On Tuesday 05 September 2017 02:55 PM, Jan Kiszka wrote: > On 2017-09-05 10:59, Lokesh Vutla wrote: >> >> >> On Tuesday 05 September 2017 11:56 AM, Jan Kiszka wrote: >>> On 2017-08-30 12:34, 'Lokesh Vutla' via Jailhouse wrote: >>>> Even though 'struct sgi' already supports for passing affinity levels, >>>> gic_handle_sgir_write() looks only for target fields and triggers sgis >>>> to its respective targets. This will fail in case of armv8 with affinity >>>> routing enabled. So parse all the affinity levels in sgi before sending >>>> sgi. >>>> >>>> Suggested-by: Nikhil Devshatwar <[email protected]> >>>> Signed-off-by: Nikhil Devshatwar <[email protected]> >>>> Signed-off-by: Lokesh Vutla <[email protected]> >>>> --- >>>> hypervisor/arch/arm-common/irqchip.c | 14 ++++++++++++-- >>>> 1 file changed, 12 insertions(+), 2 deletions(-) >>>> >>>> diff --git a/hypervisor/arch/arm-common/irqchip.c >>>> b/hypervisor/arch/arm-common/irqchip.c >>>> index 2019342..64c081d 100644 >>>> --- a/hypervisor/arch/arm-common/irqchip.c >>>> +++ b/hypervisor/arch/arm-common/irqchip.c >>>> @@ -132,6 +132,7 @@ void gic_handle_sgir_write(struct sgi *sgi, bool >>>> virt_input) >>>> struct per_cpu *cpu_data = this_cpu_data(); >>>> unsigned long targets = sgi->targets; >>>> unsigned int cpu; >>>> + u64 mpidr, clst, sgi_clst, core; >>> >>> Let's write this out: cluster. >> >> okay. >> >>> >>>> >>>> if (sgi->routing_mode == 2) { >>>> /* Route to the caller itself */ >>>> @@ -139,14 +140,23 @@ void gic_handle_sgir_write(struct sgi *sgi, bool >>>> virt_input) >>>> sgi->targets = (1 << cpu_data->cpu_id); >>>> } else { >>>> sgi->targets = 0; >>>> + sgi_clst = (u64)sgi->aff3 << MPIDR_LEVEL_SHIFT(3) | >>>> + (u64)sgi->aff2 << MPIDR_LEVEL_SHIFT(2) | >>>> + (u64)sgi->aff1 << MPIDR_LEVEL_SHIFT(1); >>> >>> I wonder if we shouldn't refactor the sgi structure and encode the >>> cluster information in it via a single field that is encoded like mpidr. >>> That may save us from juggling with the aff1..3 fields all the time. >> >> After think a bit, I do not see any issue replacing aff[123] with >> cluster_id and store the value as "mpidr && ~0xffUL"? Will try to make >> changes and repost the entire series. > > Better use MPIDR_CPUID_MASK - it filters the right bits per architecture.
This MASK include aff0 value as well. I guess the idea here is to just pass cluster id and aff0 value gets passed via targets field. Thanks and regards, Lokesh -- You received this message because you are subscribed to the Google Groups "Jailhouse" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
