On Tue, Mar 16, 2021 at 09:16:34PM +0800, Joerg Roedel wrote: > Hi Huang, > > On Thu, Mar 11, 2021 at 10:28:07PM +0800, Huang Rui wrote: > > diff --git a/drivers/iommu/amd/iommu.c b/drivers/iommu/amd/iommu.c > > index f0adbc48fd17..a08e885403b7 100644 > > --- a/drivers/iommu/amd/iommu.c > > +++ b/drivers/iommu/amd/iommu.c > > @@ -3862,7 +3862,7 @@ static int irq_remapping_select(struct irq_domain *d, struct irq_fwspec *fwspec, > > else if (x86_fwspec_is_hpet(fwspec)) > > devid = get_hpet_devid(fwspec->param[0]); > > > > - if (devid < 0) > > + if (devid < 0 || !amd_iommu_rlookup_table) > > return 0; > > The problem is deeper than this fix suggests. I prepared other fixes for > this particular problem. Please find them here: > Thanks for the comments. Could you please elaborate this? Do you mean while amd_iommu=off, we won't prepare the IVRS, and even needn't get all ACPI talbes. Because they are never be used and the next state will always goes into IOMMU_CMDLINE_DISABLED, am I right? Thanks, Ray > https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgit.kernel.org%2Fpub%2Fscm%2Flinux%2Fkernel%2Fgit%2Fjoro%2Flinux.git%2Flog%2F%3Fh%3Diommu-fixes&data=04%7C01%7Cray.huang%40amd.com%7Cce731dda3b444ac9a14108d8e87dbb3e%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637514974013915073%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=NVahf1tIfgno%2BNWPu8hY4DygiGWdKXBJ8G6OsD%2BHC14%3D&reserved=0 > > Regards, > > Joerg
