Re: PR/56434 CVS commit: src/sys/dev/pci

[Jason Thorpe <thorpej%me.com@localhost>]

The following reply was made to PR port-alpha/56434; it has been noted by GNATS.

From: Jason Thorpe <thorpej%me.com@localhost>
To: gnats-bugs%netbsd.org@localhost
Cc: martin%netbsd.org@localhost,
 gnats-admin%netbsd.org@localhost,
 netbsd-bugs%netbsd.org@localhost,
 Andreas Gustafsson <gson%gson.org@localhost>
Subject: Re: PR/56434 CVS commit: src/sys/dev/pci
Date: Mon, 20 Nov 2023 13:19:16 -0800

 > On Nov 20, 2023, at 9:20 AM, Andreas Gustafsson <gson%gson.org@localhost> wrote:
 >=20
 > There are fewer kernel messages than before (for example, the "lost
 > interrupt" message is gone), but the kernel is still reporting a DMA
 > error, and sysinst still appears to silently hang once the disk error
 > has occurred.
 
 I expected it to report a DMA error, but I did not expect it to hang.  =
 Turns out, the =E2=80=9Cwd=E2=80=9D code just retries the request on a =
 DMA error.  That seems to be not a very smart idea, but maybe it=E2=80=99s=
  a good thing to do on some systems?  Anyway, I will add some additional =
 error codes so that =E2=80=9Cwd=E2=80=9D can correctly determine that =
 this error is fatal and a retry should not occur.
 
 -- thorpej