| Message ID | 6ca7d9834079b7824f9432799cce2308aa88d159.1696833629.git.nicola.vetrini@bugseng.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | Fix or deviate various instances of missing declarations | expand |
On 09.10.2023 08:54, Nicola Vetrini wrote: > --- a/xen/include/xen/consoled.h > +++ b/xen/include/xen/consoled.h > @@ -12,7 +12,7 @@ size_t consoled_guest_tx(char c); > > #else > > -size_t consoled_guest_tx(char c) { return 0; } > +static inline size_t consoled_guest_tx(char c) { return 0; } Why inline? We do so in headers, but we generally avoid "inline" in .c files. Jan
On 16/10/2023 16:52, Jan Beulich wrote: > On 09.10.2023 08:54, Nicola Vetrini wrote: >> --- a/xen/include/xen/consoled.h >> +++ b/xen/include/xen/consoled.h >> @@ -12,7 +12,7 @@ size_t consoled_guest_tx(char c); >> >> #else >> >> -size_t consoled_guest_tx(char c) { return 0; } >> +static inline size_t consoled_guest_tx(char c) { return 0; } > > Why inline? We do so in headers, but we generally avoid "inline" in > .c files. > > Jan Yes. The file modified is in fact an header.
On 17.10.2023 17:24, Nicola Vetrini wrote: > On 16/10/2023 16:52, Jan Beulich wrote: >> On 09.10.2023 08:54, Nicola Vetrini wrote: >>> --- a/xen/include/xen/consoled.h >>> +++ b/xen/include/xen/consoled.h >>> @@ -12,7 +12,7 @@ size_t consoled_guest_tx(char c); >>> >>> #else >>> >>> -size_t consoled_guest_tx(char c) { return 0; } >>> +static inline size_t consoled_guest_tx(char c) { return 0; } >> >> Why inline? We do so in headers, but we generally avoid "inline" in >> .c files. > > Yes. The file modified is in fact an header. Hmm, how did I not pay attention? Yet then a different question arises: Without the "static inline" I'd expect this to result in a build error from any two .c files including this header. Yet we aren't aware of such a build issue, so I wonder whether the stub is needed in the first place. Jan
On 17/10/2023 18:26, Jan Beulich wrote: > On 17.10.2023 17:24, Nicola Vetrini wrote: >> On 16/10/2023 16:52, Jan Beulich wrote: >>> On 09.10.2023 08:54, Nicola Vetrini wrote: >>>> --- a/xen/include/xen/consoled.h >>>> +++ b/xen/include/xen/consoled.h >>>> @@ -12,7 +12,7 @@ size_t consoled_guest_tx(char c); >>>> >>>> #else >>>> >>>> -size_t consoled_guest_tx(char c) { return 0; } >>>> +static inline size_t consoled_guest_tx(char c) { return 0; } >>> >>> Why inline? We do so in headers, but we generally avoid "inline" in >>> .c files. >> >> Yes. The file modified is in fact an header. > > Hmm, how did I not pay attention? Yet then a different question arises: > Without the "static inline" I'd expect this to result in a build error > from any two .c files including this header. Yet we aren't aware of > such a build issue, so I wonder whether the stub is needed in the first > place. > > Jan This is a good observation. Right now I see only one caller, that is conditioned on both CONFIG_X86 and pv_shim and pv_console, making this stub unused as far as I can tell. It might indeed be a good idea to drop it.
diff --git a/xen/include/xen/consoled.h b/xen/include/xen/consoled.h index fd5d220a8aca..e943d8d48f7b 100644 --- a/xen/include/xen/consoled.h +++ b/xen/include/xen/consoled.h @@ -12,7 +12,7 @@ size_t consoled_guest_tx(char c); #else -size_t consoled_guest_tx(char c) { return 0; } +static inline size_t consoled_guest_tx(char c) { return 0; } #endif /* !CONFIG_PV_SHIM */ #endif /* __XEN_CONSOLED_H__ */
The definition of 'consoled_guest_tx' can be static inline, thereby fixing a violation of MISRA C:2012 Rule 8.4. Fixes: 5ef49f185c2d ("x86/pv-shim: shadow PV console's page for L2 DomU") Signed-off-by: Nicola Vetrini <nicola.vetrini@bugseng.com> Reviewed-by: Stefano Stabellini <sstabellini@kernel.org> --- xen/include/xen/consoled.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)