03:44 <sb0> rjo, you should make several calls to smgr.register() instead of dumping it all into Browser.save_state()/restore_state()

03:45 <sb0> like in the dashboard

03:45 <sb0> smgr will keep restoring the state of other components if one fails, your code won't

03:47 <sb0> also "sub" stands for "subscriber"; models.LocalModelManager is not really a subscriber

04:26 <GitHub135> [artiq] sbourdeauducq pushed 6 new commits to master: https://git.io/vKfIz

04:26 <GitHub135> artiq/master 4c8a835 Sebastien Bourdeauducq: worker: increase send_timeout (Windows can be really slow)

04:26 <GitHub135> artiq/master fdc2577 Sebastien Bourdeauducq: master/dataset_db: support keeping old persist flag

04:26 <GitHub135> artiq/master cc9edc1 Sebastien Bourdeauducq: frontend/client: support keeping persist flag

04:26 <GitHub159> [artiq] sbourdeauducq pushed 1 new commit to release-1: https://git.io/vKfIg

04:26 <GitHub159> artiq/release-1 bf3c3cd Sebastien Bourdeauducq: worker: increase send_timeout (Windows can be really slow)

04:30 <GitHub10> [artiq] sbourdeauducq pushed 1 new commit to release-1: https://git.io/vKfIw

04:30 <GitHub10> artiq/release-1 1946e3c Sebastien Bourdeauducq: language/environment: fix mutate_dataset with parent. Closes #498

04:39 <bb-m-labs> build #513 of artiq-kc705-nist_clock is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq-kc705-nist_clock/builds/513

04:44 <bb-m-labs> build #245 of artiq-win64-test is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq-win64-test/builds/245

04:46 <bb-m-labs> build #784 of artiq is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/784

19:45 <mumptai> Is there a reason for the testbenches in misoc still being written for legacy branch? any other than other things being more important, and the cores already having being tested before?