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=-23.3 required=3.0 tests=BAYES_00,DKIMWL_WL_MED, DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_CR_TRAILER,INCLUDES_PATCH,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS, URIBL_BLOCKED,USER_IN_DEF_DKIM_WL autolearn=ham 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 7E4E7C4338F for ; Mon, 16 Aug 2021 22:46:55 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 503CD60F22 for ; Mon, 16 Aug 2021 22:46:55 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S232495AbhHPWr0 (ORCPT ); Mon, 16 Aug 2021 18:47:26 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:39276 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S232473AbhHPWrZ (ORCPT ); Mon, 16 Aug 2021 18:47:25 -0400 Received: from mail-yb1-xb49.google.com (mail-yb1-xb49.google.com [IPv6:2607:f8b0:4864:20::b49]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 170E0C061764 for ; Mon, 16 Aug 2021 15:46:53 -0700 (PDT) Received: by mail-yb1-xb49.google.com with SMTP id i32-20020a25b2200000b02904ed415d9d84so18288878ybj.0 for ; Mon, 16 Aug 2021 15:46:53 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20161025; h=date:in-reply-to:message-id:mime-version:references:subject:from:to :cc; bh=go5gEr/LK1ikMjcMOeCWMGgc9H7TA/C/hnBeQvGa1/s=; b=uJYQ3qp1mzmZbNi0iFWt9KGYsrFJHOAvGNqOz5cdMFsESdKF13hc6bITYbqdtObkIk Jovb8zjwNW1AQQcPumXP7YQZ/aUBE3jhZHbI54iCCj/BH2VH+6ZLx96KUplPv5n4NgFS oHDovPjuhfCrhis55Bg45hZG8uBlTDBtT+lTnO40Jru5HlYYuD/liep1TKHoMd80KCh7 z7zG55c//RryYp4IL9UvcvYQimk05RVPMzQho+crhmShSUxvynVJy7tLhyErmCILhPOp IIjzVBPlauf0JH6PWsJ28HCYlIPFcE5gMewWF9OWTfeZZQHi884txCrK41W3wGuHt2VC 7a+Q== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:date:in-reply-to:message-id:mime-version :references:subject:from:to:cc; bh=go5gEr/LK1ikMjcMOeCWMGgc9H7TA/C/hnBeQvGa1/s=; b=lU6TArH5a9XOzSq1MbBl0tQv8lz9kBCLMbUFpdzRWr4sYVv+hDsPEzJuimS+XetBZU RYbEcVco6+KU+97/SQpPcu9PWYNpnPh8qfMoyLPTKV8MivOK4gfmBuGiS8e/drJCpXcB xVKxUhpiqGzGyvXMU2SwaH/LbjUhWEr0DqvKJsoEhEXbUez78cFKpWfYWGMTuPtlRtWC tTsWfIpi+xxaQMwIQ+Tz1wsvZDg+K2uUO2hY6niNope/XBGAejGEVZL1cHwOJro8IJvS dpa52dzld/sqzgqjZTNCAZLg4e4Gd9iBbXhjGqUPCCPEMDyZ1wOWfTGk18mPqbupRi5U 4vNg== X-Gm-Message-State: AOAM530izjcvyCDb4pCL+tSQpx+ICEz692U/SNOoyymbsuN56e6v32la PVjycj2F5LL39/9wjJ1lBoxtpxY= X-Google-Smtp-Source: ABdhPJwO+8jkIUXp40zgAdtCUJmwbqGpOfo9f6s4gIzmSiABgsE2Tfyk6CI/XUbV93YAlWsS6yI4T5I= X-Received: from sdf2.svl.corp.google.com ([2620:15c:2c4:201:afa2:caa8:d56a:b355]) (user=sdf job=sendgmr) by 2002:a25:c08a:: with SMTP id c132mr420862ybf.511.1629154012379; Mon, 16 Aug 2021 15:46:52 -0700 (PDT) Date: Mon, 16 Aug 2021 15:46:50 -0700 In-Reply-To: Message-Id: Mime-Version: 1.0 References: <20210816164832.1743675-1-sdf@google.com> <1b3cb059-9ecb-a0c9-3c99-805788088d09@iogearbox.net> Subject: Re: [PATCH bpf-next] bpf: use kvmalloc in map_lookup_elem From: sdf@google.com To: Andrii Nakryiko Cc: Daniel Borkmann , Networking , bpf , Alexei Starovoitov , Andrii Nakryiko Content-Type: text/plain; charset="UTF-8"; format=flowed; delsp=yes Precedence: bulk List-ID: X-Mailing-List: bpf@vger.kernel.org On 08/16, Andrii Nakryiko wrote: > On Mon, Aug 16, 2021 at 2:43 PM Daniel Borkmann > wrote: > > > > On 8/16/21 6:48 PM, Stanislav Fomichev wrote: > > > Use kvmalloc/kvfree for temporary value when looking up a map. > > > kmalloc might not be sufficient for percpu maps where the value is > big. > > > > > > Can be reproduced with netcnt test on qemu with "-smp 255". > > > > > > Signed-off-by: Stanislav Fomichev > > > --- > > > kernel/bpf/syscall.c | 4 ++-- > > > 1 file changed, 2 insertions(+), 2 deletions(-) > > > > > > diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c > > > index 9a2068e39d23..ae0b1c1c8ece 100644 > > > --- a/kernel/bpf/syscall.c > > > +++ b/kernel/bpf/syscall.c > > > @@ -1076,7 +1076,7 @@ static int map_lookup_elem(union bpf_attr *attr) > > > value_size = bpf_map_value_size(map); > > > > > > err = -ENOMEM; > > > - value = kmalloc(value_size, GFP_USER | __GFP_NOWARN); > > > + value = kvmalloc(value_size, GFP_USER | __GFP_NOWARN); > > > if (!value) > > > goto free_key; > > > > What about other cases like map_update_elem(), shouldn't they be adapted > > similarly? > And in the same vein (with keys potentially being big as well), should > we switch __bpf_copy_key() to use vmemdup_user() instead of > memdup_user()? Those are good questions :-) I'm assuming that whatever is doing kmalloc on top of bpf_map_value_size() should definitely use kvmalloc since bpf_map_value_size() can blow up the value size. (will resend) For __bpf_copy_key I'm less sure, but it might be a good idea as well. Let me try to look at bit more into it, but feels like there shouldn't be any downsides?