  Sorry for the noise.

Kind regards,
Nicolas

On Sat, Aug 17, 2024 at 5:50 PM Nicolas Schier <nicolas@fjasle.eu> wrote:
>
> oops, I missed that 'core' is delivered external.  Sorry for the noise.

No worries at all, thanks for taking a look! I appreciate the reviews.

Cheers,
Miguel