On Sat, 30 May 2020 at 11:17, Russell King - ARM Linux admin <linux@xxxxxxxxxxxxxxx> wrote: > > On Sat, May 30, 2020 at 10:51:32AM +0200, Ard Biesheuvel wrote: > > On Sat, 30 May 2020 at 10:41, Russell King - ARM Linux admin > > <linux@xxxxxxxxxxxxxxx> wrote: > > > > > > On Thu, May 28, 2020 at 09:01:55AM +0200, Ard Biesheuvel wrote: > > > > On Thu, 28 May 2020 at 01:23, Russell King - ARM Linux admin > > > > <linux@xxxxxxxxxxxxxxx> wrote: > > > > > > > > > > Ard, > > > > > > > > > > Please take a look. Obviously, whatever the resolution is going to be > > > > > needed when Linus opens the merge window. > > > > > > > > > > > > > Sorry for that. > > > > > > > > I have pushed the signed tag below to resolve it. Those changes were > > > > already in v5.7-rc2, so I wouldn't expect this to cause more trouble. > > > > If you prefer, you could merge v5.7-rc2 into your tree directly > > > > instead. > > > > > > In light of Stephen's report of a different conflict on the 29th, I > > > haven't pulled this. I don't know if that's a side effect of this > > > change having been picked up by -next or not. > > > > > > > Fair enough. Both conflicts are unambiguous and self explanatory so I > > don't think it should be a problem, right? > > I don't know - I don't have a resolution for the first one, Stephen > didn't provide a 3-way diff with his report, and I was expecting a > 3-way diff from you for it rather than another pull request. > That was not clear to me. > I now also don't know whether the conflict on the 28th still exists > or not. > Yes it does. That conflict is with v5.7-rc2, which exists because your tree is based on v5.7-rc1, and so a fix that arrived in the mean time is touching some of the lines that are being removed in v5.8. That is why I mentioned that you could preempt this by merging v5.7-rc2 into your tree as well. > I'm completely confused, and I'm considering dropping the original > EFI pull request on the grounds that the merge window opens tomorrow, > and there isn't going to be another -next before that happens, so we > don't know what's going to happen whatever action we take. > The other conflict is with the EFI changes queued up for v5.8 in the -tip tree. The pr_efi_err() function was renamed to efi_err(), which was used in a lot of places. This is slightly annoying, but rather straightforward to resolve. In any case, I am happy to go along with whatever you feel is best.
