2010/11/22 Catalin Marinas <catalin.marinas at arm.com>: > On 22 November 2010 17:03, Per Fransson <per.fransson.ml at gmail.com> wrote: >> On the other hand, do we really need to enable the interrupts before >> performing an ipi? >> The important thing must be for them to be enabled on the >> receiving/callee cores. It's not >> as if we want the crashing core to be interruptable, is it? > > That's for the situation when the other core tries to send an IPI at > the same time. Unlikely but both cores may deadlock. > Am I right that deadlock can occur only if the same IPI was sent? And there should be no situation when CRASH_DUMP IPI was sent from different processors at the same time. > I actually had a patch to allow this which was really complicated and > it still failed in a real-world scenario which I could foresee. > > -- > Catalin > > _______________________________________________ > kexec mailing list > kexec at lists.infradead.org > http://lists.infradead.org/mailman/listinfo/kexec > -- Best regards, Maxim Uvarov
