X-MimeOLE: Produced By Microsoft Exchange V6.5
Received: by onstor-exch02.onstor.net 
	id <01C7CB1E.2EAD9207@onstor-exch02.onstor.net>; Fri, 20 Jul 2007 14:34:49 -0800
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
Content-class: urn:content-classes:message
Subject: RE: review request
Date: Fri, 20 Jul 2007 14:34:49 -0800
Message-ID: <BB375AF679D4A34E9CA8DFA650E2B04E049DE767@onstor-exch02.onstor.net>
In-Reply-To: <20070720153243.662483fc@ripper.onstor.net>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: review request
Thread-Index: AcfLHeOW47MOgWpDSViPEO4CDiCqHwAAEMMA
References: <20070720145637.65039a66@ripper.onstor.net><BB375AF679D4A34E9CA8DFA650E2B04E049DE747@onstor-exch02.onstor.net> <20070720153243.662483fc@ripper.onstor.net>
From: "Maxim Kozlovsky" <maxim.kozlovsky@onstor.com>
To: "Andy Sharp" <andy.sharp@onstor.com>

ok

>-----Original Message-----
>From: Andy Sharp
>Sent: Friday, July 20, 2007 3:33 PM
>To: Maxim Kozlovsky
>Subject: Re: review request
>
>On Fri, 20 Jul 2007 15:21:57 -0700 "Maxim Kozlovsky"
><maxim.kozlovsky@onstor.com> wrote:
>
>> mpsc.c:
>>
>> 875: This forces the allocation size to be 64K aligned, not the
memory
>> region itself. Either the code or the comment is wrong.
>
>Hmm, again, originally done this way by Dan.  You're right, I can't
>think of why this would make sense, or even why the address, nevermind
>the size would need to be so aligned.
>
>I fixed the comment (?)
>
>> Why did you remove a bunch of lines from mpsc.c? Were they wrong, or
>> just unnecessary?
>
>I think the lines you are talking about were lines that were in Dan's
>version of this driver, but were not in the version that was in the
>current kernel tree.  I added them because I didn't know if they were
>changes required to make the driver work or not.  In the process of
>finding the source of the lockups I removed them and the driver works
>fine, so....
>
>And then there is the one line I removed (commented out?) that
>unnecessarily acks the interrupt, which seems to be the cause of the
>lockups.  Although the incorrect order of unmasking recv interrupts and
>then setting the receive buffer pointers couldn't have helped anything.
>
>>
>> >-----Original Message-----
>> >From: Andy Sharp
>> >Sent: Friday, July 20, 2007 2:57 PM
>> >To: Maxim Kozlovsky
>> >Subject: review request
>> >
>> >The only files that matter are the first 7, the rest are git
>> >repository files, many of which haven't changed and will be
>> >auto-reverted when I do the submit.
>>
