Re: [PATCH v3 10/18] vtd: Use pci_sbdf_t in acpi_parse_dev_scope()

[Jan Beulich <jbeulich@xxxxxxxx>]

On 29.06.2026 19:21, Teddy Astie wrote:
> Use a dedicated pci_sbdf_t struct that we update instead of recreating
> one each time we need it.
> 
> Signed-off-by: Teddy Astie <teddy.astie@xxxxxxxxxx>

Acked-by: Jan Beulich <jbeulich@xxxxxxxx>
albeit preferably with ...

> @@ -1195,8 +1177,9 @@ int cf_check intel_iommu_get_reserved_device_memory(
>  static int __init cf_check parse_rmrr_param(const char *str)
>  {
>      const char *s = str, *cur, *stmp;
> -    unsigned int seg, bus, dev, func, dev_count;
> +    unsigned int dev_count;
>      unsigned long start, end;
> +    pci_sbdf_t sbdf;

... this new variable moved to ...

> @@ -1233,7 +1216,7 @@ static int __init cf_check parse_rmrr_param(const char 
> *str)
>          do {
>              bool def_seg = false;
>  
> -            stmp = parse_pci_seg(s + 1, &seg, &bus, &dev, &func, &def_seg);
> +            stmp = parse_pci_seg(s + 1, &sbdf, &def_seg);
>              if ( !stmp )
>                  return -EINVAL;
>  
> @@ -1242,12 +1225,11 @@ static int __init cf_check parse_rmrr_param(const 
> char *str)
>               * Segment will be replaced with one from first device.
>               */
>              if ( user_rmrrs[nr_rmrr].dev_count && def_seg )
> -                seg = PCI_SEG(user_rmrrs[nr_rmrr].sbdf[0]);
> +                sbdf.seg = PCI_SEG(user_rmrrs[nr_rmrr].sbdf[0]);
>  
>              /* Keep sbdf's even if they differ and later report an error. */
>              dev_count = user_rmrrs[nr_rmrr].dev_count;
> -            user_rmrrs[nr_rmrr].sbdf[dev_count] =
> -               PCI_SBDF(seg, bus, dev, func).sbdf;
> +            user_rmrrs[nr_rmrr].sbdf[dev_count] = sbdf.sbdf;
>  
>              user_rmrrs[nr_rmrr].dev_count++;
>              s = stmp;

... the narrow-most scope it's used in.

Jan