Re: [PATCH v2 4/4] sock: Remove redundant cond of memcg pressure
From: Simon Horman <hidden>
Date: 2023-05-22 12:55:09
Also in:
lkml
On Mon, May 22, 2023 at 03:01:22PM +0800, Abel Wu wrote:
Now with the preivous patch, __sk_mem_raise_allocated() considers
nit: s/preivous/previous/
quoted hunk ↗ jump to hunk
the memory pressure of both global and the socket's memcg on a func- wide level, making the condition of memcg's pressure in question redundant. Signed-off-by: Abel Wu <redacted> --- net/core/sock.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)diff --git a/net/core/sock.c b/net/core/sock.c index 7641d64293af..baccbb58a11a 100644 --- a/net/core/sock.c +++ b/net/core/sock.c@@ -3029,9 +3029,14 @@ int __sk_mem_raise_allocated(struct sock *sk, int size, int amt, int kind) if (sk_has_memory_pressure(sk)) { u64 alloc; - if (!sk_under_memory_pressure(sk)) + if (!sk_under_global_memory_pressure(sk)) return 1; alloc = sk_sockets_allocated_read_positive(sk); + /* + * If under global pressure, allow the sockets that are below + * average memory usage to raise, trying to be fair among all + * the sockets under global constrains. + */
nit: /* Multi-line comments in networking code * look like this. */
if (sk_prot_mem_limits(sk, 2) > alloc * sk_mem_pages(sk->sk_wmem_queued + atomic_read(&sk->sk_rmem_alloc) + -- 2.37.3