6
0
mirror of https://github.com/FirebirdSQL/firebird-qa.git synced 2025-01-22 13:33:07 +01:00

Added/Updated tests\bugs\gh_8219_test.py: Confirmed problem on snapshotrs before 20-aug-2024 -- see notes. Test can be used only for ServerMode = Super or SuperClassic

This commit is contained in:
pavel-zotov 2024-09-05 12:28:07 +03:00
parent 17d8f37104
commit 1a3a9e0db7

145
tests/bugs/gh_8219_test.py Normal file
View File

@ -0,0 +1,145 @@
#coding:utf-8
"""
ID: issue-8219
ISSUE: https://github.com/FirebirdSQL/firebird/issues/8219
TITLE: Database creation in 3.0.12, 4.0.5 and 5.0.1 slower than in previous releases
DESCRIPTION:
We can estimate perfomance by comparison of time that is spent to create DB vs result of some crypt function.
Function crypt_hash(<string> using SHA512) has been selected for that because of notable CPU consumation.
Stored procedured SP_GEN_HASH is created for evaluation of crypt hash, it will run loop for N_HASH_EVALUATE_COUNT times.
Duration for each measure is difference between psutil.Process(fb_pid).cpu_times() counters.
We do <N_MEASURES> times call of SP and create_database(), with adding results to map.
Finally, we get ratio between medians of these measures (see 'median_ratio')
Test is considered as passed if median_ratio less than threshold <MAX_RATIO>.
NOTES:
[05.09.2024] pzotov.
1. Confirmed problem on snapshotrs before 20-aug-2024.
Medians ratio on Windows:
1. Before fix:
6.0.0.423: 0.39; 6.0.0.436: 0.39; 6.0.0.437: 0.35;
5.0.1.1464: 0.42; 5.0.1.1469: 0.39; 5.0.1.1479: 0.35
4.0.5.3136: 0.42; 4.0.6.3142: 0.39
2. After fix ratio reduced to ~0.25:
6.0.0.438: 0.21; 6.0.0.442: 0.21; 6.0.0.438: 0.21; 6.0.0.442: 0.21; 6.0.0.450: 0.24
5.0.2.1481: 0.25; 5.0.2.1482: 0.21; 5.0.2.1493: 0.22
4.0.6.3144: 0.25; 4.0.6.3149: 0.29
Medians ratio on Windows:
1. Before fix:
6.0.0.397-c734c96: 0.48; 6.0.0.438-088b529: 0.49
2. After fix ratio reduced to ~0.25:
6.0.0.441-75042b5: 0.23
5.0.2.1481-fc71044: 0.24
4.0.6.3144-5a3b718: 0.27
2. Test DB must NOT have charset = utf8, otherwise 'implementation limit exceeded' will raise; win1251 was selected for work.
3. Test can be used only for ServerMode = Super or SuperClassic
(because in CS a new process is made and we have no value of cpu_times() *before* DB creation).
"""
import os
import psutil
import pytest
from firebird.qa import *
from firebird.driver import driver_config, create_database, NetProtocol
from pathlib import Path
#--------------------------------------------------------------------
def median(lst):
n = len(lst)
s = sorted(lst)
return (sum(s[n//2-1:n//2+1])/2.0, s[n//2])[n % 2] if n else None
#--------------------------------------------------------------------
###########################
### S E T T I N G S ###
###########################
# How many times we create databases
N_MEASURES = 31
# How many iterations must be done for hash evaluation:
N_HASH_EVALUATE_COUNT = 3000
# Maximal value for ratio between maximal and minimal medians
#
#############################################
MAX_RATIO = 0.30 if os.name == 'nt' else 0.33
#############################################
init_script = \
f'''
set term ^;
create or alter procedure sp_gen_hash (n_cnt int) as
declare v_hash varbinary(64);
declare s varchar(32765);
begin
s = lpad('', 32765, uuid_to_char(gen_uuid()));
while (n_cnt > 0) do
begin
v_hash = crypt_hash(s using SHA512);
n_cnt = n_cnt - 1;
end
end
^
commit
^
'''
db = db_factory(init = init_script, charset = 'win1251')
act = python_act('db')
tmp_fdb = temp_file('tmp_gh_8219.tmp')
expected_stdout = """
Medians ratio: acceptable
"""
@pytest.mark.version('>=4.0.5')
def test_1(act: Action, tmp_fdb: Path, capsys):
if act.vars['server-arch'].lower() == 'classic':
pytest.skip('Can be used only for SS / SC.')
srv_cfg = driver_config.register_server(name = 'test_srv_gh_8219', config = '')
db_cfg_name = 'tmp_8219'
db_cfg_object = driver_config.register_database(name = db_cfg_name)
db_cfg_object.server.value = srv_cfg.name
db_cfg_object.protocol.value = NetProtocol.INET
db_cfg_object.database.value = str(tmp_fdb)
with act.db.connect() as con:
cur=con.cursor()
cur.execute('select mon$server_pid as p from mon$attachments where mon$attachment_id = current_connection')
fb_pid = int(cur.fetchone()[0])
times_map = {}
for i in range(0, N_MEASURES):
fb_info_init = psutil.Process(fb_pid).cpu_times()
cur.callproc( 'sp_gen_hash', (N_HASH_EVALUATE_COUNT,) )
fb_info_curr = psutil.Process(fb_pid).cpu_times()
times_map[ 'hash_eval', i ] = max(fb_info_curr.user - fb_info_init.user, 0.000001)
fb_info_init = psutil.Process(fb_pid).cpu_times()
with create_database(db_cfg_name, user = act.db.user, password = act.db.password, overwrite = True) as dbc:
pass
fb_info_curr = psutil.Process(fb_pid).cpu_times()
times_map[ 'db_create', i ] = max(fb_info_curr.user - fb_info_init.user, 0.000001)
sp_gen_hash_median = median([v for k,v in times_map.items() if k[0] == 'hash_eval'])
sp_db_create_median = median([v for k,v in times_map.items() if k[0] == 'db_create'])
median_ratio = sp_db_create_median / sp_gen_hash_median
print( 'Medians ratio: ' + ('acceptable' if median_ratio <= MAX_RATIO else '/* perf_issue_tag */ POOR: %s, more than threshold: %s' % ( '{:9g}'.format(median_ratio), '{:9g}'.format(MAX_RATIO) ) ) )
if median_ratio > MAX_RATIO:
print(f'CPU times for each of {N_MEASURES} measures:')
for sp_name in ('hash_eval', 'db_create', ):
print(f'{sp_name=}:')
for p in [v for k,v in times_map.items() if k[0] == sp_name]:
print(p)
act.expected_stdout = expected_stdout
act.stdout = capsys.readouterr().out
assert act.clean_stdout == act.clean_expected_stdout