[RESEND] Conflict between char-misc and netdev

[RESEND] Conflict between char-misc and netdev

[Manivannan Sadhasivam]

Hi Jakub, Dave,

Due to the below commit in netdev there is a conflict between char-misc
and netdev trees:

5c2c85315948 ("bus: mhi: pci-generic: configurable network interface MRU")

Jakub, I noticed that you fixed the conflict locally in netdev:

d2e11fd2b7fc ("Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net")

But the commit touches the MHI bus and it should've been merged into mhi
tree then it goes via char-misc. It was unfortunate that neither
linux-arm-msm nor me were CCed to the patch :/

Could you please revert the commit?

Thanks,
Mani

Re: [RESEND] Conflict between char-misc and netdev

[Jakub Kicinski]

On Thu, 12 Aug 2021 19:02:15 +0530 Manivannan Sadhasivam wrote:
> Due to the below commit in netdev there is a conflict between char-misc
> and netdev trees:
> 
> 5c2c85315948 ("bus: mhi: pci-generic: configurable network interface MRU")
> 
> Jakub, I noticed that you fixed the conflict locally in netdev:
> 
> d2e11fd2b7fc ("Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net")
> 
> But the commit touches the MHI bus and it should've been merged into mhi
> tree then it goes via char-misc. It was unfortunate that neither
> linux-arm-msm nor me were CCed to the patch :/
> 
> Could you please revert the commit?

Apologies for missing the lacking CC list, the extra resolution work,
etc. etc. Let me try to sharpen the "were maintainers CCed" check in
patchwork to make this less likely in the future.

About the situation at hand - is the commit buggy? Or is there work
that's pending in char-misc that's going to conflict in a major way? 
Any chance you could just merge the same patch into mhi and git will 
do its magic?

Re: [RESEND] Conflict between char-misc and netdev

[Manivannan Sadhasivam]

On Thu, Aug 12, 2021 at 06:51:13AM -0700, Jakub Kicinski wrote:
> On Thu, 12 Aug 2021 19:02:15 +0530 Manivannan Sadhasivam wrote:
> > Due to the below commit in netdev there is a conflict between char-misc
> > and netdev trees:
> > 
> > 5c2c85315948 ("bus: mhi: pci-generic: configurable network interface MRU")
> > 
> > Jakub, I noticed that you fixed the conflict locally in netdev:
> > 
> > d2e11fd2b7fc ("Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net")
> > 
> > But the commit touches the MHI bus and it should've been merged into mhi
> > tree then it goes via char-misc. It was unfortunate that neither
> > linux-arm-msm nor me were CCed to the patch :/
> > 
> > Could you please revert the commit?
> 
> Apologies for missing the lacking CC list, the extra resolution work,
> etc. etc. Let me try to sharpen the "were maintainers CCed" check in
> patchwork to make this less likely in the future.
> 

No worries. I know how loaded the netdev list is :)

> About the situation at hand - is the commit buggy? Or is there work
> that's pending in char-misc that's going to conflict in a major way? 
> Any chance you could just merge the same patch into mhi and git will 
> do its magic?

I'm not happy with the MHI change and I do have a comment about the
variable "mru_default". So it'd be good if we revert the commit
entirely!

Also I do have few more patches coming for the same driver touched by
5c2c85315948.

Thanks,
Mani

Re: [RESEND] Conflict between char-misc and netdev

[Jakub Kicinski]

On Thu, 12 Aug 2021 19:35:18 +0530 Manivannan Sadhasivam wrote:
> > About the situation at hand - is the commit buggy? Or is there work
> > that's pending in char-misc that's going to conflict in a major way? 
> > Any chance you could just merge the same patch into mhi and git will 
> > do its magic?  
> 
> I'm not happy with the MHI change and I do have a comment about the
> variable "mru_default". So it'd be good if we revert the commit
> entirely!

Would you mind rendering that comment you're referring to as a commit
message and sending a fix-up or a revert patch against net-next?
I wouldn't be able to do it justice.

Re: [RESEND] Conflict between char-misc and netdev

[Richard Laing]

On 8/13/21 4:12 AM, Jakub Kicinski wrote:
> Would you mind rendering that comment you're referring to as a commit
> message and sending a fix-up or a revert patch against net-next?
> I wouldn't be able to do it justice.
I would also be interested in the comment.

I would be happy if the patch is reverted from net-next, I have been 
looking at a more flexible alternative allowing the size to be specified 
per channel rather than as a global value.