[RFC PATCH 8/9] lib:sbi: Reset the mhpmevent value upon counter reset
Anup Patel
anup at brainfault.org
Thu Sep 23 01:46:00 PDT 2021
On Fri, Sep 10, 2021 at 2:10 AM Atish Patra <atish.patra at wdc.com> wrote:
Please use "lib: sbi:" instead of "lib:sbi:" as patch subject prefix.
>
> The hardware solely relies on the event selector value in mhpmevent
> to figure out what event to monitor using that counter. It should be
> reset when counter reset happens.
>
> Signed-off-by: Atish Patra <atish.patra at wdc.com>
Otherwise, it looks good to me.
Reviewed-by: Anup Patel <anup.patel at wdc.com>
Regards,
Anup
> ---
> lib/sbi/sbi_pmu.c | 18 +++++++++++++++++-
> 1 file changed, 17 insertions(+), 1 deletion(-)
>
> diff --git a/lib/sbi/sbi_pmu.c b/lib/sbi/sbi_pmu.c
> index 20f78c75200e..f33842052a46 100644
> --- a/lib/sbi/sbi_pmu.c
> +++ b/lib/sbi/sbi_pmu.c
> @@ -368,6 +368,20 @@ static int pmu_ctr_stop_fw(uint32_t cidx, uint32_t fw_evt_code)
> return 0;
> }
>
> +static int pmu_reset_hw_mhpmevent(int ctr_idx)
> +{
> + if (ctr_idx < 3 || ctr_idx >= SBI_PMU_HW_CTR_MAX)
> + return SBI_EFAIL;
> +#if __riscv_xlen == 32
> + csr_write_num(CSR_MCOUNTINHIBIT + ctr_idx, 0);
> + csr_write_num(CSR_MHPMEVENT3H + ctr_idx - 3, 0);
> +#else
> + csr_write_num(CSR_MCOUNTINHIBIT + ctr_idx, 0);
> +#endif
> +
> + return 0;
> +}
> +
> int sbi_pmu_ctr_stop(unsigned long cbase, unsigned long cmask,
> unsigned long flag)
> {
> @@ -391,8 +405,10 @@ int sbi_pmu_ctr_stop(unsigned long cbase, unsigned long cmask,
> else
> ret = pmu_ctr_stop_hw(cbase);
>
> - if (!ret && (flag & SBI_PMU_STOP_FLAG_RESET))
> + if (flag & SBI_PMU_STOP_FLAG_RESET) {
> active_events[hartid][cbase] = SBI_PMU_EVENT_IDX_INVALID;
> + pmu_reset_hw_mhpmevent(cbase);
> + }
> }
>
> return ret;
> --
> 2.31.1
>
>
> --
> opensbi mailing list
> opensbi at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/opensbi
More information about the opensbi
mailing list