add to Debian

add to Debian

[James Harper]

Any suggestions on the best way to get bcache onto a Debian Wheezy system?

Would the current patches apply cleanly against 3.2?

James

Re: add to Debian

[Joseph Glanville]

On 7 August 2012 15:39, James Harper <james.harper-NMzNsA1hOHcW+bLBXbPJGg@public.gmane.org> wrote:
> Any suggestions on the best way to get bcache onto a Debian Wheezy system?
>
> Would the current patches apply cleanly against 3.2?
>
> James
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-bcache" in
> the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Hi James,

Yes the current patches apply cleanly afaik. :)

There is still an outstanding bug with relation to Xen's blkback that
I haven't had time to debug further yet that you might run into.

Joseph.

-- 
CTO | Orion Virtualisation Solutions | www.orionvm.com.au
Phone: 1300 56 99 52 | Mobile: 0428 754 846

RE: add to Debian

[James Harper]

> Hi James,
> 
> Yes the current patches apply cleanly afaik. :)
> 
> There is still an outstanding bug with relation to Xen's blkback that I haven't
> had time to debug further yet that you might run into.
> 

What are the circumstances under which this bug appears? I'm planning on doing this initially:

sd[ab]
md0
lvm
bcache (per lv)

for testing, and then exposing the cached device to xen. Assuming testing goes well I'll do it like this:

sd[ab]
md0
bcache
lvm

with a bit of reworking to create a non-bcache'd /boot raid unless grub2 knows all about bcache, or if bcache is transparent to grub2.

James

Re: add to Debian

[Joseph Glanville]

On 7 August 2012 16:16, James Harper <james.harper-NMzNsA1hOHcW+bLBXbPJGg@public.gmane.org> wrote:
>> Hi James,
>>
>> Yes the current patches apply cleanly afaik. :)
>>
>> There is still an outstanding bug with relation to Xen's blkback that I haven't
>> had time to debug further yet that you might run into.
>>
>
> What are the circumstances under which this bug appears? I'm planning on doing this initially:
>
> sd[ab]
> md0
> lvm
> bcache (per lv)
>
> for testing, and then exposing the cached device to xen. Assuming testing goes well I'll do it like this:
>
> sd[ab]
> md0
> bcache
> lvm
>
> with a bit of reworking to create a non-bcache'd /boot raid unless grub2 knows all about bcache, or if bcache is transparent to grub2.
>
> James

That is precisely the setup I am currently using.

I haven't done much debugging yet unfortunately but I can't get VMs to
boot using PV disk drivers (they work fine using QEMU).

I think it has something to do with DM/blkback interactions with
bcaches bio splitting code, the bug only emerged recently and older
versions of bcache work fine.

Joseph.

-- 
CTO | Orion Virtualisation Solutions | www.orionvm.com.au
Phone: 1300 56 99 52 | Mobile: 0428 754 846

RE: add to Debian

[James Harper]

> 
> On 7 August 2012 16:16, James Harper <james.harper-NMzNsA1hOHcW+bLBXbPJGg@public.gmane.org>
> wrote:
> >> Hi James,
> >>
> >> Yes the current patches apply cleanly afaik. :)
> >>
> >> There is still an outstanding bug with relation to Xen's blkback that
> >> I haven't had time to debug further yet that you might run into.
> >>
> >
> > What are the circumstances under which this bug appears? I'm planning on
> doing this initially:
> >
> > sd[ab]
> > md0
> > lvm
> > bcache (per lv)
> >
> > for testing, and then exposing the cached device to xen. Assuming testing
> goes well I'll do it like this:
> >
> > sd[ab]
> > md0
> > bcache
> > lvm
> >
> > with a bit of reworking to create a non-bcache'd /boot raid unless grub2
> knows all about bcache, or if bcache is transparent to grub2.
> >
> > James
> 
> That is precisely the setup I am currently using.
> 
> I haven't done much debugging yet unfortunately but I can't get VMs to boot
> using PV disk drivers (they work fine using QEMU).
> 
> I think it has something to do with DM/blkback interactions with bcaches bio
> splitting code, the bug only emerged recently and older versions of bcache
> work fine.
> 

If I get time to get it all going I'll have a look. If it's blkback interaction problems I should be able to help track it down, although it might be a few days before I can actually get to look at it.

James

RE: add to Debian

[James Harper]

> 
> Hi James,
> 
> Yes the current patches apply cleanly afaik. :)
> 
> There is still an outstanding bug with relation to Xen's blkback that I haven't
> had time to debug further yet that you might run into.
> 

What is the best way to get a diff against the current linux head?

Thanks

James

Re: add to Debian

[Joseph Glanville]

On 7 August 2012 21:30, James Harper <james.harper-NMzNsA1hOHcW+bLBXbPJGg@public.gmane.org> wrote:
>>
>> Hi James,
>>
>> Yes the current patches apply cleanly afaik. :)
>>
>> There is still an outstanding bug with relation to Xen's blkback that I haven't
>> had time to debug further yet that you might run into.
>>
>
> What is the best way to get a diff against the current linux head?
>
> Thanks
>
> James
> --
> To unsubscribe from this list: send the line "unsubscribe linux-bcache" in
> the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

The bcache tree is rebased on Linus's tree every now and then.
Generally there is a bunch of branches that refer to the relevant
releases/backports of the bcache code.
Linus's tags are still intact though so you can do a git diff
<tagname> or any other <refspec>

For 3.2 it would be something along the lines of this:

git clone http://evilpiepirate.org/git/linux-bcache.git
git checkout bcache-3.2
git diff v3.2


-- 
CTO | Orion Virtualisation Solutions | www.orionvm.com.au
Phone: 1300 56 99 52 | Mobile: 0428 754 846

RE: add to Debian

[James Harper]

> 
> The bcache tree is rebased on Linus's tree every now and then.
> Generally there is a bunch of branches that refer to the relevant
> releases/backports of the bcache code.
> Linus's tags are still intact though so you can do a git diff <tagname> or any
> other <refspec>
> 
> For 3.2 it would be something along the lines of this:
> 
> git clone http://evilpiepirate.org/git/linux-bcache.git
> git checkout bcache-3.2
> git diff v3.2
> 

Is the bcache-3.2 branch as up to date (wrt bcache code) as the bcache branch?

Thanks

James

Re: add to Debian

[Joseph Glanville]

On 7 August 2012 22:15, James Harper <james.harper-NMzNsA1hOHcW+bLBXbPJGg@public.gmane.org> wrote:
>>
>> The bcache tree is rebased on Linus's tree every now and then.
>> Generally there is a bunch of branches that refer to the relevant
>> releases/backports of the bcache code.
>> Linus's tags are still intact though so you can do a git diff <tagname> or any
>> other <refspec>
>>
>> For 3.2 it would be something along the lines of this:
>>
>> git clone http://evilpiepirate.org/git/linux-bcache.git
>> git checkout bcache-3.2
>> git diff v3.2
>>
>
> Is the bcache-3.2 branch as up to date (wrt bcache code) as the bcache branch?

Atm they are not in sync, The 3.2 code is usually pretty stable
though, at some point in time Kent will rebase his changes onto 3.2.

>
> Thanks
>
> James
> --
> To unsubscribe from this list: send the line "unsubscribe linux-bcache" in
> the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html



-- 
CTO | Orion Virtualisation Solutions | www.orionvm.com.au
Phone: 1300 56 99 52 | Mobile: 0428 754 846

Re: add to Debian

[gnehzuil.lzheng-Re5JQEeQqe8AvxtiuMwx3w]

On 08/07/2012 08:12 PM, Joseph Glanville wrote:
> On 7 August 2012 21:30, James Harper <james.harper-NMzNsA1hOHcW+bLBXbPJGg@public.gmane.org> wrote:
>>>
>>> Hi James,
>>>
>>> Yes the current patches apply cleanly afaik. :)
>>>
>>> There is still an outstanding bug with relation to Xen's blkback that I haven't
>>> had time to debug further yet that you might run into.
>>>
>>
>> What is the best way to get a diff against the current linux head?
>>
>> Thanks
>>
>> James
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-bcache" in
>> the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 
> The bcache tree is rebased on Linus's tree every now and then.
> Generally there is a bunch of branches that refer to the relevant
> releases/backports of the bcache code.
> Linus's tags are still intact though so you can do a git diff
> <tagname> or any other <refspec>
> 
> For 3.2 it would be something along the lines of this:
> 
> git clone http://evilpiepirate.org/git/linux-bcache.git

Hi Joseph,

I cannot get bcache repo from this url.  Could you please tell me where
can I get it in another url?  Thanks.

Regards,
Zheng