On Tue, Feb 25, 2025 at 10:57:38AM +0100, Vlastimil Babka wrote:
> On 2/24/25 12:44, Uladzislau Rezki wrote:
> > On Fri, Feb 21, 2025 at 06:28:49PM +0100, Vlastimil Babka wrote:
> >> On 2/21/25 17:30, Keith Busch wrote:
> >> > On Wed, Aug 07, 2024 at 12:31:19PM +0200, Vlastimil Babka wrote:
> >> >> We would like to replace call_rcu() users with kfree_rcu() where the
> >> >> existing callback is just a kmem_cache_free(). However this causes
> >> >> issues when the cache can be destroyed (such as due to module unload).
> >> >>
> >> >> Currently such modules should be issuing rcu_barrier() before
> >> >> kmem_cache_destroy() to have their call_rcu() callbacks processed first.
> >> >> This barrier is however not sufficient for kfree_rcu() in flight due
> >> >> to the batching introduced by a35d16905efc ("rcu: Add basic support for
> >> >> kfree_rcu() batching").
> >> >>
> >> >> This is not a problem for kmalloc caches which are never destroyed, but
> >> >> since removing SLOB, kfree_rcu() is allowed also for any other cache,
> >> >> that might be destroyed.
> >> >>
> >> >> In order not to complicate the API, put the responsibility for handling
> >> >> outstanding kfree_rcu() in kmem_cache_destroy() itself. Use the newly
> >> >> introduced kvfree_rcu_barrier() to wait before destroying the cache.
> >> >> This is similar to how we issue rcu_barrier() for SLAB_TYPESAFE_BY_RCU
> >> >> caches, but has to be done earlier, as the latter only needs to wait for
> >> >> the empty slab pages to finish freeing, and not objects from the slab.
> >> >>
> >> >> Users of call_rcu() with arbitrary callbacks should still issue
> >> >> rcu_barrier() before destroying the cache and unloading the module, as
> >> >> kvfree_rcu_barrier() is not a superset of rcu_barrier() and the
> >> >> callbacks may be invoking module code or performing other actions that
> >> >> are necessary for a successful unload.
> >> >>
> >> >> Signed-off-by: Vlastimil Babka <vbabka@xxxxxxx>
> >> >> ---
> >> >> mm/slab_common.c | 3 +++
> >> >> 1 file changed, 3 insertions(+)
> >> >>
> >> >> diff --git a/mm/slab_common.c b/mm/slab_common.c
> >> >> index c40227d5fa07..1a2873293f5d 100644
> >> >> --- a/mm/slab_common.c
> >> >> +++ b/mm/slab_common.c
> >> >> @@ -508,6 +508,9 @@ void kmem_cache_destroy(struct kmem_cache *s)
> >> >> if (unlikely(!s) || !kasan_check_byte(s))
> >> >> return;
> >> >>
> >> >> + /* in-flight kfree_rcu()'s may include objects from our cache */
> >> >> + kvfree_rcu_barrier();
> >> >> +
> >> >> cpus_read_lock();
> >> >> mutex_lock(&slab_mutex);
> >> >
> >> > This patch appears to be triggering a new warning in certain conditions
> >> > when tearing down an nvme namespace's block device. Stack trace is at
> >> > the end.
> >> >
> >> > The warning indicates that this shouldn't be called from a
> >> > WQ_MEM_RECLAIM workqueue. This workqueue is responsible for bringing up
> >> > and tearing down block devices, so this is a memory reclaim use AIUI.
> >> > I'm a bit confused why we can't tear down a disk from within a memory
> >> > reclaim workqueue. Is the recommended solution to simply remove the WQ
> >> > flag when creating the workqueue?
> >>
> >> I think it's reasonable to expect a memory reclaim related action would
> >> destroy a kmem cache. Mateusz's suggestion would work around the issue, but
> >> then we could get another surprising warning elsewhere. Also making the
> >> kmem_cache destroys async can be tricky when a recreation happens
> >> immediately under the same name (implications with sysfs/debugfs etc). We
> >> managed to make the destroying synchronous as part of this series and it
> >> would be great to keep it that way.
> >>
> >> > ------------[ cut here ]------------
> >> > workqueue: WQ_MEM_RECLAIM nvme-wq:nvme_scan_work is flushing !WQ_MEM_RECLAIM events_unbound:kfree_rcu_work
> >>
> >> Maybe instead kfree_rcu_work should be using a WQ_MEM_RECLAIM workqueue? It
> >> is after all freeing memory. Ulad, what do you think?
> >>
> > We reclaim memory, therefore WQ_MEM_RECLAIM seems what we need.
> > AFAIR, there is an extra rescue worker, which can really help
> > under a low memory condition in a way that we do a progress.
> >
> > Do we have a reproducer of mentioned splat?
>
> I tried to create a kunit test for it, but it doesn't trigger anything. Maybe
> it's too simple, or racy, and thus we are not flushing any of the queues from
> kvfree_rcu_barrier()?
>
See some comments below. I will try to reproduce it today. But from the
first glance it should trigger it.
> ----8<----
> From 1e19ea78e7fe254034970f75e3b7cb705be50163 Mon Sep 17 00:00:00 2001
> From: Vlastimil Babka <vbabka@xxxxxxx>
> Date: Tue, 25 Feb 2025 10:51:28 +0100
> Subject: [PATCH] add test for kmem_cache_destroy in a workqueue
>
> ---
> lib/slub_kunit.c | 48 ++++++++++++++++++++++++++++++++++++++++++++++++
> 1 file changed, 48 insertions(+)
>
> diff --git a/lib/slub_kunit.c b/lib/slub_kunit.c
> index f11691315c2f..5fe9775fba05 100644
> --- a/lib/slub_kunit.c
> +++ b/lib/slub_kunit.c
> @@ -6,6 +6,7 @@
> #include <linux/module.h>
> #include <linux/kernel.h>
> #include <linux/rcupdate.h>
> +#include <linux/delay.h>
> #include "../mm/slab.h"
>
> static struct kunit_resource resource;
> @@ -181,6 +182,52 @@ static void test_kfree_rcu(struct kunit *test)
> KUNIT_EXPECT_EQ(test, 0, slab_errors);
> }
>
> +struct cache_destroy_work {
> + struct work_struct work;
> + struct kmem_cache *s;
> +};
> +
> +static void cache_destroy_workfn(struct work_struct *w)
> +{
> + struct cache_destroy_work *cdw;
> +
> + cdw = container_of(w, struct cache_destroy_work, work);
> +
> + kmem_cache_destroy(cdw->s);
> +}
> +
> +static void test_kfree_rcu_wq_destroy(struct kunit *test)
> +{
> + struct test_kfree_rcu_struct *p;
> + struct cache_destroy_work cdw;
> + struct workqueue_struct *wq;
> + struct kmem_cache *s;
> +
> + if (IS_BUILTIN(CONFIG_SLUB_KUNIT_TEST))
> + kunit_skip(test, "can't do kfree_rcu() when test is built-in");
> +
> + INIT_WORK_ONSTACK(&cdw.work, cache_destroy_workfn);
> + wq = alloc_workqueue("test_kfree_rcu_destroy_wq", WQ_UNBOUND | WQ_MEM_RECLAIM, 0);
>
Maybe it is worth to add WQ_HIGHPRI also to be ahead?
> + if (!wq)
> + kunit_skip(test, "failed to alloc wq");
> +
> + s = test_kmem_cache_create("TestSlub_kfree_rcu_wq_destroy",
> + sizeof(struct test_kfree_rcu_struct),
> + SLAB_NO_MERGE);
> + p = kmem_cache_alloc(s, GFP_KERNEL);
> +
> + kfree_rcu(p, rcu);
> +
> + cdw.s = s;
> + queue_work(wq, &cdw.work);
> + msleep(1000);
I am not sure it is needed. From the other hand it does nothing if
i do not miss anything.
> + flush_work(&cdw.work);
> +
> + destroy_workqueue(wq);
> +
> + KUNIT_EXPECT_EQ(test, 0, slab_errors);
> +}
> +
> static void test_leak_destroy(struct kunit *test)
> {
> struct kmem_cache *s = test_kmem_cache_create("TestSlub_leak_destroy",
> @@ -254,6 +301,7 @@ static struct kunit_case test_cases[] = {
> KUNIT_CASE(test_clobber_redzone_free),
> KUNIT_CASE(test_kmalloc_redzone_access),
> KUNIT_CASE(test_kfree_rcu),
> + KUNIT_CASE(test_kfree_rcu_wq_destroy),
> KUNIT_CASE(test_leak_destroy),
> KUNIT_CASE(test_krealloc_redzone_zeroing),
> {}
> --
> 2.48.1
>
>
--
Uladzislau Rezki
