Message-ID: <20130520152415.GD24598@phenom.dumpdata.com>
Date:	Mon, 20 May 2013 11:24:15 -0400
From:	Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>
To:	Stefano Stabellini <stefano.stabellini@...citrix.com>
Cc:	Zhenzhong Duan <zhenzhong.duan@...cle.com>,
	"xen-devel@...ts.xensource.com" <xen-devel@...ts.xensource.com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Thomas Gleixner <tglx@...utronix.de>,
	Ingo Molnar <mingo@...hat.com>,
	"H. Peter Anvin" <hpa@...or.com>,
	Yuval Shaia <yuval.shaia@...cle.com>,
	Feng Jin <joe.jin@...cle.com>, Chien Yen <chien.yen@...cle.com>
Subject: Re: [PATCH] xen: reuse the same pirq allocated when driver load
 first time

On Mon, May 20, 2013 at 11:24:04AM +0100, Stefano Stabellini wrote:
> On Fri, 17 May 2013, Zhenzhong Duan wrote:
> > On 2013-05-15 17:41, Stefano Stabellini wrote:
> > > On Tue, 14 May 2013, Konrad Rzeszutek Wilk wrote:
> > > > On Tue, May 14, 2013 at 02:49:50PM +0100, Stefano Stabellini wrote:
> > > > > On Mon, 13 May 2013, Konrad Rzeszutek Wilk wrote:
> > > > > > On Mon, May 13, 2013 at 06:24:46PM +0100, Stefano Stabellini wrote:
> > > > > > > On Mon, 13 May 2013, Konrad Rzeszutek Wilk wrote:
> > > > > > > > On Mon, May 13, 2013 at 03:50:52PM +0100, Stefano Stabellini
> > > > > > > > wrote:
> > > > > > > > > On Mon, 13 May 2013, Konrad Rzeszutek Wilk wrote:
> > > > > > > > > > On Mon, May 13, 2013 at 12:06:43PM +0100, Stefano Stabellini
> > > > > > > > > > wrote:
> > > > > > > > > > > On Fri, 10 May 2013, Konrad Rzeszutek Wilk wrote:
> > > > > > > > > > > > On Wed, May 08, 2013 at 04:18:24PM +0800, Zhenzhong Duan
> > > > > > > > > > > > wrote:
> > > > > > > > > > > > > When driver load and unload in a loop, pirq will exhaust
> > > > > > > > > > > > > finally.
> > > > > > > > > > > > > Try to use the same pirq which was already mapped and
> > > > > > > > > > > > > binded at first time
> > > > > > > > > > > > So what happens if I unload and reload two drivers in
> > > > > > > > > > > > random order?
> > > > > > > > > > > > 
> > > > > > > > > > > > > when driver loaded.
> > > > > > > > > > > > > 
> > > > > > > > > > > > > Read pirq from msix entry and test if data is
> > > > > > > > > > > > > XEN_PIRQ_MSI_DATA
> > > > > > > > > > > > > xen_irq_from_pirq(pirq) < 0 checking is wrong as irq
> > > > > > > > > > > > > will be freed
> > > > > > > > > > > > > when driver unload, it's always true in second load.
> > > > > > > > > > > > If my understanding is right the issue at hand is that the
> > > > > > > > > > > > caching
> > > > > > > > > > > > information about the pirq disappears once the driver has
> > > > > > > > > > > > been
> > > > > > > > > > > > unloaded b/c the event's irq-info is removed (as the
> > > > > > > > > > > > driver is
> > > > > > > > > > > > unloaded and free_irq is called).
> > > > > > > > > > > > 
> > > > > > > > > > > > Stefano,
> > > > > > > > > > > > Is there a specific write to the MSI structure that would
> > > > > > > > > > > > cause the
> > > > > > > > > > > > hypervisor to drop the PIRQ? Or a nice hypercall to "free"
> > > > > > > > > > > > an
> > > > > > > > > > > > PIRQ in usage?
> > > > > > > > > > > We already have a "free PIRQ" hypercall, it's called
> > > > > > > > > > > PHYSDEVOP_unmap_pirq and should be called by QEMU.
> > > > > > > > > > Considering that we call function that allocates
> > > > > > > > > > (PHYSDEVOP_get_free_pirq)
> > > > > > > > > > it in the Linux kernel (and not in QEMU), perhaps that should
> > > > > > > > > > be done in the
> > > > > > > > > > Linux kernel as part of xen_destroy_irq()? Or would that
> > > > > > > > > > confuse QEMU?
> > > > > > > > > I think it would confuse QEMU. It is probably better to let the
> > > > > > > > > unmap
> > > > > > > > > being handled by it.
> > > > > > > > > 
> > > > > > > > > 
> > > > > > > > > > It looks like QEMU only does that hypercall (via
> > > > > > > > > > xc_physdev_unmap_pirq)
> > > > > > > > > > unregister_real_device which is only called during pci unplug?
> > > > > > > > > You are right! I would think that this behaviour is erroneous
> > > > > > > > > unless it
> > > > > > > > > was done on purpose to avoid allocating MSIs twice.
> > > > > > > > > If that is the case we would need to do something similar in
> > > > > > > > > Linux too.
> > > > > > > > > 
> > > > > > > > > I think that the issue is the mismatch between QEMU's and
> > > > > > > > > Linux's
> > > > > > > > > behaviours: either both should be allocating MSIs once, or they
> > > > > > > > > should
> > > > > > > > > both be allocating and deallocating MSIs every time the driver
> > > > > > > > > is loaded
> > > > > > > > > and unloaded.
> > > > > > > > Right. But we also have the scenario that QEMU and Linux are going
> > > > > > > > to
> > > > > > > > be out of sync. So we need fixes in both places - I think.
> > > > > > > QEMU is the owner of the pirq, in fact it is the one that creates
> > > > > > > and
> > > > > > > destroys the mapping. I think that the right place to fix this
> > > > > > > problem
> > > > > > > is in QEMU, the ABI would be much cleaner as a result. As a side
> > > > > > > effect
> > > > > > > we don't need to make any changes in Linux.
> > > > > > You do. You need to remove the PHYSDEVOP_get_free_pirq call in that
> > > > > > case.
> > > > >   PHYSDEVOP_get_free_pirq needs to stay, because Linux needs to know the
> > > > > pirq that QEMU is going to use.
> > > > That looks like an API violation. We have an hypercall that
> > > > allocates the PIRQ in the Linux, then two hypercalls in the QEMU
> > > > layer - one to map, and the other to unmap and free.
> > > > 
> > > > > However I would let QEMU handle the mapping (it already does that in
> > > > > pt_msi_setup calling xc_physdev_map_pirq_msi) and unmapping (that is
> > > > > done by calling xc_domain_unbind_msi_irq from pt_msi_disable).
> > > > > I think the problem is that pt_msi_disable is only called on
> > > > > unregister_real_device and pt_reset_interrupt_and_io_mapping, not when
> > > > > the guest disables MSIs.
> > > > Sure, I am not disputing that. I think the fix in QEMU to call the
> > > > unmap is correct.
> > > > 
> > > > But I am also wondering whether it makes sense to do that in the Linux
> > > > kernel - as it does the alloc in the first place. Seems like a bit of
> > > > duct-tape has been used to connect this plumbing together.
> > > 
> > > I admit that it is not a great interface.
> > > I would be open to options that move the entire setup/freeing in Linux,
> > > but keep in mind that we need to retain the pirq code in QEMU for pure
> > > HVM guests.
> > Hi Stefano,
> > 
> > do you work out a patch for me to test?

I did the patch that which calls PHYSDEVOP_unmap_pirq from within
Linux kernel. And it did not fix a thing. Here is the patch (sorry about
the copy-n-paste):
diff --git a/drivers/xen/events.c b/drivers/xen/events.c
index 6a6bbe4..d122ca9 100644
--- a/drivers/xen/events.c
+++ b/drivers/xen/events.c
@@ -789,21 +789,28 @@ error_irq:
        return ret;
 }
 #endif
-
+static bool unmap_pirq = false;
+static int __init unmap_pirq_setup(char *s)
+{
+       unmap_pirq = true;
+       return 1;
+}
+__setup("unmap_pirq", unmap_pirq_setup);
 int xen_destroy_irq(int irq)
 {
        struct irq_desc *desc;
        struct physdev_unmap_pirq unmap_irq;
        struct irq_info *info = info_for_irq(irq);
        int rc = -ENOENT;
-
        mutex_lock(&irq_mapping_update_lock);

        desc = irq_to_desc(irq);
        if (!desc)
                goto out;

-       if (xen_initial_domain()) {
+       if (xen_initial_domain() || unmap_pirq) {
+               printk(KERN_INFO "irq: %d, pirq:%d unmapping\n",
+                      irq, info->u.pirq.pirq);
                unmap_irq.pirq = info->u.pirq.pirq;
                unmap_irq.domid = info->u.pirq.domid;
                rc = HYPERVISOR_physdev_op(PHYSDEVOP_unmap_pirq, &unmap_irq);


Meaning that the later call to xen_allocate_pirq_msi would still get new PIRQs
value.

On the other hand, the original patch proposed by Zhenzhong works nicely.


.. I am wondering what is going wrong here.
> 
> I'll be traveling/busy for a few weeks, maybe it's best if someone else
> picks up this work item.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists