Hi Michael, the motivation for the bug report and the fix was the "official" hpitest suite coming from SAF. I saw several test cases don't pass for the simulator and newsimulator plugin. If you're using this test suite for your regression tests I hope and think it should run as usual or better.
In my case I could reduce the failures/errors for ThresholdsSet test cases running with simulator and newsimulator plugin to 1/0. (For the 1 I will write a bug report on monday against the simulator, it's a problem to copy the threshold values without checking the IsSupport field). If you have some problems running the test cases, please let it me know and we can look together on the issue (sometimes there is a fault behind a fault ;-)). But - my test "hardware" works fine with the changes. Hope I could help Lars On Friday, 12. March 2010 22:55, Bishop, Michael (ISB Linux/Telco) wrote: > Lars - > > I have a concern which is the possibility of breaking OpenHPI regression > tests. I have not run the regression tests myself (not yet - but I will > later when I start working on the release process) - so I'm somewhat > unfamiliar with the level at which they do testing. > > Are you familiar with the test suite and whether or not it could be > affected by this change? > > Lars - is it possible for you to run the hpi test suite before/after you > make this change to see if any tests are affected? > > Anyone else have comments on this? > > Thanks for listening, > --michael > > > -----Original Message----- > From: Lars Wetzel [mailto:[email protected]] > Sent: Friday, March 12, 2010 11:49 AM > To: [email protected] > Subject: [Openhpi-devel] Plan for split oh_valid_threshold in two functions > > > > > > Hi all, > > > > due to the bugfix for 2969326 at the moment the ordering from threshold > values is only be checked with the values coming from the caller of > saHpiSensorThresholdsSet(). > > > > To check the thresholds ordering also with threshold reading values - as it > was done before the fix - I want to separate the ordering part in a new > function: > > sahpi_struct_utils: oh_valid_ordering > > > > It will be no longer part of function oh_valid_threshold due to performance > reason (else it is called twice inside safhpi.c::saHpiSensorThresholdsSet). > > > > If you have any objections, please let it me know. > > You can find a patch file which includes the change from above as > attachment on bug 2969326. > > > > If I got no objections against the plan to split the function, I will put > it into the trunk branch next Wednesday, 17.03. > > > > Thanks to Anton for the helping hand! > > > > Regards > > Lars -- ------------------------------- Dipl. Wi.ing. Lars Wetzel Uttinger Str. 13 86938 Schondorf a. Ammersee Tel.: 0179-2096845 Mail: [email protected] USt-IdNr.: DE181396006 ------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ Openhpi-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/openhpi-devel
