From 4089946d835f19eb1338c62f6eaf76f4a91da8e6 Mon Sep 17 00:00:00 2001 From: Philippe Proulx Date: Tue, 1 Aug 2017 15:27:12 -0400 Subject: [PATCH] 2.8, 2.9, 2.10: since 2.8, LTTng keeps lost sub-buffer count in overwrite mode Signed-off-by: Philippe Proulx --- 2.10/lttng-docs-2.10.txt | 14 +++++++++----- 2.8/lttng-docs-2.8.txt | 12 ++++++++---- 2.9/lttng-docs-2.9.txt | 12 ++++++++---- 3 files changed, 25 insertions(+), 13 deletions(-) diff --git a/2.10/lttng-docs-2.10.txt b/2.10/lttng-docs-2.10.txt index 37117b0..10d1f44 100644 --- a/2.10/lttng-docs-2.10.txt +++ b/2.10/lttng-docs-2.10.txt @@ -1308,15 +1308,19 @@ always keep a fixed amount of the latest data. Which mechanism you should choose depends on your context: prioritize the newest or the oldest event records in the ring buffer? -Beware that, in overwrite mode, the tracer abandons a whole sub-buffer +Beware that, in overwrite mode, the tracer abandons a _whole sub-buffer_ as soon as a there's no space left for a new event record, whereas in discard mode, the tracer only discards the event record that doesn't fit. -In discard mode, LTTng increments a count of lost event records when -an event record is lost and saves this count to the trace. In -overwrite mode, LTTng keeps no information when it overwrites a -sub-buffer before consuming it. +In discard mode, LTTng increments a count of lost event records when an +event record is lost and saves this count to the trace. In overwrite +mode, since LTTng 2.8, LTTng increments a count of lost sub-buffers when +a sub-buffer is lost and saves this count to the trace. In this mode, +the exact number of lost event records in those lost sub-buffers is not +saved to the trace. Trace analyses can use the trace's saved discarded +event record and sub-buffer counts to decide whether or not to perform +the analyses even if trace data is known to be missing. There are a few ways to decrease your probability of losing event records. diff --git a/2.8/lttng-docs-2.8.txt b/2.8/lttng-docs-2.8.txt index a90ec94..80cdf9a 100644 --- a/2.8/lttng-docs-2.8.txt +++ b/2.8/lttng-docs-2.8.txt @@ -1535,10 +1535,14 @@ as soon as a there's no space left for a new event record, whereas in discard mode, the tracer only discards the event record that doesn't fit. -In discard mode, LTTng increments a count of lost event records when -an event record is lost and saves this count to the trace. In -overwrite mode, LTTng keeps no information when it overwrites a -sub-buffer before consuming it. +In discard mode, LTTng increments a count of lost event records when an +event record is lost and saves this count to the trace. In overwrite +mode, since LTTng 2.8, LTTng increments a count of lost sub-buffers when +a sub-buffer is lost and saves this count to the trace. In this mode, +the exact number of lost event records in those lost sub-buffers is not +saved to the trace. Trace analyses can use the trace's saved discarded +event record and sub-buffer counts to decide whether or not to perform +the analyses even if trace data is known to be missing. There are a few ways to decrease your probability of losing event records. diff --git a/2.9/lttng-docs-2.9.txt b/2.9/lttng-docs-2.9.txt index 36f4963..2457b8c 100644 --- a/2.9/lttng-docs-2.9.txt +++ b/2.9/lttng-docs-2.9.txt @@ -1592,10 +1592,14 @@ as soon as a there's no space left for a new event record, whereas in discard mode, the tracer only discards the event record that doesn't fit. -In discard mode, LTTng increments a count of lost event records when -an event record is lost and saves this count to the trace. In -overwrite mode, LTTng keeps no information when it overwrites a -sub-buffer before consuming it. +In discard mode, LTTng increments a count of lost event records when an +event record is lost and saves this count to the trace. In overwrite +mode, since LTTng 2.8, LTTng increments a count of lost sub-buffers when +a sub-buffer is lost and saves this count to the trace. In this mode, +the exact number of lost event records in those lost sub-buffers is not +saved to the trace. Trace analyses can use the trace's saved discarded +event record and sub-buffer counts to decide whether or not to perform +the analyses even if trace data is known to be missing. There are a few ways to decrease your probability of losing event records. -- 2.34.1