[PATCH] clocksource/drivers/timer-riscv: Drop extra CSR write

Anup Patel apatel at ventanamicro.com
Wed Mar 13 09:56:34 PDT 2024


On Wed, Mar 13, 2024 at 1:03 AM Samuel Holland
<samuel.holland at sifive.com> wrote:
>
> On riscv32, the time comparator value is split across two CSRs. We write
> both when stopping the timer, but realistically the time is just as
> unlikely to reach 0xffffffff00000000 as 0xffffffffffffffff, so there is
> no need to write the low CSR.

Even though unlikely, there is still a theoretical possibility of
counter reaching value 0xffffffff00000000.

The good thing about value 0xffffffffffffffff is that the counter will
immediately wrap around after reaching it.

Regards,
Anup


>
> Signed-off-by: Samuel Holland <samuel.holland at sifive.com>
> ---
>
>  drivers/clocksource/timer-riscv.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/clocksource/timer-riscv.c b/drivers/clocksource/timer-riscv.c
> index e66dcbd66566..eaaf01f3c34b 100644
> --- a/drivers/clocksource/timer-riscv.c
> +++ b/drivers/clocksource/timer-riscv.c
> @@ -35,9 +35,10 @@ static bool riscv_timer_cannot_wake_cpu;
>  static void riscv_clock_event_stop(void)
>  {
>         if (static_branch_likely(&riscv_sstc_available)) {
> -               csr_write(CSR_STIMECMP, ULONG_MAX);
>                 if (IS_ENABLED(CONFIG_32BIT))
>                         csr_write(CSR_STIMECMPH, ULONG_MAX);
> +               else
> +                       csr_write(CSR_STIMECMP, ULONG_MAX);
>         } else {
>                 sbi_set_timer(U64_MAX);
>         }
> --
> 2.43.1
>
>



More information about the linux-riscv mailing list