Date: Sun, 4 Apr 2004 20:09:33 -0400
From: Ben Collins <bcollins@debian.org>
To: "Randy.Dunlap" <rddunlap@osdl.org>
Cc: benh@kernel.crashing.org, marcel.lanz@ds9.ch, linux-kernel@vger.kernel.org
Subject: Re: [PANIC] ohci1394 & copy large files
Message-ID: <20040405000933.GY13168@phunnypharm.org>
References: <20040404141600.GB10378@ds9.ch> <20040404141339.GW13168@phunnypharm.org> <1081119623.1285.121.camel@gaston> <20040404231746.GX13168@phunnypharm.org> <20040404162818.1caa25a9.rddunlap@osdl.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20040404162818.1caa25a9.rddunlap@osdl.org>
User-Agent: Mutt/1.5.5.1+cvs20040105i
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 2259
Lines: 46

On Sun, Apr 04, 2004 at 04:28:18PM -0700, Randy.Dunlap wrote:
> On Sun, 4 Apr 2004 19:17:46 -0400 Ben Collins <bcollins@debian.org> wrote:
> 
> | On Mon, Apr 05, 2004 at 09:00:24AM +1000, Benjamin Herrenschmidt wrote:
> | > On Mon, 2004-04-05 at 00:13, Ben Collins wrote:
> | > > On Sun, Apr 04, 2004 at 04:16:00PM +0200, Marcel Lanz wrote:
> | > > > Since 2.6.4 and still in 2.6.5 I get regurarly a Kernel panic if I try
> | > > > to backup large files (10-35GB) to an external attached disc (200GB/JFS) via ieee1394/sbp2.
> | > > > 
> | > > > Has anyone similar problems ?
> | > > 
> | > > Known issue, fixed in our repo. I still need to sync with Linus once I
> | > > iron one more issue and merge some more patches.
> | > 
> | > Hi Ben !
> | > 
> | > I don't want to be too critical or harsh or whatever, but why don't you
> | > just send such fixes right upstream instead of stacking patches for a
> | > while in your repo ? From my experience, such "batching" of patches is
> | > the _wrong_ thing to do, and typically, there is a major useability
> | > issue with sbp2 that could have been "right" in 2.6.5 final and will not
> | > be (so we'll have to wait what ? 1 or 2 monthes more now to have a
> | > release kernel with a reliable sbp2)
> | 
> | Because the fix was pretty extensive and needed testing. It was
> | potentially more broken that the problem it was fixing. Sending untested
> | patches to Linus is far worse than batching a few up and pushing to him.
> 
> Was (is) it already being tested more extensively in the -mm patches
> before going to Linus?  Should/could be.  E.g., that's what gregkh does,
> and ACPI, etc.

That's what generally happens in our own repo, and if I get the chance
to sync them to my bk tree, then that's what happens with -mm too.
Wasn't the case here since I've been swamped for a little over a week.

-- 
Debian     - http://www.debian.org/
Linux 1394 - http://www.linux1394.org/
Subversion - http://subversion.tigris.org/
WatchGuard - http://www.watchguard.com/
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/