Re: [patch v3 34/36] x86/smpboot: Implement a bit spinlock to protect the realmode stack

[Peter Zijlstra <peterz@xxxxxxxxxxxxx>]

On Mon, May 08, 2023 at 09:44:22PM +0200, Thomas Gleixner wrote:

> @@ -252,6 +252,17 @@ SYM_INNER_LABEL(secondary_startup_64_no_
>       movq    TASK_threadsp(%rax), %rsp
>  
>       /*
> +      * Now that this CPU is running on its own stack, drop the realmode
> +      * protection. For the boot CPU the pointer is NULL!
> +      */
> +     movq    trampoline_lock(%rip), %rax
        movl    $0, (%rax)

> +.Lsetup_gdt:
> +     /*
>        * We must switch to a new descriptor in kernel space for the GDT
>        * because soon the kernel won't have access anymore to the userspace
>        * addresses where we're currently running on. We have to do that here

> --- a/arch/x86/realmode/rm/trampoline_64.S
> +++ b/arch/x86/realmode/rm/trampoline_64.S
> @@ -37,6 +37,24 @@
>       .text
>       .code16
>  
> +.macro LOAD_REALMODE_ESP
> +     /*
> +      * Make sure only one CPU fiddles with the realmode stack
> +      */
> +.Llock_rm\@:
> +     btl     $0, tr_lock
> +     jnc     2f
> +     pause
> +     jmp     .Llock_rm\@
> +2:
> +     lock
> +     btsl    $0, tr_lock
> +     jc      .Llock_rm\@

Do we really care about performance here; or should we pick the simpler
form? Also, 'lock' is a prefix, not an instruction.

.Llock_rm\@:
        lock btsl       $0, tr_lock;
        jnc             2f
        pause
        jmp             .Llock_rm\@
2:

> +
> +     # Setup stack
> +     movl    $rm_stack_end, %esp
> +.endm
> +
>       .balign PAGE_SIZE