* [PATCH v2 net-next] drivers: net: davinci_mdio: prevent spurious timeout
@ 2018-05-09 11:00 ` Sekhar Nori
0 siblings, 0 replies; 5+ messages in thread
From: Sekhar Nori @ 2018-05-09 11:00 UTC (permalink / raw)
To: David S . Miller
Cc: Grygorii Strashko, linux-omap, netdev, Andrew Lunn, Sekhar Nori
A well timed kernel preemption in the time_after() loop
in wait_for_idle() can result in a spurious timeout
error to be returned.
Fix it by using readl_poll_timeout() which takes care of
this issue.
Signed-off-by: Sekhar Nori <nsekhar@ti.com>
---
v2: use readl_poll_timeout() per suggestion from Andrew.
The issue has not been personally observed by me, but has
been reported by users. Sending for next-next given the
non-critical nature. There is seems to be no easy way to
reproduce this.
drivers/net/ethernet/ti/davinci_mdio.c | 15 +++++++++------
1 file changed, 9 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c
index 3c33f4504d8e..d073432a5dbe 100644
--- a/drivers/net/ethernet/ti/davinci_mdio.c
+++ b/drivers/net/ethernet/ti/davinci_mdio.c
@@ -34,6 +34,7 @@
#include <linux/clk.h>
#include <linux/err.h>
#include <linux/io.h>
+#include <linux/iopoll.h>
#include <linux/pm_runtime.h>
#include <linux/davinci_emac.h>
#include <linux/of.h>
@@ -227,14 +228,16 @@ static inline int wait_for_user_access(struct davinci_mdio_data *data)
static inline int wait_for_idle(struct davinci_mdio_data *data)
{
struct davinci_mdio_regs __iomem *regs = data->regs;
- unsigned long timeout = jiffies + msecs_to_jiffies(MDIO_TIMEOUT);
+ u32 val, ret;
- while (time_after(timeout, jiffies)) {
- if (__raw_readl(®s->control) & CONTROL_IDLE)
- return 0;
+ ret = readl_poll_timeout(®s->control, val, val & CONTROL_IDLE,
+ 0, MDIO_TIMEOUT * 1000);
+ if (ret) {
+ dev_err(data->dev, "timed out waiting for idle\n");
+ return ret;
}
- dev_err(data->dev, "timed out waiting for idle\n");
- return -ETIMEDOUT;
+
+ return 0;
}
static int davinci_mdio_read(struct mii_bus *bus, int phy_id, int phy_reg)
--
2.16.2
^ permalink raw reply related [flat|nested] 5+ messages in thread* [PATCH v2 net-next] drivers: net: davinci_mdio: prevent spurious timeout
@ 2018-05-09 11:00 ` Sekhar Nori
0 siblings, 0 replies; 5+ messages in thread
From: Sekhar Nori @ 2018-05-09 11:00 UTC (permalink / raw)
To: David S . Miller
Cc: Grygorii Strashko, linux-omap, netdev, Andrew Lunn, Sekhar Nori
A well timed kernel preemption in the time_after() loop
in wait_for_idle() can result in a spurious timeout
error to be returned.
Fix it by using readl_poll_timeout() which takes care of
this issue.
Signed-off-by: Sekhar Nori <nsekhar@ti.com>
---
v2: use readl_poll_timeout() per suggestion from Andrew.
The issue has not been personally observed by me, but has
been reported by users. Sending for next-next given the
non-critical nature. There is seems to be no easy way to
reproduce this.
drivers/net/ethernet/ti/davinci_mdio.c | 15 +++++++++------
1 file changed, 9 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c
index 3c33f4504d8e..d073432a5dbe 100644
--- a/drivers/net/ethernet/ti/davinci_mdio.c
+++ b/drivers/net/ethernet/ti/davinci_mdio.c
@@ -34,6 +34,7 @@
#include <linux/clk.h>
#include <linux/err.h>
#include <linux/io.h>
+#include <linux/iopoll.h>
#include <linux/pm_runtime.h>
#include <linux/davinci_emac.h>
#include <linux/of.h>
@@ -227,14 +228,16 @@ static inline int wait_for_user_access(struct davinci_mdio_data *data)
static inline int wait_for_idle(struct davinci_mdio_data *data)
{
struct davinci_mdio_regs __iomem *regs = data->regs;
- unsigned long timeout = jiffies + msecs_to_jiffies(MDIO_TIMEOUT);
+ u32 val, ret;
- while (time_after(timeout, jiffies)) {
- if (__raw_readl(®s->control) & CONTROL_IDLE)
- return 0;
+ ret = readl_poll_timeout(®s->control, val, val & CONTROL_IDLE,
+ 0, MDIO_TIMEOUT * 1000);
+ if (ret) {
+ dev_err(data->dev, "timed out waiting for idle\n");
+ return ret;
}
- dev_err(data->dev, "timed out waiting for idle\n");
- return -ETIMEDOUT;
+
+ return 0;
}
static int davinci_mdio_read(struct mii_bus *bus, int phy_id, int phy_reg)
--
2.16.2
^ permalink raw reply related [flat|nested] 5+ messages in thread* Re: [PATCH v2 net-next] drivers: net: davinci_mdio: prevent spurious timeout
2018-05-09 11:00 ` Sekhar Nori
(?)
@ 2018-05-09 13:30 ` Andrew Lunn
2018-05-09 15:46 ` Sekhar Nori
-1 siblings, 1 reply; 5+ messages in thread
From: Andrew Lunn @ 2018-05-09 13:30 UTC (permalink / raw)
To: Sekhar Nori; +Cc: David S . Miller, Grygorii Strashko, linux-omap, netdev
On Wed, May 09, 2018 at 04:30:24PM +0530, Sekhar Nori wrote:
> A well timed kernel preemption in the time_after() loop
> in wait_for_idle() can result in a spurious timeout
> error to be returned.
>
> Fix it by using readl_poll_timeout() which takes care of
> this issue.
>
> Signed-off-by: Sekhar Nori <nsekhar@ti.com>
> ---
> v2: use readl_poll_timeout() per suggestion from Andrew.
>
> The issue has not been personally observed by me, but has
> been reported by users. Sending for next-next given the
> non-critical nature. There is seems to be no easy way to
> reproduce this.
>
> drivers/net/ethernet/ti/davinci_mdio.c | 15 +++++++++------
> 1 file changed, 9 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c
> index 3c33f4504d8e..d073432a5dbe 100644
> --- a/drivers/net/ethernet/ti/davinci_mdio.c
> +++ b/drivers/net/ethernet/ti/davinci_mdio.c
> @@ -34,6 +34,7 @@
> #include <linux/clk.h>
> #include <linux/err.h>
> #include <linux/io.h>
> +#include <linux/iopoll.h>
> #include <linux/pm_runtime.h>
> #include <linux/davinci_emac.h>
> #include <linux/of.h>
> @@ -227,14 +228,16 @@ static inline int wait_for_user_access(struct davinci_mdio_data *data)
> static inline int wait_for_idle(struct davinci_mdio_data *data)
> {
> struct davinci_mdio_regs __iomem *regs = data->regs;
> - unsigned long timeout = jiffies + msecs_to_jiffies(MDIO_TIMEOUT);
> + u32 val, ret;
>
> - while (time_after(timeout, jiffies)) {
> - if (__raw_readl(®s->control) & CONTROL_IDLE)
> - return 0;
> + ret = readl_poll_timeout(®s->control, val, val & CONTROL_IDLE,
> + 0, MDIO_TIMEOUT * 1000);
> + if (ret) {
> + dev_err(data->dev, "timed out waiting for idle\n");
> + return ret;
> }
> - dev_err(data->dev, "timed out waiting for idle\n");
> - return -ETIMEDOUT;
> +
> + return 0;
> }
Hi Sekhar
You could simplify this to:
> + if (ret)
> + dev_err(data->dev, "timed out waiting for idle\n");
> + return ret;
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Andrew
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: [PATCH v2 net-next] drivers: net: davinci_mdio: prevent spurious timeout
2018-05-09 13:30 ` Andrew Lunn
@ 2018-05-09 15:46 ` Sekhar Nori
0 siblings, 0 replies; 5+ messages in thread
From: Sekhar Nori @ 2018-05-09 15:46 UTC (permalink / raw)
To: Andrew Lunn; +Cc: David S . Miller, Grygorii Strashko, linux-omap, netdev
On Wednesday 09 May 2018 07:00 PM, Andrew Lunn wrote:
> On Wed, May 09, 2018 at 04:30:24PM +0530, Sekhar Nori wrote:
>> A well timed kernel preemption in the time_after() loop
>> in wait_for_idle() can result in a spurious timeout
>> error to be returned.
>>
>> Fix it by using readl_poll_timeout() which takes care of
>> this issue.
>>
>> Signed-off-by: Sekhar Nori <nsekhar@ti.com>
>> ---
>> v2: use readl_poll_timeout() per suggestion from Andrew.
>>
>> The issue has not been personally observed by me, but has
>> been reported by users. Sending for next-next given the
>> non-critical nature. There is seems to be no easy way to
>> reproduce this.
>>
>> drivers/net/ethernet/ti/davinci_mdio.c | 15 +++++++++------
>> 1 file changed, 9 insertions(+), 6 deletions(-)
>>
>> diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c
>> index 3c33f4504d8e..d073432a5dbe 100644
>> --- a/drivers/net/ethernet/ti/davinci_mdio.c
>> +++ b/drivers/net/ethernet/ti/davinci_mdio.c
>> @@ -34,6 +34,7 @@
>> #include <linux/clk.h>
>> #include <linux/err.h>
>> #include <linux/io.h>
>> +#include <linux/iopoll.h>
>> #include <linux/pm_runtime.h>
>> #include <linux/davinci_emac.h>
>> #include <linux/of.h>
>> @@ -227,14 +228,16 @@ static inline int wait_for_user_access(struct davinci_mdio_data *data)
>> static inline int wait_for_idle(struct davinci_mdio_data *data)
>> {
>> struct davinci_mdio_regs __iomem *regs = data->regs;
>> - unsigned long timeout = jiffies + msecs_to_jiffies(MDIO_TIMEOUT);
>> + u32 val, ret;
>>
>> - while (time_after(timeout, jiffies)) {
>> - if (__raw_readl(®s->control) & CONTROL_IDLE)
>> - return 0;
>> + ret = readl_poll_timeout(®s->control, val, val & CONTROL_IDLE,
>> + 0, MDIO_TIMEOUT * 1000);
>> + if (ret) {
>> + dev_err(data->dev, "timed out waiting for idle\n");
>> + return ret;
>> }
>> - dev_err(data->dev, "timed out waiting for idle\n");
>> - return -ETIMEDOUT;
>> +
>> + return 0;
>> }
>
> Hi Sekhar
>
> You could simplify this to:
>
>> + if (ret)
>> + dev_err(data->dev, "timed out waiting for idle\n");
>> + return ret;
>
> Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Indeed. v3 sent.
Thanks,
Sekhar
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: [PATCH v2 net-next] drivers: net: davinci_mdio: prevent spurious timeout
@ 2018-05-09 15:46 ` Sekhar Nori
0 siblings, 0 replies; 5+ messages in thread
From: Sekhar Nori @ 2018-05-09 15:46 UTC (permalink / raw)
To: Andrew Lunn; +Cc: David S . Miller, Grygorii Strashko, linux-omap, netdev
On Wednesday 09 May 2018 07:00 PM, Andrew Lunn wrote:
> On Wed, May 09, 2018 at 04:30:24PM +0530, Sekhar Nori wrote:
>> A well timed kernel preemption in the time_after() loop
>> in wait_for_idle() can result in a spurious timeout
>> error to be returned.
>>
>> Fix it by using readl_poll_timeout() which takes care of
>> this issue.
>>
>> Signed-off-by: Sekhar Nori <nsekhar@ti.com>
>> ---
>> v2: use readl_poll_timeout() per suggestion from Andrew.
>>
>> The issue has not been personally observed by me, but has
>> been reported by users. Sending for next-next given the
>> non-critical nature. There is seems to be no easy way to
>> reproduce this.
>>
>> drivers/net/ethernet/ti/davinci_mdio.c | 15 +++++++++------
>> 1 file changed, 9 insertions(+), 6 deletions(-)
>>
>> diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c
>> index 3c33f4504d8e..d073432a5dbe 100644
>> --- a/drivers/net/ethernet/ti/davinci_mdio.c
>> +++ b/drivers/net/ethernet/ti/davinci_mdio.c
>> @@ -34,6 +34,7 @@
>> #include <linux/clk.h>
>> #include <linux/err.h>
>> #include <linux/io.h>
>> +#include <linux/iopoll.h>
>> #include <linux/pm_runtime.h>
>> #include <linux/davinci_emac.h>
>> #include <linux/of.h>
>> @@ -227,14 +228,16 @@ static inline int wait_for_user_access(struct davinci_mdio_data *data)
>> static inline int wait_for_idle(struct davinci_mdio_data *data)
>> {
>> struct davinci_mdio_regs __iomem *regs = data->regs;
>> - unsigned long timeout = jiffies + msecs_to_jiffies(MDIO_TIMEOUT);
>> + u32 val, ret;
>>
>> - while (time_after(timeout, jiffies)) {
>> - if (__raw_readl(®s->control) & CONTROL_IDLE)
>> - return 0;
>> + ret = readl_poll_timeout(®s->control, val, val & CONTROL_IDLE,
>> + 0, MDIO_TIMEOUT * 1000);
>> + if (ret) {
>> + dev_err(data->dev, "timed out waiting for idle\n");
>> + return ret;
>> }
>> - dev_err(data->dev, "timed out waiting for idle\n");
>> - return -ETIMEDOUT;
>> +
>> + return 0;
>> }
>
> Hi Sekhar
>
> You could simplify this to:
>
>> + if (ret)
>> + dev_err(data->dev, "timed out waiting for idle\n");
>> + return ret;
>
> Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Indeed. v3 sent.
Thanks,
Sekhar
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2018-05-09 15:46 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-05-09 11:00 [PATCH v2 net-next] drivers: net: davinci_mdio: prevent spurious timeout Sekhar Nori
2018-05-09 11:00 ` Sekhar Nori
2018-05-09 13:30 ` Andrew Lunn
2018-05-09 15:46 ` Sekhar Nori
2018-05-09 15:46 ` Sekhar Nori
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.