On Wed, Mar 18, 2020 at 01:04:20PM +0530, Mahesh Salgaonkar wrote: > Per PAPR, it is expected to set effective address provided flag in > sub_err_type member of mc extended error log (i.e > rtas_event_log_v6_mc.sub_err_type). This somehow got missed in original > fwnmi-mce patch series. The current code just updates the effective address > but does not set the flag to indicate that it is available. Hence guest > fails to extract effective address from mce rtas log. This patch fixes > that. > > Without this patch guest MCE logs fails print DAR value: > > [ 11.933608] Disabling lock debugging due to kernel taint > [ 11.933773] MCE: CPU0: machine check (Severe) Host TLB Multihit [Recovered] > [ 11.933979] MCE: CPU0: NIP: [c000000000090b34] > radix__flush_tlb_range_psize+0x194/0xf00 > [ 11.934223] MCE: CPU0: Initiator CPU > [ 11.934341] MCE: CPU0: Unknown > > After the change: > > [ 22.454149] Disabling lock debugging due to kernel taint > [ 22.454316] MCE: CPU0: machine check (Severe) Host TLB Multihit DAR: > deadbeefdeadbeef [Recovered] > [ 22.454605] MCE: CPU0: NIP: [c0000000003e5804] kmem_cache_alloc+0x84/0x330 > [ 22.454820] MCE: CPU0: Initiator CPU > [ 22.454944] MCE: CPU0: Unknown > > > Signed-off-by: Mahesh Salgaonkar <mah...@linux.ibm.com>
Applied to ppc-for-5.0, thanks.
> ---
> Change in v2:
> - Fixed coding style issues.
> ---
> hw/ppc/spapr_events.c | 26 ++++++++++++++++++++++++++
> 1 file changed, 26 insertions(+)
>
> diff --git a/hw/ppc/spapr_events.c b/hw/ppc/spapr_events.c
> index 8b32b7eea5..cb6bfedc53 100644
> --- a/hw/ppc/spapr_events.c
> +++ b/hw/ppc/spapr_events.c
> @@ -243,6 +243,14 @@ struct rtas_event_log_v6_mc {
> #define RTAS_LOG_V6_MC_TLB_PARITY 1
> #define RTAS_LOG_V6_MC_TLB_MULTIHIT 2
> #define RTAS_LOG_V6_MC_TLB_INDETERMINATE 3
> +/*
> + * Per PAPR,
> + * For UE error type, set bit 1 of sub_err_type to indicate effective addr is
> + * provided. For other error types (SLB/ERAT/TLB), set bit 0 to indicate
> + * same.
> + */
> +#define RTAS_LOG_V6_MC_UE_EA_ADDR_PROVIDED 0x40
> +#define RTAS_LOG_V6_MC_EA_ADDR_PROVIDED 0x80
> uint8_t reserved_1[6];
> uint64_t effective_address;
> uint64_t logical_address;
> @@ -726,6 +734,22 @@ void
> spapr_hotplug_req_remove_by_count_indexed(SpaprDrcType drc_type,
> RTAS_LOG_V6_HP_ACTION_REMOVE, drc_type, &drc_id);
> }
>
> +static void spapr_mc_set_ea_provided_flag(struct mc_extended_log *ext_elog)
> +{
> + switch (ext_elog->mc.error_type) {
> + case RTAS_LOG_V6_MC_TYPE_UE:
> + ext_elog->mc.sub_err_type |= RTAS_LOG_V6_MC_UE_EA_ADDR_PROVIDED;
> + break;
> + case RTAS_LOG_V6_MC_TYPE_SLB:
> + case RTAS_LOG_V6_MC_TYPE_ERAT:
> + case RTAS_LOG_V6_MC_TYPE_TLB:
> + ext_elog->mc.sub_err_type |= RTAS_LOG_V6_MC_EA_ADDR_PROVIDED;
> + break;
> + default:
> + break;
> + }
> +}
> +
> static uint32_t spapr_mce_get_elog_type(PowerPCCPU *cpu, bool recovered,
> struct mc_extended_log *ext_elog)
> {
> @@ -751,6 +775,7 @@ static uint32_t spapr_mce_get_elog_type(PowerPCCPU *cpu,
> bool recovered,
> ext_elog->mc.sub_err_type = mc_derror_table[i].error_subtype;
> if (mc_derror_table[i].dar_valid) {
> ext_elog->mc.effective_address =
> cpu_to_be64(env->spr[SPR_DAR]);
> + spapr_mc_set_ea_provided_flag(ext_elog);
> }
>
> summary |= mc_derror_table[i].initiator
> @@ -769,6 +794,7 @@ static uint32_t spapr_mce_get_elog_type(PowerPCCPU *cpu,
> bool recovered,
> ext_elog->mc.sub_err_type = mc_ierror_table[i].error_subtype;
> if (mc_ierror_table[i].nip_valid) {
> ext_elog->mc.effective_address = cpu_to_be64(env->nip);
> + spapr_mc_set_ea_provided_flag(ext_elog);
> }
>
> summary |= mc_ierror_table[i].initiator
>
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
signature.asc
Description: PGP signature
