Re: Build warnings in Xen 5.15.y and 5.10.y with retbleed backports

[Boris Ostrovsky]

On 7/17/22 1:20 AM, Juergen Gross wrote:
> What about filling the complete hypercall page just with "int 3" or "ud2"?
>
> Any try to do a hypercall before the hypercall page has been initialized
> is a bug anyway. What good can come from calling a function which will
> return a basically random value instead of doing a privileged operation?

This is all about objtool, that's why 'ret' was added there originally by f4b4bc10b0b8 ("x86/xen: Support objtool vmlinux.o validation in xen-head.S").


Before that it was all 'nop' which is similar to what you are suggesting ('int3' or 'ud2' would of course be better)


-boris