[PATCH net v3] net: axienet: free outstanding TX buffers in axienet_dma_bd_release()

Daniel Machon daniel.machon at microchip.com
Thu Oct 15 01:03:21 PDT 2026


> On 10/11/26 10:46, Daniel Machon wrote:
> > > On Tue, 6 Oct 2026 06:44:55 +0300 Sagi Maimon wrote:
> > > > Agreed on the helper.  As on v2, I kept the fix minimal for net, and
> > > > will send the helper to net-next once this has reached it, with both
> > > > axienet_dma_bd_release() and axienet_dma_err_handler() using it.
> > > 
> > > Let's do this now, trying to keep fixes small is a bit of
> > > an anti-pattern
> > 
> > Out of curiosity, why is that an anti-pattern?
> > 
> > I'd think that small fixes are, in general, easier to backport to stable. Is this
> > more of procedural question - if the whole fix gets into net, it will get merged
> > into net-next by the end of the week anyway?
> I'm answering here because I can only guess how scarefully long is
> Jakub's email backlog, and possibly this could get lost.

I appreciate that.

> 
> I think the key here is the context. This revision is incorrect because
> in order to be minimal intentionally adds a significant chunk of duplicate
> code, but such duplication must be avoided, the 'keep is small' is not a
> good argument to allow that.
> 
> I guess a one-line alternative implementing a different solution
> (if existing) could have been ok.
> 
> /P
> 

Thanks for the clarification.

/Daniel



More information about the linux-arm-kernel mailing list