Re: [PATCH v4 2/5] vpci: use a local variable for the BAR rangeset

[Jason Andryuk <jason.andryuk@amd.com>]

On 2026-10-15 10:08, Mykyta Poturai wrote:
> From: Mykyta Poturai <Mykyta_Poturai@epam.com>
> 
> In the BAR (un)mapping paths latch the rangeset of the BAR being
> processed into a local variable, instead of dereferencing struct
> vpci_bar on every use. Preparatory work for moving the rangesets out
> of struct vpci_bar.
> 
> No functional change.
> 
> Signed-off-by: Mykyta Poturai <mykyta_poturai@epam.com>
> Signed-off-by: Stewart Hildebrand <stewart.hildebrand@amd.com>

I guess you need a second S-o-B after Stewart's since you are 
submitting?  Unless...

> ---
> v3->v4:
> * new patch, split out of "vpci: Use pervcpu ranges for BAR mapping" so
>    that the non-functional churn is separate from the functional change

... it is something totally created by you.  Then you can drop Stewart - 
I think.
> @@ -218,8 +219,7 @@ bool vpci_process_pending(struct vcpu *v)
>           {
>               spin_lock(&pdev->vpci->lock);
>               /* Disable memory decoding unconditionally on failure. */
> -            modify_decoding(pdev, v->vpci.cmd & ~PCI_COMMAND_MEMORY,
> -                            false);
> +            modify_decoding(pdev, v->vpci.cmd & ~PCI_COMMAND_MEMORY, false);

This seems like an unrelated change.  I think you should drop it and 
incorporate into the next patch touching this line.

With that:

Reviewed-by: Jason Andryuk <jason.andryuk@amd.com>