Re: [PATCH -tip v8 13/13] x86/kprobes: Fixup return address in generic trampoline handler

[Masami Hiramatsu <mhiramat@xxxxxxxxxx>]

On Mon, 5 Jul 2021 10:34:58 +0200
Ingo Molnar <mingo@xxxxxxxxxx> wrote:

> 
> * Masami Hiramatsu <mhiramat@xxxxxxxxxx> wrote:
> 
> > In x86, kretprobe trampoline address on the stack frame will
> > be replaced with the real return address after returning from
> > trampoline_handler. Before fixing the return address, the real
> > return address can be found in the current->kretprobe_instances.
> > 
> > However, since there is a window between updating the
> > current->kretprobe_instances and fixing the address on the stack,
> > if an interrupt caused at that timing and the interrupt handler
> > does stacktrace, it may fail to unwind because it can not get
> > the correct return address from current->kretprobe_instances.
> > 
> > This will minimize that window by fixing the return address
> > right before updating current->kretprobe_instances.
> 
> Is there still a window? I.e. is it "minimized" (to how big of a window?), 
> or eliminated?

Oh, this will eliminate the window, because the return address is
fixed before updating the 'current->kretprobe_instance'.


> 
> > +void arch_kretprobe_fixup_return(struct pt_regs *regs,
> > +				 unsigned long correct_ret_addr)
> > +{
> > +	unsigned long *frame_pointer;
> > +
> > +	frame_pointer = ((unsigned long *)&regs->sp) + 1;
> > +
> > +	/* Replace fake return address with real one. */
> > +	*frame_pointer = correct_ret_addr;
> 
> Firstly, why does &regs->sp have to be forced to 'unsigned long *'? 
> 
> pt_regs::sp is 'unsigned long' on both 32-bit and 64-bit kernels AFAICS.

Ah, right.

> 
> Secondly, the new code modified by your patch now looks like this:
> 
>         frame_pointer = ((unsigned long *)&regs->sp) + 1;
>  
> +       kretprobe_trampoline_handler(regs, frame_pointer);
> 
> where:
> 
> +void arch_kretprobe_fixup_return(struct pt_regs *regs,
> +                                unsigned long correct_ret_addr)
> +{
> +       unsigned long *frame_pointer;
> +
> +       frame_pointer = ((unsigned long *)&regs->sp) + 1;
> +
> +       /* Replace fake return address with real one. */
> +       *frame_pointer = correct_ret_addr;
> +}
> 
> So we first do:
> 
>         frame_pointer = ((unsigned long *)&regs->sp) + 1;
> 
> ... and pass that in to arch_kretprobe_fixup_return() as 
> 'correct_ret_addr', which does:

No, 'correct_ret_addr' is found from 'current->kretprobe_instances'

        /* Find correct address and all nodes for this frame. */
        correct_ret_addr = (void *)__kretprobe_find_ret_addr(current, &node);

> 
> +       frame_pointer = ((unsigned long *)&regs->sp) + 1;
> +	*frame_pointer = correct_ret_addr;
> 
> ... which looks like the exact same thing as:
> 
> 	*frame_pointer = frame_pointer;
> 
> ... obfuscated through a thick layer of type casts?

Thus it will be the same thing as

	*frame_pointer = __kretprobe_find_ret_addr(current, &node);

Actually, this is a bit confusing because same 'frame_pointer' is
calcurated twice from 'regs->sp'. This is because the return address
is stored at 'frame_pointer' or not depends on the architecture.


Thank you,

> 
> Thanks,
> 
> 	Ingo


-- 
Masami Hiramatsu <mhiramat@xxxxxxxxxx>