{"id":815203,"url":"http://patchwork.ozlabs.org/api/patches/815203/?format=json","web_url":"http://patchwork.ozlabs.org/project/netdev/patch/20170918230553.1624357-4-yhs@fb.com/","project":{"id":7,"url":"http://patchwork.ozlabs.org/api/projects/7/?format=json","name":"Linux network development","link_name":"netdev","list_id":"netdev.vger.kernel.org","list_email":"netdev@vger.kernel.org","web_url":null,"scm_url":null,"webscm_url":null,"list_archive_url":"","list_archive_url_format":"","commit_url_format":""},"msgid":"<20170918230553.1624357-4-yhs@fb.com>","list_archive_url":null,"date":"2017-09-18T23:05:52","name":"[net-next,v3,3/4] bpf: add helper bpf_perf_prog_read_value","commit_ref":null,"pull_url":null,"state":"superseded","archived":true,"hash":"2cb56296b4e07dbd191165ead90be43e4f9717ca","submitter":{"id":71628,"url":"http://patchwork.ozlabs.org/api/people/71628/?format=json","name":"Yonghong Song","email":"yhs@fb.com"},"delegate":{"id":34,"url":"http://patchwork.ozlabs.org/api/users/34/?format=json","username":"davem","first_name":"David","last_name":"Miller","email":"davem@davemloft.net"},"mbox":"http://patchwork.ozlabs.org/project/netdev/patch/20170918230553.1624357-4-yhs@fb.com/mbox/","series":[{"id":3748,"url":"http://patchwork.ozlabs.org/api/series/3748/?format=json","web_url":"http://patchwork.ozlabs.org/project/netdev/list/?series=3748","date":"2017-09-18T23:05:50","name":"bpf: add two helpers to read perf event enabled/running time","version":3,"mbox":"http://patchwork.ozlabs.org/series/3748/mbox/"}],"comments":"http://patchwork.ozlabs.org/api/patches/815203/comments/","check":"pending","checks":"http://patchwork.ozlabs.org/api/patches/815203/checks/","tags":{},"related":[],"headers":{"Return-Path":"","X-Original-To":"patchwork-incoming@ozlabs.org","Delivered-To":"patchwork-incoming@ozlabs.org","Authentication-Results":["ozlabs.org;\n\tspf=none (mailfrom) smtp.mailfrom=vger.kernel.org\n\t(client-ip=209.132.180.67; helo=vger.kernel.org;\n\tenvelope-from=netdev-owner@vger.kernel.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tunprotected) header.d=fb.com header.i=@fb.com header.b=\"RvSMBlAr\";\n\tdkim-atps=neutral"],"Received":["from vger.kernel.org (vger.kernel.org [209.132.180.67])\n\tby ozlabs.org (Postfix) with ESMTP id 3xx1pD4P3Lz9s4q\n\tfor ;\n\tTue, 19 Sep 2017 09:06:12 +1000 (AEST)","(majordomo@vger.kernel.org) by vger.kernel.org via listexpand\n\tid S1751431AbdIRXGG (ORCPT );\n\tMon, 18 Sep 2017 19:06:06 -0400","from mx0b-00082601.pphosted.com ([67.231.153.30]:37433 \"EHLO\n\tmx0b-00082601.pphosted.com\" rhost-flags-OK-OK-OK-OK)\n\tby vger.kernel.org with ESMTP id S1751300AbdIRXFz (ORCPT\n\t); Mon, 18 Sep 2017 19:05:55 -0400","from pps.filterd (m0109332.ppops.net [127.0.0.1])\n\tby mx0a-00082601.pphosted.com (8.16.0.21/8.16.0.21) with SMTP id\n\tv8IN3fXb031772\n\tfor ; Mon, 18 Sep 2017 16:05:55 -0700","from mail.thefacebook.com ([199.201.64.23])\n\tby mx0a-00082601.pphosted.com with ESMTP id 2d2jfw1m8g-3\n\t(version=TLSv1 cipher=ECDHE-RSA-AES256-SHA bits=256 verify=NOT)\n\tfor ; Mon, 18 Sep 2017 16:05:55 -0700","from mx-out.facebook.com (192.168.52.123) by\n\tPRN-CHUB10.TheFacebook.com (192.168.16.20) with Microsoft SMTP Server\n\tid 14.3.319.2; Mon, 18 Sep 2017 16:05:53 -0700","by devbig474.prn1.facebook.com (Postfix, from userid 128203) id\n\t446B446E04AE; Mon, 18 Sep 2017 16:05:53 -0700 (PDT)"],"DKIM-Signature":"v=1; a=rsa-sha256; c=relaxed/relaxed; d=fb.com;\n\th=from : to : cc : subject\n\t: date : message-id : in-reply-to : references : mime-version :\n\tcontent-type; s=facebook;\n\tbh=CgLc7V6LZkEWdRMKO75eestzlLIzuDS50Cn7J/PcaVw=; \n\tb=RvSMBlArpDNepFWZB+5vH0ClqJQh4Ax50ThXrKyO2AVqA+60uZ3oHjUgFy5SqsHJw0LE\n\twwnFX4ymErLV5tSMccYJBBI6JRNCJ+p3BeB4h4ccbgumcRZeDgkedOjyZLbcArMtXVZ0\n\t1xGw5AckTobbSp3Qp0kKnfNoSBhJoewQjpg= ","Smtp-Origin-Hostprefix":"devbig","From":"Yonghong Song ","Smtp-Origin-Hostname":"devbig474.prn1.facebook.com","To":", , ,\n\t, ","CC":"","Smtp-Origin-Cluster":"prn1c29","Subject":"[PATCH net-next v3 3/4] bpf: add helper bpf_perf_prog_read_value","Date":"Mon, 18 Sep 2017 16:05:52 -0700","Message-ID":"<20170918230553.1624357-4-yhs@fb.com>","X-Mailer":"git-send-email 2.9.5","In-Reply-To":"<20170918230553.1624357-1-yhs@fb.com>","References":"<20170918230553.1624357-1-yhs@fb.com>","X-FB-Internal":["Safe","Safe"],"MIME-Version":"1.0","Content-Type":"text/plain","X-Proofpoint-Spam-Reason":"safe","X-Proofpoint-Virus-Version":"vendor=fsecure engine=2.50.10432:, ,\n\tdefinitions=2017-09-18_11:, , signatures=0","Sender":"netdev-owner@vger.kernel.org","Precedence":"bulk","List-ID":"","X-Mailing-List":"netdev@vger.kernel.org"},"content":"This patch adds helper bpf_perf_prog_read_cvalue for perf event based bpf\nprograms, to read event counter and enabled/running time.\nThe enabled/running time is accumulated since the perf event open.\n\nThe typical use case for perf event based bpf program is to attach itself\nto a single event. In such cases, if it is desirable to get scaling factor\nbetween two bpf invocations, users can can save the time values in a map,\nand use the value from the map and the current value to calculate\nthe scaling factor.\n\nSigned-off-by: Yonghong Song \n---\n include/linux/perf_event.h | 1 +\n include/uapi/linux/bpf.h | 8 ++++++++\n kernel/events/core.c | 1 +\n kernel/trace/bpf_trace.c | 23 +++++++++++++++++++++++\n 4 files changed, 33 insertions(+)","diff":"diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h\nindex 13f08ee..5ff3055 100644\n--- a/include/linux/perf_event.h\n+++ b/include/linux/perf_event.h\n@@ -806,6 +806,7 @@ struct perf_output_handle {\n struct bpf_perf_event_data_kern {\n \tstruct pt_regs *regs;\n \tstruct perf_sample_data *data;\n+\tstruct perf_event *event;\n };\n \n #ifdef CONFIG_CGROUP_PERF\ndiff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h\nindex 2c68b9e..ba77022 100644\n--- a/include/uapi/linux/bpf.h\n+++ b/include/uapi/linux/bpf.h\n@@ -590,6 +590,13 @@ union bpf_attr {\n * @buf: buf to fill\n * @buf_size: size of the buf\n * Return: 0 on success or negative error code\n+ *\n+ * int bpf_perf_prog_read_value(ctx, buf, buf_size)\n+ * read perf prog attached perf event counter and enabled/running time\n+ * @ctx: pointer to ctx\n+ * @buf: buf to fill\n+ * @buf_size: size of the buf\n+ * Return : 0 on success or negative error code\n */\n #define __BPF_FUNC_MAPPER(FN)\t\t\\\n \tFN(unspec),\t\t\t\\\n@@ -647,6 +654,7 @@ union bpf_attr {\n \tFN(sk_redirect_map),\t\t\\\n \tFN(sock_map_update),\t\t\\\n \tFN(perf_event_read_value),\t\t\\\n+\tFN(perf_prog_read_value),\t\t\\\n \n /* integer value in 'imm' field of BPF_CALL instruction selects which helper\n * function eBPF program intends to call\ndiff --git a/kernel/events/core.c b/kernel/events/core.c\nindex 2d5bbe5..d039086 100644\n--- a/kernel/events/core.c\n+++ b/kernel/events/core.c\n@@ -8081,6 +8081,7 @@ static void bpf_overflow_handler(struct perf_event *event,\n \tstruct bpf_perf_event_data_kern ctx = {\n \t\t.data = data,\n \t\t.regs = regs,\n+\t\t.event = event,\n \t};\n \tint ret = 0;\n \ndiff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c\nindex 39ce5d9..596b5c9 100644\n--- a/kernel/trace/bpf_trace.c\n+++ b/kernel/trace/bpf_trace.c\n@@ -603,6 +603,18 @@ BPF_CALL_3(bpf_get_stackid_tp, void *, tp_buff, struct bpf_map *, map,\n \t\t\t flags, 0, 0);\n }\n \n+BPF_CALL_3(bpf_perf_prog_read_value_tp, void *, ctx, struct bpf_perf_event_value *,\n+\tbuf, u32, size)\n+{\n+\tstruct bpf_perf_event_data_kern *kctx = (struct bpf_perf_event_data_kern *)ctx;\n+\n+\tif (size != sizeof(struct bpf_perf_event_value))\n+\t\treturn -EINVAL;\n+\n+\treturn perf_event_read_local(kctx->event, &buf->counter, &buf->enabled,\n+\t\t\t\t &buf->running);\n+}\n+\n static const struct bpf_func_proto bpf_get_stackid_proto_tp = {\n \t.func\t\t= bpf_get_stackid_tp,\n \t.gpl_only\t= true,\n@@ -612,6 +624,15 @@ static const struct bpf_func_proto bpf_get_stackid_proto_tp = {\n \t.arg3_type\t= ARG_ANYTHING,\n };\n \n+static const struct bpf_func_proto bpf_perf_prog_read_value_proto_tp = {\n+ .func = bpf_perf_prog_read_value_tp,\n+ .gpl_only = true,\n+ .ret_type = RET_INTEGER,\n+ .arg1_type = ARG_PTR_TO_CTX,\n+ .arg2_type = ARG_PTR_TO_UNINIT_MEM,\n+ .arg3_type = ARG_CONST_SIZE,\n+};\n+\n static const struct bpf_func_proto *tp_prog_func_proto(enum bpf_func_id func_id)\n {\n \tswitch (func_id) {\n@@ -619,6 +640,8 @@ static const struct bpf_func_proto *tp_prog_func_proto(enum bpf_func_id func_id)\n \t\treturn &bpf_perf_event_output_proto_tp;\n \tcase BPF_FUNC_get_stackid:\n \t\treturn &bpf_get_stackid_proto_tp;\n+\tcase BPF_FUNC_perf_prog_read_value:\n+\t\treturn &bpf_perf_prog_read_value_proto_tp;\n \tdefault:\n \t\treturn tracing_func_proto(func_id);\n \t}\n","prefixes":["net-next","v3","3/4"]}