Am 13.09.2015 um 13:08 schrieb Mark Thomas:
Please do not add any more RV-Predict bug reports to Bugzilla until the
current set have been evaluated. To be honest it would have been better
if you had paused after adding 58367-58379.
There are two primary issues with automated tools.
1. They can report false positives. Each report needs to be looked at
evaluated by a human to determine if it is genuine or not.
2. Sometimes the tool might report something that is technically an
issue but the Tomcat developers have deliberately decided to code it
that way. BZ 58367 is an example of this. Again, each issue needs to be
evaluated by a human before adding it to Bugzilla.
3. Probably many of those reports are already included in the coverity
scans.
Regards,
Felix
The race condition in BZ 58367 can't cause any harm. At worst the code
may be slightly inefficient but the cost of the plumbing that would be
necessary to avoid the race condition was viewed as likely to add more
overhead than the problem it was trying to solve.
Many of these optimisation decisions were made several major versions
ago and Java performance and compiler optimization has changed
significantly since then. It may be time to revisit those decisions but
we are unlikely to unless the code is flagged up as causing a
performance bottleneck.
I suggest that you follow each of these bugs as we work through them to
get a sense of what the Tomcat team considers to be a bug and what it
considers not to be a bug. With that, you can do some filtering of these
reports before you add them to Bugzilla.
I don't want you to think I am completely against tools like RV-Predict.
Far from it. I know that there have been some tricky to find race
conditions in the connectors and I suspect a number of the open bugs in
Bugzilla have similar tricky to find race conditions as their root
cause. Having a tool that can find these would be a huge benefit. What
we need is to be able to filter out the noise and concentrate on the
important issues.
Mark
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
