[PATCH v7 6/8] fsl-dma: use spin_lock_bh to instead of spin_lock_irqsave

[PATCH v7 6/8] fsl-dma: use spin_lock_bh to instead of spin_lock_irqsave

[qiang.liu]

From: Qiang Liu <qiang.liu@freescale.com>

The use of spin_lock_irqsave() is a stronger locking mechanism than is
required throughout the driver. The minimum locking required should be
used instead. Interrupts will be turned off and context will be saved,
there is needless to use irqsave.

Change all instances of spin_lock_irqsave() to spin_lock_bh().
All manipulation of protected fields is done using tasklet context or
weaker, which makes spin_lock_bh() the correct choice.

Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Dan Williams <dan.j.williams@gmail.com>
Cc: Vinod Koul <vinod.koul@intel.com>
Cc: Li Yang <leoli@freescale.com>
Cc: Timur Tabi <timur@freescale.com>
Signed-off-by: Qiang Liu <qiang.liu@freescale.com>
Acked-by: Ira W. Snyder <iws@ovro.caltech.edu>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
Comments by Arnd Bergmann in v6:
"You could actually change the use of spin_lock_bh inside of the tasklet
function (dma_do_tasklet) do just spin_lock(), because softirqs are
already disabled there, but your version is also ok."

 drivers/dma/fsldma.c |   30 ++++++++++++------------------
 1 files changed, 12 insertions(+), 18 deletions(-)

diff --git a/drivers/dma/fsldma.c b/drivers/dma/fsldma.c
index b05a81f..8b9c0f7 100644
--- a/drivers/dma/fsldma.c
+++ b/drivers/dma/fsldma.c
@@ -405,10 +405,9 @@ static dma_cookie_t fsl_dma_tx_submit(struct dma_async_tx_descriptor *tx)
 	struct fsldma_chan *chan = to_fsl_chan(tx->chan);
 	struct fsl_desc_sw *desc = tx_to_fsl_desc(tx);
 	struct fsl_desc_sw *child;
-	unsigned long flags;
 	dma_cookie_t cookie;

-	spin_lock_irqsave(&chan->desc_lock, flags);
+	spin_lock_bh(&chan->desc_lock);

 	/*
 	 * assign cookies to all of the software descriptors
@@ -421,7 +420,7 @@ static dma_cookie_t fsl_dma_tx_submit(struct dma_async_tx_descriptor *tx)
 	/* put this transaction onto the tail of the pending queue */
 	append_ld_queue(chan, desc);

-	spin_unlock_irqrestore(&chan->desc_lock, flags);
+	spin_unlock_bh(&chan->desc_lock);

 	return cookie;
 }
@@ -762,15 +761,14 @@ static void fsldma_free_desc_list_reverse(struct fsldma_chan *chan,
 static void fsl_dma_free_chan_resources(struct dma_chan *dchan)
 {
 	struct fsldma_chan *chan = to_fsl_chan(dchan);
-	unsigned long flags;

 	chan_dbg(chan, "free all channel resources\n");
-	spin_lock_irqsave(&chan->desc_lock, flags);
+	spin_lock_bh(&chan->desc_lock);
 	fsldma_cleanup_descriptors(chan);
 	fsldma_free_desc_list(chan, &chan->ld_pending);
 	fsldma_free_desc_list(chan, &chan->ld_running);
 	fsldma_free_desc_list(chan, &chan->ld_completed);
-	spin_unlock_irqrestore(&chan->desc_lock, flags);
+	spin_unlock_bh(&chan->desc_lock);

 	dma_pool_destroy(chan->desc_pool);
 	chan->desc_pool = NULL;
@@ -989,7 +987,6 @@ static int fsl_dma_device_control(struct dma_chan *dchan,
 {
 	struct dma_slave_config *config;
 	struct fsldma_chan *chan;
-	unsigned long flags;
 	int size;

 	if (!dchan)
@@ -999,7 +996,7 @@ static int fsl_dma_device_control(struct dma_chan *dchan,

 	switch (cmd) {
 	case DMA_TERMINATE_ALL:
-		spin_lock_irqsave(&chan->desc_lock, flags);
+		spin_lock_bh(&chan->desc_lock);

 		/* Halt the DMA engine */
 		dma_halt(chan);
@@ -1010,7 +1007,7 @@ static int fsl_dma_device_control(struct dma_chan *dchan,
 		fsldma_free_desc_list(chan, &chan->ld_completed);
 		chan->idle = true;

-		spin_unlock_irqrestore(&chan->desc_lock, flags);
+		spin_unlock_bh(&chan->desc_lock);
 		return 0;

 	case DMA_SLAVE_CONFIG:
@@ -1052,11 +1049,10 @@ static int fsl_dma_device_control(struct dma_chan *dchan,
 static void fsl_dma_memcpy_issue_pending(struct dma_chan *dchan)
 {
 	struct fsldma_chan *chan = to_fsl_chan(dchan);
-	unsigned long flags;

-	spin_lock_irqsave(&chan->desc_lock, flags);
+	spin_lock_bh(&chan->desc_lock);
 	fsl_chan_xfer_ld_queue(chan);
-	spin_unlock_irqrestore(&chan->desc_lock, flags);
+	spin_unlock_bh(&chan->desc_lock);
 }

 /**
@@ -1069,15 +1065,14 @@ static enum dma_status fsl_tx_status(struct dma_chan *dchan,
 {
 	struct fsldma_chan *chan = to_fsl_chan(dchan);
 	enum dma_status ret;
-	unsigned long flags;

 	ret = dma_cookie_status(dchan, cookie, txstate);
 	if (ret == DMA_SUCCESS)
 		return ret;

-	spin_lock_irqsave(&chan->desc_lock, flags);
+	spin_lock_bh(&chan->desc_lock);
 	fsldma_cleanup_descriptors(chan);
-	spin_unlock_irqrestore(&chan->desc_lock, flags);
+	spin_unlock_bh(&chan->desc_lock);

 	return dma_cookie_status(dchan, cookie, txstate);
 }
@@ -1156,11 +1151,10 @@ static irqreturn_t fsldma_chan_irq(int irq, void *data)
 static void dma_do_tasklet(unsigned long data)
 {
 	struct fsldma_chan *chan = (struct fsldma_chan *)data;
-	unsigned long flags;

 	chan_dbg(chan, "tasklet entry\n");

-	spin_lock_irqsave(&chan->desc_lock, flags);
+	spin_lock_bh(&chan->desc_lock);

 	/* the hardware is now idle and ready for more */
 	chan->idle = true;
@@ -1168,7 +1162,7 @@ static void dma_do_tasklet(unsigned long data)
 	/* Run all cleanup for descriptors which have been completed */
 	fsldma_cleanup_descriptors(chan);

-	spin_unlock_irqrestore(&chan->desc_lock, flags);
+	spin_unlock_bh(&chan->desc_lock);

 	chan_dbg(chan, "tasklet exit\n");
 }
--
1.7.5.1

Re: [PATCH v7 6/8] fsl-dma: use spin_lock_bh to instead of spin_lock_irqsave

[Dan Williams]

On Thu, Aug 9, 2012 at 1:23 AM,  <qiang.liu@freescale.com> wrote:
> From: Qiang Liu <qiang.liu@freescale.com>
>
> The use of spin_lock_irqsave() is a stronger locking mechanism than is
> required throughout the driver. The minimum locking required should be
> used instead. Interrupts will be turned off and context will be saved,
> there is needless to use irqsave.
>
> Change all instances of spin_lock_irqsave() to spin_lock_bh().
> All manipulation of protected fields is done using tasklet context or
> weaker, which makes spin_lock_bh() the correct choice.

It seems you are coordinating fsl-dma copy and talitos xor operations.
 It looks like fsl-dma will be called through
talitos_process_pending()->dma_run_dependencies(), which is
potentially called in hard irq context.

This all comes back to the need to fix raid offload to manage the
channels explicitly rather than the current dependency chains.

--
Dan

RE: [PATCH v7 6/8] fsl-dma: use spin_lock_bh to instead of spin_lock_irqsave

[Liu Qiang-B32616]

> -----Original Message-----
> From: dan.j.williams@gmail.com [mailto:dan.j.williams@gmail.com] On
> Behalf Of Dan Williams
> Sent: Sunday, September 02, 2012 4:41 PM
> To: Liu Qiang-B32616
> Cc: linux-crypto@vger.kernel.org; linuxppc-dev@lists.ozlabs.org; linux-
> kernel@vger.kernel.org; vinod.koul@intel.com; Phillips Kim-R1AAHA;
> herbert@gondor.hengli.com.au; davem@davemloft.net; arnd@arndb.de;
> gregkh@linuxfoundation.org; Li Yang-R58472; Tabi Timur-B04825
> Subject: Re: [PATCH v7 6/8] fsl-dma: use spin_lock_bh to instead of
> spin_lock_irqsave
> 
> On Thu, Aug 9, 2012 at 1:23 AM,  <qiang.liu@freescale.com> wrote:
> > From: Qiang Liu <qiang.liu@freescale.com>
> >
> > The use of spin_lock_irqsave() is a stronger locking mechanism than is
> > required throughout the driver. The minimum locking required should be
> > used instead. Interrupts will be turned off and context will be saved,
> > there is needless to use irqsave.
> >
> > Change all instances of spin_lock_irqsave() to spin_lock_bh().
> > All manipulation of protected fields is done using tasklet context or
> > weaker, which makes spin_lock_bh() the correct choice.
> 
> It seems you are coordinating fsl-dma copy and talitos xor operations.
>  It looks like fsl-dma will be called through
> talitos_process_pending()->dma_run_dependencies(), which is
> potentially called in hard irq context.
> 
> This all comes back to the need to fix raid offload to manage the
> channels explicitly rather than the current dependency chains.
So you mean I must implement talitos_run_dependencies() and fsldma_run_dependencies()? Invoke async_tx->callback() respectively.
How about avoiding irq context in talitos?

> 
> --
> Dan