pm_runtime handles sdio power on and power off transitions.
An old workaround for trying to control the power explicitly from the
driver was in fact causing failures on suspend/resume as the mmc layer
already power the module on resume.
In case of resume pm_runtime_get sync returns a positive device's usage
count causing the driver to try an re-initialize an already initialized
device. This was causing sdio bus failure on resume.
Remove this manual power on/off sequence as it is in-fact not needed.
Signed-off-by: Eyal Reizer <[email protected]>
Acked-by: Tony Lindgren <[email protected]>
---
drivers/net/wireless/ti/wlcore/sdio.c | 27 ++++++---------------------
1 file changed, 6 insertions(+), 21 deletions(-)
diff --git a/drivers/net/wireless/ti/wlcore/sdio.c b/drivers/net/wireless/ti/wlcore/sdio.c
index 1f727ba..6dbe61d 100644
--- a/drivers/net/wireless/ti/wlcore/sdio.c
+++ b/drivers/net/wireless/ti/wlcore/sdio.c
@@ -155,17 +155,11 @@ static int wl12xx_sdio_power_on(struct wl12xx_sdio_glue *glue)
struct mmc_card *card = func->card;
ret = pm_runtime_get_sync(&card->dev);
- if (ret) {
- /*
- * Runtime PM might be temporarily disabled, or the device
- * might have a positive reference counter. Make sure it is
- * really powered on.
- */
- ret = mmc_power_restore_host(card->host);
- if (ret < 0) {
- pm_runtime_put_sync(&card->dev);
- goto out;
- }
+ if (ret < 0) {
+ pm_runtime_put_noidle(&card->dev);
+ dev_err(glue->dev, "%s: failed to get_sync(%d)\n",
+ __func__, ret);
+ goto out;
}
sdio_claim_host(func);
@@ -178,7 +172,6 @@ static int wl12xx_sdio_power_on(struct wl12xx_sdio_glue *glue)
static int wl12xx_sdio_power_off(struct wl12xx_sdio_glue *glue)
{
- int ret;
struct sdio_func *func = dev_to_sdio_func(glue->dev);
struct mmc_card *card = func->card;
@@ -186,16 +179,8 @@ static int wl12xx_sdio_power_off(struct wl12xx_sdio_glue *glue)
sdio_disable_func(func);
sdio_release_host(func);
- /* Power off the card manually in case it wasn't powered off above */
- ret = mmc_power_save_host(card->host);
- if (ret < 0)
- goto out;
-
/* Let runtime PM know the card is powered off */
- pm_runtime_put_sync(&card->dev);
-
-out:
- return ret;
+ return pm_runtime_put_sync(&card->dev);
}
static int wl12xx_sdio_set_power(struct device *child, bool enable)
--
2.7.4
>=20
> >>
> >> > pm_runtime handles sdio power on and power off transitions.
> >> > An old workaround for trying to control the power explicitly from th=
e
> >> > driver was in fact causing failures on suspend/resume as the mmc lay=
er
> >> > already power the module on resume.
> >> >
> >> > In case of resume pm_runtime_get sync returns a positive device's
> usage
> >> > count causing the driver to try an re-initialize an already initiali=
zed
> >> > device. This was causing sdio bus failure on resume.
> >> >
> >> > Remove this manual power on/off sequence as it is in-fact not needed=
.
> >> >
> >> > Signed-off-by: Eyal Reizer <[email protected]>
> >> > Acked-by: Tony Lindgren <[email protected]>
> >>
> >> No changelog.
> >>
> >>
> https://wireless.wiki.kernel.org/en/developers/documentation/submittingp
> >> atches#changelog_missing
> >>
> >> No need to resubmit because of this, I guess you just fixed the title
> >> and added Tony's Acked-by?
> >
> > Yes, this is correct. No change in the actual patch hence there was no
> change
> > Log.
>=20
> _Always_ include a change log, even if you didn't actually change
> anything. Otherwise the maintainer has no clue what has changed and why
> a new version was submitted.
>=20
Understood. Thanks!
BR,
Eyal
>=20
> > pm_runtime handles sdio power on and power off transitions.
> > An old workaround for trying to control the power explicitly from the
> > driver was in fact causing failures on suspend/resume as the mmc layer
> > already power the module on resume.
> >
> > In case of resume pm_runtime_get sync returns a positive device's usage
> > count causing the driver to try an re-initialize an already initialized
> > device. This was causing sdio bus failure on resume.
> >
> > Remove this manual power on/off sequence as it is in-fact not needed.
> >
> > Signed-off-by: Eyal Reizer <[email protected]>
> > Acked-by: Tony Lindgren <[email protected]>
>=20
> No changelog.
>=20
> https://wireless.wiki.kernel.org/en/developers/documentation/submittingp
> atches#changelog_missing
>=20
> No need to resubmit because of this, I guess you just fixed the title
> and added Tony's Acked-by?
Yes, this is correct. No change in the actual patch hence there was no chan=
ge=20
Log.
Best Regards,
Eyal
"Reizer, Eyal" <[email protected]> writes:
>>
>> > pm_runtime handles sdio power on and power off transitions.
>> > An old workaround for trying to control the power explicitly from the
>> > driver was in fact causing failures on suspend/resume as the mmc layer
>> > already power the module on resume.
>> >
>> > In case of resume pm_runtime_get sync returns a positive device's usage
>> > count causing the driver to try an re-initialize an already initialized
>> > device. This was causing sdio bus failure on resume.
>> >
>> > Remove this manual power on/off sequence as it is in-fact not needed.
>> >
>> > Signed-off-by: Eyal Reizer <[email protected]>
>> > Acked-by: Tony Lindgren <[email protected]>
>>
>> No changelog.
>>
>> https://wireless.wiki.kernel.org/en/developers/documentation/submittingp
>> atches#changelog_missing
>>
>> No need to resubmit because of this, I guess you just fixed the title
>> and added Tony's Acked-by?
>
> Yes, this is correct. No change in the actual patch hence there was no change
> Log.
_Always_ include a change log, even if you didn't actually change
anything. Otherwise the maintainer has no clue what has changed and why
a new version was submitted.
--
Kalle Valo
Eyal Reizer <[email protected]> writes:
> pm_runtime handles sdio power on and power off transitions.
> An old workaround for trying to control the power explicitly from the
> driver was in fact causing failures on suspend/resume as the mmc layer
> already power the module on resume.
>
> In case of resume pm_runtime_get sync returns a positive device's usage
> count causing the driver to try an re-initialize an already initialized
> device. This was causing sdio bus failure on resume.
>
> Remove this manual power on/off sequence as it is in-fact not needed.
>
> Signed-off-by: Eyal Reizer <[email protected]>
> Acked-by: Tony Lindgren <[email protected]>
No changelog.
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches#changelog_missing
No need to resubmit because of this, I guess you just fixed the title
and added Tony's Acked-by?
--
Kalle Valo
Eyal Reizer <[email protected]> wrote:
> pm_runtime handles sdio power on and power off transitions.
> An old workaround for trying to control the power explicitly from the
> driver was in fact causing failures on suspend/resume as the mmc layer
> already power the module on resume.
>
> In case of resume pm_runtime_get sync returns a positive device's usage
> count causing the driver to try an re-initialize an already initialized
> device. This was causing sdio bus failure on resume.
>
> Remove this manual power on/off sequence as it is in-fact not needed.
>
> Signed-off-by: Eyal Reizer <[email protected]>
> Acked-by: Tony Lindgren <[email protected]>
Patch applied to wireless-drivers-next.git, thanks.
60f36637bbbd wlcore: sdio: allow pm to handle sdio power
--
https://patchwork.kernel.org/patch/10364645/
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches