2015-03-17 21:15 GMT+09:00 Mark Rutland <mark.rutl...@arm.com>:
> Commit 9aabf810a67cd97e ("mm/slub: optimize alloc/free fastpath by
> removing preemption on/off") introduced an occasional hang for kernels
> built with CONFIG_PREEMPT && !CONFIG_SMP.
>
> The problem is the following loop the patch introduced to
> slab_alloc_node and slab_free:
>
> do {
> tid = this_cpu_read(s->cpu_slab->tid);
> c = raw_cpu_ptr(s->cpu_slab);
> } while (IS_ENABLED(CONFIG_PREEMPT) && unlikely(tid != c->tid));
>
> GCC 4.9 has been observed to hoist the load of c and c->tid above the
> loop for !SMP kernels (as in this case raw_cpu_ptr(x) is compile-time
> constant and does not force a reload). On arm64 the generated assembly
> looks like:
>
> ffffffc00016d3c4: f9400404 ldr x4, [x0,#8]
> ffffffc00016d3c8: f9400401 ldr x1, [x0,#8]
> ffffffc00016d3cc: eb04003f cmp x1, x4
> ffffffc00016d3d0: 54ffffc1 b.ne ffffffc00016d3c8
> <slab_alloc_node.constprop.82+0x30>
>
> If the thread is preempted between the load of c->tid (into x1) and tid
> (into x4), and an allocation or free occurs in another thread (bumping
> the cpu_slab's tid), the thread will be stuck in the loop until
> s->cpu_slab->tid wraps, which may be forever in the absence of
> allocations/frees on the same CPU.
>
> This patch changes the loop condition to access c->tid with READ_ONCE.
> This ensures that the value is reloaded even when the compiler would
> otherwise assume it could cache the value, and also ensures that the
> load will not be torn.
>
> Signed-off-by: Mark Rutland <mark.rutl...@arm.com>
> Cc: Andrew Morton <a...@linux-foundation.org>
> Cc: Catalin Marinas <catalin.mari...@arm.com>
> Cc: Christoph Lameter <c...@linux.com>
> Cc: David Rientjes <rient...@google.com>
> Cc: Jesper Dangaard Brouer <bro...@redhat.com>
> Cc: Joonsoo Kim <iamjoonsoo....@lge.com>
> Cc: Linus Torvalds <torva...@linux-foundation.org>
> Cc: Pekka Enberg <penb...@kernel.org>
> Cc: Steve Capper <steve.cap...@linaro.org>
> ---
> mm/slub.c | 6 ++++--
> 1 file changed, 4 insertions(+), 2 deletions(-)
>
> Since v1 [1]:
> * Do not erroneously remove the loop
>
> [1] lkml.kernel.org/r/1426261632-8911-1-git-send-email-mark.rutl...@arm.com
>
> diff --git a/mm/slub.c b/mm/slub.c
> index 6832c4e..82c4737 100644
> --- a/mm/slub.c
> +++ b/mm/slub.c
> @@ -2449,7 +2449,8 @@ redo:
> do {
> tid = this_cpu_read(s->cpu_slab->tid);
> c = raw_cpu_ptr(s->cpu_slab);
> - } while (IS_ENABLED(CONFIG_PREEMPT) && unlikely(tid != c->tid));
> + } while (IS_ENABLED(CONFIG_PREEMPT) &&
> + unlikely(tid != READ_ONCE(c->tid)));
>
> /*
> * Irqless object alloc/free algorithm used here depends on sequence
> @@ -2718,7 +2719,8 @@ redo:
> do {
> tid = this_cpu_read(s->cpu_slab->tid);
> c = raw_cpu_ptr(s->cpu_slab);
> - } while (IS_ENABLED(CONFIG_PREEMPT) && unlikely(tid != c->tid));
> + } while (IS_ENABLED(CONFIG_PREEMPT) &&
> + unlikely(tid != READ_ONCE(c->tid)));
>
> /* Same with comment on barrier() in slab_alloc_node() */
> barrier();
> --
Hello,
Could you show me generated code again?
What we need to check is redoing whole things in the loop.
Previous attached code seems to me that it already did
refetching c->tid in the loop and this patch looks only handle
refetching c->tid.
READ_ONCE(c->tid) will trigger redoing 'tid = this_cpu_read(s->cpu_slab->tid)'?
Thanks.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
