Re: [PATCH] KVM: PPC: Align pt_regs in kvm_vcpu_arch structure
From: Murilo Opsfelder Araújo <hidden>
Date: 2022-06-17 15:25:33
Hi, Fabiano. On 5/25/22 09:49, Fabiano Rosas wrote:
The H_ENTER_NESTED hypercall receives as second parameter the address of a region of memory containing the values for the nested guest privileged registers. We currently use the pt_regs structure contained within kvm_vcpu_arch for that end. Most hypercalls that receive a memory address expect that region to not cross a 4k page boundary. We would want H_ENTER_NESTED to follow the same pattern so this patch ensures the pt_regs structure sits within a page. Signed-off-by: Fabiano Rosas <redacted>
Is it necessary to explain in the commit message that even though the second parameter needs to be 4k-aligned, we're aligning pt_regs to 512 bytes so it can be placed within a 4k boundary because its size is below 512 bytes? The natural thinking would be aligning it to 4k bytes, which would punch a huge hole in kvm_vcpu_arch. I think having the explanation of why 512 vs. 4k is worthwhile mentioning.
quoted hunk ↗ jump to hunk
--- arch/powerpc/include/asm/kvm_host.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)diff --git a/arch/powerpc/include/asm/kvm_host.h b/arch/powerpc/include/asm/kvm_host.h index faf301d0dec0..87eba60f2920 100644 --- a/arch/powerpc/include/asm/kvm_host.h +++ b/arch/powerpc/include/asm/kvm_host.h@@ -519,7 +519,11 @@ struct kvm_vcpu_arch { struct kvmppc_book3s_shadow_vcpu *shadow_vcpu; #endif - struct pt_regs regs; + /* + * This is passed along to the HV via H_ENTER_NESTED. Align to + * prevent it crossing a real 4K page. + */ + struct pt_regs regs __aligned(512); struct thread_fp_state fp;
-- Murilo