Re: [PATCH net-next v2 2/2] net: sfp: add quirks for OEM XGSPONST2001 and FS XGS-SFP-ONT-MACI

On Mon, 10 Aug 2026 17:14:29 -0700, Jakub Kicinski wrote:
> That leaves two possibilities for this module.  Either the PN field really
> does read "XGS-SFP-ONT-MACI" and SFP_QUIRK_F() would have matched just as
> well, so the commit message statement
> [...]
> does not hold for this entry, or the field really does contain garbage,
> which then falls inside the 16 compared bytes and neither prefix nor exact
> matching can ever fire, making the new entry dead code.
>
> Which of the two is it, and could the comment and the commit message be
> adjusted to describe what the code actually does here?

It's the first one: the field really does read "XGS-SFP-ONT-MACI". The
product name is XGS-SFP-ONT-MAC-I; the last hyphen does not fit the
16-byte field, so the field is fully occupied by legitimate characters
and there is nothing left to pad, with garbage or otherwise. I re-read
the module I have in service to make sure:

  vendor  "FS" + 14 spaces (properly padded)
  PN      58 47 53 2d 53 46 50 2d 4f 4e 54 2d 4d 41 43 49  "XGS-SFP-ONT-MACI"

So prefix and exact matching are indeed identical here, and the "Both
modules fail to space-pad" paragraph overclaims. In v3 this entry
becomes a plain SFP_QUIRK_F, and the commit message makes the
garbage-padding argument only for the XGSPONST2001, where it actually
holds: that stick returns the 12 legitimate characters followed by '!'
and non-printable bytes in the PN field on cold power-up (the same
module reads back clean and space-padded after a warm reseat). The
cold boot is exactly where the quirk has to land, otherwise TX_FAULT
is honored and the state machine disables the module.

> Would it be worth either noting in the comments that the vendor field is
> matched as a prefix too, or splitting the flag so prefix semantics only
> apply to the part field?

Splitting it. v3 scopes the flag to the part field (see my reply on
1/2) and the comments follow.

Thanks,
Martino