- Subject: Re: [PATCH v2 11/14] block: Rework bio splitting
- From: Vivek Goyal <vgoyal@xxxxxxxxxx>
- Date: Fri, 25 May 2012 14:48:27 -0400
- Cc: axboe@xxxxxxxxx, dm-devel@xxxxxxxxxx, Kent Overstreet <koverstreet@xxxxxxxxxx>, linux-kernel@xxxxxxxxxxxxxxx, tj@xxxxxxxxxx, linux-bcache@xxxxxxxxxxxxxxx, mpatocka@xxxxxxxxxx, yehuda@xxxxxxxxxxxxxxx, linux-fsdevel@xxxxxxxxxxxxxxx, sage@xxxxxxxxxxxx, agk@xxxxxxxxxx, drbd-dev@xxxxxxxxxxxxxxxx
- In-reply-to: <4FBE6823.50904@panasas.com>
- References: <1337817771-25038-1-git-send-email-koverstreet@google.com> <1337817771-25038-12-git-send-email-koverstreet@google.com> <4FBE6823.50904@panasas.com>
- Reply-to: device-mapper development <dm-devel@xxxxxxxxxx>
- User-agent: Mutt/1.5.21 (2010-09-15)
On Thu, May 24, 2012 at 07:56:03PM +0300, Boaz Harrosh wrote:
[..]
> In the split you have a single bio with or without bvects allocation
> should you not let the caller make sure not to set __GFP_WAIT.
>
> For me, inspecting current->bio_list is out of context and a complete
> hack. The caller should take care of it, which has more context.
>
> For example I might want to use split from OSD code where I do
> not use an elevator at all, and current->bio_list could belong
> to a completely different device. (Maybe)
FWIW, I too think that checking for current->bio_list in bio_split() sounds
hackish and it should be left to caller to set right gfp flags. And it should
be commented well.
Thanks
Vivek
--
dm-devel mailing list
dm-devel@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/dm-devel
[DM Crypt]
[Fedora Desktop]
[ATA RAID]
[Fedora Marketing]
[Fedora Packaging]
[Fedora SELinux]
[Yosemite Discussion]
[Yosemite Photos]
[KDE Users]
[Fedora Tools]
[Fedora Docs]