public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* pxa2xx_spi: spi_bfin5xx: limit reaches -1 rather than 0
@ 2009-02-08 15:53 Roel Kluin
  2009-02-08 16:02 ` [PATCH v2] " Roel Kluin
  2009-02-08 16:07 ` David Brownell
  0 siblings, 2 replies; 4+ messages in thread
From: Roel Kluin @ 2009-02-08 15:53 UTC (permalink / raw)
  To: dbrownell; +Cc: spi-devel-general, lkml

vi drivers/spi/pxa2xx_spi.c +919 
vi drivers/spi/spi_bfin5xx.c +645

        if (flush(drv_data) == 0) {
                dev_err(&drv_data->pdev->dev, "pump_transfers: flush failed\n");
                message->status = -EIO;
                giveback(drv_data);
                return;
        }

But with a postfix decrement limit reaches -1 rather than 0.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
---
 drivers/spi/pxa2xx_spi.c  |    2 +-
 drivers/spi/spi_bfin5xx.c |    2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/spi/pxa2xx_spi.c b/drivers/spi/pxa2xx_spi.c
index d0fc4ca..3164741 100644
--- a/drivers/spi/pxa2xx_spi.c
+++ b/drivers/spi/pxa2xx_spi.c
@@ -185,7 +185,7 @@ static int flush(struct driver_data *drv_data)
 		while (read_SSSR(reg) & SSSR_RNE) {
 			read_SSDR(reg);
 		}
-	} while ((read_SSSR(reg) & SSSR_BSY) && limit--);
+	} while ((read_SSSR(reg) & SSSR_BSY) && --limit);
 	write_SSSR(SSSR_ROR, reg);
 
 	return limit;
diff --git a/drivers/spi/spi_bfin5xx.c b/drivers/spi/spi_bfin5xx.c
index 7fea3cf..ea94567 100644
--- a/drivers/spi/spi_bfin5xx.c
+++ b/drivers/spi/spi_bfin5xx.c
@@ -162,7 +162,7 @@ static int flush(struct driver_data *drv_data)
 	unsigned long limit = loops_per_jiffy << 1;
 
 	/* wait for stop and clear stat */
-	while (!(read_STAT(drv_data) & BIT_STAT_SPIF) && limit--)
+	while (!(read_STAT(drv_data) & BIT_STAT_SPIF) && --limit)
 		cpu_relax();
 
 	write_STAT(drv_data, BIT_STAT_CLR);

^ permalink raw reply related	[flat|nested] 4+ messages in thread

* [PATCH v2] pxa2xx_spi: spi_bfin5xx: limit reaches -1 rather than 0
  2009-02-08 15:53 pxa2xx_spi: spi_bfin5xx: limit reaches -1 rather than 0 Roel Kluin
@ 2009-02-08 16:02 ` Roel Kluin
  2009-02-08 16:07 ` David Brownell
  1 sibling, 0 replies; 4+ messages in thread
From: Roel Kluin @ 2009-02-08 16:02 UTC (permalink / raw)
  To: dbrownell; +Cc: spi-devel-general, lkml

vi drivers/spi/pxa2xx_spi.c +919 
vi drivers/spi/spi_bfin5xx.c +645

        if (flush(drv_data) == 0) {
                dev_err(&drv_data->pdev->dev, "pump_transfers: flush failed\n");
                message->status = -EIO;
                giveback(drv_data);
                return;
        }

and:

drivers/spi/pxa2xx_spi.c:515:   if (wait_ssp_rx_stall(drv_data->ioaddr) == 0)
drivers/spi/pxa2xx_spi.c:569:           if (wait_ssp_rx_stall(drv_data->ioaddr) == 0)
drivers/spi/pxa2xx_spi.c:695:                   if (!wait_ssp_rx_stall(reg))
drivers/spi/pxa2xx_spi.c:511:   if (wait_dma_channel_stop(drv_data->rx_channel) == 0)

But with a postfix decrement limit reaches -1 rather than 0.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
---
Sorry, I didn't catch two the first time

 drivers/spi/pxa2xx_spi.c  |    6 +++---
 drivers/spi/spi_bfin5xx.c |    2 +-
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/spi/pxa2xx_spi.c b/drivers/spi/pxa2xx_spi.c
index d0fc4ca..318e568 100644
--- a/drivers/spi/pxa2xx_spi.c
+++ b/drivers/spi/pxa2xx_spi.c
@@ -185,7 +185,7 @@ static int flush(struct driver_data *drv_data)
 		while (read_SSSR(reg) & SSSR_RNE) {
 			read_SSDR(reg);
 		}
-	} while ((read_SSSR(reg) & SSSR_BSY) && limit--);
+	} while ((read_SSSR(reg) & SSSR_BSY) && --limit);
 	write_SSSR(SSSR_ROR, reg);
 
 	return limit;
@@ -459,7 +459,7 @@ static int wait_ssp_rx_stall(void const __iomem *ioaddr)
 {
 	unsigned long limit = loops_per_jiffy << 1;
 
-	while ((read_SSSR(ioaddr) & SSSR_BSY) && limit--)
+	while ((read_SSSR(ioaddr) & SSSR_BSY) && --limit)
 		cpu_relax();
 
 	return limit;
@@ -469,7 +469,7 @@ static int wait_dma_channel_stop(int channel)
 {
 	unsigned long limit = loops_per_jiffy << 1;
 
-	while (!(DCSR(channel) & DCSR_STOPSTATE) && limit--)
+	while (!(DCSR(channel) & DCSR_STOPSTATE) && --limit)
 		cpu_relax();
 
 	return limit;
diff --git a/drivers/spi/spi_bfin5xx.c b/drivers/spi/spi_bfin5xx.c
index 7fea3cf..ea94567 100644
--- a/drivers/spi/spi_bfin5xx.c
+++ b/drivers/spi/spi_bfin5xx.c
@@ -162,7 +162,7 @@ static int flush(struct driver_data *drv_data)
 	unsigned long limit = loops_per_jiffy << 1;
 
 	/* wait for stop and clear stat */
-	while (!(read_STAT(drv_data) & BIT_STAT_SPIF) && limit--)
+	while (!(read_STAT(drv_data) & BIT_STAT_SPIF) && --limit)
 		cpu_relax();
 
 	write_STAT(drv_data, BIT_STAT_CLR);

^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: pxa2xx_spi: spi_bfin5xx: limit reaches -1 rather than 0
  2009-02-08 15:53 pxa2xx_spi: spi_bfin5xx: limit reaches -1 rather than 0 Roel Kluin
  2009-02-08 16:02 ` [PATCH v2] " Roel Kluin
@ 2009-02-08 16:07 ` David Brownell
  2009-02-08 16:59   ` roel kluin
  1 sibling, 1 reply; 4+ messages in thread
From: David Brownell @ 2009-02-08 16:07 UTC (permalink / raw)
  To: Roel Kluin; +Cc: spi-devel-general, lkml, Andrew Morton

On Sunday 08 February 2009, Roel Kluin wrote:
> vi drivers/spi/pxa2xx_spi.c +919 
> vi drivers/spi/spi_bfin5xx.c +645
> 
>         if (flush(drv_data) == 0) {
>                 dev_err(&drv_data->pdev->dev, "pump_transfers: flush failed\n");
>                 message->status = -EIO;
>                 giveback(drv_data);
>                 return;
>         }
> 
> But with a postfix decrement limit reaches -1 rather than 0.
> 
> Signed-off-by: Roel Kluin <roel.kluin@gmail.com>

Acked-by: David Brownell <dbrownell@users.sourceforge.net>

Although, note that if the *actual* count mattered here,
the correect patch would be to test "flush() < 0)" ...
except when using a "do {...} while (--limit)" idiom.

Try the thought experiment of starting with "limit = 1";
not all of these patches of your would go through the 
loop even once.  Fortunately in all these cases, the
limit is an arbitrary "don't loop forever" thing.


> ---
>  drivers/spi/pxa2xx_spi.c  |    2 +-
>  drivers/spi/spi_bfin5xx.c |    2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/spi/pxa2xx_spi.c b/drivers/spi/pxa2xx_spi.c
> index d0fc4ca..3164741 100644
> --- a/drivers/spi/pxa2xx_spi.c
> +++ b/drivers/spi/pxa2xx_spi.c
> @@ -185,7 +185,7 @@ static int flush(struct driver_data *drv_data)
>  		while (read_SSSR(reg) & SSSR_RNE) {
>  			read_SSDR(reg);
>  		}
> -	} while ((read_SSSR(reg) & SSSR_BSY) && limit--);
> +	} while ((read_SSSR(reg) & SSSR_BSY) && --limit);
>  	write_SSSR(SSSR_ROR, reg);
>  
>  	return limit;
> diff --git a/drivers/spi/spi_bfin5xx.c b/drivers/spi/spi_bfin5xx.c
> index 7fea3cf..ea94567 100644
> --- a/drivers/spi/spi_bfin5xx.c
> +++ b/drivers/spi/spi_bfin5xx.c
> @@ -162,7 +162,7 @@ static int flush(struct driver_data *drv_data)
>  	unsigned long limit = loops_per_jiffy << 1;
>  
>  	/* wait for stop and clear stat */
> -	while (!(read_STAT(drv_data) & BIT_STAT_SPIF) && limit--)
> +	while (!(read_STAT(drv_data) & BIT_STAT_SPIF) && --limit)
>  		cpu_relax();
>  
>  	write_STAT(drv_data, BIT_STAT_CLR);
> 
> 



^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: pxa2xx_spi: spi_bfin5xx: limit reaches -1 rather than 0
  2009-02-08 16:07 ` David Brownell
@ 2009-02-08 16:59   ` roel kluin
  0 siblings, 0 replies; 4+ messages in thread
From: roel kluin @ 2009-02-08 16:59 UTC (permalink / raw)
  To: David Brownell; +Cc: spi-devel-general, lkml, Andrew Morton

2009/2/8 David Brownell <david-b@pacbell.net>:
> On Sunday 08 February 2009, Roel Kluin wrote:
>> vi drivers/spi/pxa2xx_spi.c +919
>> vi drivers/spi/spi_bfin5xx.c +645
>>
>>         if (flush(drv_data) == 0) {
>>                 dev_err(&drv_data->pdev->dev, "pump_transfers: flush failed\n");
>>                 message->status = -EIO;
>>                 giveback(drv_data);
>>                 return;
>>         }
>>
>> But with a postfix decrement limit reaches -1 rather than 0.
>>
>> Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
>
> Acked-by: David Brownell <dbrownell@users.sourceforge.net>

could you please review the patch at http://lkml.org/lkml/2009/2/8/72
instead? I initially overlooked two similar issues in the same file.

> Although, note that if the *actual* count mattered here,
> the correect patch would be to test "flush() < 0)" ...
> except when using a "do {...} while (--limit)" idiom.

True. You can also increase the limit with 1 when changing post- to
prefix for the while (--limit) {...} idiom.

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2009-02-08 16:59 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-02-08 15:53 pxa2xx_spi: spi_bfin5xx: limit reaches -1 rather than 0 Roel Kluin
2009-02-08 16:02 ` [PATCH v2] " Roel Kluin
2009-02-08 16:07 ` David Brownell
2009-02-08 16:59   ` roel kluin

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox