Re: [PATCH] clk: sunxi-ng: sun50i: h6: Modify GPU clock configuration to support DFS

From: Roman Stratiienko
Date: Sun Jul 03 2022 - 12:40:52 EST


Hello Samuel,

Thanks for having a look.

вс, 3 июл. 2022 г. в 09:50, Samuel Holland <samuel@xxxxxxxxxxxx>:
>
> On 6/24/22 11:52 AM, Roman Stratiienko wrote:
> > Using simple bash script it was discovered that not all CCU registers
> > can be safely used for DFS, e.g.:
> >
> > while true
> > do
> > devmem 0x3001030 4 0xb0003e02
> > devmem 0x3001030 4 0xb0001e02
> > done
> >
> > Script above changes the GPU_PLL multiplier register value. While the
> > script is running, the user should interact with the user interface.
> >
> > Using this method the following results were obtained:
> >
> > | Register | Name | Bits | Values | Result |
> > | -- | -- | -- | -- | -- |
> > | 0x3001030 | GPU_PLL.MULT | 15..8 | 20-62 | OK |
> > | 0x3001030 | GPU_PLL.INDIV | 1 | 0-1 | OK |
> > | 0x3001030 | GPU_PLL.OUTDIV | 0 | 0-1 | FAIL |
> > | 0x3001670 | GPU_CLK.DIV | 3..0 | ANY | FAIL |
> >
> > Once bits that caused system failure disabled (kept default 0),
> > it was discovered that GPU_CLK.MUX was used during DFS for some
> > reason and was causing the failure too.
>
> The GPU module clock has only one parent declared, so it is surprising that the
> mux would get set. Did this happen while the kernel driver was changing the
> frequency?

I looked through the ccu code and didn't see anything that may cause
issues, so I tested again and DFS works with MUX this time.

I'll drop this change in v2.

>
> > After disabling GPU_PLL.OUTDIV the system started to fail during
> > booting for some reason until the maximum frequency of GPU_PLL
> > clock was limited to 756MHz.
>
> The manual lists PLL_GPU's maximum frequency as 800 MHz. I assume you chose 756
> MHz because that is the highest OPP. That should be okay, too.

Setting the frequency higher than 756 makes the GPU very unstable.

I decided to validate it again and removed the frequency limitation
and can't see any issues so far.

I'll also drop this change in v2.

>
> > After all the changes made DVFS started to work seamlessly.
> >
> > Signed-off-by: Roman Stratiienko <r.stratiienko@xxxxxxxxx>
> > ---
> > drivers/clk/sunxi-ng/ccu-sun50i-h6.c | 12 +++++-------
> > 1 file changed, 5 insertions(+), 7 deletions(-)
> >
> > diff --git a/drivers/clk/sunxi-ng/ccu-sun50i-h6.c b/drivers/clk/sunxi-ng/ccu-sun50i-h6.c
> > index 2ddf0a0da526f..d941238cd178a 100644
> > --- a/drivers/clk/sunxi-ng/ccu-sun50i-h6.c
> > +++ b/drivers/clk/sunxi-ng/ccu-sun50i-h6.c
> > @@ -95,13 +95,14 @@ static struct ccu_nkmp pll_periph1_clk = {
> > },
> > };
> >
> > +/* For GPU PLL, using an output divider for DFS causes system to fail */
> > #define SUN50I_H6_PLL_GPU_REG 0x030
> > static struct ccu_nkmp pll_gpu_clk = {
> > .enable = BIT(31),
> > .lock = BIT(28),
> > .n = _SUNXI_CCU_MULT_MIN(8, 8, 12),
> > .m = _SUNXI_CCU_DIV(1, 1), /* input divider */
> > - .p = _SUNXI_CCU_DIV(0, 1), /* output divider */
> > + .max_rate = 756000000UL,
> > .common = {
> > .reg = 0x030,
> > .hw.init = CLK_HW_INIT("pll-gpu", "osc24M",
> > @@ -294,12 +295,9 @@ static SUNXI_CCU_M_WITH_MUX_GATE(deinterlace_clk, "deinterlace",
> > static SUNXI_CCU_GATE(bus_deinterlace_clk, "bus-deinterlace", "psi-ahb1-ahb2",
> > 0x62c, BIT(0), 0);
> >
> > -static const char * const gpu_parents[] = { "pll-gpu" };
> > -static SUNXI_CCU_M_WITH_MUX_GATE(gpu_clk, "gpu", gpu_parents, 0x670,
> > - 0, 3, /* M */
> > - 24, 1, /* mux */
> > - BIT(31), /* gate */
> > - CLK_SET_RATE_PARENT);
> > +/* GPU_CLK divider kept disabled to avoid interferences with DFS */
> > +static SUNXI_CCU_GATE(gpu_clk, "gpu", "pll-gpu", 0x670,
> > + BIT(31), CLK_SET_RATE_PARENT);
>
> These changes look fine to me. You also need to set the initial value for the
> fixed fields in the driver's probe function.

Will do that in v2.

I have no idea what was causing additional issues in my previous test
session. Let's forget about them for now.

Regards,
Roman.

>
> Regards,
> Samuel