This is the mail archive of the systemtap@sourceware.org mailing list for the systemtap project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [RFC -mm][PATCH 5/6] prepare kprobes code for x86 unification


On Thu, Dec 13, 2007 at 01:47:50PM +0530, Srinivasa Ds wrote:
> Ananth N Mavinakayanahalli wrote:
> >On Wed, Dec 12, 2007 at 08:21:01PM -0500, Masami Hiramatsu wrote:
> >>Masami Hiramatsu wrote:
> >>>Finally, I reproduced it by executing a command below:
> >>>% stap -e 'probe kernel.function("migration_init"){}'
> >>>
> 
> >
> >That does seem like the right thing to do. This also needs to live in
> >common code.
> >
> >Srini,
> >Want to take a stab at it?
> 
> 
> How about this Ananth ???
>         
> 
> Signed-off-by: Srinivasa DS <srinivasa@in.ibm.com> 
> 
> 
> 
> 
> --- kernel/kprobes.c.orig       2007-12-13 13:43:45.000000000 +0530
> +++ kernel/kprobes.c    2007-12-13 13:48:32.000000000 +0530
> @@ -48,6 +48,7 @@
> #include <asm/cacheflush.h>
> #include <asm/errno.h>
> #include <asm/uaccess.h>
> +#include <asm/sections.h>
> 
> #define KPROBE_HASH_BITS 6
> #define KPROBE_TABLE_SIZE (1 << KPROBE_HASH_BITS)
> @@ -722,6 +723,11 @@ int __kprobes register_kretprobe(struct 
>                if (addr == NULL)
>                        kprobe_lookup_name(rp->kp.symbol_name, addr);
>                addr += rp->kp.offset;
> +
> +       if (addr >= (unsigned long)_sinittext && 
> +           addr <= (unsigned long)_einittext)
> +               return -EINVAL;
> +

We wouldn't want to permit registration of kprobes also, do we? Looks
like you've changed it only for kretprobes.

Have you tested the patch?

Ananth

> 
>                for (i = 0; kretprobe_blacklist[i].name != NULL; i++) {
>                        if (kretprobe_blacklist[i].addr == addr)


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]