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

[Hongmei Xu]

On Tue, Nov 10, 2026 at 10:41 PM Andrew Lunn <andrew@xxxxxxx> wrote:
>
> > > 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?
>
> The provisioning ID is used to determine USXGMII vs 10G-QXGMII. If you
> don't have the ID, they get mixed up.
Totally got your point now.
The chain is:
aqr_wait_reset_complete() returns 0 on timeout →
aqr_build_fingerprint() still runs but reads a stale/zero prov_id →
aqr_translate_interface() fingerprint won't match
AQR_G3_V4_3_C_AQR_NXP_SPF_30841_MUSX_ID40019_VER1198 → USXGMII stays
as-is instead of being translated to 10G-QXGMII for AQR412C/MUSX.
So for AQR412C with MUSX firmware, a timeout is genuinely fatal
because the interface mode ends up wrong.
Will change the commit message to describe this consequence, and add a
phydev_warn on timeout so users have a visible log hint that interface
mode detection may be incorrect. I'll still not fail the probe since
the existing caller already ignores the timeout — but the warning
makes the failure visible.

Regards,
Hongmei Xu