On Thu, Jul 30, 2015 at 06:22:02PM +0200, Peter Zijlstra wrote:
> On Thu, Jul 30, 2015 at 08:54:54AM -0700, Paul E. McKenney wrote:
> 
> > Good point, and it already does, and I clearly was confused, apologies.
> > 
> > So the real way to make this happen is (for example) to build
> > with CONFIG_RCU_FANOUT=2 and CONFIG_RCU_FANOUT_LEAF=16 (the
> > default), which could accommodate up to 128 CPUs.  Then boot with
> > rcutree.rcu_fanout_leaf=2 on a system with more than 16 CPUs, with
> > rcutree.rcu_fanout_leaf=3 on a system with more than 24 CPUs, and so on.
> 
> Ah, runtime overrides and operator error, but then we can WARN(), reset
> the arguments and try again, right? No need to panic the machine and
> fail to boot.

Good point, like the patch below?  Which also legitimizes my
example after the fact, as it previously simply prohibited having
rcutree.rcu_fanout_leaf less than CONFIG_RCU_FANOUT_LEAF.  :-/

> > Of course, the truly macho way to get this error message is to build
> > with CONFIG_RCU_FANOUT=64 and CONFIG_RCU_FANOUT_LEAF=64, then boot with
> > rcutree.rcu_fanout_leaf=63 on a system with more than 16,515,072 CPUs.
> > Of course, you get serious style points if the system manages to stay
> > up for more than 24 hours without a hardware failure.  ;-)
> 
> Yes, I'll go power up the nuclear reactor in the basement first :-)

Only one?  ;-)

                                                        Thanx, Paul

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

    rcu: Eliminate panic when silly boot-time fanout specified
    
    This commit loosens rcutree.rcu_fanout_leaf range checks
    and replaces a panic() with a fallback to compile-time values.
    This fallback is accompanied by a WARN_ON(), and both occur when the
    rcutree.rcu_fanout_leaf value is too small to accommodate the number of
    CPUs.  For example, given the current four-level limit for the rcu_node
    tree, a system with more than 16 CPUs built with CONFIG_FANOUT=2 must
    have rcutree.rcu_fanout_leaf larger than 2.
    
    Reported-by: Peter Zijlstra <[email protected]>
    Signed-off-by: Paul E. McKenney <[email protected]>

diff --git a/Documentation/kernel-parameters.txt 
b/Documentation/kernel-parameters.txt
index 01b5b68a237a..2a5d4696bdb9 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -3059,9 +3059,12 @@ bytes respectively. Such letter suffixes can also be 
entirely omitted.
                        cache-to-cache transfer latencies.
 
        rcutree.rcu_fanout_leaf= [KNL]
-                       Increase the number of CPUs assigned to each
-                       leaf rcu_node structure.  Useful for very large
-                       systems.
+                       Change the number of CPUs assigned to each
+                       leaf rcu_node structure.  Useful for very
+                       large systems, which will choose the value 64,
+                       and for NUMA systems with large remote-access
+                       latencies, which will choose a value aligned
+                       with the appropriate hardware boundaries.
 
        rcutree.jiffies_till_sched_qs= [KNL]
                        Set required age in jiffies for a
diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index ce43fac5ff91..9f8040396d3e 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -4216,13 +4216,12 @@ static void __init rcu_init_geometry(void)
                rcu_fanout_leaf, nr_cpu_ids);
 
        /*
-        * The boot-time rcu_fanout_leaf parameter is only permitted
-        * to increase the leaf-level fanout, not decrease it.  Of course,
-        * the leaf-level fanout cannot exceed the number of bits in
-        * the rcu_node masks.  Complain and fall back to the compile-
-        * time values if these limits are exceeded.
+        * The boot-time rcu_fanout_leaf parameter must be at least two
+        * and cannot exceed the number of bits in the rcu_node masks.
+        * Complain and fall back to the compile-time values if this
+        * limit is exceeded.
         */
-       if (rcu_fanout_leaf < RCU_FANOUT_LEAF ||
+       if (rcu_fanout_leaf < 2 ||
            rcu_fanout_leaf > sizeof(unsigned long) * 8) {
                rcu_fanout_leaf = RCU_FANOUT_LEAF;
                WARN_ON(1);
@@ -4239,10 +4238,13 @@ static void __init rcu_init_geometry(void)
 
        /*
         * The tree must be able to accommodate the configured number of CPUs.
-        * If this limit is exceeded than we have a serious problem elsewhere.
+        * If this limit is exceeded, fall back to the compile-time values.
         */
-       if (nr_cpu_ids > rcu_capacity[RCU_NUM_LVLS - 1])
-               panic("rcu_init_geometry: rcu_capacity[] is too small");
+       if (nr_cpu_ids > rcu_capacity[RCU_NUM_LVLS - 1]) {
+               rcu_fanout_leaf = RCU_FANOUT_LEAF;
+               WARN_ON(1);
+               return;
+       }
 
        /* Calculate the number of levels in the tree. */
        for (i = 0; nr_cpu_ids > rcu_capacity[i]; i++) {

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

Reply via email to