x86/PV: make mmio_ro_emulated_write() static

[PATCH] x86/PV: make mmio_ro_emulated_write() static

Posted by Jan Beulich 4 months, 4 weeks ago

It being non-static and it not having a declaration (anymore) violates
one or more Misra rules, iirc.

Fixes: ff3e5dfa7c2d ("x86/mm: move mmio_ro_emulated_write() to PV only file")
Signed-off-by: Jan Beulich <jbeulich@suse.com>

--- a/xen/arch/x86/pv/ro-page-fault.c
+++ b/xen/arch/x86/pv/ro-page-fault.c
@@ -337,7 +337,7 @@ static int cf_check mmcfg_intercept_writ
     return X86EMUL_OKAY;
 }
 
-int cf_check mmio_ro_emulated_write(
+static int cf_check mmio_ro_emulated_write(
     enum x86_segment seg,
     unsigned long offset,
     void *p_data,

Re: [PATCH] x86/PV: make mmio_ro_emulated_write() static

Posted by Roger Pau Monné 4 months, 4 weeks ago

On Wed, Jun 04, 2025 at 11:22:00AM +0200, Jan Beulich wrote:
> It being non-static and it not having a declaration (anymore) violates
> one or more Misra rules, iirc.
> 
> Fixes: ff3e5dfa7c2d ("x86/mm: move mmio_ro_emulated_write() to PV only file")
> Signed-off-by: Jan Beulich <jbeulich@suse.com>

Acked-by: Roger Pau Monné <roger.pau@citrix.com>

We are not checking for those rules in Eclair I think?  Otherwise this
won't have passed the CI testing.  Or maybe we are just checking
builds with !CONFIG_PV?

Thanks, Roger.

Re: [PATCH] x86/PV: make mmio_ro_emulated_write() static

Posted by Nicola Vetrini 4 months, 4 weeks ago

On 2025-06-04 11:27, Roger Pau Monné wrote:
> On Wed, Jun 04, 2025 at 11:22:00AM +0200, Jan Beulich wrote:
>> It being non-static and it not having a declaration (anymore) violates
>> one or more Misra rules, iirc.
>> 
>> Fixes: ff3e5dfa7c2d ("x86/mm: move mmio_ro_emulated_write() to PV only 
>> file")
>> Signed-off-by: Jan Beulich <jbeulich@suse.com>
> 
> Acked-by: Roger Pau Monné <roger.pau@citrix.com>
> 
> We are not checking for those rules in Eclair I think?  Otherwise this
> won't have passed the CI testing.  Or maybe we are just checking
> builds with !CONFIG_PV?
> 
> Thanks, Roger.

That seems to be the case on x86. On Arm64 there is no override for 
CONFIG_PV [1], but perhaps that is unset by default? I'm not opposed to 
having more variants under analysis, but that is not something I can 
decide on individually.

Thanks,
  Nicola

[1] 
https://gitlab.com/xen-project/hardware/xen/-/blob/staging/automation/gitlab-ci/analyze.yaml?ref_type=heads

-- 
Nicola Vetrini, B.Sc.
Software Engineer
BUGSENG (https://bugseng.com)
LinkedIn: https://www.linkedin.com/in/nicola-vetrini-a42471253