[PATCHv1,0/2] Improve RK3588 clocks and power domains support

Message

Sebastian Reichel March 31, 2023, 4:30 p.m. UTC

Hi,

After this patchset the power-domain support for RK3588 is basically on par
with the downstream implementation. More importantly the clock controller
change unblocks support for any IP blocks using linked clocks. The clock
controller change must be applied before the power-domain change to avoid
accessing unclocked registers.

-- Sebastian

Boris Brezillon (1):
  soc: rockchip: power-domain: add rk3588 mem module support

Sebastian Reichel (1):
  clk: rockchip: rk3588: make gate linked clocks ignore unused

 drivers/clk/rockchip/clk-rk3588.c |  42 +++++---
 drivers/soc/rockchip/pm_domains.c | 160 +++++++++++++++++++++++-------
 2 files changed, 151 insertions(+), 51 deletions(-)

Comments

Vincent Legoll April 1, 2023, 2:30 p.m. UTC | #1

Hello,

I'm running a QuartzPro64 with this patch set applied, everything(*)
works properly.

(*) eMMC, SD, LED, ethernet, serial, watchdog.

Are there any tests I can do to help with this ?

Regards

Heiko Stübner April 3, 2023, 10:22 a.m. UTC | #2

Hi,

Am Samstag, 1. April 2023, 16:30:53 CEST schrieb Vincent Legoll:
> Hello,
> 
> I'm running a QuartzPro64 with this patch set applied, everything(*)
> works properly.
> 
> (*) eMMC, SD, LED, ethernet, serial, watchdog.
> 
> Are there any tests I can do to help with this ?

gmac1 and sdmmc do reference power-domains in the dts, so things should
be fine and I guess if your system boots successfully this might be enough.

If you want to check more, try
# mkdir /debug
# mount none /debug -t debugfs
[or do this to a location of your choice, I guess the standard is under /proc]

# cat /debug/pm_genpd/pm_genpd_summary

and check the runtime_status field.


Also you could provide a email-reply with a

Tested-by: Your Name <your-email>

line in it :-)


Heiko

Vincent Legoll April 3, 2023, 10:38 a.m. UTC | #3

On Mon, Apr 3, 2023 at 12:23 PM Heiko Stübner <[e-mail]> wrote:
> If you want to check more, try
> # mkdir /debug
> # mount none /debug -t debugfs
> [or do this to a location of your choice, I guess the standard is under /proc]
>
> # cat /debug/pm_genpd/pm_genpd_summary
>
> and check the runtime_status field.

Here it is:

domain                          status          children
            performance
    /device                                             runtime status
----------------------------------------------------------------------------------------------
sdmmc                           off-0
            0
    /devices/platform/fe2c0000.mmc                      suspended
            0
audio                           off-0
            0
sdio                            off-0
            0
pcie                            off-0
            0
gmac                            on
            0
    /devices/platform/fe1b0000.ethernet                 active
            0
usb                             off-0
            0
rga31                           off-0
            0
fec                             off-0
            0
isp1                            off-0
            0
vi                              off-0
            0
                                                isp1, fec
vo1                             off-0
            0
vo0                             off-0
            0
vop                             off-0
            0
                                                vo0
rga30                           off-0
            0
av1                             off-0
            0
vdpu                            off-0
            0
                                                av1, rkvdec0, rkvdec1, rga30
venc1                           off-0
            0
venc0                           off-0
            0
                                                venc1
rkvdec1                         off-0
            0
rkvdec0                         off-0
            0
vcodec                          off-0
            0
                                                rkvdec0, rkvdec1, venc0
gpu                             off-0
            0
npu2                            off-0
            0
npu1                            off-0
            0
nputop                          off-0
            0
                                                npu1, npu2
npu                             off-0
            0
                                                nputop

Don't know how to interpret this though...

> Also you could provide a email-reply with a
>
> Tested-by: Your Name <your-email>

You can add my:

Tested-by: Vincent Legoll <[e-mail]>

Regards