On 13 Jan 2026 at 11:35:09 AM, Théo Lebrun [off-list ref] wrote:

On Mon Jan 12, 2026 at 7:43 PM CET, Paolo Valerio wrote:

quoted

On 12 Jan 2026 at 03:16:24 PM, Paolo Valerio [off-list ref] wrote:

quoted

On 08 Jan 2026 at 04:43:43 PM, Théo Lebrun [off-list ref] wrote:

quoted

nit: while in macb_init_rx_buffer_size(), can you tweak the debug line
from mtu & rx_buffer_size to also have rx_headroom and total? So that
we have everything available to understand what is going on buffer size
wise. Something like:

-       netdev_dbg(bp->dev, "mtu [%u] rx_buffer_size [%zu]\n",
-                  bp->dev->mtu, bp->rx_buffer_size);
+       netdev_info(bp->dev, "mtu [%u] rx_buffer_size [%zu] rx_headroom [%zu] total [%u]\n",
+                   bp->dev->mtu, bp->rx_buffer_size, bp->rx_headroom,
+                   gem_total_rx_buffer_size(bp));

I missed this before:
I assume so, but just checking, is the promotion from dbg to info also
wanted?

Ah no it was a mistake. I was lazy during my testing: rather than
`#define DEBUG` I changed netdev_dbg() to netdev_info().

I wouldn't mind but that isn't the usual kernel policy wrt to logs.
A working driver should be silent.

np, makes perfectly sense

Thanks,

--
Théo Lebrun, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com