Re: [PATCH v2] PCI/sysfs: Change read permissions for VPD attributes

[Bjorn Helgaas]

On Thu, Nov 21, 2024 at 02:13:01PM +0200, Leon Romanovsky wrote:
> On Thu, Nov 21, 2024 at 01:01:27PM +0100, Jean Delvare wrote:
> > On Wed, 13 Nov 2024 14:59:58 +0200, Leon Romanovsky wrote:
> > > --- a/drivers/pci/vpd.c
> > > +++ b/drivers/pci/vpd.c
> > > @@ -332,6 +332,14 @@ static umode_t vpd_attr_is_visible(struct kobject *kobj,
> > >  	if (!pdev->vpd.cap)
> > >  		return 0;
> > >  
> > > +	/*
> > > +	 * Mellanox devices have implementation that allows VPD read by
> > > +	 * unprivileged users, so just add needed bits to allow read.
> > > +	 */
> > > +	WARN_ON_ONCE(a->attr.mode != 0600);
> > > +	if (unlikely(pdev->vendor == PCI_VENDOR_ID_MELLANOX))
> > > +		return a->attr.mode + 0044;
> ...

> I still didn't lost hope that at some point VPD will be open for read to
> all kernel devices.
> 
> Bjorn, are you ok with this patch? If yes, I'll resend the patch with
> the suggested change after the merge window.

Reading VPD is a fairly complicated dance that only works if the VPD
data is well-formatted, and the benefit of unprivileged access seems
pretty small, so the risk/reward tradeoff for making it unprivileged
for all devices doesn't seem favorable in my mind.

This quirk seems like the least bad option, so I guess I'm ok with it.

Bjorn