Re: [PATCH 3/4] x86: save user rsp in pt_regs->sp on SYSCALL64 fastpath

[Andy Lutomirski <luto@amacapital.net>]

On Mon, Mar 9, 2015 at 1:32 PM, Denys Vlasenko <vda.linux@googlemail.com> wrote:
> On Mon, Mar 9, 2015 at 9:11 PM, Andy Lutomirski <luto@amacapital.net> wrote:
>>> @@ -253,11 +247,13 @@ GLOBAL(system_call_after_swapgs)
>>>          */
>>>         ENABLE_INTERRUPTS(CLBR_NONE)
>>>         ALLOC_PT_GPREGS_ON_STACK 8              /* +8: space for orig_ax */
>>> +       movq    %rcx,RIP(%rsp)
>>> +       movq    PER_CPU_VAR(old_rsp),%rcx
>>> +       movq    %r11,EFLAGS(%rsp)
>>> +       movq    %rcx,RSP(%rsp)
>>> +       movq_cfi rax,ORIG_RAX
>>>         SAVE_C_REGS_EXCEPT_RAX_RCX_R11
>>>         movq    $-ENOSYS,RAX(%rsp)
>>> -       movq_cfi rax,ORIG_RAX
>>> -       movq    %r11,EFLAGS(%rsp)
>>> -       movq    %rcx,RIP(%rsp)
>>
>> Why the reordering?
>
> No strong reason.
>
> iret stack is "above" the rest of pt_regs.
>
> This does not matter now, but when/if we convert to PUSHes
> for register saving, pushes which build iret frame
> will have to be before "save C-clobbered registers" part,
> exactly as in this patch.

Fair enough.

Acked-by: Andy Lutomirski <luto@amacapital.net>

-- 
Andy Lutomirski
AMA Capital Management, LLC