Date: Thu, 25 Dec 2003 13:20:47 +0100 From: Andy Polyakov <appro@fy.chalmers.se> To: Chuck Swiger <chuck@pkix.net> Cc: ports@FreeBSD.org Subject: Re: dvd+rw-tools-5.15.4.8.5 and FreeBSD 5-current Message-ID: <3FEAD61F.241FB1C0@fy.chalmers.se> References: <3FE8A7BF.CE892878@fy.chalmers.se> <3FE9D410.5080604@pkix.net>
next in thread | previous in thread | raw e-mail | index | archive | help
> >The kernel deficiency manifests itself by failure to
> >automatically pull sense data for failed ATAPI command, when application
> >most expects it to (as CAM_DIS_AUTOSENSE flag is *not* set).
> >
> Would the symptoms of this problem include returning invalid sense codes
> in the event of a problem?
Yes.
> One user of dvd+rw-tools reported that his
> burner was returning a 0xF sense code (which is reserved and presumably
> not actually being set by the device) under 5.x:
^^^ Well, the problem I'm
referring to is believed to be specific to FreeBSD 5.2 only, at least
downgrading to 5.1 reportedly helps. Meaning that if user in question
can confirm that he experienced the problem under 5.0 or 5.1, then we
might have to have another look, as it must be failing for *another*
reason. But in either case...
> brueffer@cheops:~ # growisofs -Z /dev/cd1 -R -J /mnt/disc4/some/files
> ...
> :-[ LBA=0h, SENSE KEY=Fh/ASC=FFh/ASCQ=FFh ]
F/FF/FF triple is *made up* by growisofs from value of -1, which is
returned when cam_send_ccb failes or sense data remains empty for some
reason (such as kernel fails to auto-pull it:-). It should be mentioned
that in 5.15.x update error reporting was revised and you won't see
these made up F/FF/FF triples anymore, but a generic "Input/output
error" alone. Cheers. A.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3FEAD61F.241FB1C0>