Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754060Ab2KNGpD (ORCPT ); Wed, 14 Nov 2012 01:45:03 -0500 Received: from terminus.zytor.com ([198.137.202.10]:43276 "EHLO terminus.zytor.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751434Ab2KNGpB (ORCPT ); Wed, 14 Nov 2012 01:45:01 -0500 Date: Tue, 13 Nov 2012 22:44:43 -0800 From: tip-bot for Slava Pestov Message-ID: Cc: linux-kernel@vger.kernel.org, hpa@zytor.com, mingo@kernel.org, rostedt@goodmis.org, tglx@linutronix.de, slavapestov@google.com Reply-To: mingo@kernel.org, hpa@zytor.com, linux-kernel@vger.kernel.org, rostedt@goodmis.org, tglx@linutronix.de, slavapestov@google.com In-Reply-To: <1310765038-26399-1-git-send-email-slavapestov@google.com> References: <1310765038-26399-1-git-send-email-slavapestov@google.com> To: linux-tip-commits@vger.kernel.org Subject: [tip:perf/core] ring-buffer: Add a 'dropped events' counter Git-Commit-ID: 884bfe89a462fcc85c8abd96171519cf2fe70929 X-Mailer: tip-git-log-daemon Robot-ID: Robot-Unsubscribe: Contact to get blacklisted from these emails MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=UTF-8 Content-Disposition: inline X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.2.6 (terminus.zytor.com [127.0.0.1]); Tue, 13 Nov 2012 22:44:51 -0800 (PST) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 5917 Lines: 149 Commit-ID: 884bfe89a462fcc85c8abd96171519cf2fe70929 Gitweb: http://git.kernel.org/tip/884bfe89a462fcc85c8abd96171519cf2fe70929 Author: Slava Pestov AuthorDate: Fri, 15 Jul 2011 14:23:58 -0700 Committer: Steven Rostedt CommitDate: Wed, 31 Oct 2012 16:45:27 -0400 ring-buffer: Add a 'dropped events' counter The existing 'overrun' counter is incremented when the ring buffer wraps around, with overflow on (the default). We wanted a way to count requests lost from the buffer filling up with overflow off, too. I decided to add a new counter instead of retro-fitting the existing one because it seems like a different statistic to count conceptually, and also because of how the code was structured. Link: http://lkml.kernel.org/r/1310765038-26399-1-git-send-email-slavapestov@google.com Signed-off-by: Slava Pestov Signed-off-by: Steven Rostedt --- include/linux/ring_buffer.h | 1 + kernel/trace/ring_buffer.c | 41 +++++++++++++++++++++++++++++++++++------ kernel/trace/trace.c | 3 +++ 3 files changed, 39 insertions(+), 6 deletions(-) diff --git a/include/linux/ring_buffer.h b/include/linux/ring_buffer.h index 6c8835f..2007375 100644 --- a/include/linux/ring_buffer.h +++ b/include/linux/ring_buffer.h @@ -166,6 +166,7 @@ unsigned long ring_buffer_overruns(struct ring_buffer *buffer); unsigned long ring_buffer_entries_cpu(struct ring_buffer *buffer, int cpu); unsigned long ring_buffer_overrun_cpu(struct ring_buffer *buffer, int cpu); unsigned long ring_buffer_commit_overrun_cpu(struct ring_buffer *buffer, int cpu); +unsigned long ring_buffer_dropped_events_cpu(struct ring_buffer *buffer, int cpu); u64 ring_buffer_time_stamp(struct ring_buffer *buffer, int cpu); void ring_buffer_normalize_time_stamp(struct ring_buffer *buffer, diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index b979426..0ebeb1d 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -460,9 +460,10 @@ struct ring_buffer_per_cpu { unsigned long lost_events; unsigned long last_overrun; local_t entries_bytes; - local_t commit_overrun; - local_t overrun; local_t entries; + local_t overrun; + local_t commit_overrun; + local_t dropped_events; local_t committing; local_t commits; unsigned long read; @@ -2155,8 +2156,10 @@ rb_move_tail(struct ring_buffer_per_cpu *cpu_buffer, * If we are not in overwrite mode, * this is easy, just stop here. */ - if (!(buffer->flags & RB_FL_OVERWRITE)) + if (!(buffer->flags & RB_FL_OVERWRITE)) { + local_inc(&cpu_buffer->dropped_events); goto out_reset; + } ret = rb_handle_head_page(cpu_buffer, tail_page, @@ -2995,7 +2998,8 @@ unsigned long ring_buffer_entries_cpu(struct ring_buffer *buffer, int cpu) EXPORT_SYMBOL_GPL(ring_buffer_entries_cpu); /** - * ring_buffer_overrun_cpu - get the number of overruns in a cpu_buffer + * ring_buffer_overrun_cpu - get the number of overruns caused by the ring + * buffer wrapping around (only if RB_FL_OVERWRITE is on). * @buffer: The ring buffer * @cpu: The per CPU buffer to get the number of overruns from */ @@ -3015,7 +3019,9 @@ unsigned long ring_buffer_overrun_cpu(struct ring_buffer *buffer, int cpu) EXPORT_SYMBOL_GPL(ring_buffer_overrun_cpu); /** - * ring_buffer_commit_overrun_cpu - get the number of overruns caused by commits + * ring_buffer_commit_overrun_cpu - get the number of overruns caused by + * commits failing due to the buffer wrapping around while there are uncommitted + * events, such as during an interrupt storm. * @buffer: The ring buffer * @cpu: The per CPU buffer to get the number of overruns from */ @@ -3036,6 +3042,28 @@ ring_buffer_commit_overrun_cpu(struct ring_buffer *buffer, int cpu) EXPORT_SYMBOL_GPL(ring_buffer_commit_overrun_cpu); /** + * ring_buffer_dropped_events_cpu - get the number of dropped events caused by + * the ring buffer filling up (only if RB_FL_OVERWRITE is off). + * @buffer: The ring buffer + * @cpu: The per CPU buffer to get the number of overruns from + */ +unsigned long +ring_buffer_dropped_events_cpu(struct ring_buffer *buffer, int cpu) +{ + struct ring_buffer_per_cpu *cpu_buffer; + unsigned long ret; + + if (!cpumask_test_cpu(cpu, buffer->cpumask)) + return 0; + + cpu_buffer = buffer->buffers[cpu]; + ret = local_read(&cpu_buffer->dropped_events); + + return ret; +} +EXPORT_SYMBOL_GPL(ring_buffer_dropped_events_cpu); + +/** * ring_buffer_entries - get the number of entries in a buffer * @buffer: The ring buffer * @@ -3864,9 +3892,10 @@ rb_reset_cpu(struct ring_buffer_per_cpu *cpu_buffer) local_set(&cpu_buffer->reader_page->page->commit, 0); cpu_buffer->reader_page->read = 0; - local_set(&cpu_buffer->commit_overrun, 0); local_set(&cpu_buffer->entries_bytes, 0); local_set(&cpu_buffer->overrun, 0); + local_set(&cpu_buffer->commit_overrun, 0); + local_set(&cpu_buffer->dropped_events, 0); local_set(&cpu_buffer->entries, 0); local_set(&cpu_buffer->committing, 0); local_set(&cpu_buffer->commits, 0); diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index f6928ed..36c213f 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -4385,6 +4385,9 @@ tracing_stats_read(struct file *filp, char __user *ubuf, usec_rem = do_div(t, USEC_PER_SEC); trace_seq_printf(s, "now ts: %5llu.%06lu\n", t, usec_rem); + cnt = ring_buffer_dropped_events_cpu(tr->buffer, cpu); + trace_seq_printf(s, "dropped events: %ld\n", cnt); + count = simple_read_from_buffer(ubuf, count, ppos, s->buffer, s->len); kfree(s); -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/