2023-10-16 13:29:07

by Abel Wu

[permalink] [raw]
Subject: [PATCH net-next v2 2/3] sock: Doc behaviors for pressure heurisitics

There are now two accounting infrastructures for skmem, while the
heuristics in __sk_mem_raise_allocated() were actually introduced
before memcg was born.

Add some comments to clarify whether they can be applied to both
infrastructures or not.

Suggested-by: Shakeel Butt <[email protected]>
Signed-off-by: Abel Wu <[email protected]>
---
net/core/sock.c | 14 +++++++++++++-
1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/net/core/sock.c b/net/core/sock.c
index 43842520db86..9f969e3c2ddf 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -3067,7 +3067,14 @@ int __sk_mem_raise_allocated(struct sock *sk, int size, int amt, int kind)
if (allocated > sk_prot_mem_limits(sk, 2))
goto suppress_allocation;

- /* guarantee minimum buffer size under pressure */
+ /* Guarantee minimum buffer size under pressure (either global
+ * or memcg) to make sure features described in RFC 7323 (TCP
+ * Extensions for High Performance) work properly.
+ *
+ * This rule does NOT stand when exceeds global or memcg's hard
+ * limit, or else a DoS attack can be taken place by spawning
+ * lots of sockets whose usage are under minimum buffer size.
+ */
if (kind == SK_MEM_RECV) {
if (atomic_read(&sk->sk_rmem_alloc) < sk_get_rmem0(sk, prot))
return 1;
@@ -3088,6 +3095,11 @@ int __sk_mem_raise_allocated(struct sock *sk, int size, int amt, int kind)

if (!sk_under_memory_pressure(sk))
return 1;
+
+ /* Try to be fair among all the sockets under global
+ * pressure by allowing the ones that below average
+ * usage to raise.
+ */
alloc = sk_sockets_allocated_read_positive(sk);
if (sk_prot_mem_limits(sk, 2) > alloc *
sk_mem_pages(sk->sk_wmem_queued +
--
2.37.3


2023-10-16 15:51:53

by Shakeel Butt

[permalink] [raw]
Subject: Re: [PATCH net-next v2 2/3] sock: Doc behaviors for pressure heurisitics

On Mon, Oct 16, 2023 at 6:28 AM Abel Wu <[email protected]> wrote:
>
> There are now two accounting infrastructures for skmem, while the
> heuristics in __sk_mem_raise_allocated() were actually introduced
> before memcg was born.
>
> Add some comments to clarify whether they can be applied to both
> infrastructures or not.
>
> Suggested-by: Shakeel Butt <[email protected]>
> Signed-off-by: Abel Wu <[email protected]>

Acked-by: Shakeel Butt <[email protected]>