[v2] clk: rockchip: Add pclk_peri to critical clocks on RK3066/RK3188
diff mbox

Message ID 1440322357-2621-1-git-send-email-romain.perier@gmail.com
State New
Headers show

Commit Message

Romain Perier Aug. 23, 2015, 9:32 a.m. UTC
Now that the rockchip clock subsystem does clock gating with GPIO banks,
these are no longer enabled once during probe and no longer stay enabled
for eternity. When all these clocks are disabled, the parent clock pclk_peri
might be disabled too, as no other child claims it. So, we need to add pclk_peri
to the critical clocks.

Signed-off-by: Romain Perier <romain.perier@gmail.com>
Tested-by: Michael Niewoehner <linux@mniewoehner.de>
---

changes in v2:
- Improved commit messages
- Added "Tested-by" from Michael Niewoehner, as he tested v1 and no code changed

 drivers/clk/rockchip/clk-rk3188.c | 1 +
 1 file changed, 1 insertion(+)

Comments

Romain Perier Aug. 31, 2015, 2:42 p.m. UTC | #1
ping

2015-08-23 11:32 GMT+02:00 Romain Perier <romain.perier@gmail.com>:
> Now that the rockchip clock subsystem does clock gating with GPIO banks,
> these are no longer enabled once during probe and no longer stay enabled
> for eternity. When all these clocks are disabled, the parent clock pclk_peri
> might be disabled too, as no other child claims it. So, we need to add pclk_peri
> to the critical clocks.
>
> Signed-off-by: Romain Perier <romain.perier@gmail.com>
> Tested-by: Michael Niewoehner <linux@mniewoehner.de>
> ---
>
> changes in v2:
> - Improved commit messages
> - Added "Tested-by" from Michael Niewoehner, as he tested v1 and no code changed
>
>  drivers/clk/rockchip/clk-rk3188.c | 1 +
>  1 file changed, 1 insertion(+)
>
> diff --git a/drivers/clk/rockchip/clk-rk3188.c b/drivers/clk/rockchip/clk-rk3188.c
> index bf20214..839a22a 100644
> --- a/drivers/clk/rockchip/clk-rk3188.c
> +++ b/drivers/clk/rockchip/clk-rk3188.c
> @@ -717,6 +717,7 @@ static const char *const rk3188_critical_clocks[] __initconst = {
>         "aclk_peri",
>         "hclk_peri",
>         "pclk_cpu",
> +       "pclk_peri",
>  };
>
>  static void __init rk3188_common_clk_init(struct device_node *np)
> --
> 2.1.4
>
Romain Perier Sept. 4, 2015, 6:50 a.m. UTC | #2
Hi all,

just a reminder, I sent this patch two weeks ago, rk3xxx-based
platforms still need it, it fixes a freeze during boot.

Regards,
Romain

2015-08-31 16:42 GMT+02:00 Romain Perier <romain.perier@gmail.com>:
> ping
>
> 2015-08-23 11:32 GMT+02:00 Romain Perier <romain.perier@gmail.com>:
>> Now that the rockchip clock subsystem does clock gating with GPIO banks,
>> these are no longer enabled once during probe and no longer stay enabled
>> for eternity. When all these clocks are disabled, the parent clock pclk_peri
>> might be disabled too, as no other child claims it. So, we need to add pclk_peri
>> to the critical clocks.
>>
>> Signed-off-by: Romain Perier <romain.perier@gmail.com>
>> Tested-by: Michael Niewoehner <linux@mniewoehner.de>
>> ---
>>
>> changes in v2:
>> - Improved commit messages
>> - Added "Tested-by" from Michael Niewoehner, as he tested v1 and no code changed
>>
>>  drivers/clk/rockchip/clk-rk3188.c | 1 +
>>  1 file changed, 1 insertion(+)
>>
>> diff --git a/drivers/clk/rockchip/clk-rk3188.c b/drivers/clk/rockchip/clk-rk3188.c
>> index bf20214..839a22a 100644
>> --- a/drivers/clk/rockchip/clk-rk3188.c
>> +++ b/drivers/clk/rockchip/clk-rk3188.c
>> @@ -717,6 +717,7 @@ static const char *const rk3188_critical_clocks[] __initconst = {
>>         "aclk_peri",
>>         "hclk_peri",
>>         "pclk_cpu",
>> +       "pclk_peri",
>>  };
>>
>>  static void __init rk3188_common_clk_init(struct device_node *np)
>> --
>> 2.1.4
>>
Stephen Boyd Sept. 4, 2015, 4:24 p.m. UTC | #3
On 09/04, Romain Perier wrote:
> Hi all,
> 
> just a reminder, I sent this patch two weeks ago, rk3xxx-based
> platforms still need it, it fixes a freeze during boot.

Sorry I thought that Linus W was going to pick this up? If not,
we can take it through clk tree as a fix for -rc1.
Romain Perier Sept. 5, 2015, 4:12 p.m. UTC | #4
Np,

Yes it would be nice to take it for -rc1, because I think that 4.3
won't boot without this fix


Regards,
Romain

2015-09-04 18:24 GMT+02:00 Stephen Boyd <sboyd@codeaurora.org>:
> On 09/04, Romain Perier wrote:
>> Hi all,
>>
>> just a reminder, I sent this patch two weeks ago, rk3xxx-based
>> platforms still need it, it fixes a freeze during boot.
>
> Sorry I thought that Linus W was going to pick this up? If not,
> we can take it through clk tree as a fix for -rc1.
>
> --
> Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
> a Linux Foundation Collaborative Project
Romain Perier Sept. 9, 2015, 5:58 p.m. UTC | #5
so, which of you does take the patch for -rc1 ?

Regards,
Romain

2015-09-05 18:12 GMT+02:00 Romain Perier <romain.perier@gmail.com>:
> Np,
>
> Yes it would be nice to take it for -rc1, because I think that 4.3
> won't boot without this fix
>
>
> Regards,
> Romain
>
> 2015-09-04 18:24 GMT+02:00 Stephen Boyd <sboyd@codeaurora.org>:
>> On 09/04, Romain Perier wrote:
>>> Hi all,
>>>
>>> just a reminder, I sent this patch two weeks ago, rk3xxx-based
>>> platforms still need it, it fixes a freeze during boot.
>>
>> Sorry I thought that Linus W was going to pick this up? If not,
>> we can take it through clk tree as a fix for -rc1.
>>
>> --
>> Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
>> a Linux Foundation Collaborative Project
Stephen Boyd Sept. 9, 2015, 9:47 p.m. UTC | #6
On 09/09, Romain Perier wrote:
> so, which of you does take the patch for -rc1 ?
> 

Please stop top posting. I was going to take it after -rc1 is
tagged, but I can't tell what branch this is based on. It doesn't
apply to clk-next, nor to torvalds' tree, or linux-next. Is there
some other patch out there that's also needed?
Romain Perier Sept. 10, 2015, 7:05 a.m. UTC | #7
It was based on linux-next few weeks ago. I will rebase it today and send a v3.

2015-09-09 23:47 GMT+02:00 Stephen Boyd <sboyd@codeaurora.org>:
> On 09/09, Romain Perier wrote:
>> so, which of you does take the patch for -rc1 ?
>>
> Is there some other patch out there that's also needed?

No, that's a single patch.

Romain
Michael Niewöhner Sept. 10, 2015, 4:40 p.m. UTC | #8
Am 10.09.2015 um 09:05 schrieb Romain Perier <romain.perier@gmail.com>:

> It was based on linux-next few weeks ago. I will rebase it today and send a v3.
> 
> 2015-09-09 23:47 GMT+02:00 Stephen Boyd <sboyd@codeaurora.org>:
>> On 09/09, Romain Perier wrote:
>>> so, which of you does take the patch for -rc1 ?
>>> 
>> Is there some other patch out there that's also needed?
> 
> No, that's a single patch.
> 
> Romain
> 

Hi Stephen,

Romain’s patch was applied after mine in his github tree so you’ll need to apply that first: [PATCH v3] clk: rockchip: add pclk_cpu to the list of rk3188 critical clocks

Best regards
Michael
Stephen Boyd Sept. 10, 2015, 8:51 p.m. UTC | #9
On 09/10, Michael Niewoehner wrote:
> 
> Am 10.09.2015 um 09:05 schrieb Romain Perier <romain.perier@gmail.com>:
> 
> > It was based on linux-next few weeks ago. I will rebase it today and send a v3.
> > 
> > 2015-09-09 23:47 GMT+02:00 Stephen Boyd <sboyd@codeaurora.org>:
> >> On 09/09, Romain Perier wrote:
> >>> so, which of you does take the patch for -rc1 ?
> >>> 
> >> Is there some other patch out there that's also needed?
> > 
> > No, that's a single patch.
> > 
> > Romain
> > 
> 
> Hi Stephen,
> 
> Romain’s patch was applied after mine in his github tree so you’ll need to apply that first: [PATCH v3] clk: rockchip: add pclk_cpu to the list of rk3188 critical clocks
> 

Ah ok. Consider both applied to clk-fixes.

Patch
diff mbox

diff --git a/drivers/clk/rockchip/clk-rk3188.c b/drivers/clk/rockchip/clk-rk3188.c
index bf20214..839a22a 100644
--- a/drivers/clk/rockchip/clk-rk3188.c
+++ b/drivers/clk/rockchip/clk-rk3188.c
@@ -717,6 +717,7 @@  static const char *const rk3188_critical_clocks[] __initconst = {
 	"aclk_peri",
 	"hclk_peri",
 	"pclk_cpu",
+	"pclk_peri",
 };
 
 static void __init rk3188_common_clk_init(struct device_node *np)