On Sat, 4 May 2024 13:35:26 +0000
"Dr. David Alan Gilbert" <[email protected]> wrote:
> Hi,
> I've just posted a patch 'ftrace: Remove unused list 'ftrace_direct_funcs''
> that clears out some old code, but while at it I noticed the global
> 'ftrace_direct_func_count'.
>
> As far as I can tell, it's never assigned (or initialised) but it is tested:
>
> kernel/trace/fgraph.c:
> #ifndef CONFIG_HAVE_DYNAMIC_FTRACE_WITH_ARGS
> /*
> * Skip graph tracing if the return location is served by direct trampoline,
> * since call sequence and return addresses are unpredictable anyway.
> * Ex: BPF trampoline may call original function and may skip frame
> * depending on type of BPF programs attached.
> */
> if (ftrace_direct_func_count &&
> ftrace_find_rec_direct(ret - MCOUNT_INSN_SIZE))
> return -EBUSY;
> #endif
>
> So I wasn't sure whether it was just safe to nuke that section
> or whether it really needed fixing?
Yes, after commit 8788ca164eb4bad ("ftrace: Remove the legacy
_ftrace_direct API") that variable is no longer used.
-- Steve