On Fri, Nov 13, 2020 at 10:05:56AM -0700, David Ahern wrote:
> On 11/13/20 9:57 AM, Jakub Kicinski wrote:
> > Good people of build bot, 
> > 
> > would you mind shedding some light on this one? It was also reported on
> > v1, and Andrea said it's impossible to repro. Strange that build bot
> > would make the same mistake twice, tho.
> > 
> 
> I kicked off a build this morning using Andrea's patches and the config
> from the build bot; builds fine as long as the first 3 patches are applied.
> 

I can confirm this as well with clang; if I applied the first three
patches then this one, there is no error but if you just apply this one,
there will be. If you open the GitHub URL, it shows just this patch
applied, not the first three, which explains it.

For what it's worth, b4 chokes over this series:

$ b4 am -o - 20201107153139.3552-1-andrea.mayer@uniroma2.it | git am
Looking up https://lore.kernel.org/r/20201107153139.3552-1-andrea.mayer%40uniroma2.it
Grabbing thread from lore.kernel.org/linux-kselftest
Analyzing 18 messages in the thread
---
Writing /tmp/tmp8425by7fb4-am-stdout
  [net-next,v2,3/5] seg6: add callbacks for customizing the creation/destruction of a behavior
---
Total patches: 1
---
 Link: https://lore.kernel.org/r/20201107153139.3552-1-andrea.mayer@uniroma2.it
 Base: not found
---
Applying: seg6: add callbacks for customizing the creation/destruction of a behavior
error: patch failed: net/ipv6/seg6_local.c:1015
error: net/ipv6/seg6_local.c: patch does not apply
Patch failed at 0001 seg6: add callbacks for customizing the creation/destruction of a behavior
hint: Use 'git am --show-current-patch=diff' to see the failed patch
When you have resolved this problem, run "git am --continue".
If you prefer to skip this patch, run "git am --skip" instead.
To restore the original branch and stop patching, run "git am --abort".

Even if I grab the mbox from lore.kernel.org, it tries to do the same
thing and apply the 3rd patch first, which might explain why the 0day
bot got confused.

Cheers,
Nathan