diff options
| -rw-r--r-- | erts/doc/src/erlang.xml | 47 |
1 files changed, 45 insertions, 2 deletions
diff --git a/erts/doc/src/erlang.xml b/erts/doc/src/erlang.xml index d85dff2c0c..dba34cc3f2 100644 --- a/erts/doc/src/erlang.xml +++ b/erts/doc/src/erlang.xml @@ -6137,7 +6137,7 @@ ok <v>MonSettings -> {MonitorPid, Options} | undefined</v> <v> MonitorPid = pid()</v> <v> Options = [Option]</v> - <v> Option = {long_gc, Time} | {large_heap, Size} | busy_port | busy_dist_port</v> + <v> Option = {long_gc, Time} | {long_schedule, Time} | {large_heap, Size} | busy_port | busy_dist_port</v> <v> Time = Size = integer()</v> </type> <desc> @@ -6171,7 +6171,7 @@ ok <fsummary>Set system performance monitoring options</fsummary> <type> <v>MonitorPid = pid()</v> - <v>Option = {long_gc, Time} | {large_heap, Size} | busy_port | busy_dist_port</v> + <v>Option = {long_gc, Time} | {long_schedule, Time} | {large_heap, Size} | busy_port | busy_dist_port</v> <v> Time = Size = integer()</v> <v>MonSettings = {OldMonitorPid, [Option]}</v> <v> OldMonitorPid = pid()</v> @@ -6204,6 +6204,49 @@ ok notice. </p> </item> + <tag><c>{long_schedule, Time}</c></tag> + <item> + <p>If a process or port in the system runs uninterrupted + for at least <c>Time</c> wall clock milliseconds, a + message <c>{monitor, PidOrPort, long_schedule, Info}</c> + is sent to <c>MonitorPid</c>. <c>PidOrPort</c> is the + process or port that was running and <c>Info</c> is a + list of two-element tuples describing the event. In case + of a <c>pid()</c>, the tuples <c>{timeout, Millis}</c>, + <c>{in, Location}</c> and <c>{out, Location}</c> will be + present, where <c>Location</c> is either an MFA + (<c>{Module, Function, Arity}</c>) describing the + function where the process was scheduled in/out, or the + atom <c>undefined</c>. In case of a <c>port()</c>, the + tuples <c>{timeout, Millis}</c> and <c>{port_op,Op}</c> + will be present. <c>Op</c> will be one of <c>free</c>, + <c>timeout</c>, <c>input</c>, <c>output</c>, + <c>event</c> or <c>dist_cmd</c>, depending on which + driver callback was executing. <c>free</c> is an + internal operation and should never appear, while the + others represent the corresponding driver callbacks + <c>timeout</c>, <c>ready_input</c>, <c>ready_output</c>, + <c>event</c> and finally <c>outputv</c> (when the port + is used by distribution). The <c>Millis</c> value in + the <c>timeout</c> tuple will tell you the actual + uninterrupted execution time of the process or port, + which will always be <c>>=</c> the <c>Time</c> value + supplied when starting the trace. New tuples may be + added to the <c>Info</c> lists in the future, and the + order of the tuples in the list may be changed at any + time without prior notice. + </p> + <p>This can be used to detect problems with NIF's or + drivers that takes too long to execute. Generally, 1 ms + is considered a good maximum time for a driver callback + or a NIF. However, a time sharing system should usually + consider everything below 100 ms as "possible" and + fairly "normal". Schedule times above that might however + indicate swapping or a NIF/driver that is + misbehaving. Misbehaving NIF's and drivers could cause + bad resource utilization and bad overall performance of + the system.</p> + </item> <tag><c>{large_heap, Size}</c></tag> <item> <p>If a garbage collection in the system results in |