diff mbox series

[v2] edac: fix period calculation in edac_device_reset_delay_period()

Message ID 20221020124458.22153-1-farbere@amazon.com
State New
Headers
Series [v2] edac: fix period calculation in edac_device_reset_delay_period() |

Commit Message

Farber, Eliav Oct. 20, 2022, 12:44 p.m. UTC 
  Fix period calculation in case user sets a value of 1000.
The input of round_jiffies_relative() should be in jiffies and not in
milli-seconds.

Signed-off-by: Eliav Farber <farbere@amazon.com>
---
v2 --> v1:
- Fix the bug without modifying jiffs which is used to set
  edac_dev->delay.

 drivers/edac/edac_device.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

Comments

Borislav Petkov Dec. 29, 2022, 3:23 p.m. UTC | #1 
On Thu, Oct 20, 2022 at 12:44:58PM +0000, Eliav Farber wrote:
> Fix period calculation in case user sets a value of 1000.
> The input of round_jiffies_relative() should be in jiffies and not in
> milli-seconds.
> 
> Signed-off-by: Eliav Farber <farbere@amazon.com>

Fixes: c4cf3b454eca ("EDAC: Rework workqueue handling")

I guess.

Also, I think the one-liner below does the same, no?

---

diff --git a/drivers/edac/edac_device.c b/drivers/edac/edac_device.c
index 19522c568aa5..88942a6edc2c 100644
--- a/drivers/edac/edac_device.c
+++ b/drivers/edac/edac_device.c
@@ -399,7 +399,7 @@ void edac_device_reset_delay_period(struct edac_device_ctl_info *edac_dev,
 	unsigned long jiffs = msecs_to_jiffies(value);
 
 	if (value == 1000)
-		jiffs = round_jiffies_relative(value);
+		jiffs = round_jiffies_relative(jiffs);
 
 	edac_dev->poll_msec = value;
 	edac_dev->delay	    = jiffs;
Farber, Eliav Dec. 29, 2022, 8:17 p.m. UTC | #2 
On 12/29/2022 5:23 PM, Borislav Petkov wrote:
> On Thu, Oct 20, 2022 at 12:44:58PM +0000, Eliav Farber wrote:
>> Fix period calculation in case user sets a value of 1000.
>> The input of round_jiffies_relative() should be in jiffies and not in
>> milli-seconds.
>>
>> Signed-off-by: Eliav Farber <farbere@amazon.com>
>
> Fixes: c4cf3b454eca ("EDAC: Rework workqueue handling")
>
> I guess.
>
> Also, I think the one-liner below does the same, no?

The one-liner below will not work.
See the comment in edac_device_workq_setup() that explains why round
is used:
"
optimize here for the 1 second case, which will be normal value, to
fire ON the 1 second time event. This helps reduce all sorts of
timers firing on sub-second basis, while they are happy
to fire together on the 1 second exactly
"
So only the first schedule should be rounded.
But all other schedules after that should be 1000ms.
When rounding jiffs and saving it in edac_dev->delay it means that all
future schedules will not be with the correct delay.

The fix I suggested is the same logic used in:
  - edac_device_workq_setup()
  - edac_device_workq_function()

--
Regards, Eliav


> ---
>
> diff --git a/drivers/edac/edac_device.c b/drivers/edac/edac_device.c
> index 19522c568aa5..88942a6edc2c 100644
> --- a/drivers/edac/edac_device.c
> +++ b/drivers/edac/edac_device.c
> @@ -399,7 +399,7 @@ void edac_device_reset_delay_period(struct 
> edac_device_ctl_info *edac_dev,
>        unsigned long jiffs = msecs_to_jiffies(value);
>
>        if (value == 1000)
> -               jiffs = round_jiffies_relative(value);
> +               jiffs = round_jiffies_relative(jiffs);
>
>        edac_dev->poll_msec = value;
>        edac_dev->delay     = jiffs;
>
> -- 
> Regards/Gruss,
>    Boris.
>
> https://people.kernel.org/tglx/notes-about-netiquette
Borislav Petkov Dec. 30, 2022, 2:48 p.m. UTC | #3 
On Thu, Dec 29, 2022 at 10:17:48PM +0200, Farber, Eliav wrote:
> The one-liner below will not work. See the comment in
> edac_device_workq_setup() that explains why round is used:

Bah, right you are - there's that "optimization".

Ok, I've applied this:

---
From: Eliav Farber <farbere@amazon.com>
Date: Thu, 20 Oct 2022 12:44:58 +0000
Subject: [PATCH] EDAC/device: Fix period calculation in
 edac_device_reset_delay_period()

Fix period calculation in case user sets a value of 1000.  The input of
round_jiffies_relative() should be in jiffies and not in milli-seconds.

  [ bp: Use the same code pattern as in edac_device_workq_setup() for
    clarity. ]

Fixes: c4cf3b454eca ("EDAC: Rework workqueue handling")
Signed-off-by: Eliav Farber <farbere@amazon.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20221020124458.22153-1-farbere@amazon.com
---
 drivers/edac/edac_device.c | 17 ++++++++---------
 drivers/edac/edac_module.h |  2 +-
 2 files changed, 9 insertions(+), 10 deletions(-)

diff --git a/drivers/edac/edac_device.c b/drivers/edac/edac_device.c
index 19522c568aa5..878deb4880cd 100644
--- a/drivers/edac/edac_device.c
+++ b/drivers/edac/edac_device.c
@@ -394,17 +394,16 @@ static void edac_device_workq_teardown(struct edac_device_ctl_info *edac_dev)
  *	Then restart the workq on the new delay
  */
 void edac_device_reset_delay_period(struct edac_device_ctl_info *edac_dev,
-					unsigned long value)
+				    unsigned long msec)
 {
-	unsigned long jiffs = msecs_to_jiffies(value);
-
-	if (value == 1000)
-		jiffs = round_jiffies_relative(value);
-
-	edac_dev->poll_msec = value;
-	edac_dev->delay	    = jiffs;
+	edac_dev->poll_msec = msec;
+	edac_dev->delay	    = msecs_to_jiffies(msec);
 
-	edac_mod_work(&edac_dev->work, jiffs);
+	/* See comment in edac_device_workq_setup() above */
+	if (edac_dev->poll_msec == 1000)
+		edac_mod_work(&edac_dev->work, round_jiffies_relative(edac_dev->delay));
+	else
+		edac_mod_work(&edac_dev->work, edac_dev->delay);
 }
 
 int edac_device_alloc_index(void)
diff --git a/drivers/edac/edac_module.h b/drivers/edac/edac_module.h
index 763c076d96f2..47593afdc234 100644
--- a/drivers/edac/edac_module.h
+++ b/drivers/edac/edac_module.h
@@ -53,7 +53,7 @@ bool edac_stop_work(struct delayed_work *work);
 bool edac_mod_work(struct delayed_work *work, unsigned long delay);
 
 extern void edac_device_reset_delay_period(struct edac_device_ctl_info
-					   *edac_dev, unsigned long value);
+					   *edac_dev, unsigned long msec);
 extern void edac_mc_reset_delay_period(unsigned long value);
 
 /*
diff mbox series

Patch

diff --git a/drivers/edac/edac_device.c b/drivers/edac/edac_device.c
index 19522c568aa5..e944dd9b3593 100644
--- a/drivers/edac/edac_device.c
+++ b/drivers/edac/edac_device.c
@@ -398,13 +398,13 @@  void edac_device_reset_delay_period(struct edac_device_ctl_info *edac_dev,
 {
 	unsigned long jiffs = msecs_to_jiffies(value);
 
-	if (value == 1000)
-		jiffs = round_jiffies_relative(value);
-
 	edac_dev->poll_msec = value;
 	edac_dev->delay	    = jiffs;
 
-	edac_mod_work(&edac_dev->work, jiffs);
+	if (value == 1000)
+		edac_mod_work(&edac_dev->work, round_jiffies_relative(jiffs));
+	else
+		edac_mod_work(&edac_dev->work, jiffs);
 }
 
 int edac_device_alloc_index(void)