Skip to content

B
bcc
Commit 6af7b844 authored by Brendan Gregg's avatar Brendan Gregg
Browse files
Options

cachestat: bring back HITRATIO column

parent 5f5e118d
Hide whitespace changes
Inline Side-by-side
Showing with 125 additions and 100 deletions
man/man8/cachestat.8
View file @ 6af7b844
......@@ -18,18 +18,14 @@ Since this uses BPF, only the root user can use this tool.
CONFIG_BPF and bcc.
.SH EXAMPLES
.TP
Print summaries every five second:
Print summaries every second:
#
.B cachestat
.TP
Print summaries every five seconds with timestamp:
Print summaries every second with timestamp:
#
.B cachestat -T
.TP
Print summaries each second:
#
.B cachestat 1
.TP
Print output every five seconds, three times:
#
.B cachestat 5 3
......@@ -51,6 +47,9 @@ Number of page cache misses.
DIRTIES
Number of dirty pages added to the page cache.
.TP
HITRATIO
The hit ratio as a percentage.
.TP
READ_HIT%
Read hit percent of page cache usage.
.TP
......
tools/cachestat.py
View file @ 6af7b844
......@@ -15,6 +15,7 @@
# 09-Sep-2015 Brendan Gregg Created this.
# 06-Nov-2015 Allan McAleavy
# 13-Jan-2016 Allan McAleavy run pep8 against program
# 02-Feb-2019 Brendan Gregg Column shuffle, bring back %ratio
from __future__ import print_function
from bcc import BPF
......@@ -55,7 +56,7 @@ parser = argparse.ArgumentParser(
formatter_class=argparse.RawDescriptionHelpFormatter)
parser.add_argument("-T", "--timestamp", action="store_true",
help="include timestamp on output")
parser.add_argument("interval", nargs="?", default=5,
parser.add_argument("interval", nargs="?", default=1,
help="output interval, in seconds")
parser.add_argument("count", nargs="?", default=-1,
help="number of outputs")
......@@ -102,7 +103,7 @@ b.attach_kprobe(event="mark_buffer_dirty", fn_name="do_count")
if tstamp:
print("%-8s " % "TIME", end="")
print("%8s %8s %8s %8s %12s %10s" %
("TOTAL", "MISSES", "HITS", "DIRTIES", "BUFFERS_MB", "CACHED_MB"))
("HITS", "MISSES", "DIRTIES", "HITRATIO", "BUFFERS_MB", "CACHED_MB"))
loop = 0
exiting = 0
......@@ -121,38 +122,36 @@ while 1:
counts = b["counts"]
for k, v in sorted(counts.items(), key=lambda counts: counts[1].value):
if re.match(b'mark_page_accessed', b.ksym(k.ip)) is not None:
func = b.ksym(k.ip)
# partial string matches in case of .isra (necessary?)
if func.find("mark_page_accessed") == 0:
mpa = max(0, v.value)
if re.match(b'mark_buffer_dirty', b.ksym(k.ip)) is not None:
if func.find("mark_buffer_dirty") == 0:
mbd = max(0, v.value)
if re.match(b'add_to_page_cache_lru', b.ksym(k.ip)) is not None:
if func.find("add_to_page_cache_lru") == 0:
apcl = max(0, v.value)
if re.match(b'account_page_dirtied', b.ksym(k.ip)) is not None:
if func.find("account_page_dirtied") == 0:
apd = max(0, v.value)
# total = total cache accesses without counting dirties
# misses = total of add to lru because of read misses
total = (mpa - mbd)
misses = (apcl - apd)
if total < 0:
total = 0
if misses < 0:
misses = 0
hits = total - misses
# If hits are < 0, then its possible misses are overestimated
# due to possibly page cache read ahead adding more pages than
# needed. In this case just assume misses as total and reset hits.
if hits < 0:
misses = total
hits = 0
# total = total cache accesses without counting dirties
# misses = total of add to lru because of read misses
total = mpa - mbd
misses = apcl - apd
if misses < 0:
misses = 0
if total < 0:
total = 0
hits = total - misses
# If hits are < 0, then its possible misses are overestimated
# due to possibly page cache read ahead adding more pages than
# needed. In this case just assume misses as total and reset hits.
if hits < 0:
misses = total
hits = 0
ratio = 0
if total > 0:
ratio = float(hits) / total
if debug:
print("%d %d %d %d %d %d %d\n" %
......@@ -167,18 +166,10 @@ while 1:
if tstamp:
print("%-8s " % strftime("%H:%M:%S"), end="")
print("%8d %8d %8d %8d %12.0f %10.0f" %
(total, misses, hits, mbd, buff, cached))
mpa = 0
mbd = 0
apcl = 0
apd = 0
total = 0
misses = 0
hits = 0
cached = 0
buff = 0
print("%8d %8d %8d %7.2f%% %12.0f %10.0f" %
(hits, misses, mbd, 100 * ratio, buff, cached))
mpa = mbd = apcl = apd = total = misses = hits = cached = buff = 0
if exiting:
print("Detaching...")
......
tools/cachestat_example.txt
View file @ 6af7b844
# ./cachestat -h
USAGE: ./cachestat [-T] [ interval [count] ]
Demonstrations of cachestat, the Linux eBPF/bcc version.
show Linux page cache hit/miss statistics including read and write hit %
cachestat shows hits and misses to the file system page cache. For example:
# cachestat
HITS MISSES DIRTIES HITRATIO BUFFERS_MB CACHED_MB
1132 0 4 100.00% 277 4367
161 0 36 100.00% 277 4372
16 0 28 100.00% 277 4372
17154 13750 15 55.51% 277 4422
19 0 1 100.00% 277 4422
83 0 83 100.00% 277 4421
16 0 1 100.00% 277 4423
^C 0 -19 360 0.00% 277 4423
Detaching...
While tracing, there was a burst of misses in the fourth second, bringing
the hit ration down to 55%.
This shows a 1 Gbyte uncached file that is read twice:
(root) ~ # ./cachestat.py
HITS MISSES DIRTIES HITRATIO BUFFERS_MB CACHED_MB
1 0 0 100.00% 5 191
198 12136 0 1.61% 5 238
1 11007 3 0.01% 5 281
0 6384 0 0.00% 5 306
1 14464 0 0.01% 5 363
0 11776 0 0.00% 5 409
1 11712 0 0.01% 5 454
32 13184 0 0.24% 5 506
0 11232 0 0.00% 5 550
1 13056 0 0.01% 5 601
16 14720 0 0.11% 5 658
33 9920 0 0.33% 5 697
0 13248 0 0.00% 5 749
4 14144 0 0.03% 5 804
0 9728 0 0.00% 5 842
1 10816 0 0.01% 5 885
808 13504 1 5.65% 5 938
0 11409 0 0.00% 5 982
0 11520 0 0.00% 5 1027
0 15616 0 0.00% 5 1088
1 9792 0 0.01% 5 1126
0 8256 0 0.00% 5 1158
1 9600 0 0.01% 5 1196
599 4804 0 11.09% 5 1215
1 0 0 100.00% 5 1215
0 0 0 0.00% 5 1215
3 1 0 75.00% 5 1215
79536 34 0 99.96% 5 1215
87693 274 4 99.69% 6 1214
89018 3546 0 96.17% 7 1227
33531 201 4 99.40% 7 1228
22 44 0 33.33% 8 1228
0 0 0 0.00% 8 1228
73 21 2 77.66% 8 1228
It took 24 seconds to read the 1 Gbyte file the first time, shown in the output
by the high MISSES rate and low HITRATIO. The second time it took 4 seconds,
and the HITRATIO was around 99%.
This output shows a 1 Gbyte file being created and added to the page cache:
(root) ~ # ./cachestat.py
HITS MISSES DIRTIES HITRATIO BUFFERS_MB CACHED_MB
1 0 0 100.00% 8 209
0 0 165584 0.00% 8 856
0 0 96505 0.00% 8 1233
0 0 0 0.00% 8 1233
Note the high rate of DIRTIES, and the CACHED_MD size increases by 1024 Mbytes.
USAGE message:
# cachestat -h
usage: cachestat.py [-h] [-T] [interval] [count]
Count cache kernel function calls
positional arguments:
interval output interval, in seconds
count number of outputs
optional arguments:
-T include timestamp on output
examples:
./cachestat # run with default option of 5 seconds delay
./cachestat -T # run with default option of 5 seconds delay with timestamps
./cachestat 1 # print every second hit/miss stats
./cachestat -T 1 # include timestamps with one second samples
./cachestat 1 5 # run with interval of one second for five iterations
./cachestat -T 1 5 # include timestamps with interval of one second for five iterations
Following commands show a 2GB file being read into the page cache.
Command used to generate activity:
# dd if=/root/tmpfile of=/dev/null bs=8192
Output from cachestat running simultatenously:
# ./tools/cachestat.py 1
TOTAL MISSES HITS DIRTIES BUFFERS_MB CACHED_MB
1 0 1 0 8 283
0 0 0 0 8 283
0 0 0 2 8 283
0 0 0 0 8 283
10009 9173 836 2 9 369
152032 152032 0 0 9 1028
157408 157405 3 0 9 1707
150432 150432 0 0 9 2331
0 0 0 0 9 2331
1 1 0 1 9 2331
0 0 0 0 9 2331
0 0 0 0 9 2331
0 0 0 0 9 2331
The misses counter reflects a 2GB file being read and almost everything being
a page cache miss.
Below shows an example of a new 100MB file added to page cache, by using
the command: dd if=/dev/zero of=/root/tmpfile2 bs=4k count=$((256*100))
# ./tools/cachestat.py 1
TOTAL MISSES HITS DIRTIES BUFFERS_MB CACHED_MB
0 0 0 0 15 2440
0 0 0 0 15 2440
0 0 0 0 15 2440
1758 0 1758 25603 15 2540
0 0 0 0 15 2540
0 0 0 0 15 2541
~25600 pages are being dirtied (writes) which corresponds to the 100MB file
added to the page cache.
-h, --help show this help message and exit
-T, --timestamp include timestamp on output
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7