[8846] in linux-scsi channel archive
Re: Request splits
daemon@ATHENA.MIT.EDU (Eric Youngdale)
Thu May 18 10:52:19 2000
Message-ID: <00c201bfc0d7$56664db0$4d0310ac@fairfax.datafocus.com>
From: "Eric Youngdale" <eric@andante.org>
To: "Alan Cox" <alan@lxorguk.ukuu.org.uk>
Cc: "Douglas Gilbert" <dgilbert@interlog.com>,
"Brian Pomerantz" <bapper@piratehaven.org>,
<linux-scsi@vger.rutgers.edu>
Date: Thu, 18 May 2000 10:42:48 -0400
MIME-Version: 1.0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
----- Original Message -----
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
To: Eric Youngdale <eric@andante.org>
Cc: Douglas Gilbert <dgilbert@interlog.com>; Brian Pomerantz
<bapper@piratehaven.org>; <linux-scsi@vger.rutgers.edu>
Sent: Thursday, May 18, 2000 9:50 AM
Subject: Re: Request splits
> > I don't think there are other issues with increasing the limit. For
> > that matter, I am not entirely sure why there is this arbitrary limit in
the
> > code here. In other words, I am not sure if I even added it - it may
have
> > gotten added by someone else as part of the ll_rw_blk cleanups.
>
> The sglist itself presumably has a size limit and we would need to keep
the
> sg list allocation under a page.
Even this is correctly handled, I think. It looks like we allocate from
the DMA safe pool, and handle arbitrary lengths.
SCpnt->sglist_len = (SCpnt->use_sg * sizeof(struct scatterlist) + 511) &
~511;
sgpnt = (struct scatterlist *) scsi_malloc(SCpnt->sglist_len);
-Eric
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.rutgers.edu
