On 10/19/22 22:28, Uwe Kleine-König wrote:
If for whatever reasons pm_runtime_resume_and_get() failed, .remove() is
exited early, the clock isn't freed and runtime PM state isn't reset.
The right thing to do however is to free all resources that don't need
HW access after a problem with runtime PM. Also issue a warning in that
case and return 0 to suppress a less helpful warning by the driver core.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@xxxxxxxxxxxxxx>
---
drivers/i2c/busses/i2c-xiic.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/drivers/i2c/busses/i2c-xiic.c b/drivers/i2c/busses/i2c-xiic.c
index b3fe6b2aa3ca..809d6eadae2d 100644
--- a/drivers/i2c/busses/i2c-xiic.c
+++ b/drivers/i2c/busses/i2c-xiic.c
@@ -858,11 +858,14 @@ static int xiic_i2c_remove(struct platform_device *pdev)
/* remove adapter & data */
i2c_del_adapter(&i2c->adap);
- ret = pm_runtime_resume_and_get(i2c->dev);
+ ret = pm_runtime_get_sync(i2c->dev);
+
if (ret < 0)
- return ret;
+ dev_warn(&pdev->dev, "Failed to activate device for removal (%pe)\n",
+ ERR_PTR(ret));
+ else
+ xiic_deinit(i2c);
- xiic_deinit(i2c);
pm_runtime_put_sync(i2c->dev);
clk_disable_unprepare(i2c->clk);
pm_runtime_disable(&pdev->dev);
base-commit: 9abf2313adc1ca1b6180c508c25f22f9395cc780
--
2.37.2
Acked-by: Michal Simek <michal.simek@xxxxxxx>
Thanks,
Michal
