On 2/11/21 10:59 PM, Aneesh Kumar K.V wrote: > A read syscall do fail with EFAULT. But we allow read via io_uring > syscalls. Is that ok? In short, yes. As much as I'd like to apply pkey permissions to all accesses, when we don't have the CPU registers around, we don't have a choice: we have to let the access through. The same basic thing is done for accesses via the IOMMU and for things like ptrace() where the ptracer's registers don't have anything to do with the ptracee's address space. We could *probably* be a bit pickier at io_uring_submit() time. But, I'm not sure it's worth it.
