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.
Note: the pt_regs structure is currently 384 bytes in size, so
aligning to 512 is sufficient to ensure it will not cross a 4K page
and avoids punching too big a hole in struct kvm_vcpu_arch.
Signed-off-by: Fabiano Rosas <farosas@xxxxxxxxxxxxx>
Signed-off-by: Murilo Opsfelder Araújo <muriloo@xxxxxxxxxxxxx>
---
v2:
- updated commit message to inform the rationale for aligning to 512;
- added Murilo's sign-off which I had forgotten, we worked on this
together.
---
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 2909a88acd16..2c7219cef4ec 100644
--- a/arch/powerpc/include/asm/kvm_host.h
+++ b/arch/powerpc/include/asm/kvm_host.h
@@ -523,7 +523,11 @@ 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;
--
2.35.3
