Re: [PATCH] ppc/pnv: Fix NMI system reset SRR1 value

[Cédric Le Goater]

On 5/7/20 1:48 PM, Nicholas Piggin wrote:
> Commit a77fed5bd926 ("ppc/pnv: Add support for NMI interface") got the
> SRR1 setting wrong for sresets that hit outside of power-save states.
> 
> Fix this, better documenting the source for the bit definitions.
> 
> Fixes: a77fed5bd926 ("ppc/pnv: Add support for NMI interface") got the
> Cc: Cédric Le Goater <address@hidden>
> Cc: David Gibson <address@hidden>
> Signed-off-by: Nicholas Piggin <address@hidden>

We should introduce some defines like the SRR1_WAKE ones in Linux and 
cleanup powerpc_reset_wakeup(). This function uses cryptic values. 
That can be done later on as a followup.

Reviewed-by: Cédric Le Goater <address@hidden>

> ---
> 
> Thanks to Cedric for pointing out concerns with a previous MCE patch
> that unearthed this as well. Linux does not actually care what these
> SRR1[42:45] bits look like for non-powersave sresets, but we should
> follow documented behaviour as far as possible.

We should introduce some defines like the SRR1_WAKE ones in Linux and 
cleanup powerpc_reset_wakeup(). This function uses cryptic values. 
That can be done later on as a followup.


I am currently after a bug which results in a CPU hard lockup because 
of a pending interrupt. It occurs on a SMP PowerNV machine when it is 
stressed with IO, such as scp of a big file. 

I am suspecting more and more an issue with an interrupt being handled 
when the CPU is coming out of idle. I haven't seen anything wrong in
the models. Unless this maybe :

    /* Pretend to be returning from doze always as we don't lose state */
    *msr |= (0x1ull << (63 - 47));

I am not sure how in sync it is with PSSCR.


Thanks, 

C.

>  hw/ppc/pnv.c | 26 ++++++++++++++++++++------
>  1 file changed, 20 insertions(+), 6 deletions(-)
> 
> diff --git a/hw/ppc/pnv.c b/hw/ppc/pnv.c
> index a3b7a8d0ff..1b4748ce6d 100644
> --- a/hw/ppc/pnv.c
> +++ b/hw/ppc/pnv.c
> @@ -1986,12 +1986,26 @@ static void pnv_cpu_do_nmi_on_cpu(CPUState *cs, 
> run_on_cpu_data arg)
>  
>      cpu_synchronize_state(cs);
>      ppc_cpu_do_system_reset(cs);
> -    /*
> -     * SRR1[42:45] is set to 0100 which the ISA defines as implementation
> -     * dependent. POWER processors use this for xscom triggered interrupts,
> -     * which come from the BMC or NMI IPIs.
> -     */
> -    env->spr[SPR_SRR1] |= PPC_BIT(43);
> +    if (env->spr[SPR_SRR1] & PPC_BITMASK(46, 47)) {
> +        /*
> +      * Power-save wakeups, as indicated by non-zero SRR1[46:47] put the
> +      * wakeup reason in SRR1[42:45], system reset is indicated with 0b0100
> +      * (PPC_BIT(43)).
> +      */
> +        if (!(env->spr[SPR_SRR1] & PPC_BIT(43))) {
> +            warn_report("ppc_cpu_do_system_reset does not set system reset 
> wakeup reason");
> +            env->spr[SPR_SRR1] |= PPC_BIT(43);
> +        }
> +    } else {
> +        /*
> +      * For non-powersave system resets, SRR1[42:45] are defined to be
> +      * implementation-dependent. The POWER9 User Manual specifies that
> +      * an external (SCOM driven, which may come from a BMC nmi command or
> +      * another CPU requesting a NMI IPI) system reset exception should be
> +      * 0b0010 (PPC_BIT(44)).
> +         */
> +        env->spr[SPR_SRR1] |= PPC_BIT(44);
> +    }
>  }
>  
>  static void pnv_nmi(NMIState *n, int cpu_index, Error **errp)
>