mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Peter Ujfalusi <peter.ujfalusi@ti.com>
To: <vinod.koul@intel.com>
Cc: <linux-arm-kernel@lists.infradead.org>,
	<dmaengine@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
	<lars@metafoo.de>
Subject: [PATCH 02/10] dmaengine: virt-dma: Support for race free transfer termination
Date: Tue, 14 Nov 2017 16:32:04 +0200	[thread overview]
Message-ID: <20171114143212.8311-3-peter.ujfalusi@ti.com> (raw)
In-Reply-To: <20171114143212.8311-1-peter.ujfalusi@ti.com>

Even with the introduced vchan_synchronize() we can face race when
terminating a cyclic transfer.

If the terminate_all is called after the interrupt handler called
vchan_cyclic_callback(), but before the vchan_complete tasklet is called:
vc->cyclic is set to the cyclic descriptor, but the descriptor itself was
freed up in the driver's terminate_all() callback.
When the vhan_complete() is executed it will try to fetch the vc->cyclic
vdesc, but the pointer is pointing now to uninitialized memory leading to
(hard to reproduce) kernel crash.

In order to fix this, drivers should:
- call vchan_terminate_vdesc() from their terminate_all callback instead
calling their free_desc function to free up the descriptor.
- implement device_synchronize callback and call vchan_synchronize().

This way we can make sure that the descriptor is only going to be freed up
after the vchan_callback was executed in a safe manner.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
---
 drivers/dma/virt-dma.h | 30 ++++++++++++++++++++++++++++++
 1 file changed, 30 insertions(+)

diff --git a/drivers/dma/virt-dma.h b/drivers/dma/virt-dma.h
index 2edb05505102..b09b75ab0751 100644
--- a/drivers/dma/virt-dma.h
+++ b/drivers/dma/virt-dma.h
@@ -35,6 +35,7 @@ struct virt_dma_chan {
 	struct list_head desc_completed;
 
 	struct virt_dma_desc *cyclic;
+	struct virt_dma_desc *vd_terminated;
 };
 
 static inline struct virt_dma_chan *to_virt_chan(struct dma_chan *chan)
@@ -129,6 +130,25 @@ static inline void vchan_cyclic_callback(struct virt_dma_desc *vd)
 	tasklet_schedule(&vc->task);
 }
 
+/**
+ * vchan_terminate_vdesc - Disable pending cyclic callback
+ * @vd: virtual descriptor to be terminated
+ *
+ * vc.lock must be held by caller
+ */
+static inline void vchan_terminate_vdesc(struct virt_dma_desc *vd)
+{
+	struct virt_dma_chan *vc = to_virt_chan(vd->tx.chan);
+
+	/* free up stuck descriptor */
+	if (vc->vd_terminated)
+		vchan_vdesc_fini(vc->vd_terminated);
+
+	vc->vd_terminated = vd;
+	if (vc->cyclic == vd)
+		vc->cyclic = NULL;
+}
+
 /**
  * vchan_next_desc - peek at the next descriptor to be processed
  * @vc: virtual channel to obtain descriptor from
@@ -182,10 +202,20 @@ static inline void vchan_free_chan_resources(struct virt_dma_chan *vc)
  * Makes sure that all scheduled or active callbacks have finished running. For
  * proper operation the caller has to ensure that no new callbacks are scheduled
  * after the invocation of this function started.
+ * Free up the terminated cyclic descriptor to prevent memory leakage.
  */
 static inline void vchan_synchronize(struct virt_dma_chan *vc)
 {
+	unsigned long flags;
+
 	tasklet_kill(&vc->task);
+
+	spin_lock_irqsave(&vc->lock, flags);
+	if (vc->vd_terminated) {
+		vchan_vdesc_fini(vc->vd_terminated);
+		vc->vd_terminated = NULL;
+	}
+	spin_unlock_irqrestore(&vc->lock, flags);
 }
 
 #endif
-- 
Peter

Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki.
Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki

  parent reply	other threads:[~2017-11-14 14:36 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-11-14 14:32 [PATCH 00/10] dmaengine: fix race with vchan_complete Peter Ujfalusi
2017-11-14 14:32 ` [PATCH 01/10] dmaengine: virt-dma: Add helper to free/reuse a descriptor Peter Ujfalusi
2017-11-15  7:49   ` Linus Walleij
2017-11-14 14:32 ` Peter Ujfalusi [this message]
2017-11-15  7:50   ` [PATCH 02/10] dmaengine: virt-dma: Support for race free transfer termination Linus Walleij
2017-11-14 14:32 ` [PATCH 03/10] dmaengine: omap-dma: Use vchan_terminate_vdesc() instead of desc_free Peter Ujfalusi
2017-11-14 14:32 ` [PATCH 04/10] dmaengine: edma: " Peter Ujfalusi
2017-11-14 14:32 ` [PATCH 05/10] dmaengine: bcm2835-dma: " Peter Ujfalusi
2017-11-15 18:53   ` Eric Anholt
2017-11-21 11:42     ` Peter Ujfalusi
2017-11-14 14:32 ` [PATCH 06/10] dmaengine: dma-jz4780: " Peter Ujfalusi
2017-11-14 14:32 ` [PATCH 07/10] dmaengine: amba-pl08x: " Peter Ujfalusi
2017-11-15  7:48   ` Linus Walleij
2017-11-14 14:32 ` [PATCH 08/10] dmaengine: img-mdc-dma: " Peter Ujfalusi
2017-11-14 14:32 ` [PATCH 09/10] dmaengine: k3dma: " Peter Ujfalusi
2017-11-15  4:06   ` zhangfei
2017-11-14 14:32 ` [PATCH 10/10] dmaengine: s3c24xx-dma: " Peter Ujfalusi
2017-11-14 14:47 ` [PATCH 00/10] dmaengine: fix race with vchan_complete Peter Ujfalusi
2017-12-04 17:07 ` Vinod Koul

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=20171114143212.8311-3-peter.ujfalusi@ti.com \
    --to=peter.ujfalusi@ti.com \
    --cc=dmaengine@vger.kernel.org \
    --cc=lars@metafoo.de \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=vinod.koul@intel.com \
    /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

all inboxes | Powered by JetHome®