On Mon, Sep 07, 2026 at 02:49:32PM +0200, Andrew Lunn wrote:
> The same thing twice. Same capitalisation. Is one could actually be
> correct. Please see if you can find the data sheet for this device and
> see if it gives any clues.
I could not get hold of the ASIX data sheet itself, so this is from the
register definitions in the driver rather than from the vendor document.
The register is named in asix.h:
/* AX88772 & AX88178 Medium Mode Register */
and it is written by AX_CMD_WRITE_MEDIUM_MODE (0x1b). So the register
is the "Medium Mode" register, and neither word is a stray duplicate of
the other: the message is naming the register and then repeating "mode"
for it.
The sibling read function words it the way I would expect:
netdev_err(dev->net, "Error reading Medium Status register: %02x\n", ret);
which names the register and stops there.
The other thing that pushed me towards it being redundant is what the
value actually is:
int asix_write_medium_mode(struct usbnet *dev, u16 mode, int in_pm)
...
netdev_err(dev->net, "Failed to write Medium Mode mode to 0x%04x: %02x\n",
mode, ret);
The 0x%04x is 'mode', the value being written, not an address. So the
sentence is "failed to write the Medium Mode to <value>", and the second
"mode" has nothing left to refer to.
The same string is in u-boot's copy of the driver, but that was copied
from Linux, so it is not independent confirmation.
If you would rather not take my reading of it without the data sheet, I
am happy to just drop this one. Otherwise I will resend it as-is in a
later batch of the split series; it is not in the batch I am sending
now.
Thanks,
Hemanth
