Lines Matching +full:system +full:- +full:trace

1 .TH btrfsslower 8  "2016-02-15" "USER COMMANDS"
3 btrfsslower \- Trace slow btrfs file operations, with per-event details.
5 .B btrfsslower [\-h] [\-j] [\-p PID] [min_ms] [\-d DURATION]
23 \-p PID
24 Trace this PID only.
27 Minimum I/O latency (duration) to trace, in milliseconds. Default is 10 ms.
29 \-d DURATION
30 Total duration of trace in seconds.
33 Trace synchronous file reads and writes slower than 10 ms:
37 Trace slower than 1 ms:
41 Trace slower than 1 ms, and output just the fields in parsable format (csv):
43 .B btrfsslower \-j 1
45 Trace all file reads and writes (warning: the output will be verbose):
49 Trace slower than 1 ms, for PID 181 only:
51 .B btrfsslower \-p 181 1
53 Trace for 10 seconds only:
55 .B btrfsslower \-d 10
79 file system CPU cycles, file system locks, run queue latency, etc. It's a more
81 system I/O, than to measure this down at the block device interface.
84 A cached kernel file name (comes from dentry->d_name.name).
87 Completion timestamp, microseconds (\-j only).
90 File offset, bytes (\-j only).
93 Latency (duration) of the I/O, in microseconds (\-j only).
95 This adds low-overhead instrumentation to btrfs writes and fsyncs, as well
96 as all system reads and opens (due to the current implementation of the
98 the file system cache will incur extra overhead while tracing. Such reads and
102 continues to be a problem, consider switching to a tool that prints in-kernel
118 Unstable - in development.