linux-pci.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jiang Liu <jiang.liu@linux.intel.com>
To: Joerg Roedel <joro@8bytes.org>,
	David Woodhouse <dwmw2@infradead.org>,
	Dan Williams <dan.j.williams@intel.com>,
	Vinod Koul <vinod.koul@intel.com>,
	Ashok Raj <ashok.raj@intel.com>,
	Yijing Wang <wangyijing@huawei.com>,
	iommu@lists.linux-foundation.org, linux-kernel@vger.kernel.org
Cc: Jiang Liu <jiang.liu@linux.intel.com>,
	Tony Luck <tony.luck@intel.com>, Yinghai Lu <yinghai@kernel.org>,
	linux-pci@vger.kernel.org, dmaengine@vger.kernel.org
Subject: [Patch Part1 V2 15/20] iommu/vt-d: fix access after free issue in function free_dmar_iommu()
Date: Fri,  6 Dec 2013 11:21:18 +0800	[thread overview]
Message-ID: <1386300083-6882-16-git-send-email-jiang.liu@linux.intel.com> (raw)
In-Reply-To: <1386300083-6882-1-git-send-email-jiang.liu@linux.intel.com>

Function free_dmar_iommu() may access domain->iommu_lock by
	spin_unlock_irqrestore(&domain->iommu_lock, flags);
after freeing corresponding domain structure.

Sample stack dump:
[    8.912818] =========================
[    8.917072] [ BUG: held lock freed! ]
[    8.921335] 3.13.0-rc1-gerry+ #12 Not tainted
[    8.926375] -------------------------
[    8.930629] swapper/0/1 is freeing memory ffff880c23b56040-ffff880c23b5613f, with a lock still held there!
[    8.941675]  (&(&domain->iommu_lock)->rlock){......}, at: [<ffffffff81dc775c>] init_dmars+0x72c/0x95b
[    8.952582] 1 lock held by swapper/0/1:
[    8.957031]  #0:  (&(&domain->iommu_lock)->rlock){......}, at: [<ffffffff81dc775c>] init_dmars+0x72c/0x95b
[    8.968487]
[    8.968487] stack backtrace:
[    8.973602] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 3.13.0-rc1-gerry+ #12
[    8.981556] Hardware name: Intel Corporation LH Pass ........../SVRBD-ROW_T, BIOS SE5C600.86B.99.99.x059.091020121352 09/10/2012
[    8.994742]  ffff880c23b56040 ffff88042dd33c98 ffffffff815617fd ffff88042dd38b28
[    9.003566]  ffff88042dd33cd0 ffffffff810a977a ffff880c23b56040 0000000000000086
[    9.012403]  ffff88102c4923c0 ffff88042ddb4800 ffffffff81b1e8c0 ffff88042dd33d28
[    9.021240] Call Trace:
[    9.024138]  [<ffffffff815617fd>] dump_stack+0x4d/0x66
[    9.030057]  [<ffffffff810a977a>] debug_check_no_locks_freed+0x15a/0x160
[    9.037723]  [<ffffffff811aa1c2>] kmem_cache_free+0x62/0x5b0
[    9.044225]  [<ffffffff81465e27>] domain_exit+0x197/0x1c0
[    9.050418]  [<ffffffff81dc7788>] init_dmars+0x758/0x95b
[    9.056527]  [<ffffffff81dc7dfa>] intel_iommu_init+0x351/0x438
[    9.063207]  [<ffffffff81d8a711>] ? iommu_setup+0x27d/0x27d
[    9.069601]  [<ffffffff81d8a739>] pci_iommu_init+0x28/0x52
[    9.075910]  [<ffffffff81000342>] do_one_initcall+0x122/0x180
[    9.082509]  [<ffffffff81077738>] ? parse_args+0x1e8/0x320
[    9.088815]  [<ffffffff81d850e8>] kernel_init_freeable+0x1e1/0x26c
[    9.095895]  [<ffffffff81d84833>] ? do_early_param+0x88/0x88
[    9.102396]  [<ffffffff8154f580>] ? rest_init+0xd0/0xd0
[    9.108410]  [<ffffffff8154f58e>] kernel_init+0xe/0x130
[    9.114423]  [<ffffffff81574a2c>] ret_from_fork+0x7c/0xb0
[    9.120612]  [<ffffffff8154f580>] ? rest_init+0xd0/0xd0

Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com>
---
 drivers/iommu/intel-iommu.c |    7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index 1a20171..fc3473b 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -1266,7 +1266,7 @@ static void vm_domain_exit(struct dmar_domain *domain);
 static void free_dmar_iommu(struct intel_iommu *iommu)
 {
 	struct dmar_domain *domain;
-	int i;
+	int i, count;
 	unsigned long flags;
 
 	if ((iommu->domains) && (iommu->domain_ids)) {
@@ -1275,13 +1275,14 @@ static void free_dmar_iommu(struct intel_iommu *iommu)
 			clear_bit(i, iommu->domain_ids);
 
 			spin_lock_irqsave(&domain->iommu_lock, flags);
-			if (--domain->iommu_count == 0) {
+			count = --domain->iommu_count;
+			spin_unlock_irqrestore(&domain->iommu_lock, flags);
+			if (count == 0) {
 				if (domain->flags & DOMAIN_FLAG_VIRTUAL_MACHINE)
 					vm_domain_exit(domain);
 				else
 					domain_exit(domain);
 			}
-			spin_unlock_irqrestore(&domain->iommu_lock, flags);
 		}
 	}
 
-- 
1.7.10.4


  parent reply	other threads:[~2013-12-06  3:21 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-12-06  3:21 [Patch Part1 V2 00/20] Bugfixes and improvements for Intel IOMMU drivers Jiang Liu
2013-12-06  3:21 ` [Patch Part1 V2 01/20] iommu/vt-d: use dedicated bitmap to track remapping entry allocation status Jiang Liu
2013-12-06  3:21 ` [Patch Part1 V2 02/20] iommu/vt-d: fix PCI device reference leakage on error recovery path Jiang Liu
2013-12-06  3:21 ` [Patch Part1 V2 03/20] iommu/vt-d: fix a race window in allocating domain ID for virtual machines Jiang Liu
2013-12-06  3:21 ` [Patch Part1 V2 04/20] iommu/vt-d: fix resource leakage on error recovery path in iommu_init_domains() Jiang Liu
2013-12-06  3:21 ` [Patch Part1 V2 05/20] iommu/vt-d, trivial: refine support of 64bit guest address Jiang Liu
2013-12-06  3:21 ` [Patch Part1 V2 06/20] iommu/vt-d, trivial: print correct domain id of static identity domain Jiang Liu
2013-12-06  3:21 ` [Patch Part1 V2 07/20] iommu/vt-d, trivial: check suitable flag in function detect_intel_iommu() Jiang Liu
2013-12-06  3:21 ` [Patch Part1 V2 08/20] iommu/vt-d, trivial: clean up unused code Jiang Liu
2013-12-06  3:21 ` [Patch Part1 V2 09/20] iommu/vt-d: mark internal functions as static Jiang Liu
2013-12-06  3:21 ` [Patch Part1 V2 10/20] iommu/vt-d, trivial: use defined macro instead of hardcoding Jiang Liu
2013-12-06  3:21 ` [Patch Part1 V2 11/20] iommu/vt-d, trivial: simplify code with existing macros Jiang Liu
2013-12-06  3:21 ` [Patch Part1 V2 12/20] iommu/vt-d: fix invalid memory access when freeing DMAR irq Jiang Liu
2013-12-06  3:21 ` [Patch Part1 V2 13/20] iommu/vt-d: keep shared resources when failed to initialize iommu devices Jiang Liu
2013-12-06  3:21 ` [Patch Part1 V2 14/20] iommu/vt-d: avoid double free in error recovery path Jiang Liu
2013-12-06  3:21 ` Jiang Liu [this message]
2013-12-06  3:21 ` [Patch Part1 V2 16/20] iommu/vt-d: release invalidation queue when destroying IOMMU unit Jiang Liu
2013-12-06  3:21 ` [Patch Part1 V2 17/20] iommu/vt-d: fix wrong return value of dmar_table_init() Jiang Liu
2013-12-06  3:21 ` [Patch Part1 V2 18/20] iommu/vt-d, PCI, trivial: use dev_is_pci() instead of hardcoding Jiang Liu
2013-12-06  3:21 ` [Patch Part1 V2 19/20] iommu/vt-d, trivial: clean sparse warnings Jiang Liu
2013-12-06  3:21 ` [Patch Part1 V2 20/20] iommu/vt-d: free all resources if failed to initialize DMARs Jiang Liu

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=1386300083-6882-16-git-send-email-jiang.liu@linux.intel.com \
    --to=jiang.liu@linux.intel.com \
    --cc=ashok.raj@intel.com \
    --cc=dan.j.williams@intel.com \
    --cc=dmaengine@vger.kernel.org \
    --cc=dwmw2@infradead.org \
    --cc=iommu@lists.linux-foundation.org \
    --cc=joro@8bytes.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=tony.luck@intel.com \
    --cc=vinod.koul@intel.com \
    --cc=wangyijing@huawei.com \
    --cc=yinghai@kernel.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).