On Tue, 27 Jun 2017 14:59:48 +0300 "Kirill A. Shutemov" <[email protected]> wrote:
> We've moved most logic of early boot up from startup_64() written in > assembly to __startup_64() written in C. > > Fengguang reported breakage due to the change. It was tracked down to > CONFIG_FUNCTION_TRACER being enabled. Is this due to code that is in linux-next or is this issue currently in mainline? If it is in linux-next, I'll give you: Acked-by: Steven Rostedt (VMware) <[email protected]> and you can take it. If the issue happens in current mainline, then I'll take it and push it out. -- Steve > > Looks like function-tracing of the new code is harmful. > > Let's not comple the code with -pg compiler option. > > Signed-off-by: Kirill A. Shutemov <[email protected]> > Reported-by: Fengguang Wu <[email protected]> > Cc: Steven Rostedt <[email protected]> > --- > arch/x86/kernel/Makefile | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile > index 4b994232cb57..0db512092f37 100644 > --- a/arch/x86/kernel/Makefile > +++ b/arch/x86/kernel/Makefile > @@ -18,6 +18,7 @@ CFLAGS_REMOVE_pvclock.o = -pg > CFLAGS_REMOVE_kvmclock.o = -pg > CFLAGS_REMOVE_ftrace.o = -pg > CFLAGS_REMOVE_early_printk.o = -pg > +CFLAGS_REMOVE_head64.o = -pg > endif > > KASAN_SANITIZE_head$(BITS).o := n
