On Mon, 30 Nov 2015, Julien Grall wrote: > Hi Stefano, > > On 25/11/15 16:40, Stefano Stabellini wrote: > > Implement GICD_ICACTIVER and GICD_ISACTIVER reads by looking for the > > GIC_IRQ_GUEST_ACTIVE bit in the relevant struct pending_irq. However > > given that the pending to active transaction for irqs in LRs in done in > > hardware, the GIC_IRQ_GUEST_ACTIVE bit might be out of date. We'll have > > to live with that. > > Looking to the Linux tree again, I found this commit: > > commit 1c7d4dd46ee048afe19e1294634df6fa66862519 > Author: Marc Zyngier <marc.zyng...@arm.com> > Date: Mon Nov 16 19:13:28 2015 +0000 > > irqchip/gic: Add save/restore of the active state > > When using EOImode==1, we may mark interrupts as being forwarded > to a virtual machine. In that case, the interrupt is left active > while being passed to the VM. > > If we suspend the system before the VM has deactivated the interrupt, > the active state will be lost (which may be very annoying, as this > may result in spurious interrupts and a confused guest). > > To avoid this, save and restore the active state together with the > rest of the GIC registers. > > Signed-off-by: Marc Zyngier <marc.zyng...@arm.com> > Cc: <linux-arm-ker...@lists.infradead.org> > Cc: Jason Cooper <ja...@lakedaemon.net> > Cc: Russell King <li...@arm.linux.org.uk> > Link: > http://lkml.kernel.org/r/1447701208-18150-5-git-send-email-marc.zyng...@arm.com > Signed-off-by: Thomas Gleixner <t...@linutronix.de> > > So, we have to handle properly active/deactivate in order to get > suspend/resume working correctly.
We won't have any active interrupts: suspending the guest is done in cooperation with the guest kernel (see drivers/xen/manage.c), which will deactive the interrupts. In any case we only have evtchn_irq and the timers. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
