Fixes for SMOG2 HLT1 sequence
The goal is to fix two issues that currently prevent HLT1 SMOG2 lines from running from Moore:
-
Some lines in
lowMult
function give errors such as those. The error can be traced back to three lines in the
lowMult
algorithm and one in theSMOG2_minimum_bias_line
line builder that call properties that are not defined. The errors disappear after removing four conflicting lines. -
A "control flow" error appears when running the SMOG minimum bias lines, as it is shown here (control_flow_error.log). The error is related with lhcb/Allen!939 and can be fixed by reverting the order of
prefilters = gec + odin_err_filter
toprefilters = odin_err_filter + gec
in this line.
Merge request reports
Activity
assigned to @oboenteg
added RTA label
requested review from @dovombru
- Resolved by Dorothea Vom Bruch
The fixes look fine to me. However I don't understand why this was not observed in the Allen standalone CI tests running the
hlt1_pp_SMOG2_default
sequence. Maybe @samarian you have an idea?
- Resolved by Dorothea Vom Bruch
Let's test it:
/ci-test
added ci-test-triggered label
- [2022-10-12 19:11] Validation started with lhcb-master-mr#5901
- [2022-10-13 10:34] Validation started with lhcb-master-mr#5906
Edited by Software for LHCbadded 30 commits
-
07da95d1...74a0ad38 - 28 commits from branch
master
- 7a9e6e57 - Commented conflicting lines in smog min bias lines
- 9b98a117 - Fix 'Dependency circle in control flow' by reverse prefilters order
-
07da95d1...74a0ad38 - 28 commits from branch
added hlt1-throughput-increased label
mentioned in issue #376
removed review request for @dovombru
unassigned @oboenteg
mentioned in commit 571488be
mentioned in issue Moore#485 (closed)