On Mon, Oct 07, 2019 at 10:27:08AM +0200, Peter Zijlstra wrote:
> This series, which depends on the previous two, introduces static_call().
>
> static_call(), is the idea of static_branch() applied to indirect function
> calls. Remove a data load (indirection) by modifying the text.
>
> These patches are still based on the work Josh did earlier, but incorporated
> feedback from the last posting and have a lot of extra patches which resulted
> from me trying to actually use static_call().
>
> This still relies on objtool to generate the .static_call_sites section,
> mostly
> because this is a natural place for x86_64 and works for both GCC and LLVM.
> Other architectures can pick other means if/when they implement the inline
> patching. The out-of-line (aka. trampoline) variant doesn't require this.
>
FWIW, Steve, if we were to do something like:
CFLAGS += -mfentry -mfentry_name=____static_call_fentry_tramp
And have:
struct static_call_key ____static_call_fentry = {
.func = NULL,
.next = 1,
};
asm(".pushsection .static_call.text, \"ax\" \n"
".align 4 \n"
".globl ____static_call_fentry_tramp \n"
"____static_call_fentry_tramp: \n"
" ret \n"
".type ____static_call_fentry_tramp, @function \n"
".size ____static_call_fentry_tramp, . - ____static_call_fentry_tramp \n"
".popsection \n");
Then the whole function entry thing would automagicaly turn into
something like static_cond_call(fentry)(...);
Not saying we should do that, but we could ;-)
