Re: [PATCH v3 0/2] s390 vs. kprobes on ftrace

2014-10-22 Thread Vojtech Pavlik
On Wed, Oct 22, 2014 at 10:26:25AM +0200, Heiko Carstens wrote: > > I can confirm that kGraft works well on top of current mainline with > > this patch added. > > > > Another reason for a performance impact when kGraft is enabled is that > > kGraft still adds two instructions to the syscall path

Re: [PATCH v3 0/2] s390 vs. kprobes on ftrace

2014-10-22 Thread Heiko Carstens
On Tue, Oct 21, 2014 at 09:58:31PM +0200, Vojtech Pavlik wrote: > Hello Heiko, > > I can confirm that kGraft works well on top of current mainline with > this patch added. > > Another reason for a performance impact when kGraft is enabled is that > kGraft still adds two instructions to the syscal

Re: [PATCH v3 0/2] s390 vs. kprobes on ftrace

2014-10-21 Thread Vojtech Pavlik
Hello Heiko, I can confirm that kGraft works well on top of current mainline with this patch added. Another reason for a performance impact when kGraft is enabled is that kGraft still adds two instructions to the syscall path on s390x, as there is no space left for a kgraft TIF in the first eight

[PATCH v3 0/2] s390 vs. kprobes on ftrace

2014-10-21 Thread Heiko Carstens
v3: Changed patch 1/2 to incorporate feedback from Steven Rostedt and Masami Hiramatsu: rename helper function check_ftrace_location() to arch_check_ftrace_location() and convert it to a weak function, so architectures can override it without the need for new config option. v2: Changed patch 1/2 t