StsDigi: Add more error handling against timestamp overflows.
Merge request reports
Activity
We have a problem with one of the mCBM CI tests currently. Not sure how it managed to pass the chain when it was itself in the MR CI, but I try to fix it asap.
Edit: just saw that the failing tests are actually due to what we discuss in issue #2301 , so this problem is actually hidden right now
Edited by Pierre-Alain Loizeauadded 3 commits
-
34bc44a4...3459653c - 2 commits from branch
computing:master - d494e2ac - StsDigi: Add more error handling against timestamp overflows.
-
34bc44a4...3459653c - 2 commits from branch
Resolved by Felix WeiglhoferThere are more problems than the time stamp - for instance the address. The current packing scheme strips the information on sensor and side - rightly so, because they are of no relevance for the digis. However, that means that an address comprising sensor number > 0 or side = 1 will not be reproduced when stored in the digi. Similarly, there is no protection on the range of the number of units, ladders, haldladders and modules.
- Last reply by Felix Weiglhofer
added 1 commit
- b141c3d4 - StsDigi: Add more error handling against timestamp overflows.
added 11 commits
-
b141c3d4...926569d1 - 10 commits from branch
computing:master - 0091a7cc - StsDigi: Add more error handling against timestamp overflows.
-
b141c3d4...926569d1 - 10 commits from branch
- Resolved by Felix Weiglhofer
Dear @fweig ,
Could you please add the following new commit with as comment
Remove tests for old beamtime data broken by time convention change Central support for these data is now deprecated (silently broken since 28a68d9bfeaa67150daa1f72008badd61bfaf24a). Redmine issue Refs #2301.and as patch:
diff --git a/macro/CMakeLists.txt b/macro/CMakeLists.txt index e20ca475..26f792a0 100644 --- a/macro/CMakeLists.txt +++ b/macro/CMakeLists.txt @@ -3,7 +3,6 @@ add_subdirectory(run) add_subdirectory(mcbm) add_subdirectory(mvd) add_subdirectory(much) -add_subdirectory(beamtime) add_subdirectory(include) @@ -54,4 +53,5 @@ Install(CODE "FILE(MAKE_DIRECTORY ${CMAKE_INSTALL_PREFIX}/share/cbmroot/macro/di #--- Deprecated (not running) #add_subdirectory(trd) #add_subdirectory(littrack) +#add_subdirectory(beamtime)If I did not do a mistake this should allow your MR to go through the pipelines, as we discussed in the Software meeting of this morning
Edited by Pierre-Alain Loizeau - Last reply by Felix Weiglhofer
added 6 commits
-
0091a7cc...ee11f32b - 4 commits from branch
computing:master - 78504ba0 - StsDigi: Add more error handling against timestamp overflows.
- d3ecaf94 - Remove tests for old beamtime data broken by time convention change
-
0091a7cc...ee11f32b - 4 commits from branch
- Resolved by Pierre-Alain Loizeau
Unfortunately it seems that the address we provide in the unpacker parameter is containing too many "newly unused" bits and now triggering the "address stability" assert.
I will try to compute the mSTS base addresses in the new scheme so that @a.toia can change them in the same MR as the one correcting the mSTS ladders swap
Edited by Pierre-Alain Loizeau - Last reply by Pierre-Alain Loizeau
added 9 commits
-
d3ecaf94...54df0e6d - 7 commits from branch
computing:master - efdffe85 - StsDigi: Add more error handling against timestamp overflows.
- 7c701561 - Remove tests for old beamtime data broken by time convention change
-
d3ecaf94...54df0e6d - 7 commits from branch
added 6 commits
-
7c701561...f0373ba0 - 4 commits from branch
computing:master - 37d47d48 - StsDigi: Add more error handling against timestamp overflows.
- d104c019 - Remove tests for old beamtime data broken by time convention change
-
7c701561...f0373ba0 - 4 commits from branch
Dear @v.friese, @f.uhlig, @p.-a.loizeau,
you have been identified as code owner of at least one file which was changed with this merge request.
Please check the changes and approve them or request changes.
added CodeOwners label
assigned to @v.friese