I discovered the problem addressed by the first patch while I was
investigating a Kernel CI regression: [0]
It is unrelated to that regression but should still be fixed.
The second patch adresses the actual regression. Testing was focussed
on GXL and GXM for the previous patches. Unfortunately one of them
regresses USB on G12A, G12B, SM1 and A1 SoCs.
Please queue these for v5.8 so we don't end up with broken USB on
some boards.
[0] https://lore.kernel.org/linux-usb/[email protected]/T/#u
Martin Blumenstingl (2):
usb: dwc3: meson-g12a: fix error path when fetching the reset line
fails
usb: dwc3: meson-g12a: fix USB2 PHY initialization on G12A and A1 SoCs
drivers/usb/dwc3/dwc3-meson-g12a.c | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
--
2.26.2
dwc3_meson_g12a_usb2_init_phy() crashes with NULL pointer on an SM1
board (which uses the same USB setup as G12A) dereference as reported
by the Kernel CI bot. This is because of the following call flow:
dwc3_meson_g12a_probe
priv->drvdata->setup_regmaps
dwc3_meson_g12a_setup_regmaps
priv->usb2_ports is still 0 so priv->u2p_regmap[i] will be NULL
dwc3_meson_g12a_get_phys
initializes priv->usb2_ports
priv->drvdata->usb_init
dwc3_meson_g12a_usb_init
dwc3_meson_g12a_usb_init_glue
dwc3_meson_g12a_usb2_init
priv->drvdata->usb2_init_phy
dwc3_meson_g12a_usb2_init_phy
dereferences priv->u2p_regmap[i]
Call priv->drvdata->setup_regmaps only after dwc3_meson_g12a_get_phys so
priv->usb2_ports is initialized and the regmaps will be set up
correctly. This fixes the NULL dereference later on.
Fixes: 013af227f58a97 ("usb: dwc3: meson-g12a: handle the phy and glue registers separately")
Reported-by: "kernelci.org bot" <[email protected]>
Signed-off-by: Martin Blumenstingl <[email protected]>
---
drivers/usb/dwc3/dwc3-meson-g12a.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/usb/dwc3/dwc3-meson-g12a.c b/drivers/usb/dwc3/dwc3-meson-g12a.c
index ce5388338389..1f7f4d88ed9d 100644
--- a/drivers/usb/dwc3/dwc3-meson-g12a.c
+++ b/drivers/usb/dwc3/dwc3-meson-g12a.c
@@ -708,11 +708,7 @@ static int dwc3_meson_g12a_probe(struct platform_device *pdev)
return PTR_ERR(base);
priv->drvdata = of_device_get_match_data(&pdev->dev);
-
priv->dev = dev;
- ret = priv->drvdata->setup_regmaps(priv, base);
- if (ret)
- return ret;
priv->vbus = devm_regulator_get_optional(dev, "vbus");
if (IS_ERR(priv->vbus)) {
@@ -749,6 +745,10 @@ static int dwc3_meson_g12a_probe(struct platform_device *pdev)
if (ret)
goto err_disable_clks;
+ ret = priv->drvdata->setup_regmaps(priv, base);
+ if (ret)
+ return ret;
+
if (priv->vbus) {
ret = regulator_enable(priv->vbus);
if (ret)
--
2.26.2
Disable and unprepare the clocks when devm_reset_control_get_shared()
fails. This fixes the error path as this must disable the clocks which
were previously enabled.
Fixes: 1e355f21d3fb96 ("usb: dwc3: Add Amlogic A1 DWC3 glue")
Cc: Yue Wang <[email protected]>
Cc: Hanjie Lin <[email protected]>
Signed-off-by: Martin Blumenstingl <[email protected]>
---
drivers/usb/dwc3/dwc3-meson-g12a.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/usb/dwc3/dwc3-meson-g12a.c b/drivers/usb/dwc3/dwc3-meson-g12a.c
index bd744e82cad4..ce5388338389 100644
--- a/drivers/usb/dwc3/dwc3-meson-g12a.c
+++ b/drivers/usb/dwc3/dwc3-meson-g12a.c
@@ -738,7 +738,7 @@ static int dwc3_meson_g12a_probe(struct platform_device *pdev)
if (IS_ERR(priv->reset)) {
ret = PTR_ERR(priv->reset);
dev_err(dev, "failed to get device reset, err=%d\n", ret);
- return ret;
+ goto err_disable_clks;
}
ret = reset_control_reset(priv->reset);
--
2.26.2
On 26/05/2020 22:29, Martin Blumenstingl wrote:
> Disable and unprepare the clocks when devm_reset_control_get_shared()
> fails. This fixes the error path as this must disable the clocks which
> were previously enabled.
>
> Fixes: 1e355f21d3fb96 ("usb: dwc3: Add Amlogic A1 DWC3 glue")
> Cc: Yue Wang <[email protected]>
> Cc: Hanjie Lin <[email protected]>
> Signed-off-by: Martin Blumenstingl <[email protected]>
> ---
> drivers/usb/dwc3/dwc3-meson-g12a.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/usb/dwc3/dwc3-meson-g12a.c b/drivers/usb/dwc3/dwc3-meson-g12a.c
> index bd744e82cad4..ce5388338389 100644
> --- a/drivers/usb/dwc3/dwc3-meson-g12a.c
> +++ b/drivers/usb/dwc3/dwc3-meson-g12a.c
> @@ -738,7 +738,7 @@ static int dwc3_meson_g12a_probe(struct platform_device *pdev)
> if (IS_ERR(priv->reset)) {
> ret = PTR_ERR(priv->reset);
> dev_err(dev, "failed to get device reset, err=%d\n", ret);
> - return ret;
> + goto err_disable_clks;
> }
>
> ret = reset_control_reset(priv->reset);
>
Acked-by: Neil Armstrong <[email protected]>
Hi Martin,
On 26/05/2020 22:29, Martin Blumenstingl wrote:
> dwc3_meson_g12a_usb2_init_phy() crashes with NULL pointer on an SM1
> board (which uses the same USB setup as G12A) dereference as reported
> by the Kernel CI bot. This is because of the following call flow:
> dwc3_meson_g12a_probe
> priv->drvdata->setup_regmaps
> dwc3_meson_g12a_setup_regmaps
> priv->usb2_ports is still 0 so priv->u2p_regmap[i] will be NULL
> dwc3_meson_g12a_get_phys
> initializes priv->usb2_ports
> priv->drvdata->usb_init
> dwc3_meson_g12a_usb_init
> dwc3_meson_g12a_usb_init_glue
> dwc3_meson_g12a_usb2_init
> priv->drvdata->usb2_init_phy
> dwc3_meson_g12a_usb2_init_phy
> dereferences priv->u2p_regmap[i]
>
> Call priv->drvdata->setup_regmaps only after dwc3_meson_g12a_get_phys so
> priv->usb2_ports is initialized and the regmaps will be set up
> correctly. This fixes the NULL dereference later on.
>
> Fixes: 013af227f58a97 ("usb: dwc3: meson-g12a: handle the phy and glue registers separately")
> Reported-by: "kernelci.org bot" <[email protected]>
> Signed-off-by: Martin Blumenstingl <[email protected]>
> ---
> drivers/usb/dwc3/dwc3-meson-g12a.c | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/usb/dwc3/dwc3-meson-g12a.c b/drivers/usb/dwc3/dwc3-meson-g12a.c
> index ce5388338389..1f7f4d88ed9d 100644
> --- a/drivers/usb/dwc3/dwc3-meson-g12a.c
> +++ b/drivers/usb/dwc3/dwc3-meson-g12a.c
[...]
Fixes regression reported at [1] on SEI510 board based on Amlogic G12A.
Felipe, Greg, can this be queued on uxb-next for 5.8 ?
Acked-by: Neil Armstrong <[email protected]>
Thanks,
Neil
[1] http://lore.kernel.org/r/[email protected]
On Wed, May 27, 2020 at 10:17:31AM +0200, Neil Armstrong wrote:
> Hi Martin,
>
> On 26/05/2020 22:29, Martin Blumenstingl wrote:
> > dwc3_meson_g12a_usb2_init_phy() crashes with NULL pointer on an SM1
> > board (which uses the same USB setup as G12A) dereference as reported
> > by the Kernel CI bot. This is because of the following call flow:
> > dwc3_meson_g12a_probe
> > priv->drvdata->setup_regmaps
> > dwc3_meson_g12a_setup_regmaps
> > priv->usb2_ports is still 0 so priv->u2p_regmap[i] will be NULL
> > dwc3_meson_g12a_get_phys
> > initializes priv->usb2_ports
> > priv->drvdata->usb_init
> > dwc3_meson_g12a_usb_init
> > dwc3_meson_g12a_usb_init_glue
> > dwc3_meson_g12a_usb2_init
> > priv->drvdata->usb2_init_phy
> > dwc3_meson_g12a_usb2_init_phy
> > dereferences priv->u2p_regmap[i]
> >
> > Call priv->drvdata->setup_regmaps only after dwc3_meson_g12a_get_phys so
> > priv->usb2_ports is initialized and the regmaps will be set up
> > correctly. This fixes the NULL dereference later on.
> >
> > Fixes: 013af227f58a97 ("usb: dwc3: meson-g12a: handle the phy and glue registers separately")
> > Reported-by: "kernelci.org bot" <[email protected]>
> > Signed-off-by: Martin Blumenstingl <[email protected]>
> > ---
> > drivers/usb/dwc3/dwc3-meson-g12a.c | 8 ++++----
> > 1 file changed, 4 insertions(+), 4 deletions(-)
> >
> > diff --git a/drivers/usb/dwc3/dwc3-meson-g12a.c b/drivers/usb/dwc3/dwc3-meson-g12a.c
> > index ce5388338389..1f7f4d88ed9d 100644
> > --- a/drivers/usb/dwc3/dwc3-meson-g12a.c
> > +++ b/drivers/usb/dwc3/dwc3-meson-g12a.c
>
> [...]
>
> Fixes regression reported at [1] on SEI510 board based on Amlogic G12A.
>
> Felipe, Greg, can this be queued on uxb-next for 5.8 ?
>
> Acked-by: Neil Armstrong <[email protected]>
I can take this and patch 1/2 here if Felipe acks them.
thanks,
greg k-h
Greg KH <[email protected]> writes:
> On Wed, May 27, 2020 at 10:17:31AM +0200, Neil Armstrong wrote:
>> Hi Martin,
>>
>> On 26/05/2020 22:29, Martin Blumenstingl wrote:
>> > dwc3_meson_g12a_usb2_init_phy() crashes with NULL pointer on an SM1
>> > board (which uses the same USB setup as G12A) dereference as reported
>> > by the Kernel CI bot. This is because of the following call flow:
>> > dwc3_meson_g12a_probe
>> > priv->drvdata->setup_regmaps
>> > dwc3_meson_g12a_setup_regmaps
>> > priv->usb2_ports is still 0 so priv->u2p_regmap[i] will be NULL
>> > dwc3_meson_g12a_get_phys
>> > initializes priv->usb2_ports
>> > priv->drvdata->usb_init
>> > dwc3_meson_g12a_usb_init
>> > dwc3_meson_g12a_usb_init_glue
>> > dwc3_meson_g12a_usb2_init
>> > priv->drvdata->usb2_init_phy
>> > dwc3_meson_g12a_usb2_init_phy
>> > dereferences priv->u2p_regmap[i]
>> >
>> > Call priv->drvdata->setup_regmaps only after dwc3_meson_g12a_get_phys so
>> > priv->usb2_ports is initialized and the regmaps will be set up
>> > correctly. This fixes the NULL dereference later on.
>> >
>> > Fixes: 013af227f58a97 ("usb: dwc3: meson-g12a: handle the phy and glue registers separately")
>> > Reported-by: "kernelci.org bot" <[email protected]>
>> > Signed-off-by: Martin Blumenstingl <[email protected]>
>> > ---
>> > drivers/usb/dwc3/dwc3-meson-g12a.c | 8 ++++----
>> > 1 file changed, 4 insertions(+), 4 deletions(-)
>> >
>> > diff --git a/drivers/usb/dwc3/dwc3-meson-g12a.c b/drivers/usb/dwc3/dwc3-meson-g12a.c
>> > index ce5388338389..1f7f4d88ed9d 100644
>> > --- a/drivers/usb/dwc3/dwc3-meson-g12a.c
>> > +++ b/drivers/usb/dwc3/dwc3-meson-g12a.c
>>
>> [...]
>>
>> Fixes regression reported at [1] on SEI510 board based on Amlogic G12A.
>>
>> Felipe, Greg, can this be queued on uxb-next for 5.8 ?
>>
>> Acked-by: Neil Armstrong <[email protected]>
>
> I can take this and patch 1/2 here if Felipe acks them.
Sure thing, Greg. Thanks.
Acked-by: Felipe Balbi <[email protected]>
--
balbi