Now there are two indicators of socket memory pressure sit inside
struct mem_cgroup, socket_pressure and tcpmem_pressure, indicating
memory reclaim pressure in memcg->memory and ->tcpmem respectively.
When in legacy mode (cgroupv1), the socket memory is charged into
->tcpmem which is independent of ->memory, so socket_pressure has
nothing to do with socket's pressure at all. Things could be worse
by taking socket_pressure into consideration in legacy mode, as a
pressure in ->memory can lead to premature reclamation/throttling
in socket.
While for the default mode (cgroupv2), the socket memory is charged
into ->memory, and ->tcpmem/->tcpmem_pressure are simply not used.
So {socket,tcpmem}_pressure are only used in default/legacy mode
respectively for indicating socket memory pressure. This patch fixes
the pieces of code that make mixed use of both.
Fixes: 8e8ae645249b ("mm: memcontrol: hook up vmpressure to socket pressure")
Signed-off-by: Abel Wu <wuyun.abel@xxxxxxxxxxxxx>
---
v2:
- Add a comment to clarify the scope of indicators (Roman)
- Drop code cleanup on tcpmem_pressure (Roman)
- Fix the lack of a Fixes: tag (Eric)
---
include/linux/memcontrol.h | 9 +++++++--
mm/vmpressure.c | 8 ++++++++
2 files changed, 15 insertions(+), 2 deletions(-)
diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 5818af8eca5a..dbf26bc89dd4 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -284,6 +284,11 @@ struct mem_cgroup {
atomic_long_t memory_events[MEMCG_NR_MEMORY_EVENTS];
atomic_long_t memory_events_local[MEMCG_NR_MEMORY_EVENTS];
+ /*
+ * Hint of reclaim pressure for socket memroy management. Note
+ * that this indicator should NOT be used in legacy cgroup mode
+ * where socket memory is accounted/charged separately.
+ */
unsigned long socket_pressure;
/* Legacy tcp memory accounting */
@@ -1727,8 +1732,8 @@ void mem_cgroup_sk_alloc(struct sock *sk);
void mem_cgroup_sk_free(struct sock *sk);
static inline bool mem_cgroup_under_socket_pressure(struct mem_cgroup *memcg)
{
- if (!cgroup_subsys_on_dfl(memory_cgrp_subsys) && memcg->tcpmem_pressure)
- return true;
+ if (!cgroup_subsys_on_dfl(memory_cgrp_subsys))
+ return !!memcg->tcpmem_pressure;
do {
if (time_before(jiffies, READ_ONCE(memcg->socket_pressure)))
return true;
diff --git a/mm/vmpressure.c b/mm/vmpressure.c
index b52644771cc4..22c6689d9302 100644
--- a/mm/vmpressure.c
+++ b/mm/vmpressure.c
@@ -244,6 +244,14 @@ void vmpressure(gfp_t gfp, struct mem_cgroup *memcg, bool tree,
if (mem_cgroup_disabled())
return;
+ /*
+ * The in-kernel users only care about the reclaim efficiency
+ * for this @memcg rather than the whole subtree, and there
+ * isn't and won't be any in-kernel user in a legacy cgroup.
+ */
+ if (!cgroup_subsys_on_dfl(memory_cgrp_subsys) && !tree)
+ return;
+
vmpr = memcg_to_vmpressure(memcg);
/*
--
2.37.3
