[bpf-next,35/35] perf: don't touch RLIMIT_MEMLOCK

Message ID	20200725000410.3566700-36-guro@fb.com
State	Changes Requested
Delegated to:	BPF Maintainers
Headers	show Return-Path: <bpf-owner@vger.kernel.org> Smtp-Origin-Hostprefix: devvm From: Roman Gushchin <guro@fb.com> Smtp-Origin-Hostname: devvm1096.prn0.facebook.com To: <bpf@vger.kernel.org> CC: <netdev@vger.kernel.org>, Alexei Starovoitov <ast@kernel.org>, Daniel Borkmann <daniel@iogearbox.net>, <kernel-team@fb.com>, <linux-kernel@vger.kernel.org>, Roman Gushchin <guro@fb.com> Smtp-Origin-Cluster: prn0c01 Subject: [PATCH bpf-next 35/35] perf: don't touch RLIMIT_MEMLOCK Date: Fri, 24 Jul 2020 17:04:10 -0700 Message-ID: <20200725000410.3566700-36-guro@fb.com> In-Reply-To: <20200725000410.3566700-1-guro@fb.com> References: <20200725000410.3566700-1-guro@fb.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain Sender: bpf-owner@vger.kernel.org Precedence: bulk
Series	bpf: switch to memcg-based memory accounting \| expand [bpf-next,00/35] bpf: switch to memcg-based memory accounting [bpf-next,01/35] bpf: memcg-based memory accounting for bpf progs [bpf-next,02/35] bpf: memcg-based memory accounting for bpf maps [bpf-next,03/35] bpf: refine memcg-based memory accounting for arraymap maps [bpf-next,04/35] bpf: refine memcg-based memory accounting for cpumap maps [bpf-next,05/35] bpf: memcg-based memory accounting for cgroup storage maps [bpf-next,06/35] bpf: refine memcg-based memory accounting for devmap maps [bpf-next,07/35] bpf: refine memcg-based memory accounting for hashtab maps [bpf-next,08/35] bpf: memcg-based memory accounting for lpm_trie maps [bpf-next,09/35] bpf: memcg-based memory accounting for bpf ringbuffer [bpf-next,10/35] bpf: memcg-based memory accounting for socket storage maps [bpf-next,11/35] bpf: refine memcg-based memory accounting for sockmap maps [bpf-next,12/35] bpf: refine memcg-based memory accounting for xskmap maps [bpf-next,13/35] bpf: eliminate rlimit-based memory accounting for arraymap maps [bpf-next,14/35] bpf: eliminate rlimit-based memory accounting for bpf_struct_ops maps [bpf-next,15/35] bpf: eliminate rlimit-based memory accounting for cpumap maps [bpf-next,16/35] bpf: eliminate rlimit-based memory accounting for cgroup storage maps [bpf-next,17/35] bpf: eliminate rlimit-based memory accounting for devmap maps [bpf-next,18/35] bpf: eliminate rlimit-based memory accounting for hashtab maps [bpf-next,19/35] bpf: eliminate rlimit-based memory accounting for lpm_trie maps [bpf-next,20/35] bpf: eliminate rlimit-based memory accounting for queue_stack_maps maps [bpf-next,21/35] bpf: eliminate rlimit-based memory accounting for reuseport_array maps [bpf-next,22/35] bpf: eliminate rlimit-based memory accounting for bpf ringbuffer [bpf-next,23/35] bpf: eliminate rlimit-based memory accounting for sock_map maps [bpf-next,24/35] bpf: eliminate rlimit-based memory accounting for stackmap maps [bpf-next,25/35] bpf: eliminate rlimit-based memory accounting for socket storage maps [bpf-next,26/35] bpf: eliminate rlimit-based memory accounting for xskmap maps [bpf-next,27/35] bpf: eliminate rlimit-based memory accounting infra for bpf maps [bpf-next,28/35] bpf: eliminate rlimit-based memory accounting for bpf progs [bpf-next,29/35] bpf: libbpf: cleanup RLIMIT_MEMLOCK usage [bpf-next,30/35] bpf: bpftool: do not touch RLIMIT_MEMLOCK [bpf-next,31/35] bpf: runqslower: don't touch RLIMIT_MEMLOCK [bpf-next,32/35] bpf: selftests: delete bpf_rlimit.h [bpf-next,33/35] bpf: selftests: don't touch RLIMIT_MEMLOCK [bpf-next,34/35] bpf: samples: do not touch RLIMIT_MEMLOCK [bpf-next,35/35] perf: don't touch RLIMIT_MEMLOCK

Message ID

20200725000410.3566700-36-guro@fb.com

State

Changes Requested

Delegated to:

BPF Maintainers

Headers

Smtp-Origin-Hostprefix: devvm
From: Roman Gushchin <guro@fb.com>
Smtp-Origin-Hostname: devvm1096.prn0.facebook.com
To: <bpf@vger.kernel.org>
CC: <netdev@vger.kernel.org>, Alexei Starovoitov <ast@kernel.org>,
	Daniel Borkmann <daniel@iogearbox.net>, <kernel-team@fb.com>,
	<linux-kernel@vger.kernel.org>, Roman Gushchin <guro@fb.com>
Smtp-Origin-Cluster: prn0c01
Subject: [PATCH bpf-next 35/35] perf: don't touch RLIMIT_MEMLOCK
Date: Fri, 24 Jul 2020 17:04:10 -0700
Message-ID: <20200725000410.3566700-36-guro@fb.com>
In-Reply-To: <20200725000410.3566700-1-guro@fb.com>
References: <20200725000410.3566700-1-guro@fb.com>
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain
Sender: bpf-owner@vger.kernel.org
Precedence: bulk

Series

bpf: switch to memcg-based memory accounting | expand

Commit Message

Roman Gushchin July 25, 2020, 12:04 a.m. UTC

Since bpf stopped using memlock rlimit to limit the memory usage,
there is no more reason for perf to alter its own limit.

Signed-off-by: Roman Gushchin <guro@fb.com>
---
 tools/perf/builtin-trace.c      | 10 ----------
 tools/perf/tests/builtin-test.c |  6 ------
 tools/perf/util/Build           |  1 -
 tools/perf/util/rlimit.c        | 29 -----------------------------
 tools/perf/util/rlimit.h        |  6 ------
 5 files changed, 52 deletions(-)
 delete mode 100644 tools/perf/util/rlimit.c
 delete mode 100644 tools/perf/util/rlimit.h

diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index 4cbb64edc998..3d6a98a12537 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -19,7 +19,6 @@ 
 #include <api/fs/tracing_path.h>
 #include <bpf/bpf.h>
 #include "util/bpf_map.h"
-#include "util/rlimit.h"
 #include "builtin.h"
 #include "util/cgroup.h"
 #include "util/color.h"
@@ -4838,15 +4837,6 @@  int cmd_trace(int argc, const char **argv)
 		goto out;
 	}
 
-	/*
-	 * Parsing .perfconfig may entail creating a BPF event, that may need
-	 * to create BPF maps, so bump RLIM_MEMLOCK as the default 64K setting
-	 * is too small. This affects just this process, not touching the
-	 * global setting. If it fails we'll get something in 'perf trace -v'
-	 * to help diagnose the problem.
-	 */
-	rlimit__bump_memlock();
-
 	err = perf_config(trace__config, &trace);
 	if (err)
 		goto out;
diff --git a/tools/perf/tests/builtin-test.c b/tools/perf/tests/builtin-test.c
index da5b6cc23f25..e4efbba8202b 100644
--- a/tools/perf/tests/builtin-test.c
+++ b/tools/perf/tests/builtin-test.c
@@ -22,7 +22,6 @@ 
 #include <subcmd/parse-options.h>
 #include "string2.h"
 #include "symbol.h"
-#include "util/rlimit.h"
 #include <linux/kernel.h>
 #include <linux/string.h>
 #include <subcmd/exec-cmd.h>
@@ -794,11 +793,6 @@  int cmd_test(int argc, const char **argv)
 
 	if (skip != NULL)
 		skiplist = intlist__new(skip);
-	/*
-	 * Tests that create BPF maps, for instance, need more than the 64K
-	 * default:
-	 */
-	rlimit__bump_memlock();
 
 	return __cmd_test(argc, argv, skiplist);
 }
diff --git a/tools/perf/util/Build b/tools/perf/util/Build
index 8d18380ecd10..4902cd3b3b58 100644
--- a/tools/perf/util/Build
+++ b/tools/perf/util/Build
@@ -26,7 +26,6 @@  perf-y += parse-events.o
 perf-y += perf_regs.o
 perf-y += path.o
 perf-y += print_binary.o
-perf-y += rlimit.o
 perf-y += argv_split.o
 perf-y += rbtree.o
 perf-y += libstring.o
diff --git a/tools/perf/util/rlimit.c b/tools/perf/util/rlimit.c
deleted file mode 100644
index 13521d392a22..000000000000
--- a/tools/perf/util/rlimit.c
+++ /dev/null
@@ -1,29 +0,0 @@ 
-/* SPDX-License-Identifier: LGPL-2.1 */
-
-#include "util/debug.h"
-#include "util/rlimit.h"
-#include <sys/time.h>
-#include <sys/resource.h>
-
-/*
- * Bump the memlock so that we can get bpf maps of a reasonable size,
- * like the ones used with 'perf trace' and with 'perf test bpf',
- * improve this to some specific request if needed.
- */
-void rlimit__bump_memlock(void)
-{
-	struct rlimit rlim;
-
-	if (getrlimit(RLIMIT_MEMLOCK, &rlim) == 0) {
-		rlim.rlim_cur *= 4;
-		rlim.rlim_max *= 4;
-
-		if (setrlimit(RLIMIT_MEMLOCK, &rlim) < 0) {
-			rlim.rlim_cur /= 2;
-			rlim.rlim_max /= 2;
-
-			if (setrlimit(RLIMIT_MEMLOCK, &rlim) < 0)
-				pr_debug("Couldn't bump rlimit(MEMLOCK), failures may take place when creating BPF maps, etc\n");
-		}
-	}
-}
diff --git a/tools/perf/util/rlimit.h b/tools/perf/util/rlimit.h
deleted file mode 100644
index 9f59d8e710a3..000000000000
--- a/tools/perf/util/rlimit.h
+++ /dev/null
@@ -1,6 +0,0 @@ 
-#ifndef __PERF_RLIMIT_H_
-#define __PERF_RLIMIT_H_
-/* SPDX-License-Identifier: LGPL-2.1 */
-
-void rlimit__bump_memlock(void);
-#endif // __PERF_RLIMIT_H_

[bpf-next,35/35] perf: don't touch RLIMIT_MEMLOCK

Commit Message

Patch