Re: RCU callbacks and TREE_PREEMPT_RCU

From: Paul E. McKenney
Date: Thu Sep 17 2009 - 18:21:29 EST

On Thu, Sep 17, 2009 at 10:29:02AM +0200, Eric Sesterhenn wrote:
> On Wed, 2009-09-16 at 16:26 -0700, Paul E. McKenney wrote:
> > On Thu, Sep 17, 2009 at 01:19:46AM +0200, Eric Sesterhenn wrote:
> > > On Wed, 2009-09-16 at 08:57 -0700, Paul E. McKenney wrote:
> > > > On Wed, Sep 16, 2009 at 08:47:16AM -0700, Paul E. McKenney wrote:
> > > > > On Wed, Sep 16, 2009 at 04:34:15PM +0100, Catalin Marinas wrote:
> > > > > > On Wed, 2009-09-16 at 08:29 -0700, Paul E. McKenney wrote:
> > > > > > > On Wed, Sep 16, 2009 at 03:17:21PM +0100, Catalin Marinas wrote:
> > > > > > > > When TREE_PREEMPT_RCU is enabled, the rcu list traversing above fails
> > > > > > > > with access to 0x6b6b6b6b but it is fine with TREE_PREEMPT_RCU=n and
> > > > > > > > TREE_RCU=y. During clean-up, kmemleak objects should no longer be freed
> > > > > > > > by other means since kmemleak was disabled and all callbacks are
> > > > > > > > ignored. The system is a 900Mhz P3, 256MB RAM, CONFIG_SMP=n.
> > > > > > > >
> > > > > > > > Is there something I'm doing wrong in kmemleak or a bug with RCU
> > > > > > > > preemption? The kernel oops looks like this:
> > > > > > >
> > > > > > > From your description and the code above, I must suspect a bug with
> > > > > > > RCU preemption. A new one, as the only bugs I am currently chasing
> > > > > > > involve NR_CPUS>32 (>64 on 64-bit systems).
> > > > > > >
> > > > > > > CONFIG_SMP=n implies NR_CPUS==1 in your build, correct?
> > > > > >
> > > > > > CONFIG_NR_CPUS=1.
> > > > >
> > > > > I was afraid of that. ;-)
> > > >
> > > > PS to previous -- there -is- a bug in mainline for TREE_PREEMPT_RCU for
> > > > single-CPU operation, but it is with synchronize_rcu() rather than
> > > > call_rcu(). The fix is in tip/core/urgent, commit #366b04ca. Or see
> > > > the following patch.
> > > >
> > > > So, could you please give the following patch a try?
> > >
> > > Sadly this does not fix the issue, is there any further information I
> > > can provide to you?
> >
> > :-(
> >
> > Would you be willing to give the attached diagnostic patch a go?
> >
> > Thanx, Paul
>
> It does not apply cleanly against current -git
> (rcu_preempt_check_blocked_tasks is missing in my rcutree_plugin.h for
> example) I tried to apply it by hand as good as possible, and will test
> it today.
>
> root@whiterabbit:/usr/src/linux# patch -p1 < ~/RCU_callbacks_and_TREE_PREEMPT_RCU-debug
> patching file kernel/rcutree.c
> Hunk #1 FAILED at 623.
> Hunk #2 FAILED at 657.
> Hunk #3 succeeded at 722 (offset 19 lines).
> Hunk #4 succeeded at 740 (offset 19 lines).
> Hunk #5 succeeded at 765 (offset 19 lines).
> Hunk #6 succeeded at 877 (offset 19 lines).
> Hunk #7 succeeded at 886 (offset 19 lines).
> 2 out of 7 hunks FAILED -- saving rejects to file kernel/rcutree.c.rej
> patching file kernel/rcutree_plugin.h
> Hunk #1 FAILED at 206.
> Hunk #2 succeeded at 206 (offset -10 lines).
> Hunk #3 FAILED at 270.
> Hunk #4 succeeded at 283 (offset -22 lines).
> Hunk #5 succeeded at 296 (offset -22 lines).
> Hunk #6 succeeded at 473 (offset -23 lines).
> 2 out of 6 hunks FAILED -- saving rejects to file
> kernel/rcutree_plugin.h.rej

Sigh!!! I lost track of what was in mainline vs. -tip. You certainly
need the following patch from -tip as well.

Please accept apologies for my confusion!!!

Thanx, Paul

------------------------------------------------------------------------

Commit-ID: de078d875cc7fc709f7818f26d38389c04369826
Gitweb: http://git.kernel.org/tip/de078d875cc7fc709f7818f26d38389c04369826
Author: Paul E. McKenney <paulmck@xxxxxxxxxxxxxxxxxx>
AuthorDate: Tue, 8 Sep 2009 15:54:36 -0700
Committer: Ingo Molnar <mingo@xxxxxxx>
CommitDate: Fri, 18 Sep 2009 00:04:54 +0200

rcu: Need to update rnp->gpnum if preemptable RCU is to be reliable

Without this patch, tasks preempted in RCU read-side critical
sections can fail to block the grace period, given that
rnp->gpnum is used to determine which rnp->blocked_tasks[]
element the preempted task is enqueued on.

Before the patch, rnp->gpnum is always zero, so preempted tasks
are always enqueued on rnp->blocked_tasks[0], which is correct
only when the current CPU has not checked into the current
grace period and the grace-period number is even, or,
similarly, if the current CPU -has- checked into the current
grace period and the grace-period number is odd.

Signed-off-by: Paul E. McKenney <paulmck@xxxxxxxxxxxxxxxxxx>
Acked-by: Steven Rostedt <rostedt@xxxxxxxxxxx>
Cc: laijs@xxxxxxxxxxxxxx
Cc: dipankar@xxxxxxxxxx
Cc: akpm@xxxxxxxxxxxxxxxxxxxx
Cc: mathieu.desnoyers@xxxxxxxxxx
Cc: josht@xxxxxxxxxxxxxxxxxx
Cc: dvhltc@xxxxxxxxxx
Cc: niv@xxxxxxxxxx
Cc: peterz@xxxxxxxxxxxxx
LKML-Reference: <12524504771622-git-send-email->
Signed-off-by: Ingo Molnar <mingo@xxxxxxx>


---
kernel/rcutree.c | 6 +++++-
1 files changed, 5 insertions(+), 1 deletions(-)

diff --git a/kernel/rcutree.c b/kernel/rcutree.c
index 6b11b07..c634a92 100644
--- a/kernel/rcutree.c
+++ b/kernel/rcutree.c
@@ -632,6 +632,7 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags)
/* Special-case the common single-level case. */
if (NUM_RCU_NODES == 1) {
rnp->qsmask = rnp->qsmaskinit;
+ rnp->gpnum = rsp->gpnum;
rsp->signaled = RCU_SIGNAL_INIT; /* force_quiescent_state OK. */
spin_unlock_irqrestore(&rnp->lock, flags);
return;
@@ -657,8 +658,10 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags)
*/

rnp_end = rsp->level[NUM_RCU_LVLS - 1];
- for (rnp_cur = &rsp->node[0]; rnp_cur < rnp_end; rnp_cur++)
+ for (rnp_cur = &rsp->node[0]; rnp_cur < rnp_end; rnp_cur++) {
rnp_cur->qsmask = rnp_cur->qsmaskinit;
+ rnp->gpnum = rsp->gpnum;
+ }

/*
* Now set up the leaf nodes. Here we must be careful. First,
@@ -679,6 +682,7 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags)
for (; rnp_cur < rnp_end; rnp_cur++) {
spin_lock(&rnp_cur->lock); /* irqs already disabled. */
rnp_cur->qsmask = rnp_cur->qsmaskinit;
+ rnp->gpnum = rsp->gpnum;
spin_unlock(&rnp_cur->lock); /* irqs already disabled. */
}

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/