Hi Paul,The remove function was removed because it would otherwise be an empty
On 30 December 2017 at 19:21, Paul Cercueil <paul@xxxxxxxxxxxxxxx> wrote:
When the watchdog was configured for nowayout, and after the
userspace watchdog daemon closed the dev node without sending the
magic character, unloading this module stopped the watchdog
hardware, which was clearly a problem.
Besides, unloading the module is not possible when the userspace
watchdog daemon is running, so it's safe to assume that we don't
need to stop the watchdog hardware in the jz4740_wdt_remove()
function.
For this reason, the jz4740_wdt_remove() function can then be
dropped alltogether.
Signed-off-by: Paul Cercueil <paul@xxxxxxxxxxxxxxx>
---
drivers/watchdog/jz4740_wdt.c | 8 --------
1 file changed, 8 deletions(-)
v2: New patch in this series
diff --git a/drivers/watchdog/jz4740_wdt.c b/drivers/watchdog/jz4740_wdt.c
index fa7f49a3212c..02b9b8e946a2 100644
--- a/drivers/watchdog/jz4740_wdt.c
+++ b/drivers/watchdog/jz4740_wdt.c
@@ -205,16 +205,8 @@ static int jz4740_wdt_probe(struct platform_device *pdev)
return 0;
}
-static int jz4740_wdt_remove(struct platform_device *pdev)
-{
- struct jz4740_wdt_drvdata *drvdata = platform_get_drvdata(pdev);
-
- return jz4740_wdt_stop(&drvdata->wdt);
-}
-
static struct platform_driver jz4740_wdt_driver = {
.probe = jz4740_wdt_probe,
- .remove = jz4740_wdt_remove,
.driver = {
.name = "jz4740-wdt",
.of_match_table = of_match_ptr(jz4740_wdt_of_matches),
--
2.11.0
As ".remove" is removed and wdt is required for restarting the system
I am thinking that stop callback is also not required. If so does it
makes sense to remove the stop callback? I can submit a patch for the
same once this patch series goes in.