Re: SFENCE.VMA/SINVAL.VMA and page faults

Andrew Waterman

The spec mentions that implementations are permitted to ignore rs1 altogether, so trapping is certainly not required.

On Tue, Feb 7, 2023 at 6:24 AM Ved Shanbhogue <ved@...> wrote:
Greetings !

I understand that the SFENCE.VMA/SINVAL.VMA instructions have unspecified behavior if a non-canonical form virtual address is specified and specifically no page fault is expected. I understand similar unspecified behavior applies HFENCE.GVMA if a too wide GPA is specified and no guest page fault is expected.

Is this correct? 


Join { to automatically receive all group messages.