From: Keir Fraser <keir.fraser@citrix.com>
Date: Wed, 2 Jul 2008 16:28:27 +0000 (+0100)
Subject: iommu: Fix hvm_dpci_eoi for smp guest in ioapic interrupt mode
X-Git-Tag: archive/raspbian/4.8.0-1+rpi1~1^2~14188^2~126
X-Git-Url: https://dgit.raspbian.org/%22http:/www.example.com/cgi/%22https:/%22bookmarks:/%22man:///%22http:/www.example.com/cgi/%22https:/%22bookmarks:/%22man:/?a=commitdiff_plain;h=fec35adc8693a574a5fa77f06c56916a6e185043;p=xen.git

iommu: Fix hvm_dpci_eoi for smp guest in ioapic interrupt mode

My testing shows, for some smp guests, ent->fields.mask will not be
unmasked when hvm_dpci_eoi() is invoked at the first time. In that
case, we should rely on timeout call back to do pirq_guest_eoi() and
should not stop the per-irq timer too early.

From: Wei Wang <wei.wang2@amd.com>
Signed-off-by: Keir Fraser <keir.fraser@citrix.com>
---

diff --git a/xen/drivers/passthrough/io.c b/xen/drivers/passthrough/io.c
index 7e38237907..422ae18a0e 100644
--- a/xen/drivers/passthrough/io.c
+++ b/xen/drivers/passthrough/io.c
@@ -258,10 +258,12 @@ void hvm_dpci_eoi(struct domain *d, unsigned int guest_gsi,
     {
         spin_unlock(&hvm_irq_dpci->dirq_lock);
 
-        stop_timer(&hvm_irq_dpci->hvm_timer[
-            domain_irq_to_vector(d, machine_gsi)]);
         if ( (ent == NULL) || !ent->fields.mask )
+        {
+            stop_timer(&hvm_irq_dpci->hvm_timer[
+                domain_irq_to_vector(d, machine_gsi)]);
             pirq_guest_eoi(d, machine_gsi);
+        }
     }
     else
         spin_unlock(&hvm_irq_dpci->dirq_lock);