vlsi commented on code in PR #5788:
URL: https://github.com/apache/jmeter/pull/5788#discussion_r1148544589
##########
xdocs/usermanual/properties_reference.xml:
##########
@@ -1339,6 +1339,14 @@ JMETER-SERVER</source>
ORO PatternCacheLRU size.<br/>
Defaults to: <code>1000</code>
</property>
+<property name="function.cache.per.iteration">
+ <p>Cache function execution during test execution.</p>
+ <p>By default, JMeter caches function properties during a test iteration,
however,
+ it might cause unexpected results when a component is shared across
threads and the expression depends on
+ the thread variables.</p>
+ <note>The caching behaviour would likely be disabled in the upcoming
versions</note>
Review Comment:
I believe the cache creates functional issues: it caches the values, and it
does not invalidate the cache. In other words, it assumes "the computed value
depends only on the iteration number", which, in my opinion, is quite poor
approximation.
So my main goal here is to establish correctness, and if there are any
performance implications, then they should be healed separately
--
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.
To unsubscribe, e-mail: dev-unsubscr...@jmeter.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
