"div4" DIVs which divide PLLs by 4 are actually dividing "div2" DIVs by
2 to achieve a by 4 division, thus their parents are the respective
"div2" DIVs. These DIVs were mistakenly set to have the PLLs as parents.
This leads to the kernel thinking "div4"s and everything under them run
at 2x the clock speed. Fix this.
Fixes: 45bd8166a1d8 ("clk: samsung: Add initial Exynos7885 clock driver")
Signed-off-by: David Virag <[email protected]>
---
drivers/clk/samsung/clk-exynos7885.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/clk/samsung/clk-exynos7885.c b/drivers/clk/samsung/clk-exynos7885.c
index a7b106302706..368c50badd15 100644
--- a/drivers/clk/samsung/clk-exynos7885.c
+++ b/drivers/clk/samsung/clk-exynos7885.c
@@ -182,7 +182,7 @@ static const struct samsung_div_clock top_div_clks[] __initconst = {
CLK_CON_DIV_PLL_SHARED0_DIV2, 0, 1),
DIV(CLK_DOUT_SHARED0_DIV3, "dout_shared0_div3", "fout_shared0_pll",
CLK_CON_DIV_PLL_SHARED0_DIV3, 0, 2),
- DIV(CLK_DOUT_SHARED0_DIV4, "dout_shared0_div4", "fout_shared0_pll",
+ DIV(CLK_DOUT_SHARED0_DIV4, "dout_shared0_div4", "dout_shared0_div2",
CLK_CON_DIV_PLL_SHARED0_DIV4, 0, 1),
DIV(CLK_DOUT_SHARED0_DIV5, "dout_shared0_div5", "fout_shared0_pll",
CLK_CON_DIV_PLL_SHARED0_DIV5, 0, 3),
@@ -190,7 +190,7 @@ static const struct samsung_div_clock top_div_clks[] __initconst = {
CLK_CON_DIV_PLL_SHARED1_DIV2, 0, 1),
DIV(CLK_DOUT_SHARED1_DIV3, "dout_shared1_div3", "fout_shared1_pll",
CLK_CON_DIV_PLL_SHARED1_DIV3, 0, 2),
- DIV(CLK_DOUT_SHARED1_DIV4, "dout_shared1_div4", "fout_shared1_pll",
+ DIV(CLK_DOUT_SHARED1_DIV4, "dout_shared1_div4", "dout_shared1_div2",
CLK_CON_DIV_PLL_SHARED1_DIV4, 0, 1),
/* CORE */
--
2.35.1
On 26/05/2022 07:58, David Virag wrote:
> "div4" DIVs which divide PLLs by 4 are actually dividing "div2" DIVs by
> 2 to achieve a by 4 division, thus their parents are the respective
> "div2" DIVs. These DIVs were mistakenly set to have the PLLs as parents.
> This leads to the kernel thinking "div4"s and everything under them run
> at 2x the clock speed. Fix this.
>
> Fixes: 45bd8166a1d8 ("clk: samsung: Add initial Exynos7885 clock driver")
> Signed-off-by: David Virag <[email protected]>
> ---
> drivers/clk/samsung/clk-exynos7885.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
Acked-by: Krzysztof Kozlowski <[email protected]>
Best regards,
Krzysztof
On 26/05/2022 09:21, Krzysztof Kozlowski wrote:
> On 26/05/2022 07:58, David Virag wrote:
>> "div4" DIVs which divide PLLs by 4 are actually dividing "div2" DIVs by
>> 2 to achieve a by 4 division, thus their parents are the respective
>> "div2" DIVs. These DIVs were mistakenly set to have the PLLs as parents.
>> This leads to the kernel thinking "div4"s and everything under them run
>> at 2x the clock speed. Fix this.
>>
>> Fixes: 45bd8166a1d8 ("clk: samsung: Add initial Exynos7885 clock driver")
>> Signed-off-by: David Virag <[email protected]>
>> ---
>> drivers/clk/samsung/clk-exynos7885.c | 4 ++--
>> 1 file changed, 2 insertions(+), 2 deletions(-)
Sylwester,
This goes to v5.20?
Best regards,
Krzysztof
On Fri, 2022-06-24 at 09:59 +0200, Krzysztof Kozlowski wrote:
> On 26/05/2022 09:21, Krzysztof Kozlowski wrote:
> > On 26/05/2022 07:58, David Virag wrote:
> > > "div4" DIVs which divide PLLs by 4 are actually dividing "div2"
> > > DIVs by
> > > 2 to achieve a by 4 division, thus their parents are the
> > > respective
> > > "div2" DIVs. These DIVs were mistakenly set to have the PLLs as
> > > parents.
> > > This leads to the kernel thinking "div4"s and everything under
> > > them run
> > > at 2x the clock speed. Fix this.
> > >
> > > Fixes: 45bd8166a1d8 ("clk: samsung: Add initial Exynos7885 clock
> > > driver")
> > > Signed-off-by: David Virag <[email protected]>
> > > ---
> > > drivers/clk/samsung/clk-exynos7885.c | 4 ++--
> > > 1 file changed, 2 insertions(+), 2 deletions(-)
>
> Sylwester,
>
> This goes to v5.20?
>
> Best regards,
> Krzysztof
Hi Krzysztof and Sylwester,
What is going on with this patch? What will happen to it? From what
I've seen, Sylwester doesn't seem to be too active lately. I just don't
want it to get lost. With only one of the patches applied, UART is
partially broken on 7885. Don't want to make unnecessary noise, but
this patch should probably be applied sooner than later.
Best regards,
David