Date: Fri, 17 Dec 1999 12:16:38 +0100 From: Hauke Fath <hf@Melog.DE> To: mjacob@feral.com Cc: Hauke Fath <hf@Melog.DE>, Joerg Wunsch <joerg_wunsch@uriah.heep.sax.de>, freebsd-scsi@FreeBSD.ORG Subject: Re: filemarks? Message-ID: <4.2.2.19991217120438.00c89250@meloghost.melog.de> In-Reply-To: <Pine.BSF.4.05.9912160932120.23583-100000@semuta.feral.com> References: <4.2.2.19991216175352.00b83250@meloghost.melog.de>
next in thread | previous in thread | raw e-mail | index | archive | help
At 09:34 16.12.99 -0800, Matthew Jacob wrote:
>In my opinion, oft stated, it's only devices that cannot determine
>physical eot that need two filemarks. As far as I know, only 1/2" Reel
>tape drives have this property.
Sounds quite sane to me. Better to quirk half-inch reel tape than the rest
of the tape world. ;)
What were the counter arguments? Like you said, userland does not (should
not) care about the details of EOD detection. How does an application learn
about EOD? By an error code?
A while back, I saw you advocating fixed block sizes for QIC drives. Can
you explain your reasons? I have seen many problems with this default
(interoperability) and have hacked the NetBSD st.c to default to variable
blocksize.
hauke
--
Hauke Fath Tangro Software Components GmbH
D-69115 Heidelberg
hf@Tangro.DE Ruf +49-6221-13336-35, Fax -21
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-scsi" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4.2.2.19991217120438.00c89250>