Re: [PATCH net-next] net: phy: aquantia: Poll global status register for provision ID

[Hongmei Xu]

On Thu, Oct 29, 2026 at 9:35 PM Andrew Lunn <andrew@xxxxxxx> wrote:
>
> On Thu, Oct 29, 2026 at 05:28:10PM +0800, Hongmei Xu wrote:
> > After the firmware reset sequence completes, the global reserved status
> > register (VEND1_GLOBAL_RSVD_STAT1) may not yet contain a valid provision
> > ID. This can cause subsequent firmware operations to proceed with stale
> > state.
> >
> > Poll VEND1_GLOBAL_RSVD_STAT1 until the provision ID field is non-zero
> > before returning from aqr_wait_reset_complete(). A timeout is treated as
> > a non-fatal warning since the PHY can still function without a valid
> > provision ID.
>
> If it can still work without the provisioning ID, why bother reading
> it?
This is real issue we observed when working on AQR113 bring up.
Below is the basis why this change is needed.
In a word, it is required accorfing to the AQR113 vendor phy databook.
The reset sequence is describled in the databook:
1.When all of the conditions for release from Reset are true, a 20 ms
timer engages, the purpose
of which is to allow the supplies to settle prior to allowing the PHY to boot.
2.After the 20 ms timer has completed, hardware state machines
designed to guarantee PLL and
band-gap stability engage.
3. After the PLLs and band-gap are locked and functioning properly,
the processor and digital
circuitry are released from reset.
4. The PHY image is then loaded through MDIO boot-load and the processor boots.
5. After the processor boots, any provisioned register values are set,
and the PHY enters the
provisioned operating state.

And the definition for Provisiong id in the phy databook is:
Customers may receive multiple ROM images that differ only in their
provisioning.
This field is used to differentiate those images. This field is used
in conjunction
with the firmware major and minor revision numbers to uniquely
identify ROM images.

Since we flash our own FW images on the AQR113 . Prov id cannot be 0x0 .
What we observed is if we didn't wait for Prov id, the id would coming as 0x0.
That was causing link issues. After this change is applied, link never
fails then.

>
> Maybe extend the commit message saying what features will be missing
> if there is no ID.
>
Sure, will add more detail information in the commit message.

> It looks like USXGMII and 10G-QXGMII might be mixed up? Is such a mix
> up not fatal?
>
>         Andrew
Is this comment mis-posted? No USXGMII and 10G-QXGMII related inforamtion
in this patch, isn't it? Please help to confirm if I have any
misunsterdaning here?

Regards,
Hongmei Xu