On Mon, May 5, 2025 at 11:34 AM Michael Chan [off-list ref] wrote:
The patch works. I forced a smaller TX_MAX_FRAGS to easily test it.
But we now skip the warn_once() above since we intercept it earlier.
Should we add a warn_once() here also? Thanks.
I do not think a warn_once() is needed here, GSO is pretty fast, in
the unlikely case this would be hit.
You could add a per tx queue counter, that ethtool -S could fetch for
curious minds.