The hdmi-cec clock must be 32khz in order for cec to work correctly.
Ensure after enabling the clock we set it in order for the hardware to
work as expected.
Warn on failure, in case this is a static clock that is slighty off.
Fixes hdmi-cec support on Rockchip devices.
Fixes: ebe32c3e282a ("drm/bridge: synopsys/dw-hdmi: Enable cec clock")
Signed-off-by: Peter Geis <[email protected]>
---
drivers/gpu/drm/bridge/synopsys/dw-hdmi.c | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
index 54d8fdad395f..1a96da60e357 100644
--- a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
+++ b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
@@ -48,6 +48,9 @@
#define HDMI14_MAX_TMDSCLK 340000000
+/* HDMI CEC needs a clock rate of 32khz */
+#define HDMI_CEC_CLK_RATE 32768
+
enum hdmi_datamap {
RGB444_8B = 0x01,
RGB444_10B = 0x03,
@@ -3347,6 +3350,10 @@ struct dw_hdmi *dw_hdmi_probe(struct platform_device *pdev,
ret);
goto err_iahb;
}
+
+ ret = clk_set_rate(hdmi->cec_clk, HDMI_CEC_CLK_RATE);
+ if (ret)
+ dev_warn(hdmi->dev, "Cannot set HDMI cec clock rate: %d\n", ret);
}
/* Product and revision IDs */
--
2.25.1
On Wed, Jan 26, 2022 at 5:25 PM Peter Geis <[email protected]> wrote:
> +
> + ret = clk_set_rate(hdmi->cec_clk, HDMI_CEC_CLK_RATE);
> + if (ret)
> + dev_warn(hdmi->dev, "Cannot set HDMI cec clock rate: %d\n", ret);
You are setting the cec clock rate after it has been enabled, which
can be glitchy.
Better to set the rate prior to enabling the clock.
Hi Peter,
On Wed, Jan 26, 2022 at 03:24:26PM -0500, Peter Geis wrote:
> The hdmi-cec clock must be 32khz in order for cec to work correctly.
> Ensure after enabling the clock we set it in order for the hardware to
> work as expected.
> Warn on failure, in case this is a static clock that is slighty off.
> Fixes hdmi-cec support on Rockchip devices.
You removed this sentence in v2, but I just wanted to mention that
clk_set_rate() won't fail when the desired clock rate can't be
archieved. Instead, you will get the best rate that actually can be
reached. If you want to check that you are happy with the rate you'll
have to do a clk_round_rate() before setting the rate or a
clk_get_rate() afterwards.
Sascha
--
Pengutronix e.K. | |
Steuerwalder Str. 21 | http://www.pengutronix.de/ |
31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
On Thu, Jan 27, 2022 at 4:33 AM Sascha Hauer <[email protected]> wrote:
>
> Hi Peter,
>
> On Wed, Jan 26, 2022 at 03:24:26PM -0500, Peter Geis wrote:
> > The hdmi-cec clock must be 32khz in order for cec to work correctly.
> > Ensure after enabling the clock we set it in order for the hardware to
> > work as expected.
> > Warn on failure, in case this is a static clock that is slighty off.
> > Fixes hdmi-cec support on Rockchip devices.
>
> You removed this sentence in v2, but I just wanted to mention that
> clk_set_rate() won't fail when the desired clock rate can't be
> archieved. Instead, you will get the best rate that actually can be
> reached. If you want to check that you are happy with the rate you'll
> have to do a clk_round_rate() before setting the rate or a
> clk_get_rate() afterwards.
Thanks, the behavior in v2 is actually what I'm looking for.
I dug into clk_set_rate while checking into its interaction with
clk_prepare and I came to this conclusion.
>
> Sascha
>
> --
> Pengutronix e.K. | |
> Steuerwalder Str. 21 | http://www.pengutronix.de/ |
> 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
> Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
> Wiadomość napisana przez Peter Geis <[email protected]> w dniu 13.03.2022, o godz. 13:56:
>
>>
>
> I was worried about that, thanks for testing it.
> Can you send me the cec_clk rate before and after this patch?
>
Here it is:
working:
enable prepare protect duty hardware
clock count count count rate accuracy phase cycle enable
-------------------------------------------------------------------------------------------------------
clk_hdmi_cec 1 2 0 32743 0 0 50000 Y
non-working:
enable prepare protect duty hardware
clock count count count rate accuracy phase cycle enable
-------------------------------------------------------------------------------------------------------
clk_hdmi_cec 1 2 0 32768 0 0 50000 Y
dmesg
[ 52.745988] cec-dw_hdmi: message 44 timed out
[ 54.879307] cec-dw_hdmi: message 44 timed out
[ 57.012654] cec-dw_hdmi: message 88 timed out
[ 59.145973] cec-dw_hdmi: message 88 timed out
player:
2022-03-13 11:04:40.938355 E CECAdapter: CLinuxCECAdapterCommunication::Write - ioctl CEC_TRANSMIT failed - tx_status=00 errno=22
2022-03-13 11:04:40.938366 E CECAdapter: CLinuxCECAdapterCommunication::Write - ioctl CEC_TRANSMIT failed - tx_status=00 errno=22
On 2022-03-13 12:56, Peter Geis wrote:
> On Sun, Mar 13, 2022 at 6:13 AM Piotr Oniszczuk
> <[email protected]> wrote:
>>
>>
>>
>>> Wiadomość napisana przez Peter Geis <[email protected]> w dniu 26.01.2022, o godz. 21:24:
>>>
>>> The hdmi-cec clock must be 32khz in order for cec to work correctly.
>>> Ensure after enabling the clock we set it in order for the hardware to
>>> work as expected.
>>> Warn on failure, in case this is a static clock that is slighty off.
>>> Fixes hdmi-cec support on Rockchip devices.
>>>
>>> Fixes: ebe32c3e282a ("drm/bridge: synopsys/dw-hdmi: Enable cec clock")
>>>
>>> Signed-off-by: Peter Geis <[email protected]>
>>> ---
>>> drivers/gpu/drm/bridge/synopsys/dw-hdmi.c | 7 +++++++
>>> 1 file changed, 7 insertions(+)
>>>
>>> diff --git a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
>>> index 54d8fdad395f..1a96da60e357 100644
>>> --- a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
>>> +++ b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
>>> @@ -48,6 +48,9 @@
>>>
>>> #define HDMI14_MAX_TMDSCLK 340000000
>>>
>>> +/* HDMI CEC needs a clock rate of 32khz */
>>> +#define HDMI_CEC_CLK_RATE 32768
>>> +
>>> enum hdmi_datamap {
>>> RGB444_8B = 0x01,
>>> RGB444_10B = 0x03,
>>> @@ -3347,6 +3350,10 @@ struct dw_hdmi *dw_hdmi_probe(struct platform_device *pdev,
>>> ret);
>>> goto err_iahb;
>>> }
>>> +
>>> + ret = clk_set_rate(hdmi->cec_clk, HDMI_CEC_CLK_RATE);
>>> + if (ret)
>>> + dev_warn(hdmi->dev, "Cannot set HDMI cec clock rate: %d\n", ret);
>>> }
>>>
>>> /* Product and revision IDs */
>>> --
>>> 2.25.1
>>>
>>>
>>> _______________________________________________
>>> Linux-rockchip mailing list
>>> [email protected]
>>> http://lists.infradead.org/mailman/listinfo/linux-rockchip
>>
>> Peter,
>>
>> On my 5.17-rc7 with applied rk356x VOP2 v8 series - this patch makes CEC working on rk3566.
>> Unfortunately it breaks working ok CEC on rk3399 rockpi-4b.
>>
>> Reverting this patch brings back CEC on rk3399 - but rk3366 becomes with non working CEC
>>
>> I'm not sure how to move forward with this....
>
> I was worried about that, thanks for testing it.
> Can you send me the cec_clk rate before and after this patch?
Hmm, looks like there might be a bug in the RK3399 clock driver -
although it's treated as having mux_pll_p as parents, according to the
public TRM the CEC clock appears unique in being backwards compared to
every other mux of those two inputs. I'm now tempted to test this on my
board tonight and see if I can see 24MHz on the CEC pin... :)
Robin.
> Wiadomość napisana przez Peter Geis <[email protected]> w dniu 26.01.2022, o godz. 21:24:
>
> The hdmi-cec clock must be 32khz in order for cec to work correctly.
> Ensure after enabling the clock we set it in order for the hardware to
> work as expected.
> Warn on failure, in case this is a static clock that is slighty off.
> Fixes hdmi-cec support on Rockchip devices.
>
> Fixes: ebe32c3e282a ("drm/bridge: synopsys/dw-hdmi: Enable cec clock")
>
> Signed-off-by: Peter Geis <[email protected]>
> ---
> drivers/gpu/drm/bridge/synopsys/dw-hdmi.c | 7 +++++++
> 1 file changed, 7 insertions(+)
>
> diff --git a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
> index 54d8fdad395f..1a96da60e357 100644
> --- a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
> +++ b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
> @@ -48,6 +48,9 @@
>
> #define HDMI14_MAX_TMDSCLK 340000000
>
> +/* HDMI CEC needs a clock rate of 32khz */
> +#define HDMI_CEC_CLK_RATE 32768
> +
> enum hdmi_datamap {
> RGB444_8B = 0x01,
> RGB444_10B = 0x03,
> @@ -3347,6 +3350,10 @@ struct dw_hdmi *dw_hdmi_probe(struct platform_device *pdev,
> ret);
> goto err_iahb;
> }
> +
> + ret = clk_set_rate(hdmi->cec_clk, HDMI_CEC_CLK_RATE);
> + if (ret)
> + dev_warn(hdmi->dev, "Cannot set HDMI cec clock rate: %d\n", ret);
> }
>
> /* Product and revision IDs */
> --
> 2.25.1
>
>
> _______________________________________________
> Linux-rockchip mailing list
> [email protected]
> http://lists.infradead.org/mailman/listinfo/linux-rockchip
Peter,
On my 5.17-rc7 with applied rk356x VOP2 v8 series - this patch makes CEC working on rk3566.
Unfortunately it breaks working ok CEC on rk3399 rockpi-4b.
Reverting this patch brings back CEC on rk3399 - but rk3366 becomes with non working CEC
I'm not sure how to move forward with this....
br
On Sun, Mar 13, 2022 at 6:13 AM Piotr Oniszczuk
<[email protected]> wrote:
>
>
>
> > Wiadomość napisana przez Peter Geis <[email protected]> w dniu 26.01.2022, o godz. 21:24:
> >
> > The hdmi-cec clock must be 32khz in order for cec to work correctly.
> > Ensure after enabling the clock we set it in order for the hardware to
> > work as expected.
> > Warn on failure, in case this is a static clock that is slighty off.
> > Fixes hdmi-cec support on Rockchip devices.
> >
> > Fixes: ebe32c3e282a ("drm/bridge: synopsys/dw-hdmi: Enable cec clock")
> >
> > Signed-off-by: Peter Geis <[email protected]>
> > ---
> > drivers/gpu/drm/bridge/synopsys/dw-hdmi.c | 7 +++++++
> > 1 file changed, 7 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
> > index 54d8fdad395f..1a96da60e357 100644
> > --- a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
> > +++ b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
> > @@ -48,6 +48,9 @@
> >
> > #define HDMI14_MAX_TMDSCLK 340000000
> >
> > +/* HDMI CEC needs a clock rate of 32khz */
> > +#define HDMI_CEC_CLK_RATE 32768
> > +
> > enum hdmi_datamap {
> > RGB444_8B = 0x01,
> > RGB444_10B = 0x03,
> > @@ -3347,6 +3350,10 @@ struct dw_hdmi *dw_hdmi_probe(struct platform_device *pdev,
> > ret);
> > goto err_iahb;
> > }
> > +
> > + ret = clk_set_rate(hdmi->cec_clk, HDMI_CEC_CLK_RATE);
> > + if (ret)
> > + dev_warn(hdmi->dev, "Cannot set HDMI cec clock rate: %d\n", ret);
> > }
> >
> > /* Product and revision IDs */
> > --
> > 2.25.1
> >
> >
> > _______________________________________________
> > Linux-rockchip mailing list
> > [email protected]
> > http://lists.infradead.org/mailman/listinfo/linux-rockchip
>
> Peter,
>
> On my 5.17-rc7 with applied rk356x VOP2 v8 series - this patch makes CEC working on rk3566.
> Unfortunately it breaks working ok CEC on rk3399 rockpi-4b.
>
> Reverting this patch brings back CEC on rk3399 - but rk3366 becomes with non working CEC
>
> I'm not sure how to move forward with this....
I was worried about that, thanks for testing it.
Can you send me the cec_clk rate before and after this patch?
>
> br
On 2022-03-14 11:31, Robin Murphy wrote:
> On 2022-03-13 12:56, Peter Geis wrote:
>> On Sun, Mar 13, 2022 at 6:13 AM Piotr Oniszczuk
>> <[email protected]> wrote:
>>>
>>>
>>>
>>>> Wiadomość napisana przez Peter Geis <[email protected]> w dniu
>>>> 26.01.2022, o godz. 21:24:
>>>>
>>>> The hdmi-cec clock must be 32khz in order for cec to work correctly.
>>>> Ensure after enabling the clock we set it in order for the hardware to
>>>> work as expected.
>>>> Warn on failure, in case this is a static clock that is slighty off.
>>>> Fixes hdmi-cec support on Rockchip devices.
>>>>
>>>> Fixes: ebe32c3e282a ("drm/bridge: synopsys/dw-hdmi: Enable cec clock")
>>>>
>>>> Signed-off-by: Peter Geis <[email protected]>
>>>> ---
>>>> drivers/gpu/drm/bridge/synopsys/dw-hdmi.c | 7 +++++++
>>>> 1 file changed, 7 insertions(+)
>>>>
>>>> diff --git a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
>>>> b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
>>>> index 54d8fdad395f..1a96da60e357 100644
>>>> --- a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
>>>> +++ b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
>>>> @@ -48,6 +48,9 @@
>>>>
>>>> #define HDMI14_MAX_TMDSCLK 340000000
>>>>
>>>> +/* HDMI CEC needs a clock rate of 32khz */
>>>> +#define HDMI_CEC_CLK_RATE 32768
>>>> +
>>>> enum hdmi_datamap {
>>>> RGB444_8B = 0x01,
>>>> RGB444_10B = 0x03,
>>>> @@ -3347,6 +3350,10 @@ struct dw_hdmi *dw_hdmi_probe(struct
>>>> platform_device *pdev,
>>>> ret);
>>>> goto err_iahb;
>>>> }
>>>> +
>>>> + ret = clk_set_rate(hdmi->cec_clk, HDMI_CEC_CLK_RATE);
>>>> + if (ret)
>>>> + dev_warn(hdmi->dev, "Cannot set HDMI cec clock
>>>> rate: %d\n", ret);
>>>> }
>>>>
>>>> /* Product and revision IDs */
>>>> --
>>>> 2.25.1
>>>>
>>>>
>>>> _______________________________________________
>>>> Linux-rockchip mailing list
>>>> [email protected]
>>>> http://lists.infradead.org/mailman/listinfo/linux-rockchip
>>>
>>> Peter,
>>>
>>> On my 5.17-rc7 with applied rk356x VOP2 v8 series - this patch makes
>>> CEC working on rk3566.
>>> Unfortunately it breaks working ok CEC on rk3399 rockpi-4b.
>>>
>>> Reverting this patch brings back CEC on rk3399 - but rk3366 becomes
>>> with non working CEC
>>>
>>> I'm not sure how to move forward with this....
>>
>> I was worried about that, thanks for testing it.
>> Can you send me the cec_clk rate before and after this patch?
>
> Hmm, looks like there might be a bug in the RK3399 clock driver -
> although it's treated as having mux_pll_p as parents, according to the
> public TRM the CEC clock appears unique in being backwards compared to
> every other mux of those two inputs. I'm now tempted to test this on my
> board tonight and see if I can see 24MHz on the CEC pin... :)
Nope, turns out that's an error in the TRM and the mux is fine. The bug
is between the clock driver blindly assuming xin32k is usable, and
nearly all RK3399 boards forgetting to claim the pinctrl to actually
enable the input :/
Robin.