getPrbHT spends half of its cpu time retrieving a MsgStream for messages rarely if ever printed (DEBUG)

Normalized to parent so in this case the function appears as 100%.... Took me a bit to understand why some seemingly simple math bring in a pthread_mutex_lock in the main loop etc etc

If you are to need them again (why?) please read https://indico.cern.ch/event/252634/contributions/564148/attachments/440055/610507/2013-06-10-swtut.pdf by @ssnyder slides 27-28. You would prb be better of passing it ,and replicating the unlikely from the std macros.

This merge request affects 1 package:

• InnerDetector/InDetConditions/TRT_ConditionsData

added InnerDetector master review-pending-level-1 labels

changed title from getPrbHT spends 30% of its time (all the time) retrieving a MsgStream for messages rarely if ever printed (DEBUG) to getPrbHT spends half of its cpu time retrieving a MsgStream for messages rarely if ever printed (DEBUG)

changed the description

removed InnerDetector master review-pending-level-1 labels

CI Result SUCCESS

	Athena	AthSimulation
externals
cmake
make
required tests
optional tests

Full details available at NICOS MR-24800-2019-07-11-18-35
Athena: number of compilation errors 0, warnings 0
AthSimulation: number of compilation errors 0, warnings 0
For experts only: Jenkins output [CI-MERGE-REQUEST-CC7 1073]

changed the description

Looks OK.

Tadej (L2)

added InnerDetector master review-approved labels

merged

mentioned in commit 26d5f4d4

added sweep:ignore label