This typo prevented jump_label_text_reserved from checking for any
conflicts in the core kernel.
Signed-off-by: Anders Kaseorg <[email protected]>
---
kernel/jump_label.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/kernel/jump_label.c b/kernel/jump_label.c
index 3b79bd9..4aedde1 100644
--- a/kernel/jump_label.c
+++ b/kernel/jump_label.c
@@ -250,7 +250,7 @@ int jump_label_text_reserved(void *start, void *end)
{
struct jump_entry *iter;
struct jump_entry *iter_start = __start___jump_table;
- struct jump_entry *iter_stop = __start___jump_table;
+ struct jump_entry *iter_stop = __stop___jump_table;
int conflict = 0;
iter = iter_start;
--
1.7.5
On Thu, Apr 28, 2011 at 01:53:43PM -0400, Anders Kaseorg wrote:
> This typo prevented jump_label_text_reserved from checking for any
> conflicts in the core kernel.
>
> Signed-off-by: Anders Kaseorg <[email protected]>
> ---
> kernel/jump_label.c | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/kernel/jump_label.c b/kernel/jump_label.c
> index 3b79bd9..4aedde1 100644
> --- a/kernel/jump_label.c
> +++ b/kernel/jump_label.c
> @@ -250,7 +250,7 @@ int jump_label_text_reserved(void *start, void *end)
> {
> struct jump_entry *iter;
> struct jump_entry *iter_start = __start___jump_table;
> - struct jump_entry *iter_stop = __start___jump_table;
> + struct jump_entry *iter_stop = __stop___jump_table;
> int conflict = 0;
>
> iter = iter_start;
> --
> 1.7.5
>
This is fixed in the jump label re-write patch, which is queued for
2.6.40 in the -tip tree. See: http://marc.info/?l=linux-kernel&m=130391786701337&w=2
I believe only kprobes is relying on this check. Did you hit this in
practive using kprobes?
Thanks,
-Jason
On Thu, Apr 28, 2011 at 14:42, Jason Baron <[email protected]> wrote:
> This is fixed in the jump label re-write patch, which is queued for
> 2.6.40 in the -tip tree. See: http://marc.info/?l=linux-kernel&m=130391786701337&w=2
Okay. It would be nice if the bug fixes had been separated out into
commits that could be picked onto the -stable kernels. Should I send
my patch to [email protected]?
> I believe only kprobes is relying on this check. Did you hit this in
> practive using kprobes?
Ksplice needs this check too.
Anders
On Thu, 2011-04-28 at 15:10 -0400, Anders Kaseorg wrote:
> On Thu, Apr 28, 2011 at 14:42, Jason Baron <[email protected]> wrote:
> > This is fixed in the jump label re-write patch, which is queued for
> > 2.6.40 in the -tip tree. See: http://marc.info/?l=linux-kernel&m=130391786701337&w=2
>
> Okay. It would be nice if the bug fixes had been separated out into
I didn't realize this was a bug fix that was noticed by current users,
otherwise I would have done so.
> commits that could be picked onto the -stable kernels. Should I send
> my patch to [email protected]?
>
> > I believe only kprobes is relying on this check. Did you hit this in
> > practive using kprobes?
>
If you do, note that it is pending in tip and give the commit id for it.
It will make stable as soon as the commit makes it in mainline in
the .40 merge window.
If possible we could cherry pick the patch for mainline to get it in
sooner, and it would just be a wash when it goes in with the current tip
branch.
-- Steve