Re: [GIT PULL] tracing: Updates for 4.4

[Steven Rostedt <rostedt@goodmis.org>]

On Fri, 6 Nov 2015 14:15:48 -0800
Linus Torvalds <torvalds@linux-foundation.org> wrote:

> On Fri, Nov 6, 2015 at 2:09 PM, Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> >
> > This was reported against linux-next on Nov 2 by Sergey Senozhatsky who
> > supplied a fix patch that I then used as the merge conflict resolution
> > from Nov 3 onward (which everyone involved knew about) ...
> 
> Ok, so the problem is that even though a maintainer is aware of the
> semantic conflict, the pull request doesn't talk about it.

Damn, that was my fault. I simply forgot to mention it :-(


> 
> A lot of maintainers *do* let me know, which I really appreciate, not
> only because it avoids the surprise, but because it happens that I
> miss these semantic conflicts.
> 
> Sometimes the semantic conflict is something that simply doesn't show
> up on x86-64 build I do.  Or I was on the road and didn't do a full
> allmodconfig build. Or it needs very specific config options etc. Or I
> just screw up. When the pull message warns me about it, that will
> help.
> 
> But maybe it's one of those things that isn't written down, and people
> don't think about as part of doing their pull request, so they forget.
> 

Yep, I was going to add that, but I've been working on internal Red Hat
Bugzilla's all week that my mind was not fully focused on that pull
request.

I'll have to add something to my scripts that pulls in any merge
breakage, such that when I run the scripts to generate the pull
request, it adds them in automatically. Because obviously, the manual
approach isn't working for me.

-- Steve