[PATCH 2/7] riscv: Match memory barriers between send_ipi_many and handle_ipi

Bin Meng bmeng.cn at gmail.com
Fri Sep 11 09:45:05 CEST 2020


On Tue, Sep 8, 2020 at 2:17 AM Sean Anderson <seanga2 at gmail.com> wrote:
>
> Without a matching barrier on the write side, the barrier in handle_ipi
> does nothing. It was entirely possible for the boot hart to write to addr,
> arg0, and arg1 *after* sending the IPI, because there was no barrier on the
> sending side.
>
> Fixes: 90ae28143700bae4edd23930a7772899ad259058

nits: wrong format of Fixes tag

> Signed-off-by: Sean Anderson <seanga2 at gmail.com>
> ---
>
>  arch/riscv/lib/smp.c | 2 ++
>  1 file changed, 2 insertions(+)
>
> diff --git a/arch/riscv/lib/smp.c b/arch/riscv/lib/smp.c
> index ac22136314..ab6d8bd7fa 100644
> --- a/arch/riscv/lib/smp.c
> +++ b/arch/riscv/lib/smp.c
> @@ -54,6 +54,8 @@ static int send_ipi_many(struct ipi_data *ipi, int wait)
>                 gd->arch.ipi[reg].arg0 = ipi->arg0;
>                 gd->arch.ipi[reg].arg1 = ipi->arg1;
>
> +               __smp_mb();
> +
>                 ret = riscv_send_ipi(reg);
>                 if (ret) {
>                         pr_err("Cannot send IPI to hart %d\n", reg);

Reviewed-by: Bin Meng <bin.meng at windriver.com>

Regards,
Bin


More information about the U-Boot mailing list