[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1291928497-5599-3-git-send-email-konrad.wilk@oracle.com>
Date: Thu, 9 Dec 2010 16:01:37 -0500
From: Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>
To: linux-kernel@...r.kernel.org,
Stefano Stabellini <stefano.stabellini@...citrix.com>
Cc: xen-devel@...ts.xensource.com,
Jeremy Fitzhardinge <jeremy@...p.org>,
Konrad Rzeszutek Wilk <konrad@...nel.org>,
Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>
Subject: [PATCH 2/2] xen/irq: Don't fall over when nr_irqs_gsi > nr_irqs.
This scenario where the nr_irq_gsi is greater than nr_irqs
is rather strange but lets still try to survive. Make sure
to print a warning so the user wouldn't be surprised in case
things don't work.
Solves a bootup-crash when booting Xen and Linux under QEMU.
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>
---
drivers/xen/events.c | 9 +++++++++
1 files changed, 9 insertions(+), 0 deletions(-)
diff --git a/drivers/xen/events.c b/drivers/xen/events.c
index 4d4a23d..11687dd 100644
--- a/drivers/xen/events.c
+++ b/drivers/xen/events.c
@@ -411,6 +411,7 @@ static int find_unbound_irq(void)
if (bottom == nr_irqs)
goto no_irqs;
+retry:
/* This loop starts from the top of IRQ space and goes down.
* We need this b/c if we have a PCI device in a Xen PV guest
* we do not have an IO-APIC (though the backend might have them)
@@ -434,6 +435,14 @@ static int find_unbound_irq(void)
goto no_irqs;
res = irq_alloc_desc_at(irq, -1);
+ if (res == -EEXIST) {
+ top--;
+ if (bottom > top)
+ printk(KERN_ERR "Eating in GSI/MSI space (%ld)!" \
+ " Your PCI device might not work!\n", top);
+ if (top > NR_IRQS_LEGACY)
+ goto retry;
+ }
if (WARN_ON(res != irq))
return -1;
--
1.7.1
--
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