Re: kern/55965: st(4) on adaptec fails to 'mt erase'

[Manuel Bouyer <bouyer%antioche.eu.org@localhost>]

The following reply was made to PR kern/55965; it has been noted by GNATS.

From: Manuel Bouyer <bouyer%antioche.eu.org@localhost>
To: Hauke Fath <hauke%Espresso.Rhein-Neckar.DE@localhost>
Cc: gnats-bugs%NetBSD.org@localhost, kern-bug-people%NetBSD.org@localhost, gnats-admin%NetBSD.org@localhost
Subject: Re: kern/55965: st(4) on adaptec fails to 'mt erase'
Date: Fri, 29 Jan 2021 22:43:11 +0100

 On Fri, Jan 29, 2021 at 10:00:34PM +0100, Hauke Fath wrote:
 > On Fri, 29 Jan 2021 19:55:10 +0100, Manuel Bouyer wrote:
 > > What's strange here is that the timeout seems to happen way too fast
 > > (the command should have a 3h timeout).
 > 
 > Four hours for the full erase, three minutes for the short one (which 
 > works).
 > 
 > The full erase errors out within less than a minute, though.
 > 
 > > The first thing to check is the timeout value, and if it matches what the
 > > driver did ask.
 > 
 > Where would I look? I got a mental stack overflow trying to trace the 
 > scsipi call sequence from an st 'erase' ioctl...
 
 I would start from the ahc driver.
 In aic7xxx_osm.c, ahc_timeout(), you could print the value of
 scb->xs->timeout. If the value looks good, the issue is probably in the ahc
 driver and you'll have to find where the callout has been scheduled.
 If the value is wrong, then maybe a Debugger() call (to get a stack trace)
 when this specific command appears in ADAPTER_REQ_RUN_XFER can help.
 
 -- 
 Manuel Bouyer <bouyer%antioche.eu.org@localhost>
      NetBSD: 26 ans d'experience feront toujours la difference
 --