xen/5396d818-avoid-crash-on-HVM-domain-destroy-with-PCI-passthrough.patch

27 lines
969 B
Diff
Raw Normal View History

- bnc#882127 - Xen kernel panics on booting SLES12 Beta 8 53a199d7-x86-EFI-allow-FPU-XMM-use-in-runtime-service-functions.patch - Upstream patches from Jan 538c338f-x86-amd_ucode-flip-revision-numbers-in-printk.patch 538ee637-ACPI-Prevent-acpi_table_entries-from-falling-into-a-infinite-loop.patch 5390917a-VT-d-honor-APEI-firmware-first-mode-in-XSA-59-workaround-code.patch 53909259-x86-domctl-two-functional-fixes-to-XEN_DOMCTL_-gs-etvcpuextstate.patch 5390927f-x86-fix-reboot-shutdown-with-running-HVM-guests.patch 5396d818-avoid-crash-on-HVM-domain-destroy-with-PCI-passthrough.patch 5396e805-x86-HVM-refine-SMEP-test-in-HVM_CR4_GUEST_RESERVED_BITS.patch 539ebe62-x86-EFI-improve-boot-time-diagnostics.patch 539ec004-x86-mce-don-t-spam-the-console-with-CPUx-Temperature-z.patch 53a040c6-page-alloc-scrub-pages-used-by-hypervisor-upon-freeing.patch (replaces xsa100.patch) 53a1990a-IOMMU-prevent-VT-d-device-IOTLB-operations-on-wrong-IOMMU.patch - Replace 'domUloader' with 'pygrub' when converting or importing Xen domains into libvirt with xen2libvirt. domUloader is no longer provided in xen-tools. Modified: xen2libvirt.py Thu Jun 13 15:50:19 MDT 2014 - cyliu@suse.com - fate#310956: Support Direct Kernel Boot for FV guests patches would go to upstream: qemu side: qemu-support-xen-hvm-direct-kernel-boot.patch xen side: xen-pass-kernel-initrd-to-qemu.patch - bnc#880751 - VUL-0: xen: Hypervisor heap contents leaked to guests xsa100.patch OBS-URL: https://build.opensuse.org/package/show/Virtualization/xen?expand=0&rev=320
2014-07-01 03:36:17 +00:00
# Commit b9ae60907e6dbc686403e52a7e61a6f856401a1b
# Date 2014-06-10 12:04:08 +0200
# Author Juergen Gross <jgross@suse.com>
# Committer Jan Beulich <jbeulich@suse.com>
avoid crash on HVM domain destroy with PCI passthrough
c/s bac6334b5 "move domain to cpupool0 before destroying it" introduced a
problem when destroying a HVM domain with PCI passthrough enabled. The
moving of the domain to cpupool0 includes moving the pirqs to the cpupool0
cpus, but the event channel infrastructure already is unusable for the
domain. So just avoid moving pirqs for dying domains.
Signed-off-by: Juergen Gross <jgross@suse.com>
--- a/xen/common/schedule.c
+++ b/xen/common/schedule.c
@@ -297,7 +297,8 @@ int sched_move_domain(struct domain *d,
spin_unlock_irq(lock);
v->sched_priv = vcpu_priv[v->vcpu_id];
- evtchn_move_pirqs(v);
+ if ( !d->is_dying )
+ evtchn_move_pirqs(v);
new_p = cpumask_cycle(new_p, c->cpu_valid);