From: Jakub Kicinski Subject: Re: [PATCH net-next] net-memcg: pass in gfp_t mask to mem_cgroup_charge_skmem() Date: Wed, 12 Oct 2022 17:38:25 -0700 Message-ID: <20221012173825.45d6fbf2@kernel.org> References: <20210817194003.2102381-1-weiwan@google.com> <20221012163300.795e7b86@kernel.org> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1665621507; bh=fIBx0fk/i8Gefd9ZrizLeDWC35/uKkbdvMp4L+ilGNU=; h=Date:From:To:Cc:Subject:In-Reply-To:References:From; b=eNs9F9rFjafI66l1WW5TTwHbmhFJfSVZnqqopbkioUXtDKPK3ZXJ5vS3L3IWP7hEg 3mpiVdYUZhVOW1o8sirKSi/2D8i355P6m6fVomdolTfp+Flio7KfrTYPtJqpGDQUtB y0SIq1IsGnlPN9et7Rern9nsc3exKuFw0aqCtCtPxQF5M86EPfKdeJgu78+jctC4RH kViRiN6Elm0D4SWR4uRwsgsX/ufs0VoLz7wF8QlKNo8ooSjpPJy4eNXoW5d3oelkpz v2nNtb23iJq83ftiFPOcfb/QX7ToN8YGHxfj+llW+wIaCbm8TEY8mGrSXoP6tGT3Xc EjpdQTw0eq/CA== In-Reply-To: List-ID: Content-Type: text/plain; charset="us-ascii" To: Shakeel Butt Cc: Wei Wang , Eric Dumazet , netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, "David S . Miller" , cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org, Roman Gushchin On Wed, 12 Oct 2022 17:17:38 -0700 Shakeel Butt wrote: > Did the revert of this patch fix the issue you are seeing? The reason > I am asking is because this patch should not change the behavior. > Actually someone else reported the similar issue for UDP RX at [1] and > they tested the revert as well. The revert did not fix the issue for > them. > > Wei has a better explanation at [2] why this patch is not the cause > for these issues. We're talking TCP here, to be clear. I haven't tested a revert, yet (not that easy to test with a real workload) but I'm relatively confident the change did introduce an "unforced" call, specifically this bit: @@ -2728,10 +2728,12 @@ int __sk_mem_raise_allocated(struct sock *sk, int size, int amt, int kind) { struct proto *prot = sk->sk_prot; long allocated = sk_memory_allocated_add(sk, amt); + bool memcg_charge = mem_cgroup_sockets_enabled && sk->sk_memcg; bool charged = true; - if (mem_cgroup_sockets_enabled && sk->sk_memcg && - !(charged = mem_cgroup_charge_skmem(sk->sk_memcg, amt))) + if (memcg_charge && + !(charged = mem_cgroup_charge_skmem(sk->sk_memcg, amt, + gfp_memcg_charge()))) where gfp_memcg_charge() is GFP_NOWAIT in softirq. The above gets called from (inverted stack): tcp_data_queue() tcp_try_rmem_schedule(sk, skb, skb->truesize) tcp_try_rmem_schedule() sk_rmem_schedule() __sk_mem_schedule() __sk_mem_raise_allocated() Is my confidence unjustified? :)