Re: next-20250627: IOMMU DMA warning during NVMe I/O completion after 06cae0e3f61c

[Ben Copeland]

On Thu, 3 Jul 2025 at 15:29, Keith Busch <kbusch@xxxxxxxxxx> wrote:
>
> On Thu, Jul 03, 2025 at 11:30:42AM +0200, Christoph Hellwig wrote:
> > I think the idea to reconstruct the dma addresses from PRPs should
> > be considered a failure by now.  It works fine for SGLs, but for
> > PRPs we're better off just stashing them away.  Bob, can you try
>
> s/Bob/Ben
>
> > something like the patch below?  To be fully safe it needs a mempool,
> > and it could use some cleanups, but it does pass testing on my setup
> > here, so I'd love to see if if fixes your issue.

I have tested it on my system and can no longer see the regression.

Happy to retest when the patch goes through.

Tested-by: Ben Copeland <ben.copeland@xxxxxxxxxx>

Thank you!


Ben

>
> Thanks for confirming.
>
> While this is starting to look a bit messy, I believe it's still an
> overall win: you've cut down the vector walking in the setup path from 3
> to 1, which reduces a non-trivial amount of overhead for even moderately
> sized IO.