Re: [RFC][PATCH v2] ftrace/x86: Emulate call function while updating in breakpoint handler

2019-05-01 Thread Steven Rostedt
On Wed, 01 May 2019 10:26:32 +0200 Nicolai Stange wrote: > > +extern asmlinkage void ftrace_emulate_call_irqon(void); > > +extern asmlinkage void ftrace_emulate_call_irqoff(void); > > +extern asmlinkage void ftrace_emulate_call_nmi(void); > > +extern asmlinkage void

Re: [RFC][PATCH v2] ftrace/x86: Emulate call function while updating in breakpoint handler

2019-05-01 Thread Nicolai Stange
Hi Steve, many thanks for moving this forward! Steven Rostedt writes: > > diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c > index ef49517f6bb2..9160f5cc3b6d 100644 > --- a/arch/x86/kernel/ftrace.c > +++ b/arch/x86/kernel/ftrace.c > @@ -17,6 +17,7 @@ > #include > #include

Re: [RFC][PATCH v2] ftrace/x86: Emulate call function while updating in breakpoint handler

2019-04-30 Thread Linus Torvalds
On Tue, Apr 30, 2019 at 6:35 PM Steven Rostedt wrote: > > > Probably easier to move it from inline asm to ftrace_X.S and use the > lockdep TRACE_ON/OFF macros. Yeah, that should clean up the percpu stuff too since we have helper macros for it for *.S files anyway. I only did the asm() in C

Re: [RFC][PATCH v2] ftrace/x86: Emulate call function while updating in breakpoint handler

2019-04-30 Thread Steven Rostedt
On Tue, 30 Apr 2019 17:53:34 -0400 Steven Rostedt wrote: > + if (ftrace_location(ip)) { > + /* A breakpoint at the beginning of the function was hit */ > + if (in_nmi()) { > + /* NMIs have their own trampoline */ > +