diff mbox series

clk: rockchip: fix frac settings of GPLL clock for rk3328

Message ID 20181222164249.25620-1-katsuhiro@katsuster.net (mailing list archive)
State Mainlined, archived
Commit a0e447b0c50240a90ab84b7126b3c06b0bab4adc
Headers show
Series clk: rockchip: fix frac settings of GPLL clock for rk3328 | expand

Commit Message

Katsuhiro Suzuki Dec. 22, 2018, 4:42 p.m. UTC
This patch fixes settings of GPLL frequency in fractional mode for
rk3328. In this mode, FOUTVCO is calcurated by following formula:
  FOUTVCO = FREF * FBDIV / REFDIV + ((FREF * FRAC / REFDIV) >> 24)

The problem is in FREF * FRAC >> 24 term. This result always lacks
one from target value is specified by rate member. For example first
itme of rk3328_pll_frac_rate originally has
  - rate  : 1016064000
  - refdiv: 3
  - fbdiv : 127
  - frac  : 134217
  - FREF * FBDIV / REFDIV        = 1016000000
  - (FREF * FRAC / REFDIV) >> 24 = 63999
Thus calculated rate is 1016063999. It seems wrong.

If frac has 134218 (it is increased 1 from original value), second
term is 64000. All other items have same situation. So this patch
adds 1 to frac member in all items of rk3328_pll_frac_rate.

Signed-off-by: Katsuhiro Suzuki <katsuhiro@katsuster.net>
---
 drivers/clk/rockchip/clk-rk3328.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

Comments

Heiko Stübner Dec. 24, 2018, 7:34 a.m. UTC | #1
Hi,

Am Samstag, 22. Dezember 2018, 17:42:49 CET schrieb Katsuhiro Suzuki:
> This patch fixes settings of GPLL frequency in fractional mode for
> rk3328. In this mode, FOUTVCO is calcurated by following formula:
>   FOUTVCO = FREF * FBDIV / REFDIV + ((FREF * FRAC / REFDIV) >> 24)
> 
> The problem is in FREF * FRAC >> 24 term. This result always lacks
> one from target value is specified by rate member. For example first
> itme of rk3328_pll_frac_rate originally has
>   - rate  : 1016064000
>   - refdiv: 3
>   - fbdiv : 127
>   - frac  : 134217
>   - FREF * FBDIV / REFDIV        = 1016000000
>   - (FREF * FRAC / REFDIV) >> 24 = 63999
> Thus calculated rate is 1016063999. It seems wrong.
> 
> If frac has 134218 (it is increased 1 from original value), second
> term is 64000. All other items have same situation. So this patch
> adds 1 to frac member in all items of rk3328_pll_frac_rate.
> 
> Signed-off-by: Katsuhiro Suzuki <katsuhiro@katsuster.net>

so while this sounds all quite right to me, I've added some Rockchip
people that have clock experience to hopefully get an Ack on the
change :-)

@Elaine + Finley: does this explanation and the below change look right?


Thanks
Heiko

> ---
>  drivers/clk/rockchip/clk-rk3328.c | 12 ++++++------
>  1 file changed, 6 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/clk/rockchip/clk-rk3328.c b/drivers/clk/rockchip/clk-rk3328.c
> index faa94adb2a37..65ab5c2f48b0 100644
> --- a/drivers/clk/rockchip/clk-rk3328.c
> +++ b/drivers/clk/rockchip/clk-rk3328.c
> @@ -78,17 +78,17 @@ static struct rockchip_pll_rate_table rk3328_pll_rates[] = {
>  
>  static struct rockchip_pll_rate_table rk3328_pll_frac_rates[] = {
>  	/* _mhz, _refdiv, _fbdiv, _postdiv1, _postdiv2, _dsmpd, _frac */
> -	RK3036_PLL_RATE(1016064000, 3, 127, 1, 1, 0, 134217),
> +	RK3036_PLL_RATE(1016064000, 3, 127, 1, 1, 0, 134218),
>  	/* vco = 1016064000 */
> -	RK3036_PLL_RATE(983040000, 24, 983, 1, 1, 0, 671088),
> +	RK3036_PLL_RATE(983040000, 24, 983, 1, 1, 0, 671089),
>  	/* vco = 983040000 */
> -	RK3036_PLL_RATE(491520000, 24, 983, 2, 1, 0, 671088),
> +	RK3036_PLL_RATE(491520000, 24, 983, 2, 1, 0, 671089),
>  	/* vco = 983040000 */
> -	RK3036_PLL_RATE(61440000, 6, 215, 7, 2, 0, 671088),
> +	RK3036_PLL_RATE(61440000, 6, 215, 7, 2, 0, 671089),
>  	/* vco = 860156000 */
> -	RK3036_PLL_RATE(56448000, 12, 451, 4, 4, 0, 9797894),
> +	RK3036_PLL_RATE(56448000, 12, 451, 4, 4, 0, 9797895),
>  	/* vco = 903168000 */
> -	RK3036_PLL_RATE(40960000, 12, 409, 4, 5, 0, 10066329),
> +	RK3036_PLL_RATE(40960000, 12, 409, 4, 5, 0, 10066330),
>  	/* vco = 819200000 */
>  	{ /* sentinel */ },
>  };
>
Heiko Stübner Dec. 24, 2018, 7:54 a.m. UTC | #2
Hi Elaine,

Am Montag, 24. Dezember 2018, 08:48:05 CET schrieb zhangqing@rock-chips.com:
> Hi: Heiko:
> 
> This explanation and the change is OK. Thanks for Katsuhiro's correction.

thanks for your quick confirmation. Can I make this an

Acked-by: Elaine Zhang <zhangqing@rock-chips.com>

when applying the patch?

Thanks
Heiko


> 张晴
> 福州瑞芯微电子股份有限公司
> Fuzhou Rockchip Electronics Co.,Ltd
> 地址:福建省福州市铜盘路软件大道89号软件园A区21号楼
> Add:No.21 Building, A District, No.89 Software Boulevard Fuzhou, Fujian 350003, P.R.China
> Tel:+86-0591-83991906-8601
> 邮编:350003
> E-mail:elaine.zhang@rock-chips.com
>  
> From: Heiko Stuebner
> Date: 2018-12-24 15:34
> To: Katsuhiro Suzuki; Finley Xiao; Elaine Zhang
> CC: linux-rockchip; linux-arm-kernel; linux-kernel
> Subject: Re: [PATCH] clk: rockchip: fix frac settings of GPLL clock for rk3328
> Hi,
>  
> Am Samstag, 22. Dezember 2018, 17:42:49 CET schrieb Katsuhiro Suzuki:
> > This patch fixes settings of GPLL frequency in fractional mode for
> > rk3328. In this mode, FOUTVCO is calcurated by following formula:
> >   FOUTVCO = FREF * FBDIV / REFDIV + ((FREF * FRAC / REFDIV) >> 24)
> > 
> > The problem is in FREF * FRAC >> 24 term. This result always lacks
> > one from target value is specified by rate member. For example first
> > itme of rk3328_pll_frac_rate originally has
> >   - rate  : 1016064000
> >   - refdiv: 3
> >   - fbdiv : 127
> >   - frac  : 134217
> >   - FREF * FBDIV / REFDIV        = 1016000000
> >   - (FREF * FRAC / REFDIV) >> 24 = 63999
> > Thus calculated rate is 1016063999. It seems wrong.
> > 
> > If frac has 134218 (it is increased 1 from original value), second
> > term is 64000. All other items have same situation. So this patch
> > adds 1 to frac member in all items of rk3328_pll_frac_rate.
> > 
> > Signed-off-by: Katsuhiro Suzuki <katsuhiro@katsuster.net>
>  
> so while this sounds all quite right to me, I've added some Rockchip
> people that have clock experience to hopefully get an Ack on the
> change :-)
>  
> @Elaine + Finley: does this explanation and the below change look right?
>  
>  
> Thanks
> Heiko
>  
> > ---
> >  drivers/clk/rockchip/clk-rk3328.c | 12 ++++++------
> >  1 file changed, 6 insertions(+), 6 deletions(-)
> > 
> > diff --git a/drivers/clk/rockchip/clk-rk3328.c b/drivers/clk/rockchip/clk-rk3328.c
> > index faa94adb2a37..65ab5c2f48b0 100644
> > --- a/drivers/clk/rockchip/clk-rk3328.c
> > +++ b/drivers/clk/rockchip/clk-rk3328.c
> > @@ -78,17 +78,17 @@ static struct rockchip_pll_rate_table rk3328_pll_rates[] = {
> >  
> >  static struct rockchip_pll_rate_table rk3328_pll_frac_rates[] = {
> >  /* _mhz, _refdiv, _fbdiv, _postdiv1, _postdiv2, _dsmpd, _frac */
> > - RK3036_PLL_RATE(1016064000, 3, 127, 1, 1, 0, 134217),
> > + RK3036_PLL_RATE(1016064000, 3, 127, 1, 1, 0, 134218),
> >  /* vco = 1016064000 */
> > - RK3036_PLL_RATE(983040000, 24, 983, 1, 1, 0, 671088),
> > + RK3036_PLL_RATE(983040000, 24, 983, 1, 1, 0, 671089),
> >  /* vco = 983040000 */
> > - RK3036_PLL_RATE(491520000, 24, 983, 2, 1, 0, 671088),
> > + RK3036_PLL_RATE(491520000, 24, 983, 2, 1, 0, 671089),
> >  /* vco = 983040000 */
> > - RK3036_PLL_RATE(61440000, 6, 215, 7, 2, 0, 671088),
> > + RK3036_PLL_RATE(61440000, 6, 215, 7, 2, 0, 671089),
> >  /* vco = 860156000 */
> > - RK3036_PLL_RATE(56448000, 12, 451, 4, 4, 0, 9797894),
> > + RK3036_PLL_RATE(56448000, 12, 451, 4, 4, 0, 9797895),
> >  /* vco = 903168000 */
> > - RK3036_PLL_RATE(40960000, 12, 409, 4, 5, 0, 10066329),
> > + RK3036_PLL_RATE(40960000, 12, 409, 4, 5, 0, 10066330),
> >  /* vco = 819200000 */
> >  { /* sentinel */ },
> >  };
> > 
>  
>  
>  
>  
>  
>
Heiko Stübner Dec. 24, 2018, 9:18 a.m. UTC | #3
Am Samstag, 22. Dezember 2018, 17:42:49 CET schrieb Katsuhiro Suzuki:
> This patch fixes settings of GPLL frequency in fractional mode for
> rk3328. In this mode, FOUTVCO is calcurated by following formula:
>   FOUTVCO = FREF * FBDIV / REFDIV + ((FREF * FRAC / REFDIV) >> 24)
> 
> The problem is in FREF * FRAC >> 24 term. This result always lacks
> one from target value is specified by rate member. For example first
> itme of rk3328_pll_frac_rate originally has
>   - rate  : 1016064000
>   - refdiv: 3
>   - fbdiv : 127
>   - frac  : 134217
>   - FREF * FBDIV / REFDIV        = 1016000000
>   - (FREF * FRAC / REFDIV) >> 24 = 63999
> Thus calculated rate is 1016063999. It seems wrong.
> 
> If frac has 134218 (it is increased 1 from original value), second
> term is 64000. All other items have same situation. So this patch
> adds 1 to frac member in all items of rk3328_pll_frac_rate.
> 
> Signed-off-by: Katsuhiro Suzuki <katsuhiro@katsuster.net>

applied for 4.22 with Elaine's Acked-by

Thanks
Heiko
diff mbox series

Patch

diff --git a/drivers/clk/rockchip/clk-rk3328.c b/drivers/clk/rockchip/clk-rk3328.c
index faa94adb2a37..65ab5c2f48b0 100644
--- a/drivers/clk/rockchip/clk-rk3328.c
+++ b/drivers/clk/rockchip/clk-rk3328.c
@@ -78,17 +78,17 @@  static struct rockchip_pll_rate_table rk3328_pll_rates[] = {
 
 static struct rockchip_pll_rate_table rk3328_pll_frac_rates[] = {
 	/* _mhz, _refdiv, _fbdiv, _postdiv1, _postdiv2, _dsmpd, _frac */
-	RK3036_PLL_RATE(1016064000, 3, 127, 1, 1, 0, 134217),
+	RK3036_PLL_RATE(1016064000, 3, 127, 1, 1, 0, 134218),
 	/* vco = 1016064000 */
-	RK3036_PLL_RATE(983040000, 24, 983, 1, 1, 0, 671088),
+	RK3036_PLL_RATE(983040000, 24, 983, 1, 1, 0, 671089),
 	/* vco = 983040000 */
-	RK3036_PLL_RATE(491520000, 24, 983, 2, 1, 0, 671088),
+	RK3036_PLL_RATE(491520000, 24, 983, 2, 1, 0, 671089),
 	/* vco = 983040000 */
-	RK3036_PLL_RATE(61440000, 6, 215, 7, 2, 0, 671088),
+	RK3036_PLL_RATE(61440000, 6, 215, 7, 2, 0, 671089),
 	/* vco = 860156000 */
-	RK3036_PLL_RATE(56448000, 12, 451, 4, 4, 0, 9797894),
+	RK3036_PLL_RATE(56448000, 12, 451, 4, 4, 0, 9797895),
 	/* vco = 903168000 */
-	RK3036_PLL_RATE(40960000, 12, 409, 4, 5, 0, 10066329),
+	RK3036_PLL_RATE(40960000, 12, 409, 4, 5, 0, 10066330),
 	/* vco = 819200000 */
 	{ /* sentinel */ },
 };