diff mbox

[v2,04/13] clk: davinci: psc-dm646x: use two lookup entries for the aemif clock

Message ID 20180626092537.6737-5-brgl@bgdev.pl (mailing list archive)
State New, archived
Headers show

Commit Message

Bartosz Golaszewski June 26, 2018, 9:25 a.m. UTC
From: Bartosz Golaszewski <bgolaszewski@baylibre.com>

We want to be able to get the clock both from the board file by its
con_id and from the aemif driver by dev_id.

Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
---
 drivers/clk/davinci/psc-dm646x.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

Comments

David Lechner June 26, 2018, 4:17 p.m. UTC | #1
On 06/26/2018 04:25 AM, Bartosz Golaszewski wrote:
> From: Bartosz Golaszewski <bgolaszewski@baylibre.com>
> 
> We want to be able to get the clock both from the board file by its
> con_id and from the aemif driver by dev_id.
> 
> Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
> ---

Reviewed-by: David Lechner <david@lechnology.com>
Sekhar Nori June 27, 2018, 10:01 a.m. UTC | #2
On Tuesday 26 June 2018 02:55 PM, Bartosz Golaszewski wrote:
> From: Bartosz Golaszewski <bgolaszewski@baylibre.com>
> 
> We want to be able to get the clock both from the board file by its
> con_id and from the aemif driver by dev_id.
> 
> Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>

I see the DM365 and DM644x EVM board files getting reference to aemif
clock using just the con_id, but don't see that for DM646x. Am I
overlooking something?

Thanks,
Sekhar
Bartosz Golaszewski June 27, 2018, 10:03 a.m. UTC | #3
2018-06-27 12:01 GMT+02:00 Sekhar Nori <nsekhar@ti.com>:
> On Tuesday 26 June 2018 02:55 PM, Bartosz Golaszewski wrote:
>> From: Bartosz Golaszewski <bgolaszewski@baylibre.com>
>>
>> We want to be able to get the clock both from the board file by its
>> con_id and from the aemif driver by dev_id.
>>
>> Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
>
> I see the DM365 and DM644x EVM board files getting reference to aemif
> clock using just the con_id, but don't see that for DM646x. Am I
> overlooking something?
>
> Thanks,
> Sekhar

It's being done indirectly from mach-davinci/aemif.c. I guess we can
remove it again after last patch.

Bart
Sekhar Nori June 27, 2018, 10:26 a.m. UTC | #4
On Wednesday 27 June 2018 03:33 PM, Bartosz Golaszewski wrote:
> 2018-06-27 12:01 GMT+02:00 Sekhar Nori <nsekhar@ti.com>:
>> On Tuesday 26 June 2018 02:55 PM, Bartosz Golaszewski wrote:
>>> From: Bartosz Golaszewski <bgolaszewski@baylibre.com>
>>>
>>> We want to be able to get the clock both from the board file by its
>>> con_id and from the aemif driver by dev_id.
>>>
>>> Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
>>
>> I see the DM365 and DM644x EVM board files getting reference to aemif
>> clock using just the con_id, but don't see that for DM646x. Am I
>> overlooking something?
>>
>> Thanks,
>> Sekhar
> 
> It's being done indirectly from mach-davinci/aemif.c. I guess we can
> remove it again after last patch.

I see. One way to avoid that would be to merge this into 11/13. I prefer
that, but it prevents clean separation of clock and machine patches. So
okay with current way too.

Thanks,
Sekhar
David Lechner June 27, 2018, 4:54 p.m. UTC | #5
On 06/27/2018 05:26 AM, Sekhar Nori wrote:
> On Wednesday 27 June 2018 03:33 PM, Bartosz Golaszewski wrote:
>> 2018-06-27 12:01 GMT+02:00 Sekhar Nori <nsekhar@ti.com>:
>>> On Tuesday 26 June 2018 02:55 PM, Bartosz Golaszewski wrote:
>>>> From: Bartosz Golaszewski <bgolaszewski@baylibre.com>
>>>>
>>>> We want to be able to get the clock both from the board file by its
>>>> con_id and from the aemif driver by dev_id.
>>>>
>>>> Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
>>>
>>> I see the DM365 and DM644x EVM board files getting reference to aemif
>>> clock using just the con_id, but don't see that for DM646x. Am I
>>> overlooking something?
>>>
>>> Thanks,
>>> Sekhar
>>
>> It's being done indirectly from mach-davinci/aemif.c. I guess we can
>> remove it again after last patch.
> 
> I see. One way to avoid that would be to merge this into 11/13. I prefer
> that, but it prevents clean separation of clock and machine patches. So
> okay with current way too.

I think da850 has a con_id-only lookup that can be removed as well. I'm
in favor of doing another series to clean these up later.
diff mbox

Patch

diff --git a/drivers/clk/davinci/psc-dm646x.c b/drivers/clk/davinci/psc-dm646x.c
index c3f82ed70a80..20012dc7471a 100644
--- a/drivers/clk/davinci/psc-dm646x.c
+++ b/drivers/clk/davinci/psc-dm646x.c
@@ -18,7 +18,8 @@ 
 LPSC_CLKDEV1(ide_clkdev,	NULL,		"palm_bk3710");
 LPSC_CLKDEV2(emac_clkdev,	NULL,		"davinci_emac.1",
 				"fck",		"davinci_mdio.0");
-LPSC_CLKDEV1(aemif_clkdev,	"aemif",	NULL);
+LPSC_CLKDEV2(aemif_clkdev,	"aemif",	NULL,
+				NULL,		"ti-aemif");
 LPSC_CLKDEV1(mcasp0_clkdev,	NULL,		"davinci-mcasp.0");
 LPSC_CLKDEV1(mcasp1_clkdev,	NULL,		"davinci-mcasp.1");
 LPSC_CLKDEV1(uart0_clkdev,	NULL,		"serial8250.0");