Index | Thread | Search

From:
Crystal Kolipe <kolipe.c@exoticsilicon.com>
Subject:
Re: mbuf dma 64 bit
To:
Alexander Bluhm <bluhm@openbsd.org>
Cc:
Mark Kettenis <mark.kettenis@xs4all.nl>, tech@openbsd.org
Date:
Wed, 4 Mar 2026 22:48:41 +0000

Download raw body.

Thread
On Wed, Mar 04, 2026 at 11:02:54PM +0100, Alexander Bluhm wrote:
> What options do we have to get this in?
> 1. Ignore these machines, devices won't work.
> 2. Detect the problem during ifconfig up and print an error.
>    Maybe it works after a reboot if the device is attached during boot.
> 3. Wait for mbuf bounce buffer diff for unsupported devices.
> 4. Wait until someone implements IOMMU.  But this is not
>    available for all machines.
> 
> Question is how relevant are problematic setups for our user base.
> Can we risk to do 1 or 2?

Can't we put some loud diagnostic messages in snaps to test for these possible
conditions and see if anybody speaks up to at least get an idea if it's going
to cause widespread breakage?

This change to enable 64-bit DMA is one of those things that's very difficult
to test on a wide enough variety and combinations of hardware to avoid
breakage finding it's way into the next release.