Bug#850425: Debian bug #850425 - mpt3sas "swiotlb buffer is full" problem only under Xen

2017-01-16 Thread Ben Hutchings
On Mon, 2017-01-16 at 11:47 +, Ian Jackson wrote:
> Andy Smith writes ("Re: Bug#850425: Debian bug #850425 - mpt3sas "swiotlb 
> buffer is full" problem only under Xen"):
> > On Mon, Jan 09, 2017 at 06:01:14PM +, Ian Jackson wrote:
> > > Patches only make it into Linux upstream stable releases if someone
> > > pushes to get them in.
> > 
> > Do you have any advice how I could push to get this to happen?
> > 
> > Obviously I need this in order for Xen to be usable on my hardware
> > but we're talking mpt3sas driver here which is pretty common, so I
> > think it is quite desirable.
> 
> Right.  Well, you need to post to linux-kernel saying "this patch
> fixes such-and-such".  You should:
> 
> 1. git clone or git fetch the current master, which is
>  git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git#master
> and check that it doesn't have the patch.

It is better to use the subsystem's master branch.  For many subsystems
this is listed in MAINTAINERS; otherwise 'git log --merges directory/'
should provide a clue.

[...]
> Backports to Debian are requested via the Debian BTS but you ideally
> the patch would be in upstream stable trees.

We're happy to take fixes once they've been accepted by the subsystem
maintainer.

> To get the patch into upstream stable trees, you need to send a
> > similar kind of email, but to <sta...@vger.kernel.org> (and the
> maintainers for the subsystems, I think).
[...]

It is easier to do this by putting a 'Cc: sta...@vger.kernel.org'
pseudo-header in the patch (along with Signed-off-by etc).

Ben.

-- 
Ben Hutchings
Every program is either trivial or else contains at least one bug



signature.asc
Description: This is a digitally signed message part


Bug#850425: Debian bug #850425 - mpt3sas "swiotlb buffer is full" problem only under Xen

2017-01-16 Thread Ian Jackson
Andrew Cooper writes ("Re: Bug#850425: Debian bug #850425 - mpt3sas "swiotlb 
buffer is full" problem only under Xen"):
> On 16/01/17 13:00, Ian Jackson wrote:
> > FYI David Vrabel doesn't work for Citrix any more but I'm hoping if
> > there are questions Andrew Cooper will be able to help.
> 
> I am not sure how best to proceed.  IIRC, they were essentially
> rejected, without any suitable alternative proposed by the maintainers.

Do you have a reference ?

Ian.



Bug#850425: Debian bug #850425 - mpt3sas "swiotlb buffer is full" problem only under Xen

2017-01-16 Thread Andrew Cooper
On 16/01/17 13:00, Ian Jackson wrote:
> Andy Smith writes ("Re: Bug#850425: Debian bug #850425 - mpt3sas "swiotlb 
> buffer is full" problem only under Xen"):
>> On Mon, Jan 16, 2017 at 11:47:54AM +, Ian Jackson wrote:
>>> Right.  Well, you need to post to linux-kernel saying "this patch
>>> fixes such-and-such".  You should:
> ...
>> I see. This is not something I've done before, but I'd be willing to
>> give it a go.
>>
>> But, I am not the author of these two patches. That's David Vrabel,
>> and they're from something Andrew Cooper referred to as "the
>> XenServer Patch queue", here:
> Indeed.
>
>> At the present time I don't actually know in detail what these
>> patches do, nor how they have been tested, etc.; I only know that
>> they fix the problem I was seeing.
>>
>> Is it appropriate for me to be the one presenting them to
>> linux-kernel and the maintainers and so on?
> Yes, it is normal to be passing on other people's patches.  Although
> of course if there are questions about them then you may not be able
> to answer.
>
> You should CC the patch authors (generally, everyone named with a
> Signed-off-by or a CC) on your emails.
>
> FYI David Vrabel doesn't work for Citrix any more but I'm hoping if
> there are questions Andrew Cooper will be able to help.

I am not sure how best to proceed.  IIRC, they were essentially
rejected, without any suitable alternative proposed by the maintainers.

~Andrew



Bug#850425: Debian bug #850425 - mpt3sas "swiotlb buffer is full" problem only under Xen

2017-01-16 Thread Ian Jackson
Andy Smith writes ("Re: Bug#850425: Debian bug #850425 - mpt3sas "swiotlb 
buffer is full" problem only under Xen"):
> On Mon, Jan 16, 2017 at 11:47:54AM +, Ian Jackson wrote:
> > Right.  Well, you need to post to linux-kernel saying "this patch
> > fixes such-and-such".  You should:
...
> I see. This is not something I've done before, but I'd be willing to
> give it a go.
> 
> But, I am not the author of these two patches. That's David Vrabel,
> and they're from something Andrew Cooper referred to as "the
> XenServer Patch queue", here:

Indeed.

> At the present time I don't actually know in detail what these
> patches do, nor how they have been tested, etc.; I only know that
> they fix the problem I was seeing.
> 
> Is it appropriate for me to be the one presenting them to
> linux-kernel and the maintainers and so on?

Yes, it is normal to be passing on other people's patches.  Although
of course if there are questions about them then you may not be able
to answer.

You should CC the patch authors (generally, everyone named with a
Signed-off-by or a CC) on your emails.

FYI David Vrabel doesn't work for Citrix any more but I'm hoping if
there are questions Andrew Cooper will be able to help.

Regards,
Ian.



Bug#850425: Debian bug #850425 - mpt3sas "swiotlb buffer is full" problem only under Xen

2017-01-16 Thread Andy Smith
Hi Ian,

On Mon, Jan 16, 2017 at 11:47:54AM +, Ian Jackson wrote:
> Right.  Well, you need to post to linux-kernel saying "this patch
> fixes such-and-such".  You should:

[…]

I see. This is not something I've done before, but I'd be willing to
give it a go.

But, I am not the author of these two patches. That's David Vrabel,
and they're from something Andrew Cooper referred to as "the
XenServer Patch queue", here:


https://github.com/xenserver/linux-3.x.pg/blob/master/master/series#L613-L614

At the present time I don't actually know in detail what these
patches do, nor how they have been tested, etc.; I only know that
they fix the problem I was seeing.

Is it appropriate for me to be the one presenting them to
linux-kernel and the maintainers and so on?

Thanks,
Andy



Bug#850425: Debian bug #850425 - mpt3sas "swiotlb buffer is full" problem only under Xen

2017-01-16 Thread Ian Jackson
Andy Smith writes ("Re: Bug#850425: Debian bug #850425 - mpt3sas "swiotlb 
buffer is full" problem only under Xen"):
> On Mon, Jan 09, 2017 at 06:01:14PM +, Ian Jackson wrote:
> > Patches only make it into Linux upstream stable releases if someone
> > pushes to get them in.
> 
> Do you have any advice how I could push to get this to happen?
> 
> Obviously I need this in order for Xen to be usable on my hardware
> but we're talking mpt3sas driver here which is pretty common, so I
> think it is quite desirable.

Right.  Well, you need to post to linux-kernel saying "this patch
fixes such-and-such".  You should:

1. git clone or git fetch the current master, which is
 git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git#master
and check that it doesn't have the patch.

2. Post the patch to linux-kernel.  You need to CC all the relevant
maintainers and mailing lists.  The kernel source tree has a script
get_maintainer.pl which will produce a list.  Your message should
include all the relevant information, including:
  * A copy of the patch (eg as produced by git-format-patch)
  * A description of why it is needed, how it has been tested,
or whatever

If you don't get a sensible response you may need to repeat yourself.
There is no bug tracker or upstream work queue tracker; the retry
token is with you at all times.

When the patch is in the upstream tree (or perhaps, when it has been
queued by one of the driver maintainers for the next merge window, or
whatever), you should consider asking for backports.

Backports to Debian are requested via the Debian BTS but you ideally
the patch would be in upstream stable trees.

To get the patch into upstream stable trees, you need to send a
similar kind of email, but to <sta...@vger.kernel.org> (and the
maintainers for the subsystems, I think).  You will again need to
chase this until you get a response.  The corresponding trees are
something like
  
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git#linux-3.19.y

Getting a respone may take some time.  When you get a response saying
`queued' that does not mean it's actually in the public branch; only
that it is in the maintainers' private tree for the next push.  I'm
told that that may again take several weeks.  If it takes much longer
you should probably chase it.

Now that I have written all of this down, I hope you can see why we're
not volunteering to do all this donkey work :-/.  But, I would be
happy to review drafts of your mails if you like, and to be CC'd.

Good luck.

Ian.



Bug#850425: Debian bug #850425 - mpt3sas "swiotlb buffer is full" problem only under Xen

2017-01-15 Thread Andy Smith
Hi Ian,

On Mon, Jan 09, 2017 at 06:01:14PM +, Ian Jackson wrote:
> Patches only make it into Linux upstream stable releases if someone
> pushes to get them in.

Do you have any advice how I could push to get this to happen?

Obviously I need this in order for Xen to be usable on my hardware
but we're talking mpt3sas driver here which is pretty common, so I
think it is quite desirable.

Cheers,
Andy



Bug#850425: Debian bug #850425 - mpt3sas "swiotlb buffer is full" problem only under Xen

2017-01-09 Thread Ben Hutchings
On Mon, 2017-01-09 at 18:01 +, Ian Jackson wrote:
> Andy Smith writes ("Debian bug #850425 - mpt3sas "swiotlb buffer is
> full" problem only under Xen"):
> > I'm trying to get two Xen patches applied in Debian as I need it
> > for
> > Xen to work on my hardware. The bug report is here:
> > 
> > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=850425
> 
> Hi.  (Bounced this to my work account where I do Xen stuff.)
> 
> > Ben Hutchings is asking why the two patches are not yet applied
> > upstream and doesn't seem to have got an answer. I was wondering if
> > you could take a look.
> 
> I'm afraid I don't know.
> 
> Patches only make it into Linux upstream stable releases if someone
> pushes to get them in.  I don't know whether this specific patch is
> even in the current Linux tip, which would normally be the route to
> getting them into stable.

It's not (unless that's changed since my previous mail).

Ben.

-- 
Ben Hutchings
In a hierarchy, every employee tends to rise to his level of
incompetence.



signature.asc
Description: This is a digitally signed message part


Bug#850425: Debian bug #850425 - mpt3sas "swiotlb buffer is full" problem only under Xen

2017-01-09 Thread Ian Jackson
Andy Smith writes ("Debian bug #850425 - mpt3sas "swiotlb buffer is full" 
problem only under Xen"):
> I'm trying to get two Xen patches applied in Debian as I need it for
> Xen to work on my hardware. The bug report is here:
> 
> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=850425

Hi.  (Bounced this to my work account where I do Xen stuff.)

> Ben Hutchings is asking why the two patches are not yet applied
> upstream and doesn't seem to have got an answer. I was wondering if
> you could take a look.

I'm afraid I don't know.

Patches only make it into Linux upstream stable releases if someone
pushes to get them in.  I don't know whether this specific patch is
even in the current Linux tip, which would normally be the route to
getting them into stable.

Ian.