On 2/21/24 06:40, Sumit Garg wrote:
> On Tue, 20 Feb 2024 at 21:02, Marek Vasut <marex at denx.de> wrote:
>>
>> On 2/20/24 14:10, Sumit Garg wrote:
>>> Pre-requisite to enable PCIe support on iMX8MP SoC.
>>
>> Please write a proper commit message .
>>
>
> How about the following?
>
> Add support for i.MX8MP reset controller. It is required
> to enable PCIe support on the iMX8MP SoC.
The part about PCIe is not relevant, it is a reset controller and it is
needed to reset various IPs. See e.g. Linux:
e08672c03981 ("reset: imx7: Add support for i.MX8MP SoC")
>>> Signed-off-by: Sumit Garg <sumit.garg at linaro.org>
>>> ---
>>> drivers/reset/reset-imx7.c | 114 +++++++++++++++++++++++++++++++++++++
>>> 1 file changed, 114 insertions(+)
>>>
>>> diff --git a/drivers/reset/reset-imx7.c b/drivers/reset/reset-imx7.c
>>> index eaef2cc2cdf..c1de84dea8b 100644
>>> --- a/drivers/reset/reset-imx7.c
>>> +++ b/drivers/reset/reset-imx7.c
>>> @@ -10,6 +10,7 @@
>>> #include <dm.h>
>>> #include <dt-bindings/reset/imx7-reset.h>
>>> #include <dt-bindings/reset/imx8mq-reset.h>
>>> +#include <dt-bindings/reset/imx8mp-reset.h>
>>> #include <reset-uclass.h>
>>> #include <linux/bitops.h>
>>> #include <linux/delay.h>
>>> @@ -252,6 +253,115 @@ static int imx7_reset_assert_imx8mq(struct reset_ctl *rst)
>>> return 0;
>>> }
>>>
>>> +enum imx8mp_src_registers {
>>> + SRC_SUPERMIX_RCR = 0x0018,
>>> + SRC_AUDIOMIX_RCR = 0x001c,
>>> + SRC_MLMIX_RCR = 0x0028,
>>> + SRC_GPU2D_RCR = 0x0038,
>>> + SRC_GPU3D_RCR = 0x003c,
>>> + SRC_VPU_G1_RCR = 0x0048,
>>> + SRC_VPU_G2_RCR = 0x004c,
>>> + SRC_VPUVC8KE_RCR = 0x0050,
>>> + SRC_NOC_RCR = 0x0054,
>>> +};
>>
>> This seems copied from Linux, include Linux commit ID as of which this
>> was imported from in commit message.
>
> Do you expect imx8mp_src_registers to change? I expect them to be the
> same, so does Linux commit ID add any value here?
The import commit ID makes it easier to cherry-pick follow up patches or
check for changes. If that commit ID is missing, that task is harder.