[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-users] rebased opensuse 2.6.31 xen patches


  • To: Bruce Edge <bruce.edge@xxxxxxxxx>, xen-users List <xen-users@xxxxxxxxxxxxxxxxxxx>
  • From: Andrew Lyon <andrew.lyon@xxxxxxxxx>
  • Date: Fri, 16 Oct 2009 08:31:49 +0100
  • Cc:
  • Delivery-date: Fri, 16 Oct 2009 00:34:09 -0700
  • Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type; b=it05xtdHPWO10zYx2OljXzB/1KN9xCh92f+6F4fBArWTDpPne4392Td4bRvtiihe1F skaTNp+gnkvdMqgCP8Rg+zUnz1HDQBJYjOr7bmShDiNLSuohut4Z4yyvJjzWnVka8mZN KCgSIn/S2immocQ9THgpzPvKKYxunq/Q28VzI=
  • List-id: Xen user discussion <xen-users.lists.xensource.com>

On Thu, Oct 15, 2009 at 11:01 PM, Bruce Edge <bruce.edge@xxxxxxxxx> wrote:
> On Thu, Oct 15, 2009 at 2:56 AM, Andrew Lyon <andrew.lyon@xxxxxxxxx> wrote:
>> On Thu, Oct 15, 2009 at 9:32 AM, Jan Beulich <JBeulich@xxxxxxxxxx> wrote:
>>>>>> Andrew Lyon <andrew.lyon@xxxxxxxxx> 15.10.09 10:22 >>>
>>>>Thanks Bruce, I've CC'ed Jan in case he wants to merge this fix into his 
>>>>tree.
>>>
>>> No, I won't - this is due to you dropping all our other patches (in the 
>>> given
>>> case, the reason why this works for us is patches.fixes/use-totalram_pages.
>>>
>>> If you drop all our non-Xen patches, you generally will want to also drop
>>> all of patches.xen/xen3-* except for the ones stating "Automatically
>>> created from "patches.kernel.org/*" by xen-port-patches.py", otherwise
>>> you will always have the risk of those depending on something the
>>> respective native patch introduced.
>>>
>>> Jan
>>>
>>>
>>
>> Sorry Jan, I have a script that filters the xen patches exactly as you
>> describe but I had made a small mistake which allowed
>> xen3-use-totalram_pages.patch to slip through.
>>
>> Will be more careful in future.
>>
>> Andy
>>
>
> Andy,
>
> Apologies, I'm not up on OpenSuse kernel patch mechanisms.
>
> Any chance you could post your filter script so we can see what
> patches should actually be applied?
> Or just a list or what needs to skipped.
>

Its very simple, many of the patches that are named xen3* e.g.
xen3-x86_64-unwind-annotations have a comment in them that explains
they were created by a script:

xen3-x86_64-unwind-annotations:Automatically created from
"patches.arch/x86_64-unwind-annotations" by xen-port-patches.py

So unless I include the corresponding patches.arch patch as well we
need to exclude this patch.

Those that are created from a kernel.org patch are ok, for example:

xen3-patch-2.6.18:Automatically created from
"patches.kernel.org/patch-2.6.18" by xen-port-patches.py

I also exclude some other patches that are OpenSuse specific or for
experimental features that most users do not need:

patches.xen/tmem
patches.xen/xen-modular-blktap
patches.xen/xen-blkback-bimodal-suse
patches.xen/xen-blkif-protocol-fallback-hack
patches.xen/xen-swiotlb-heuristics

So far 2.6.31 seems to be extremely stable and performs very well on
my test system, all credit must go to Jan Beulich as he does all the
hard work forward porting the patches, the changes required to make
them apply to vanilla kernel.org are extremely minor and the total
diff is under 6k.

Andy

> What's the purpose of including these additional patches?
>
> Thanks
>
> -Bruce
>
> _______________________________________________
> Xen-users mailing list
> Xen-users@xxxxxxxxxxxxxxxxxxx
> http://lists.xensource.com/xen-users
>

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.