On 03/02/2016 10:38 AM, Will Deacon wrote: > On Tue, Mar 01, 2016 at 05:17:55PM -0800, Geoff Levand wrote: >> On Tue, 2016-03-01 at 10:54 -0700, Azriel Samson wrote: >> >>> We have been using the kexec/kdump patches on ARM64 and would like to >>> use them from upstream. >>> >>> Assuming these are not going to be in Linux-4.5, a rebased set on >>> Linux-4.5 would be very useful. >> >> I'm waiting for one of two things to happen, either Marc Zyngier gives >> an ack for the patch 'arm64: kvm: allows kvm cpu hotplug', or the arm64 >> maintainers decide they will accept my proposal to merge kexec with a >> conditional in the Kconfig that only allows one of KVM or kexec to be >> selected. > > Depending on !KVM isn't even an option as far as I'm concerned. It basically > makes this all dead code. > > As for awaiting an ack on the kvm hotplug patch, afaict, Akashi was going > to address James's comments on the last posting [1]. Come on! I have already fixed this issue, but still waiting for Geoff rebasing his patch as he is now a submitter of the whole patch series including kexec/kdump. (Please note, as James suggested, some (minor) modification need be made for rebasing onto v4.5) >> The kexec patches have been in review for 20 months now, with only >> minor changes in the past 15. I don't expect much would be gained in >> posting the kexec series again unless there is a chance they would be >> merged. > > Why do you say that? v13 attracted lots of comments from a variety of > reviewers, so the onus is on you to address that feedback and post a > new version. Alternatively, declare that you no longer to intend to work > on the series, and perhaps then somebody else can pick it up and finish > the job. *If* Geoff is no longer willing to work on upstreaming kexec stuff, I will probably take over his task. -Takahiro AKASHI > Will > > [1] http://lkml.kernel.org/r/56A873C0.6090406 at linaro.org >
