Re: [BUG] x86_64 pci_map_sg modifies sg list - fails multiple map/unmaps

[Andi Kleen]

On 05 Jan 2004 18:05:47 -0600
James Bottomley <James.Bottomley@xxxxxxxxxxxx> wrote:

> On Mon, 2004-01-05 at 15:31, Andi Kleen wrote:
> > For the sake of bug-to-bug compatibility to the SCSI layer this patch may
> > work. I haven't tested it so no guarantees if it won't eat your file systems.
> > Feedback welcome anyways.
> 
> This isn't a bug in SCSI, it's a deliberate design feature.  SCSI has
> certain events, like QUEUE full that cause us to re-queue the pending
> I/O.  Other block layer drivers that can get these EAGAIN type queueing
> problems from the device also follow this model.

It's ok. I fixed the code now[1] If you have other undocumented requirements
you should document them though, otherwise there may be more problems.
Since merging is disabled by default now it won't trigger anyways.

[1] will be in next merge after testing, the last patch i posted still
had one bug.

> As to the idempotence of map/unmap: I'm ambivalent.  If it's going to be
> a performance hit to return the sg list to its prior state in unmap,
> then it does seem a waste given that for most of our I/O transactions we
> simply free the sg list after the unmap.

With the evil dma_length trick it is actually near zero cost.

> 1. Fix the x86_64 mapping layer as your patch proposes (how much of a
> performance hit on every transaction will this be)?

I don't expect a significant performance hit.

-Andi
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/