From: Julien Grall <julien.grall@linaro.org>
To: xen-devel@lists.xen.org
Cc: patches@linaro.org, ian.campbell@citrix.com,
Julien Grall <julien.grall@linaro.org>,
Stefano.Stabellini@eu.citrix.com
Subject: [PATCH 3/5] xen/arm: Don't reinject the IRQ if it's already in LRs
Date: Tue, 25 Jun 2013 00:04:24 +0100 [thread overview]
Message-ID: <1372115067-17071-4-git-send-email-julien.grall@citrix.com> (raw)
In-Reply-To: <1372115067-17071-1-git-send-email-julien.grall@citrix.com>
From: Julien Grall <julien.grall@linaro.org>
When an IRQ, marked as IRQS_ONESHOT, is injected Linux will:
- Disable the IRQ
- Call the interrupt handler
- Conditionnally enable the IRQ
- EOI the IRQ
When Linux will re-enable the IRQ, Xen will inject again the IRQ because it's
still inflight. Therefore, LRs will contains duplicated IRQs and Xen will
EOI it twice if it's a physical IRQ.
Signed-off-by: Julien Grall <julien.grall@linaro.org>
---
xen/arch/arm/gic.c | 27 +++++++++++++++++++++++++++
xen/arch/arm/vgic.c | 3 ++-
xen/include/asm-arm/gic.h | 3 +++
3 files changed, 32 insertions(+), 1 deletion(-)
diff --git a/xen/arch/arm/gic.c b/xen/arch/arm/gic.c
index 21575df..bf05716 100644
--- a/xen/arch/arm/gic.c
+++ b/xen/arch/arm/gic.c
@@ -570,6 +570,33 @@ int __init setup_dt_irq(const struct dt_irq *irq, struct irqaction *new)
return rc;
}
+/* Check if an IRQ was already injected to the current VCPU */
+bool_t gic_irq_injected(unsigned int irq)
+{
+ bool_t found = 0;
+ int i = 0;
+ unsigned int virq;
+
+ spin_lock_irq(&gic.lock);
+
+ while ( (i = find_next_bit((unsigned long *)&this_cpu(lr_mask),
+ nr_lrs, i)) < nr_lrs )
+ {
+ virq = GICH[GICH_LR + i] & GICH_LR_VIRTUAL_MASK;
+
+ if ( virq == irq )
+ {
+ found = 1;
+ break;
+ }
+ i++;
+ }
+
+ spin_unlock_irq(&gic.lock);
+
+ return found;
+}
+
static inline void gic_set_lr(int lr, unsigned int virtual_irq,
unsigned int state, unsigned int priority)
{
diff --git a/xen/arch/arm/vgic.c b/xen/arch/arm/vgic.c
index 2d91dce..cea9233 100644
--- a/xen/arch/arm/vgic.c
+++ b/xen/arch/arm/vgic.c
@@ -369,8 +369,9 @@ static void vgic_enable_irqs(struct vcpu *v, uint32_t r, int n)
while ( (i = find_next_bit((const long unsigned int *) &r, 32, i)) < 32 ) {
irq = i + (32 * n);
p = irq_to_pending(v, irq);
- if ( !list_empty(&p->inflight) )
+ if ( !list_empty(&p->inflight) && !gic_irq_injected(irq) )
gic_set_guest_irq(v, irq, GICH_LR_PENDING, p->priority);
+
i++;
}
}
diff --git a/xen/include/asm-arm/gic.h b/xen/include/asm-arm/gic.h
index 513c1fc..f9e9ef1 100644
--- a/xen/include/asm-arm/gic.h
+++ b/xen/include/asm-arm/gic.h
@@ -197,6 +197,9 @@ extern unsigned int gic_number_lines(void);
int gic_irq_xlate(const u32 *intspec, unsigned int intsize,
unsigned int *out_hwirq, unsigned int *out_type);
+/* Check if an IRQ was already injected to the current VCPU */
+bool_t gic_irq_injected(unsigned int irq);
+
#endif /* __ASSEMBLY__ */
#endif
--
1.7.10.4
next prev parent reply other threads:[~2013-06-24 23:04 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-06-24 23:04 [PATCH 0/5] Fix multiple issues with the interrupts on ARM Julien Grall
2013-06-24 23:04 ` [PATCH 1/5] xen/arm: Physical IRQ is not always equal to virtual IRQ Julien Grall
2013-06-25 13:16 ` Stefano Stabellini
2013-06-25 15:21 ` Julien Grall
2013-06-25 16:06 ` Ian Campbell
2013-06-24 23:04 ` [PATCH 2/5] xen/arm: Keep count of inflight interrupts Julien Grall
2013-06-25 16:12 ` Ian Campbell
2013-06-25 16:58 ` Stefano Stabellini
2013-06-25 17:46 ` Julien Grall
2013-06-25 18:38 ` Stefano Stabellini
2013-06-26 10:59 ` Ian Campbell
2013-06-26 11:10 ` Stefano Stabellini
2013-06-26 11:16 ` Ian Campbell
2013-06-26 10:58 ` Ian Campbell
2013-06-26 11:08 ` Stefano Stabellini
2013-06-26 11:15 ` Ian Campbell
2013-06-26 11:23 ` Stefano Stabellini
2013-06-26 11:41 ` Ian Campbell
2013-06-26 11:50 ` Stefano Stabellini
2013-06-26 11:57 ` Ian Campbell
2013-06-26 14:02 ` Stefano Stabellini
2013-06-24 23:04 ` Julien Grall [this message]
2013-06-25 13:24 ` [PATCH 3/5] xen/arm: Don't reinject the IRQ if it's already in LRs Stefano Stabellini
2013-06-25 13:55 ` Julien Grall
2013-06-25 16:36 ` Stefano Stabellini
2013-06-25 16:46 ` Ian Campbell
2013-06-25 17:05 ` Stefano Stabellini
2013-06-26 10:53 ` Ian Campbell
2013-06-26 11:19 ` Stefano Stabellini
2013-06-25 16:48 ` Julien Grall
2013-06-25 16:59 ` Stefano Stabellini
2013-06-25 16:14 ` Ian Campbell
2013-06-24 23:04 ` [PATCH 4/5] xen/arm: Rename gic_irq_{startup, shutdown} to gic_irq_{mask, unmask} Julien Grall
2013-06-24 23:04 ` [PATCH 5/5] xen/arm: Only enable physical IRQs when the guest asks Julien Grall
2013-06-25 16:19 ` Stefano Stabellini
2013-06-25 16:55 ` Julien Grall
2013-06-25 17:07 ` Stefano Stabellini
2013-12-02 17:26 ` Ian Campbell
2013-12-02 17:37 ` Stefano Stabellini
2013-06-25 16:28 ` Ian Campbell
2013-06-25 17:38 ` Julien Grall
2013-06-25 18:27 ` Stefano Stabellini
2013-06-26 10:55 ` Ian Campbell
2013-06-26 13:03 ` Julien Grall
2013-07-31 13:08 ` [PATCH 0/5] Fix multiple issues with the interrupts on ARM Andrii Anisov
2013-07-31 14:00 ` Julien Grall
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1372115067-17071-4-git-send-email-julien.grall@citrix.com \
--to=julien.grall@linaro.org \
--cc=Stefano.Stabellini@eu.citrix.com \
--cc=ian.campbell@citrix.com \
--cc=patches@linaro.org \
--cc=xen-devel@lists.xen.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).