sparc64: Export flush_ptrace_access() (needed by lustre)
Subsystems (2)
PATCH 1/1 sparc64: Export flush_ptrace_access() (needed by lustre)
ERROR: "flush_ptrace_access" [drivers/staging/lustre/lustre/libcfs/libcfs.ko] undefined!
--- arch/sparc/kernel/ptrace_64.c | 2 ++ 1 file changed, 2 insertions(+)
arch/sparc/kernel/ptrace_64.c +2 -0
@@ -14,6 +14,7 @@#include <linux/sched.h> #include <linux/mm.h> #include <linux/errno.h> +#include <linux/export.h> #include <linux/ptrace.h> #include <linux/user.h> #include <linux/smp.h>
@@ -116,6 +117,7 @@ void flush_ptrace_access(struct vm_area_struct *vma, struct page *page,preempt_enable (); } +EXPORT_SYMBOL (flush_ptrace_access);static int get_from_target (struct task_struct *target,unsigned long uaddr,void *kbuf,int len)
Signature · 1 line
1.7.9.5
Quoted 2 lines · Geert Uytterhoeven
ERROR: "
flush_ptrace_access" [drivers/staging/lustre/lustre/libcfs/libcfs.ko] undefined!
This seems to be more copy_to_user_page fallout, so instead of all these arch patches we should figure out why lustre absolutely wants to use this API that so far has been for core MM code only.
Quoted 7 lines · Christoph Hellwig
Quoted 2 lines · Geert Uytterhoeven
ERROR: "
flush_ptrace_access" [drivers/staging/lustre/lustre/libcfs/libcfs.ko] undefined!This seems to be more
copy_to_user_pagefallout, so instead of all these arch patches we should figure out why lustre absolutely wants to use this API that so far has been for core MM code only.
Looking at the code is seems to be for a reimplementation of __access_remote_vm. So at very least it should use that, but I still requiestion why it's doing that at all.
Geert, please don't just export stuff for staging/ use only. Code
wouldn't be in staging if it weren't a steaming pile of junk, so adjusting anyting in the real tree for it needs proper discussion first.