From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-5.8 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI, SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED autolearn=no autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 57281C433E0 for ; Thu, 14 Jan 2021 13:08:01 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 1C33623A57 for ; Thu, 14 Jan 2021 13:08:01 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1728963AbhANNH1 (ORCPT ); Thu, 14 Jan 2021 08:07:27 -0500 Received: from mail-40134.protonmail.ch ([185.70.40.134]:11385 "EHLO mail-40134.protonmail.ch" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726704AbhANNH0 (ORCPT ); Thu, 14 Jan 2021 08:07:26 -0500 Date: Thu, 14 Jan 2021 13:06:30 +0000 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=pm.me; s=protonmail; t=1610629598; bh=7AdvX5kGbWrQn/5ozYyBEsZ2bK/E+Czp98mj5nYiQBw=; h=Date:To:From:Cc:Reply-To:Subject:In-Reply-To:References:From; b=jAx7Or27E1NLXwQiwr0kUXI+vh3zGPsz835szUUoZ4ZkQZKJtEh8PeZj/d3sQC2GJ mHgle+HVufNWfn1D7WUd5jxQLOVvJAYal7rCsQZqdnCbwzp1BqECZagkoce6RHNQ08 +0wogp1vXtWIIvsP/l4yukM2pXTcjS4/4wmVuSTmlcFpU66qa8W+aRcVN1oyWpkMnI wyvG0s0IXbKQv2Tb4MdFWp5m8aOaC3SwxX66ktQWjH6J4D1g38yBDD/FgRYNKpQAlh GFoeA/k2M8Htjr3yQPnGOhRewp1eVr2EOl/Xu68Qp/FWQbpWoPAdlYiXac+jZFgQXs SFiWyZiMn9M7A== To: Dmitry Vyukov From: Alexander Lobakin Cc: Alexander Lobakin , Eric Dumazet , "David S. Miller" , Jakub Kicinski , Edward Cree , Jonathan Lemon , Willem de Bruijn , Miaohe Lin , Steffen Klassert , Guillaume Nault , Yadu Kishore , Al Viro , netdev , LKML Reply-To: Alexander Lobakin Subject: Re: [PATCH v2 net-next 2/3] skbuff: (re)use NAPI skb cache on allocation path Message-ID: <20210114130615.9885-1-alobakin@pm.me> In-Reply-To: References: <20210113133523.39205-1-alobakin@pm.me> <20210113133635.39402-1-alobakin@pm.me> <20210113133635.39402-2-alobakin@pm.me> <20210114114046.7272-1-alobakin@pm.me> <20210114124406.9049-1-alobakin@pm.me> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Dmitry Vyukov Date: Thu, 14 Jan 2021 13:51:44 +0100 > On Thu, Jan 14, 2021 at 1:50 PM Dmitry Vyukov wrote: >> >> On Thu, Jan 14, 2021 at 1:44 PM Alexander Lobakin wrote= : >>> >>> From: Dmitry Vyukov >>> Date: Thu, 14 Jan 2021 12:47:31 +0100 >>> >>>> On Thu, Jan 14, 2021 at 12:41 PM Alexander Lobakin wr= ote: >>>>> >>>>> From: Eric Dumazet >>>>> Date: Wed, 13 Jan 2021 15:36:05 +0100 >>>>> >>>>>> On Wed, Jan 13, 2021 at 2:37 PM Alexander Lobakin w= rote: >>>>>>> >>>>>>> Instead of calling kmem_cache_alloc() every time when building a NA= PI >>>>>>> skb, (re)use skbuff_heads from napi_alloc_cache.skb_cache. Previous= ly >>>>>>> this cache was only used for bulk-freeing skbuff_heads consumed via >>>>>>> napi_consume_skb() or __kfree_skb_defer(). >>>>>>> >>>>>>> Typical path is: >>>>>>> - skb is queued for freeing from driver or stack, its skbuff_head >>>>>>> goes into the cache instead of immediate freeing; >>>>>>> - driver or stack requests NAPI skb allocation, an skbuff_head is >>>>>>> taken from the cache instead of allocation. >>>>>>> >>>>>>> Corner cases: >>>>>>> - if it's empty on skb allocation, bulk-allocate the first half; >>>>>>> - if it's full on skb consuming, bulk-wipe the second half. >>>>>>> >>>>>>> Also try to balance its size after completing network softirqs >>>>>>> (__kfree_skb_flush()). >>>>>> >>>>>> I do not see the point of doing this rebalance (especially if we do = not change >>>>>> its name describing its purpose more accurately). >>>>>> >>>>>> For moderate load, we will have a reduced bulk size (typically one o= r two). >>>>>> Number of skbs in the cache is in [0, 64[ , there is really no risk = of >>>>>> letting skbs there for a long period of time. >>>>>> (32 * sizeof(sk_buff) =3D 8192) >>>>>> I would personally get rid of this function completely. >>>>> >>>>> When I had a cache of 128 entries, I had worse results without this >>>>> function. But seems like I forgot to retest when I switched to the >>>>> original size of 64. >>>>> I also thought about removing this function entirely, will test. >>>>> >>>>>> Also it seems you missed my KASAN support request ? >>>>> I guess this is a matter of using kasan_unpoison_range(), we can ask= for help. >>>>> >>>>> I saw your request, but don't see a reason for doing this. >>>>> We are not caching already freed skbuff_heads. They don't get >>>>> kmem_cache_freed before getting into local cache. KASAN poisons >>>>> them no earlier than at kmem_cache_free() (or did I miss someting?). >>>>> heads being cached just get rid of all references and at the moment >>>>> of dropping to the cache they are pretty the same as if they were >>>>> allocated. >>>> >>>> KASAN should not report false positives in this case. >>>> But I think Eric meant preventing false negatives. If we kmalloc 17 >>>> bytes, KASAN will detect out-of-bounds accesses beyond these 17 bytes. >>>> But we put that data into 128-byte blocks, KASAN will miss >>>> out-of-bounds accesses beyond 17 bytes up to 128 bytes. >>>> The same holds for "logical" use-after-frees when object is free, but >>>> not freed into slab. >>>> >>>> An important custom cache should use annotations like >>>> kasan_poison_object_data/kasan_unpoison_range. >>> >>> As I understand, I should >>> kasan_poison_object_data(skbuff_head_cache, skb) and then >>> kasan_unpoison_range(skb, sizeof(*skb)) when putting it into the >>> cache? >> >> I think it's the other way around. It should be _un_poisoned when used. >> If it's fixed size, then unpoison_object_data should be a better fit: >> https://elixir.bootlin.com/linux/v5.11-rc3/source/mm/kasan/common.c#L253 > > Variable-size poisoning/unpoisoning would be needed for the skb data itse= lf: > https://bugzilla.kernel.org/show_bug.cgi?id=3D199055 This cache is for skbuff_heads only, not for the entire skbs. All linear data and frags gets freed before head hits the cache. The cache will store skbuff_heads as if they were freshly allocated by kmem_cache_alloc(). Al