Re: [PATCHv4] kprobes: Introduce kprobe_handle_fault()

[Harvey Harrison <harvey.harrison@gmail.com>]

On Thu, 2008-01-10 at 00:16 +0100, Heiko Carstens wrote:
> >  arch/avr32/mm/fault.c   |   21 +--------------------
> >  arch/ia64/mm/fault.c    |   24 +-----------------------
> >  arch/powerpc/mm/fault.c |   25 +------------------------
> >  arch/s390/mm/fault.c    |   25 +------------------------
> >  arch/sparc64/mm/fault.c |   23 +----------------------
> >  arch/x86/mm/fault_64.c  |   23 ++---------------------
> >  include/linux/kprobes.h |   17 +++++++++++++++++
> >  7 files changed, 24 insertions(+), 134 deletions(-)
> 
> Somehow I think you missed arch/x86/mm/fault_32.c :)
> 

X86_32 _needs_ !user_mode_vm() as opposed to all of the others needing
!user_mode(), I was planning to deal with this in the unification of
fault_32|64.c at a later date.  So for now X86_32 will not use this.

> > This uncovered a possible bug in the s390 version as that purely
> > copied the x86 version unconditionally passing 14 as the trapnr
> > rather than the error_code parameter.

> 
> Uhm.. yes. 14 is HFP-Significance exception. That doesn't make too much
> sense. Passing error_code here would be the right thing to do.
> Also I just checked with David Wilder: system tap itself doesn't have any
> fault handlers. So it should be safe to change this.

OK, I will send a two patch series as per Christoph's request and will
include the s390 error_code fix instead of 14.

Cheers,

Harvey