With the boot parameter "xen_nopvspin" specified a Xen guest should not
make use of paravirt spinlocks, but behave as if running on bare
metal. This is not true, however, as the qspinlock code will fall back
to a test-and-set scheme when it is detecting a hypervisor.
In order to avoid this set the virt_spin_lock pvops function to
_paravirt_false() in case xen_nopvspin has been specified.
Signed-off-by: Juergen Gross <jgross@xxxxxxxx>
---
arch/x86/xen/spinlock.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/arch/x86/xen/spinlock.c b/arch/x86/xen/spinlock.c
index 25a7c4302ce7..c01cedfa745d 100644
--- a/arch/x86/xen/spinlock.c
+++ b/arch/x86/xen/spinlock.c
@@ -129,6 +129,8 @@ void __init xen_init_spinlocks(void)
if (!xen_pvspin) {
printk(KERN_DEBUG "xen: PV spinlocks disabled\n");
+ pv_lock_ops.virt_spin_lock =
+ __PV_IS_CALLEE_SAVE(_paravirt_false);
return;
}
printk(KERN_DEBUG "xen: PV spinlocks enabled\n");
--
2.12.3
_______________________________________________
Virtualization mailing list
Virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx
https://lists.linuxfoundation.org/mailman/listinfo/virtualization
