linux-kernel - Re: [PATCH v2 14/15] scsi: ncr53c8xx: replace CCB

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20221109194048.hcs65yzpsgvek73m@tarta.nabijaczleweli.xyz>
Date:   Wed, 9 Nov 2022 20:40:48 +0100
From:   Ahelenia Ziemiańska 
        <nabijaczleweli@...ijaczleweli.xyz>
To:     James Bottomley <jejb@...ux.ibm.com>
Cc:     Jonathan Corbet <corbet@....net>,
        Federico Vaga <federico.vaga@...a.pv.it>,
        Alex Shi <alexs@...nel.org>,
        Yanteng Si <siyanteng@...ngson.cn>,
        Hu Haowen <src.res@...il.cn>,
        "Martin K. Petersen" <martin.petersen@...cle.com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Jiri Slaby <jirislaby@...nel.org>, linux-doc@...r.kernel.org,
        linux-kernel@...r.kernel.org,
        linux-doc-tw-discuss@...ts.sourceforge.net,
        linux-scsi@...r.kernel.org
Subject: Re: [PATCH v2 14/15] scsi: ncr53c8xx: replace CCB_MAGIC with bool
 busy

On Fri, Nov 04, 2022 at 03:57:20PM -0400, James Bottomley wrote:
> On Wed, 2022-11-02 at 00:06 +0100, Ahelenia Ziemiańska wrote:
> [...]
> > @@ -4356,7 +4347,7 @@ static int ncr_queue_command (struct ncb *np,
> > struct scsi_cmnd *cmd)
> >  	*/
> >  
> >  	/* activate this job.  */
> > -	cp->magic		= CCB_MAGIC;
> > +	cp->busy		= true;
> >  
> >  	/*
> >  	**	insert next CCBs into start queue.
> 
> Are you sure this is just an internal magic number?  The way these old
> scripts engines used to work is that the CCB array forms a mailbox and
> the card continually scans the mailbox to see if it has any work to do.
> These magic values are often a signal to the card that the CCB array
> contains a valid entry it hasn't seen and it should work on it, so the
> card might be expecting to see this full 32 bit value.

AFAICT this is an internal flag: in FreeBSD, whence this is copied, this
driver is introduced in 37bd2c9c33d9684c0bf650e02579ebb329724d6a[1],
and AFAICT it's used in exactly the same way as it is here; the manual[2]
doesn't mention it at all; neither do the scripts; and the layout is
vastly different between there and current Linux; so I'm happy to
conclude it's not used by the device, I think?

1: https://cgit.freebsd.org/src/commit/?id=37bd2c9c33d9684c0bf650e02579ebb329724d6a
2: http://bitsavers.trailing-edge.com/components/ncr_symbios/scsi/53C810/53C810_PCI-SCSI_IO_Processor_Data_Manual_May1993.pdf

Download attachment "signature.asc" of type "application/pgp-signature" (834 bytes)