Re: [FTRACE] Enabling function_graph causes OOPS

2009-09-13 Thread Steven Rostedt
On Sun, 2009-09-13 at 14:37 +1000, Benjamin Herrenschmidt wrote: On Sun, 2009-09-13 at 00:07 -0400, Steven Rostedt wrote: 982: R_PPC64_ADDR16_HIGHEST ftrace_return_to_handler 984: 60 84 00 00 ori r4,r4,0 986: R_PPC64_ADDR16_HIGHER

Re: MPC5200/BestComm functions question

2009-09-13 Thread Albrecht Dreß
Am 12.09.09 16:01 schrieb(en) Grant Likely: I've actually got a better driver which exports an API for doing local plus bus FIFO transfers. It's been posted to the mailing list a couple of times. I'll cc: you the next time I post it (real soon now). Thanks a lot, that will be really

Re: [FTRACE] Enabling function_graph causes OOPS

2009-09-13 Thread Benjamin Herrenschmidt
.../... Something is totally messed up here. Could it be that we don't handle R_PPC64_ADDR16_* relocs in arch/powerpc/kernel/modules/module_64.c ? Sachin, do you see a bunch of Unknown ADD relocation in your dmesg ? Ben, The thing is, this is kernel proper. This code is

Re: [2/5] Cleanup management of kmem_caches for pagetables

2009-09-13 Thread Benjamin Herrenschmidt
On Wed, 2009-09-09 at 15:59 +1000, David Gibson wrote: 6 files changed, 73 insertions(+), 108 deletions(-) That's a pretty good start :-) +struct kmem_cache *pgtable_cache[PGF_SHIFT_MASK]; + +void pgtable_cache_add(unsigned shift, void (*ctor)(void *)) +{ + char *name; + unsigned

Re: [2/5] Cleanup management of kmem_caches for pagetables

2009-09-13 Thread David Gibson
On Mon, Sep 14, 2009 at 08:28:11AM +1000, Benjamin Herrenschmidt wrote: On Wed, 2009-09-09 at 15:59 +1000, David Gibson wrote: 6 files changed, 73 insertions(+), 108 deletions(-) That's a pretty good start :-) Isn't it :) +struct kmem_cache *pgtable_cache[PGF_SHIFT_MASK]; + +void