[PATCH 2/5] firmware: zynqmp: Load config overlay for core0 to pmufw
Stefan Herbrechtsmeier
stefan at herbrechtsmeier.net
Wed Jul 20 18:02:03 CEST 2022
Hi,
Am 19.07.22 um 06:44 schrieb Soma, Ashok Reddy:
> Hi Stefan,
>
>> -----Original Message-----
>> From: Stefan Herbrechtsmeier <stefan.herbrechtsmeier-oss at weidmueller.com>
>> Sent: Saturday, July 16, 2022 4:48 PM
>> To: Simek, Michal <michal.simek at amd.com>; Ashok Reddy Soma <ashok.reddy.soma at xilinx.com>; u-boot at lists.denx.de
>> Cc: adrian.fiergolski at fastree3d.com; jh80.chung at samsung.com; sven at svenpeter.dev; kettenis at openbsd.org; sjg at chromium.org; git at xilinx.com; git (AMD-Xilinx) <git at amd.com>
>> Subject: Re: [PATCH 2/5] firmware: zynqmp: Load config overlay for core0 to pmufw
>>
>> CAUTION: This message has originated from an External Source. Please use proper judgment and caution when opening attachments, clicking links, or responding to this email.
>>
>>
>> Am 15.07.2022 um 18:34 schrieb Michal Simek:
>>>
>>>
>>> On 7/15/22 18:13, Stefan Herbrechtsmeier wrote:
>>>> Am 15.07.2022 um 11:39 schrieb Ashok Reddy Soma:
>>>>> Try loading pmufw config overlay for core0, if it doesn't return any
>>>>> error it means pmufw is accepting nodes for other IP's. Otherwise
>>>>> dont try to load config object for any other IP, just return from
>>>>> zynqmp_pmufw_node function.
>>>>>
>>>>> Signed-off-by: Ashok Reddy Soma <ashok.reddy.soma at xilinx.com>
>>>>> ---
>>>>>
>>>>> drivers/firmware/firmware-zynqmp.c | 14 ++++++++++++++
>>>>> 1 file changed, 14 insertions(+)
>>>>>
>>>>> diff --git a/drivers/firmware/firmware-zynqmp.c
>>>>> b/drivers/firmware/firmware-zynqmp.c
>>>>> index 34d9b47003..288151533e 100644
>>>>> --- a/drivers/firmware/firmware-zynqmp.c
>>>>> +++ b/drivers/firmware/firmware-zynqmp.c
>>>>> @@ -68,8 +68,13 @@ int zynqmp_pmufw_config_close(void)
>>>>> return 0;
>>>>> }
>>>>> +static bool config_enabled;
>>>>> +
>>>>
>>>> Please move the variable inside the function.
>>>
>>> How can this work? When you move it to zynqmp_pmufw_node() then won't
>>> be visible in zynqmp_power_probe() and vice-versa.
>
>> If you reuse the zynqmp_pmufw_node function in zynqmp_power_probe function you can check the id parameter to update the config_enabled variable in zynqmp_pmufw_node.
>
> are you suggesting to change like this ? this works fine for me. Shall i send V2 with this >
> --- a/drivers/firmware/firmware-zynqmp.c
> +++ b/drivers/firmware/firmware-zynqmp.c
> @@ -29,6 +29,7 @@ struct zynqmp_power {
> } zynqmp_power;
>
> #define NODE_ID_LOCATION 5
> +#define APU0_ID NODE_APU_0
Why don't you use the NODE_APU_0 define direct?
> static unsigned int xpm_configobject[] = {
> /**********************************************************************/
> @@ -68,18 +69,22 @@ int zynqmp_pmufw_config_close(void)
> return 0;
> }
>
> -static bool config_enabled;
>
> int zynqmp_pmufw_node(u32 id)
> {
> - if (!config_enabled)
> + static bool config_enabled;
I would invert the meaning from enabled to skip ...
> + int ret;
> +
> + if (!config_enabled && id != APU0_ID)
to simplify the check.
if (skip)
> return 0;
>
> /* Record power domain id */
> xpm_configobject[NODE_ID_LOCATION] = id;
>
> - zynqmp_pmufw_load_config_object(xpm_configobject,
> - sizeof(xpm_configobject));
> + ret = zynqmp_pmufw_load_config_object(xpm_configobject,
> + sizeof(xpm_configobject));
> + if(!ret && id == APU0_ID)
> + config_enabled = true;
if(ret && id == APU0_ID)
skip = true;
>
> return 0;
> }
> @@ -272,14 +277,8 @@ static int zynqmp_power_probe(struct udevice *dev)
> ret >> ZYNQMP_PM_VERSION_MAJOR_SHIFT,
> ret & ZYNQMP_PM_VERSION_MINOR_MASK);
>
> - if (IS_ENABLED(CONFIG_ARCH_ZYNQMP)) {
> - xpm_configobject[NODE_ID_LOCATION] = NODE_APU_0;
> -
> - ret = zynqmp_pmufw_load_config_object(xpm_configobject,
> - sizeof(xpm_configobject));
> - if (!ret)
> - config_enabled = true;
> - }
> + if (IS_ENABLED(CONFIG_ARCH_ZYNQMP))
> + zynqmp_pmufw_node(APU0_ID);
>
> return 0;
>
Regards
Stefan
More information about the U-Boot
mailing list