From: Fenghua Yu <fenghua.yu@intel.com>
To: "'David Woodhouse'" <dwmw2@infradead.org>,
"'Ingo Molnar'" <mingo@elte.hu>,
"'Andrew Morton'" <akpm@linux-foundation.org>
Cc: "'LKML'" <linux-kernel@vger.kernel.org>,
"'IOMMU'" <iommu@lists.linux-foundation.org>
Subject: [PATCH] Time out for possible dead loops during queued invalidation wait
Date: Wed, 20 May 2009 10:42:59 -0700 [thread overview]
Message-ID: <20090520174259.GA10646@linux-os.sc.intel.com> (raw)
In-Reply-To: <A6AD88C3F2289247BE726C37303E1EB8A4DF7DB1@orsmsx505.amr.corp.intel.com>
Two loops in qi_submit_sync() do not have time out. If hardware can not finish
the queued invalidation for some reason, the loops could end up in dead loops
without any hint for what is going on. I add time out for the loops and report
warning when time out happens.
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
---
dmar.c | 12 ++++++++++++
1 files changed, 12 insertions(+)
diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index fa3a113..95baacd 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -637,6 +637,7 @@ int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu)
struct qi_desc *hw, wait_desc;
int wait_index, index;
unsigned long flags;
+ cycles_t start_time;
if (!qi)
return 0;
@@ -644,8 +645,13 @@ int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu)
hw = qi->desc;
spin_lock_irqsave(&qi->q_lock, flags);
+ start_time = get_cycles();
while (qi->free_cnt < 3) {
spin_unlock_irqrestore(&qi->q_lock, flags);
+ if (DMAR_OPERATION_TIMEOUT < (get_cycles() - start_time)) {
+ WARN(1, "No space in invalidation queue.\n");
+ return -ENOSPC;
+ }
cpu_relax();
spin_lock_irqsave(&qi->q_lock, flags);
}
@@ -675,6 +681,7 @@ int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu)
*/
writel(qi->free_head << 4, iommu->reg + DMAR_IQT_REG);
+ start_time = get_cycles();
while (qi->desc_status[wait_index] != QI_DONE) {
/*
* We will leave the interrupts disabled, to prevent interrupt
@@ -687,6 +694,11 @@ int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu)
if (rc)
goto out;
+ if (DMAR_OPERATION_TIMEOUT < (get_cycles() - start_time)) {
+ WARN(1, "Queued invalidation can not complete.\n");
+ goto out;
+ }
+
spin_unlock(&qi->q_lock);
cpu_relax();
spin_lock(&qi->q_lock);
next prev parent reply other threads:[~2009-05-20 17:43 UTC|newest]
Thread overview: 61+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20090327212241.234500000@intel.com>
2009-03-28 14:24 ` [patch 0/4] Intel IOMMU Supspend/Resume Support Andrew Lutomirski
2009-03-30 23:01 ` David Woodhouse
[not found] ` <20090327212321.520992000@intel.com>
2009-04-03 12:37 ` [patch 4/4] Intel IOMMU Suspend/Resume Support - Code Clean Up David Woodhouse
[not found] ` <20090327212321.070229000@intel.com>
2009-04-16 0:19 ` [PATCH] Intel IOMMU Pass Through Support Fenghua Yu
2009-04-16 2:13 ` Han, Weidong
2009-04-19 10:05 ` David Woodhouse
2009-04-20 17:27 ` Yu, Fenghua
2009-05-13 23:13 ` [PATCH] Fix Intel IOMMU Compilation Warnings on IA64 Fenghua Yu
2009-05-14 15:17 ` David Woodhouse
2009-05-14 15:31 ` Matthew Wilcox
2009-05-14 17:59 ` Fenghua Yu
2009-06-18 18:05 ` [PATCH 1/2] IOMMU Identity Mapping Support: iommu_identity_mapping definition Fenghua Yu
2009-06-18 18:08 ` Muli Ben-Yehuda
2009-06-18 18:13 ` Chris Wright
2009-06-18 18:14 ` Yu, Fenghua
2009-06-18 18:25 ` Muli Ben-Yehuda
2009-06-18 18:31 ` Chris Wright
2009-06-18 18:41 ` Muli Ben-Yehuda
2009-06-18 18:50 ` Yu, Fenghua
2009-06-18 18:51 ` Chris Wright
2009-06-18 19:09 ` Yu, Fenghua
2009-06-25 0:38 ` [PATCH] IA64 Compilation Error Fix for Intel IOMMU Identity Mapping Support Fenghua Yu
2009-06-25 1:00 ` FUJITA Tomonori
2009-06-25 4:16 ` [PATCH v2] " Fenghua Yu
2009-06-25 4:48 ` FUJITA Tomonori
2009-06-25 7:11 ` David Woodhouse
2009-06-25 21:52 ` David Woodhouse
2009-06-25 21:56 ` Yu, Fenghua
2009-06-26 18:21 ` David Woodhouse
2009-06-25 22:00 ` Linus Torvalds
2009-06-25 22:46 ` Tony Luck
2009-06-25 23:43 ` Chris Wright
2009-06-26 1:35 ` Linus Torvalds
2009-06-26 1:52 ` Chris Wright
2009-06-26 2:00 ` Linus Torvalds
2009-06-26 2:08 ` Chris Wright
2009-06-26 11:15 ` David Woodhouse
2009-06-27 0:03 ` Chris Wright
2009-06-27 11:44 ` David Woodhouse
2009-06-18 18:13 ` [PATCH 1/2] IOMMU Identity Mapping Support: iommu_identity_mapping definition Chris Wright
2009-06-18 18:28 ` Yu, Fenghua
2009-06-18 18:34 ` Chris Wright
2009-07-04 18:40 ` David Woodhouse
2009-05-20 17:42 ` Fenghua Yu [this message]
2009-05-27 5:51 ` [PATCH] Time out for possible dead loops during queued invalidation wait Andrew Morton
2009-05-27 22:40 ` Yu, Fenghua
2009-05-27 22:48 ` Andrew Morton
2009-05-27 23:25 ` Yu, Fenghua
2009-05-27 23:51 ` Andrew Morton
2009-05-28 0:47 ` Yu, Fenghua
2009-06-18 18:05 ` [PATCH 2/2] IOMMU Identity Mapping Support: Intel IOMMU implementation Fenghua Yu
2009-06-18 19:15 ` Chris Wright
2009-06-18 19:40 ` Yu, Fenghua
2009-06-18 20:02 ` Chris Wright
2009-06-19 20:47 ` [PATCH v2] IOMMU Identity Mapping Support (drivers/pci/intel_iommu.c) Fenghua Yu
2009-04-30 23:29 ` [PATCH] Intel IOMMU Pass Through Support Andrew Morton
2009-04-30 23:37 ` Randy Dunlap
2009-05-01 0:00 ` Andrew Morton
2009-05-01 0:57 ` Fenghua Yu
2009-05-01 0:05 ` Fenghua Yu
2009-05-01 0:14 ` Andrew Morton
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=20090520174259.GA10646@linux-os.sc.intel.com \
--to=fenghua.yu@intel.com \
--cc=akpm@linux-foundation.org \
--cc=dwmw2@infradead.org \
--cc=iommu@lists.linux-foundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
/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