Regression in next caused by 95026658c46e i2c: do not use internal..

Regression in next caused by 95026658c46e i2c: do not use internal..

[Tony Lindgren]

Hi Sudip & Wolfram,

Looks like commit 95026658c46e ("i2c: do not use internal data from
driver core") in Linux next introduced a regression breaking booting
on most ARM boards according to kernelci.org [0].

Having something like this happen just as the merge window opens is
not nice. You guys should already know.. The patches in next should
be sitting there for a few weeks before the merge window so issues
like this are found out early.

Regards,

Tony

[0] https://kernelci.org/boot/all/job/next/

Regression in next caused by 95026658c46e i2c: do not use internal..

[Wolfram Sang]

Hi,

> Having something like this happen just as the merge window opens is
> not nice. You guys should already know..

I already applied the fix.

> The patches in next should be sitting there for a few weeks before the
> merge window so issues like this are found out early.

I understand this was a major hickup and I am sorry for this. I should
have known better.

I am not taking 100% of the blame, though: The amount of submissions is
rising faster than the amount of non-maintainer review being done (at
least in the subsystems I work on). I took a wrong step in trying to
handle the gap. I am really sorry about the fuzz, yet I can't deny there
is also something inevitable in that to me. That being said, I'll of
course try harder to not let this happen again.

Regards,

   Wolfram

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20160314/4f3c058b/attachment.sig>

Regression in next caused by 95026658c46e i2c: do not use internal..

[Tony Lindgren]

* Wolfram Sang <wsa@the-dreams.de> [160314 09:52]:
> Hi,
> 
> > Having something like this happen just as the merge window opens is
> > not nice. You guys should already know..
> 
> I already applied the fix.

OK great thanks!

> > The patches in next should be sitting there for a few weeks before the
> > merge window so issues like this are found out early.
> 
> I understand this was a major hickup and I am sorry for this. I should
> have known better.

Thanks for understanding my suffering :) I've been chasing various
regressions constantly over past months..  So I've started to keep my
eye on Linux next hoping to prevent being stuck chasing regressions
forever..

> I am not taking 100% of the blame, though: The amount of submissions is
> rising faster than the amount of non-maintainer review being done (at
> least in the subsystems I work on). I took a wrong step in trying to
> handle the gap. I am really sorry about the fuzz, yet I can't deny there
> is also something inevitable in that to me. That being said, I'll of
> course try harder to not let this happen again.

Yeah the same applies to may subsystems for sure. Probably a good idea
to be extra careful about not causing regressions with core subsystem
changes.

Regards,

Tony

Regression in next caused by 95026658c46e i2c: do not use internal..

[Sudip Mukherjee]

On Monday 14 March 2016 04:09 PM, Tony Lindgren wrote:
> Hi Sudip & Wolfram,
>
> Looks like commit 95026658c46e ("i2c: do not use internal data from
> driver core") in Linux next introduced a regression breaking booting
> on most ARM boards according to kernelci.org [0].
>
> Having something like this happen just as the merge window opens is
> not nice. You guys should already know.. The patches in next should
> be sitting there for a few weeks before the merge window so issues
> like this are found out early.

Yes, sorry about the mess.

regards
sudip