Re: [PATCH v3 1/4] clk: qcom: clk-alpha-pll: fix Zonda L value rounding

[Konrad Dybcio]

On 10/27/26 7:08 PM, Taniya Das wrote:
> zonda_pll_adjust_l_val() obtains the integer quotient of the requested
> rate and parent rate in quotient, but writes the requested rate to L_VAL
> when the fractional alpha value requires rounding.  This programs a
> nonsensical integer divider.  For example, a 1.2 GHz rate with a 19.2 MHz
> parent requires L = 63, but programs L = 1200000001.
> 
> Use quotient when calculating the rounded L value.
> 
> A GPT-assisted review reported this issue.
> 
> Fixes: a4d89b11aca3 ("clk: qcom: clk-alpha-pll: Simplify the zonda_pll_adjust_l_val()")
> Cc: stable@xxxxxxxxxxxxxxx
> Reported-by: Konrad Dybcio <konrad.dybcio@xxxxxxxxxxxxxxxx>
> Link: https://lore.kernel.org/9dc66fa5-47a9-417b-b8ff-b3a61ea75678@xxxxxxxxxxxxxxxx
> Signed-off-by: Taniya Das <taniya.das@xxxxxxxxxxxxxxxx>
> ---

Reviewed-by: Konrad Dybcio <konrad.dybcio@xxxxxxxxxxxxxxxx>

Konrad