On Mon, Dec 01, 2014 at 11:11:43AM +0000, David Vrabel wrote: > On 27/11/14 18:36, Luis R. Rodriguez wrote: > > On Thu, Nov 27, 2014 at 07:36:31AM +0100, Juergen Gross wrote: > >> On 11/26/2014 11:26 PM, Luis R. Rodriguez wrote: > >>> From: "Luis R. Rodriguez" <mcgrof@xxxxxxxx> > >>> > >>> Some folks had reported that some xen hypercalls take a long time > >>> to complete when issued from the userspace private ioctl mechanism, > >>> this can happen for instance with some hypercalls that have many > >>> sub-operations, this can happen for instance on hypercalls that use > [...] > >>> --- a/drivers/xen/privcmd.c > >>> +++ b/drivers/xen/privcmd.c > >>> @@ -60,6 +60,9 @@ static long privcmd_ioctl_hypercall(void __user *udata) > >>> hypercall.arg[0], hypercall.arg[1], > >>> hypercall.arg[2], hypercall.arg[3], > >>> hypercall.arg[4]); > >>> +#ifndef CONFIG_PREEMPT > >>> + schedule(); > >>> +#endif > > As Juergen points out, this does nothing. You need to schedule while in > the middle of the hypercall. > > Remember that Xen's hypercall preemption only preempts the hypercall to > run interrupts in the guest. How is it ensured that when the kernel preempts on this code path on CONFIG_PREEMPT=n kernel that only interrupts in the guest are run? > >>> > >>> return ret; > >>> } > >>> > >> > >> Sorry, I don't think this will solve anything. You're calling schedule() > >> right after the long running hypercall just nanoseconds before returning > >> to the user. > > > > Yeah, well that is what [1] tried as well only it tried using > > preempt_schedule_irq() on the hypercall callback... > > No. My patch added a schedule point in the middle of a hypercall on the > return from an interrupt (e.g., the timer interrupt). OK that provides much better context and given that I do see the above hunk as pointless. I was completely misrepresenting what the callback was for. Now -- just to address my issues with the use of preempt_schedule_irq(). If the above is addressed that I think should address most of my concerns, if we can figure out a way to not deal with it to be arch specific that'd be neat, and if we could not have to ifdef around stuff even better. Luis -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
