Skip to content

Fix bug in lumi summary decoding (sizes and offsets of counters were swapped)

Daniel Charles Craik requested to merge dcraik_lumiSummaryFix into master

!3837 (merged) introduced a bug to HltLumiSummaryDecoder that swapped over the size and offset values for each lumi counter read from the schema. This presented as the encoding key of the decoded object always being set to 0 (read from 0 bits starting at offset 32) which caused the LumiSummaryWriter to fail in the test_lumi test of MooreOnline. Any banks decoded using versions of LHCb that include that MR and not this one will have nonsense counter values, however, the raw un-decoded banks are unaffected.

This MR fixes the bug.

FYI @graven @sponce @sesen

Merge request reports