Choose timezone
Your profile timezone:
ROOT Review planning
Present The ROOT team, Peter, Aleandro, Valeri (R), Bill (R), Federico, John
o Axel:
- 101 issue to go for ROOT 6
- Philippe is working std::__1 issue
- Olivier is working on the "dump as macro" feature for canvases which
results in code which takes a lot of time to compile (minutes!)
o Alice:
- Alice going in background mode: R6 it was working already some time ago
- Run the comparison at a reasonable scale R5 and R6 after the grid services
upgrade
o Atlas:
- Valeri on vacation until next week
- Next step: migrate reflexless layer to Athena 19 gcc47
- Rolf: testing R 5.13 + plugin mgr against slightly different Atlas
release with success
- News about the R6 with Atlas to be expected in 2 weeks
o LHCb: Marco absent, this is his mail:
Gaudi:
waiting for tomorrow to see if Axel's fix to https://sft.its.cern.ch/jira/browse/ROOT-5838 works (last show-stopper in Gaudi).
workarounds enabled:
https://sft.its.cern.ch/jira/browse/ROOT-5836
duplicated dictionaries <https://sft.its.cern.ch/jira/browse/SPI-385>
issue with python.h included too late: at least cppyy should include it ASAP
probably not urgent but in debug builds the call of a C++ function is extremely slow (orders of magnitude wrt ROOT5), while in
the optimized build is still much slower than ROOT5, but more bearable (a factor 2-5, if I remember correctly)
LHCb:
regression between Monday and Tuesday (probably already fixed, but today's LHCb nightlies suffered from a different internal problem)
segfault in GetBaseOffset https://sft.its.cern.ch/jira/browse/ROOT-5840
LHCb lokinumbers0: wrong base exception spec https://sft.its.cern.ch/jira/browse/ROOT-5721
genreflex exclusion not working properly https://sft.its.cern.ch/jira/browse/ROOT-5911
warnings printed during dictionary load https://sft.its.cern.ch/jira/browse/ROOT-5841
I just started to analyze the problems with other projects in the LHCb Software stack. Most of them fail because of ROOT-5841 and
ROOT-5721, but there are other failure that I'll post on JIRA in the coming days (e.g. a segfault in quags, called by
ROOT::Math::GSLIntegrator::Integral).
o CMS:
- CMS has a fully compiling build!
- Fwk tests pass
- Rootfiles can be written and read back
- BUT a confirmation: exceptions do not propagate out of the interpreter
- Validation suite is too slow (hours VS minutes) - First order issue for
Bill to look at The suite ran was RunTheMatrix
- We don't have info about the flags used for the build: dbg, LLVMDEV etc.
o Cool: Andrea absent, news next week.
o Axel:
- 101 issue to go for ROOT 6
- Philippe is working std::__1 issue
- Olivier is working on the "dump as macro" feature for canvases which
results in code which takes a lot of time to compile (minutes!)
o Alice:
- Alice going in background mode: R6 it was working already some time ago
- Run the comparison at a reasonable scale R5 and R6 after the grid services
upgrade
o Atlas:
- Valeri on vacation until next week
- Next step: migrate reflexless layer to Athena 19 gcc47
- Rolf: testing R 5.13 + plugin mgr against slightly different Atlas
release with success
- News about the R6 with Atlas to be expected in 2 weeks
o LHCb: Marco absent, this is his mail:
Gaudi:
waiting for tomorrow to see if Axel's fix to https://sft.its.cern.ch/jira/browse/ROOT-5838 works (last show-stopper in Gaudi).
workarounds enabled:
https://sft.its.cern.ch/jira/browse/ROOT-5836
duplicated dictionaries <https://sft.its.cern.ch/jira/browse/SPI-385>
issue with python.h included too late: at least cppyy should include it ASAP
probably not urgent but in debug builds the call of a C++ function is extremely slow (orders of magnitude wrt ROOT5), while in
the optimized build is still much slower than ROOT5, but more bearable (a factor 2-5, if I remember correctly)
LHCb:
regression between Monday and Tuesday (probably already fixed, but today's LHCb nightlies suffered from a different internal problem)
segfault in GetBaseOffset https://sft.its.cern.ch/jira/browse/ROOT-5840
LHCb lokinumbers0: wrong base exception spec https://sft.its.cern.ch/jira/browse/ROOT-5721
genreflex exclusion not working properly https://sft.its.cern.ch/jira/browse/ROOT-5911
warnings printed during dictionary load https://sft.its.cern.ch/jira/browse/ROOT-5841
I just started to analyze the problems with other projects in the LHCb Software stack. Most of them fail because of ROOT-5841 and
ROOT-5721, but there are other failure that I'll post on JIRA in the coming days (e.g. a segfault in quags, called by
ROOT::Math::GSLIntegrator::Integral).
o CMS:
- CMS has a fully compiling build!
- Fwk tests pass
- Rootfiles can be written and read back
- BUT a confirmation: exceptions do not propagate out of the interpreter
- Validation suite is too slow (hours VS minutes) - First order issue for
Bill to look at The suite ran was RunTheMatrix
- We don't have info about the flags used for the build: dbg, LLVMDEV etc.
o Cool: Andrea absent, news next week.
There are minutes attached to this event.
Show them.