linux-kernel - Re: [PATCH]: PCI Error Recovery: Symbios SCSI device driver

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 for Android: free password hash cracker in your pocket

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

Message-ID: <20061031222631.GQ6360@austin.ibm.com>
Date:	Tue, 31 Oct 2006 16:26:32 -0600
From:	linas@...tin.ibm.com (Linas Vepstas)
To:	James Bottomley <James.Bottomley@...elEye.com>
Cc:	Matthew Wilcox <matthew@....cx>, linux-scsi@...r.kernel.org,
	linux-pci@...ey.karlin.mff.cuni.cz, linux-kernel@...r.kernel.org
Subject: Re: [PATCH]: PCI Error Recovery: Symbios SCSI device driver

On Tue, Oct 31, 2006 at 02:24:01PM -0500, James Bottomley wrote:
> 
> Just for my own edification, what happens on the dual function (dual
> channel) boards?  We have two threads there and two separate I/O
> processors.  I assume a PCI error will kill both, 

Yes.

> do we need to do
> something about this?

I'm not sure, and actually, I have not thought about 
or tested this case for the symbios. 

The answer depends on the h/w design.  On PCI
multi-function cards, the PCI reset callbacks will 
get called for each PCI function. (Each function 
gets to vote/veto how it wants te reset to proceed).

If the hardware supports completely independent scsi 
host initialization for each scsi i/o processor, 
then things should just work. 

If some part of the card init sequence needs to run 
only once, even when there are two i/o processors, then
this needs to be protected against.  I presume that
using  if(0 == PCI_FUNC(pdev->devfn)) is enough to
make sure the hardware initilization is called only once 
for the card -- i.e. by calling it only for PCI
function zero. If there needs to be some additional 
locking to make sure that the card initialization
completes before the i/o processor initialization 
starts ... well, I don't know about that.

--linas

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/