-
Notifications
You must be signed in to change notification settings - Fork 4.4k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Make SwitchProducer work with ConditionalTask #37563
Make SwitchProducer work with ConditionalTask #37563
Conversation
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-37563/29295
|
A new Pull Request was created by @makortel (Matti Kortelainen) for master. It involves the following packages:
@cmsbuild, @smuzaffar, @Dr15Jones, @makortel can you please review it and eventually sign? Thanks. cms-bot commands are listed here |
enable gpu |
if (not aliasModulesToProcess.empty() and | ||
aliasModulesToProcess.find(moduleLabel) == aliasModulesToProcess.end()) { | ||
continue; | ||
} |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I added the aliasModulesToProcess
argument and these lines when extracting the processEDAliases()
function into a separate file.
@@ -1298,7 +1298,7 @@ def _insertPaths(self, processPSet, nodeVisitor): | |||
l.sort() | |||
decoratedList.extend(l) | |||
decoratedList.append("@") | |||
iPath.insertInto(processPSet, triggername, decoratedList) | |||
iPath.insertInto(processPSet, triggername, decoratedList[:]) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
decoratedList
is copied because otherwise the same list
object would be shared among the Paths, and the contents would be that of the last Path.
-1 Failed Tests: UnitTests RelVals Unit TestsI found errors in the following unit tests: ---> test testTauEmbeddingProducers had ERRORS RelVals
GPU Comparison SummarySummary:
|
a964d75
to
20a4549
Compare
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-37563/29335
|
20a4549
to
ecdb11a
Compare
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-37563/29402
|
Pull request #37563 was updated. @cmsbuild, @smuzaffar, @Dr15Jones, @makortel can you please check and sign again. |
@cmsbuild, please test Addressed Chris' comments. |
-1 Failed Tests: UnitTests Unit TestsI found errors in the following unit tests: ---> test TestDQMOnlineClient-l1tstage2_dqm_sourceclient had ERRORS GPU Comparison SummarySummary:
Comparison SummarySummary:
|
+1 The unit test fails already in the IBs |
This pull request is fully signed and it will be integrated in one of the next master IBs (but tests are reportedly failing). This pull request will now be reviewed by the release team before it's merged. @perrotta, @dpiparo, @qliphy (and backports should be raised in the release meeting by the corresponding L2) |
merge |
PR description:
Further testing of
ConditionalTask
(#37305) for HLT revealed that it andSwitchProducer
mechanism didn't work together at all (as discussed in #36938 (comment) onwards). This PR makes theSwitchProducer
to work withConditionalTask
with the following changesSwitchProducer
case module objects get now their module labels set (previously they were not).SwitchProducerCUDA
and the case modules were always either defined inline or cloned from other module objectsSwitchProducer
is on aConditionalTask
or in aPath
, the case modules get treated as if they were in an anonymousConditionalTask
associated with theConditionalTask
/Path
SwitchProducer
with anEDAlias
as a chosen case required adding thoseEDAlias
es into theProductRegistry
so that thecallWhenNewProductsRegistered()
callbacks inSwitchProducer
get called, that lead to data product consumption being properly registered, that is needed for the module dependence checks inStreamSchedule::tryToPlaceConditionalModules()
processEDAliases()
function out fromSchedule.cc
SwitchProducer
withEDAlias
also revealed that theEDAlias
treatment was generally broken forConditionalTask
.PR validation:
Framework unit tests run. The HLT test in #36938 (comment) runs as well.