X-MimeOLE: Produced By Microsoft Exchange V6.5
Received: by onstor-exch02.onstor.net 
	id <01C842A2.7463F512@onstor-exch02.onstor.net>; Wed, 19 Dec 2007 16:51:28 -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 cl#27051
Date: Wed, 19 Dec 2007 16:51:28 -0800
Message-ID: <BB375AF679D4A34E9CA8DFA650E2B04E072E0972@onstor-exch02.onstor.net>
In-Reply-To: <20071219115735.75b53120@ripper.onstor.net>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: review cl#27051
Thread-Index: AchCeWYGP6hHzbT3ROqqyD0G2rQnlwAKPSJw
References: <20071218155128.1af30098@ripper.onstor.net><BB375AF679D4A34E9CA8DFA650E2B04E072E057F@onstor-exch02.onstor.net><20071219114623.1cd93b72@ripper.onstor.net><BB375AF679D4A34E9CA8DFA650E2B04E072E0659@onstor-exch02.onstor.net> <20071219115735.75b53120@ripper.onstor.net>
From: "Maxim Kozlovsky" <maxim.kozlovsky@onstor.com>
To: "Andy Sharp" <andy.sharp@onstor.com>

Ok=20

>-----Original Message-----
>From: Andy Sharp
>Sent: Wednesday, December 19, 2007 11:58 AM
>To: Maxim Kozlovsky
>Subject: Re: review cl#27051
>
>On Wed, 19 Dec 2007 11:47:33 -0800 "Maxim Kozlovsky"
><maxim.kozlovsky@onstor.com> wrote:
>
>> I mean if it is not aligned there is the code which reads the last
>> byte, you need to flush the cache for the last byte as well.
>
>Let me put it another way, at this level, buflen is always 512 when we
>get here from the paging subsystem, so this will always flush the right
>amount, as long as you agree that "the right amount" might be too much.
>If we get here from any other path, we don't care, and would really
>rather not do the flushing, but, well, it ain't great, I know.
>
>> >-----Original Message-----
>> >From: Andy Sharp
>> >Sent: Wednesday, December 19, 2007 11:46 AM
>> >To: Maxim Kozlovsky
>> >Subject: Re: review cl#27051
>> >
>> >On Wed, 19 Dec 2007 10:20:40 -0800 "Maxim Kozlovsky"
>> ><maxim.kozlovsky@onstor.com> wrote:
>> >
>> >> Libata-core.c:
>> >>
>> >> This looks like completely wrong place to do cache flush. The
cache
>> is
>> >> not flushed for the remaining byte for some reason.
>> >
>> >No this is not the best, but it is a work around.  If you have a
>> >better place to flush, please let us all in on it.
>> >
>> >What do you mean "the remaining byte"?
>> >
>> >>
>> >> >-----Original Message-----
>> >> >From: Andy Sharp
>> >> >Sent: Tuesday, December 18, 2007 3:51 PM
>> >> >To: Maxim Kozlovsky
>> >> >Subject: review cl#27051
>> >> >
>> >> >Change 27051 by andys@ripper on 2007/12/18 13:57:40 *pending*
>> >> >
>> >> >        Changes to make CF cards pageable and bootable.
>> >> >        Fix minor whitespace problem in cougar/prom.c.
>> >> >        Add 1125 to list of processors to check for proper
>> >> >        revision setting in config file.
>> >> >
>> >> >Affected files ...
>> >> >
>> >> >... //depot/dev/linux/kernel/linux-mips-
>> >> >2.6/arch/mips/onstor/cougar/prom.c#2 edit
>> >> >... //depot/dev/linux/kernel/linux-mips-
>> >> >2.6/arch/mips/sibyte/sb1250/setup.c#2 edit
>> >> >... //depot/dev/linux/kernel/linux-mips-2.6/cougar-config#4 edit
>> >> >...
>> //depot/dev/linux/kernel/linux-mips-2.6/drivers/ata/libata-core.c#2
>> >> >edit
>> >>
