Re: [PATCH v12 06/12] iommu/arm-smmu-v3: Disable the EVTQ and the PRIQ in a kdump kernel
From: Nicolin Chen
Date: Mon Nov 16 2026 - 14:08:33 EST
On Mon, Nov 16, 2026 at 11:08:49AM +0000, John Garry wrote:
> On 11/2/26 05:19, Nicolin Chen wrote:
> > A kdump kernel cannot use either queue.
>
> I'd say a "kdump kernel cannot use either queue without potential problems".
OK.
> > @@ -5573,6 +5570,15 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
> > if (arm_smmu_sva_supported(smmu))
> > smmu->features |= ARM_SMMU_FEAT_SVA;
> > + /*
> > + * A kdump kernel wants neither queue: the crashed kernel's CDs and page
> > + * tables might be corrupted, spamming events, and page requests cannot
> > + * be served. A disabled queue discards new records without raising any
> > + * global error.
> > + */
> > + if (is_kdump_kernel())
> > + smmu->features &= ~(ARM_SMMU_FEAT_EVTQ | ARM_SMMU_FEAT_PRI);
>
> It's less than ideal that we unconditionally set ARM_SMMU_FEAT_EVTQ at the
> beginning of this function, and then here conditionally unset it. Can we
> just conditionally set it once?
I see the opposite. Conditionally setting would need two pieces of
comments. It's neater to have it here right before:
if (is_kdump_kernel())
arm_smmu_device_kdump_probe()
Nicolin