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 Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id F24D6C433F5 for ; Thu, 6 Oct 2022 10:09:53 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230511AbiJFKJw (ORCPT ); Thu, 6 Oct 2022 06:09:52 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:34384 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231179AbiJFKJv (ORCPT ); Thu, 6 Oct 2022 06:09:51 -0400 Received: from szxga01-in.huawei.com (szxga01-in.huawei.com [45.249.212.187]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 85031BC0E; Thu, 6 Oct 2022 03:09:48 -0700 (PDT) Received: from kwepemi500013.china.huawei.com (unknown [172.30.72.54]) by szxga01-in.huawei.com (SkyGuard) with ESMTP id 4MjnB64ZwCzlXKL; Thu, 6 Oct 2022 18:05:18 +0800 (CST) Received: from [10.67.111.192] (10.67.111.192) by kwepemi500013.china.huawei.com (7.221.188.120) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.31; Thu, 6 Oct 2022 18:09:44 +0800 Message-ID: Date: Thu, 6 Oct 2022 18:09:44 +0800 MIME-Version: 1.0 User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:91.0) Gecko/20100101 Thunderbird/91.9.0 Subject: Re: [PATCH bpf-next v2 0/4] Add ftrace direct call for arm64 Content-Language: en-US To: Steven Rostedt , Florent Revest CC: Mark Rutland , Catalin Marinas , Daniel Borkmann , , , , Will Deacon , Jean-Philippe Brucker , Ingo Molnar , Oleg Nesterov , Alexei Starovoitov , Andrii Nakryiko , Martin KaFai Lau , Song Liu , Yonghong Song , John Fastabend , KP Singh , Stanislav Fomichev , Hao Luo , Jiri Olsa , Zi Shen Lim , Pasha Tatashin , Ard Biesheuvel , Marc Zyngier , Guo Ren , Masami Hiramatsu References: <20220913162732.163631-1-xukuohai@huaweicloud.com> <970a25e4-9b79-9e0c-b338-ed1a934f2770@huawei.com> <2cb606b4-aa8b-e259-cdfd-1bfc61fd7c44@huawei.com> <7f34d333-3b2a-aea5-f411-d53be2c46eee@huawei.com> <20221005110707.55bd9354@gandalf.local.home> <20221005113019.18aeda76@gandalf.local.home> From: Xu Kuohai In-Reply-To: <20221005113019.18aeda76@gandalf.local.home> Content-Type: text/plain; charset="UTF-8"; format=flowed Content-Transfer-Encoding: 7bit X-Originating-IP: [10.67.111.192] X-ClientProxiedBy: dggems702-chm.china.huawei.com (10.3.19.179) To kwepemi500013.china.huawei.com (7.221.188.120) X-CFilter-Loop: Reflected Precedence: bulk List-ID: X-Mailing-List: bpf@vger.kernel.org On 10/5/2022 11:30 PM, Steven Rostedt wrote: > On Wed, 5 Oct 2022 17:10:33 +0200 > Florent Revest wrote: > >> On Wed, Oct 5, 2022 at 5:07 PM Steven Rostedt wrote: >>> >>> On Wed, 5 Oct 2022 22:54:15 +0800 >>> Xu Kuohai wrote: >>> >>>> 1.3 attach bpf prog with with direct call, bpftrace -e 'kfunc:vfs_write {}' >>>> >>>> # dd if=/dev/zero of=/dev/null count=1000000 >>>> 1000000+0 records in >>>> 1000000+0 records out >>>> 512000000 bytes (512 MB, 488 MiB) copied, 1.72973 s, 296 MB/s >>>> >>>> >>>> 1.4 attach bpf prog with with indirect call, bpftrace -e 'kfunc:vfs_write {}' >>>> >>>> # dd if=/dev/zero of=/dev/null count=1000000 >>>> 1000000+0 records in >>>> 1000000+0 records out >>>> 512000000 bytes (512 MB, 488 MiB) copied, 1.99179 s, 257 MB/s >> >> Thanks for the measurements Xu! >> >>> Can you show the implementation of the indirect call you used? >> >> Xu used my development branch here >> https://github.com/FlorentRevest/linux/commits/fprobe-min-args > > That looks like it could be optimized quite a bit too. > > Specifically this part: > > static bool bpf_fprobe_entry(struct fprobe *fp, unsigned long ip, struct ftrace_regs *regs, void *private) > { > struct bpf_fprobe_call_context *call_ctx = private; > struct bpf_fprobe_context *fprobe_ctx = fp->ops.private; > struct bpf_tramp_links *links = fprobe_ctx->links; > struct bpf_tramp_links *fentry = &links[BPF_TRAMP_FENTRY]; > struct bpf_tramp_links *fmod_ret = &links[BPF_TRAMP_MODIFY_RETURN]; > struct bpf_tramp_links *fexit = &links[BPF_TRAMP_FEXIT]; > int i, ret; > > memset(&call_ctx->ctx, 0, sizeof(call_ctx->ctx)); > call_ctx->ip = ip; > for (i = 0; i < fprobe_ctx->nr_args; i++) > call_ctx->args[i] = ftrace_regs_get_argument(regs, i); > > for (i = 0; i < fentry->nr_links; i++) > call_bpf_prog(fentry->links[i], &call_ctx->ctx, call_ctx->args); > > call_ctx->args[fprobe_ctx->nr_args] = 0; > for (i = 0; i < fmod_ret->nr_links; i++) { > ret = call_bpf_prog(fmod_ret->links[i], &call_ctx->ctx, > call_ctx->args); > > if (ret) { > ftrace_regs_set_return_value(regs, ret); > ftrace_override_function_with_return(regs); > > bpf_fprobe_exit(fp, ip, regs, private); > return false; > } > } > > return fexit->nr_links; > } > > There's a lot of low hanging fruit to speed up there. I wouldn't be too > fast to throw out this solution if it hasn't had the care that direct calls > have had to speed that up. > > For example, trampolines currently only allow to attach to functions with 6 > parameters or less (3 on x86_32). You could make 7 specific callbacks, with > zero to 6 parameters, and unroll the argument loop. > > Would also be interesting to run perf to see where the overhead is. There > may be other locations to work on to make it almost as fast as direct > callers without the other baggage. > There is something wrong with my pi4 perf, I'll send the perf report after I fix it. > -- Steve > >> >> As it stands, the performance impact of the fprobe based >> implementation would be too high for us. I wonder how much Mark's idea >> here https://git.kernel.org/pub/scm/linux/kernel/git/mark/linux.git/log/?h=arm64/ftrace/per-callsite-ops >> would help but it doesn't work right now. > > > .