Re: [PULL REQUEST]: bcache stability patches

[Johannes Thumshirn <jthumshirn@suse.de>]

On Sun, 2015-11-29 at 20:13 -0800, Eric Wheeler wrote:
> Hello all,
> 
> Please pull this bcache branch with stability patches and see if it fixes 
> any issue that you might have!  If you have stability patches (*that you 
> have tested*) which are not part of this branch, then please forward them 
> so they can be included.
> 
> We have been using bcache with great performance and stability for over a 
> year; it is the best SSD caching mechanism that we have found for our 
> workload.  
> 
> I've been collecting these fixes for a while now and just put them up on 
> github so they can be easily merged.  I continue to see mailing list 
> discussions about problems that users are having which these patches 
> address, so hopefully we can get these merged:
>  
>   https://github.com/ewheelerinc/linux/commits/bcache-patches-for-3.17
>   git add remote ewheelerinc https://github.com/ewheelerinc/linux.git
>   git fetch ewheelerinc
>   git merge ewheelerinc/bcache-patches-for-3.17
> 
> This branch is from a clone of Linus's tree circa 3.17-rc so git merge 
> should bring this in cleanly to any later branch.  This will merge cleanly 
> into kernel.org longerm versions v4.1.13 and v3.18.24.  The original 
> author is cited in each commit; you are being Cc'ed in this email if you 
> are an author on any of these patches or are part of a mailing list thread 
> indicating problems that this pull may fix.

I just did a test merge on the patches against today's master and I had no
merge conflicts. The overall diff looks reasonable as well and it builds w/o
any warnings.

Just pulling them straight in won't be a good idea though, as the From: always
points to Eric, that needs to be fixed up (but that's all just mechanics).

> 
> Each commit includes `Cc: stable@vger.kernel.org` so if they get pulled 
> into mainline we will see bcache stable on the newer LTS branches of 
> kernel.org.
> 
> This git branch includes patches that have been collected and tested by 
> Denis Bychkov from his post on 2015-09-16 with one exception: the 
> refill_dirty() patch has been updated to the latest stable version which 
> Kent posted the next day.  The refill_dirty patch addresses issues when 
> bcache is backed by raid5/6 md's causing 
> queue_limits->partial_stripes_expensive to be nonzero.
> 
> 
> Kent,
> 
> Can this be sent to Jens for a pull into mainline?  If you can give your 
> OK on this patchset then I'll approach Jens and see if he can pull this 
> in.  Maybe Cc him in your response.
> 
> If there's anything in here that you would like to omit or change from 
> this pull then let me know and I'll rebase.
> 
> Thanks!
> 
> -Eric
> 

Kent, can you please merge these patches? There are people out there actually
using bcache and these patches aren't just cosmetics.

Thanks a lot for collecting these patches Eric.
	Johannes

> 
> 
> --
> Eric Wheeler, President           eWheeler, Inc. dba Global Linux Security
> 888-LINUX26 (888-546-8926)        Fax: 503-716-3878           PO Box 25107
> www.GlobalLinuxSecurity.pro      ; Linux since 1996!     Portland, OR 97298
> --
> To unsubscribe from this list: send the line "unsubscribe linux-bcache" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html