> > +_GLOBAL(__setup_cpu_power9)
> > + mflr r11
> > + bl __init_FSCR
> > + bl __init_PMU
> Just to keep in mind, I am not sure whether
> powerisa 3.0 support MMCRS spr, so we
> will need a feature check in __init_PMU()
> for power9.
Yeah, I'm not expecting this to work.
I'm trying to lay down a common base we can start working on. There
are lots of people working a bunch different bases. I want to avoid
that and we can do that by upstreaming.
> > + bl __init_hvmode_206
> > + mtlr r11
> > + beqlr
> > + li r0,0
> > + mtspr SPRN_LPID,r0
> > + mfspr r3,SPRN_LPCR
> > + ori r3, r3, LPCR_PECEDH
> > + bl __init_LPCR
> > + bl __init_HFSCR
> > + bl __init_tlb_power9
> > + bl __init_PMU_HV
>
> Again, need to check whether powerisa 3.0 support MMCRH spr
> which is used in __init_PMU_HV()
Same here.
> > + { /* Hacked up Power9 */
/me reviews his own patch...
Oops
> > + .pvr_mask = 0xffff0000,
> > + .pvr_value = 0x004e0000,
> > + .cpu_name = "POWER9 (raw)",
> > + .cpu_features = CPU_FTRS_POWER9,
> > + .cpu_user_features = COMMON_USER_POWER9,
> > + .cpu_user_features2 = COMMON_USER2_POWER9,
> > + .mmu_features = MMU_FTRS_POWER9,
> > + .icache_bsize = 128,
> > + .dcache_bsize = 128,
> > + .num_pmcs = 6,
> > + .pmc_type = PPC_PMC_IBM,
> > + .oprofile_cpu_type = "ppc64/power8",
>
> This should be ppc64/power9. We use "oprofile_cpu_type" in PMU init.
Yep, we can fix that up when we post PMU patches, but if I repost I'll
change so it doesn't match with old one.
Mikey
_______________________________________________
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
