Re: [PATCH v9 01/36] tracing: Add a comment about ftrace_regs definition

[Florent Revest]

On Mon, Apr 15, 2024 at 2:49 PM Masami Hiramatsu (Google)
<mhiramat@xxxxxxxxxx> wrote:
>
> From: Masami Hiramatsu (Google) <mhiramat@xxxxxxxxxx>
>
> To clarify what will be expected on ftrace_regs, add a comment to the
> architecture independent definition of the ftrace_regs.
>
> Signed-off-by: Masami Hiramatsu (Google) <mhiramat@xxxxxxxxxx>
> Acked-by: Mark Rutland <mark.rutland@xxxxxxx>
> ---
>  Changes in v8:
>   - Update that the saved registers depends on the context.
>  Changes in v3:
>   - Add instruction pointer
>  Changes in v2:
>   - newly added.
> ---
>  include/linux/ftrace.h |   26 ++++++++++++++++++++++++++
>  1 file changed, 26 insertions(+)
>
> diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
> index 54d53f345d14..b81f1afa82a1 100644
> --- a/include/linux/ftrace.h
> +++ b/include/linux/ftrace.h
> @@ -118,6 +118,32 @@ extern int ftrace_enabled;
>
>  #ifndef CONFIG_HAVE_DYNAMIC_FTRACE_WITH_ARGS
>
> +/**
> + * ftrace_regs - ftrace partial/optimal register set
> + *
> + * ftrace_regs represents a group of registers which is used at the
> + * function entry and exit. There are three types of registers.
> + *
> + * - Registers for passing the parameters to callee, including the stack
> + *   pointer. (e.g. rcx, rdx, rdi, rsi, r8, r9 and rsp on x86_64)
> + * - Registers for passing the return values to caller.
> + *   (e.g. rax and rdx on x86_64)

Ooc, have we ever considered skipping argument registers that are not
return value registers in the exit code paths ? For example, why would
we want to save rdi in a return handler ?

But if we want to avoid the situation of having "sparse ftrace_regs"
all over again, we'd have to split ftrace_regs into a ftrace_args_regs
and a ftrace_ret_regs which would make this refactoring even more
painful, just to skip a few instructions. :|

I don't necessarily think it's worth it, I just wanted to make sure
this was considered.

> + * - Registers for hooking the function call and return including the
> + *   frame pointer (the frame pointer is architecture/config dependent)
> + *   (e.g. rip, rbp and rsp for x86_64)
> + *
> + * Also, architecture dependent fields can be used for internal process.
> + * (e.g. orig_ax on x86_64)
> + *
> + * On the function entry, those registers will be restored except for
> + * the stack pointer, so that user can change the function parameters
> + * and instruction pointer (e.g. live patching.)
> + * On the function exit, only registers which is used for return values
> + * are restored.
> + *
> + * NOTE: user *must not* access regs directly, only do it via APIs, because
> + * the member can be changed according to the architecture.
> + */
>  struct ftrace_regs {
>         struct pt_regs          regs;
>  };
>