bryancall edited a comment on issue #7546:
URL: https://github.com/apache/trafficserver/issues/7546#issuecomment-786356094
Here is output from the package using the gcc compiler. This is using the
http/2 benchmark. It isn't as bad as AOCC, but still it is an expensive
operation:
```
**http2load**
finished in 120.34s, 24928.55 req/s, 390.48MB/s
requests: 3000000 total, 3000000 started, 3000000 done, 3000000 succeeded, 0
failed, 0 errored, 0 timeout
status codes: 3000000 2xx, 0 3xx, 0 4xx, 0 5xx
traffic: 45.89GB (49274125891) total, 39.21MB (41119867) headers (space
savings 96.56%), 45.78GB (49152000000) data
min max mean sd +/- sd
time for request: 297us 629.51ms 8.01ms 8.71ms 88.32%
time for connect: 33.66ms 91.14ms 53.87ms 14.13ms 55.50%
time to 1st byte: 35.44ms 123.39ms 69.10ms 16.27ms 63.50%
req/s : 121.07 183.79 140.32 22.49 72.00%
**dstat**
You did not select any stats, using -cdngy by default.
ESC[7l----total-cpu-usage---- -dsk/total- ---net/lo-- -net/total-
---paging-- ---system--
usr sys idl wai hiq siq| read writ| recv send: recv send| in out | int
csw
1 0 99 0 0 0|2614B 211k| 0 0 : 0 0 | 0 0 |
28k 84k
44 34 22 0 0 0|3123B 1209k| 338M 338M: 11M 640k| 0 0
|1045k 443k
46 34 20 0 0 0| 0 11M| 355M 355M: 11M 727k| 0 0
|1073k 440k
46 34 20 0 0 0| 0 54M| 347M 347M: 11M 589k| 0 0
|1069k 419k
45 35 19 0 0 0| 819B 688k| 386M 386M: 11M 1408k| 0 0
|1089k 509k
48 32 19 0 0 0| 0 5896k| 368M 368M: 11M 1487k| 0 0
|1084k 499k
43 37 20 0 0 1| 0 55M| 378M 378M: 11M 1542k| 0 0
|1096k 494k
45 36 19 1 0 1|2048B 57M| 384M 384M: 11M 1689k| 0 0
|1095k 512k
46 34 19 0 0 0| 0 4642k| 385M 385M: 11M 1547k| 0 0
|1089k 518k
33 38 28 0 0 1| 0 4176k| 465M 465M: 11M 1877k| 0 0
|1001k 652k
22 36 41 0 0 1|3686B 47M| 503M 503M: 11M 2015k| 0 0 |
858k 666k
22 34 41 2 0 1|3277B 47M| 439M 439M: 11M 1637k| 0 0
|1122k 561k
22 26 50 1 0 1|2867B 47M| 554M 554M: 11M 1814k| 0 0 |
806k 692k
**perf stat**
Performance counter stats for process id '17983':
10,466,445.26 msec task-clock # 86.920 CPUs utilized
28,088,848 context-switches # 0.003 M/sec
1,237,496 cpu-migrations # 0.118 K/sec
567,537 page-faults # 0.054 K/sec
33,958,594,177,446 cycles # 3.245 GHz
(66.69%)
977,798,939,842 stalled-cycles-frontend # 2.88% frontend cycles
idle (66.68%)
675,897,082,577 stalled-cycles-backend # 1.99% backend cycles
idle (66.69%)
4,346,448,211,697 instructions # 0.13 insn per cycle
# 0.22 stalled cycles
per insn (66.68%)
916,758,375,688 branches # 87.590 M/sec
(66.68%)
27,001,398,959 branch-misses # 2.95% of all branches
(66.70%)
120.414481542 seconds time elapsed
**perf report**
# Total Lost Samples: 0
#
# Samples: 48M of event 'cycles'
# Event count (approx.): 36487625633965
#
# Overhead Shared Object Symbol
# .......... ....................
..................................................
#
39.17% [kernel.kallsyms] [k] native_queued_spin_lock_slowpath
31.72% traffic_server [.] LogObject::_checkout_write
1.79% libpthread-2.17.so [.] pthread_mutex_trylock
1.67% libjemalloc.so.2 [.] je_malloc_mutex_lock_slow
0.54% [kernel.kallsyms] [k] copy_user_generic_string
0.46% [kernel.kallsyms] [k] acpi_processor_ffh_cstate_enter
0.43% libc-2.17.so [.] __memcpy_ssse3
0.41% [kernel.kallsyms] [k] try_to_wake_up
0.38% [kernel.kallsyms] [k] native_write_msr_safe
```
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
