Closed fknorr closed 1 month ago
Check-perf-impact results: (7b849de16ff11660b98988ab0b032db7)
:question: No new benchmark data submitted. :question:
Please re-run the microbenchmarks and include the results if your commit could potentially affect performance.
Totals | |
---|---|
Change from base Build 9992874606: | 0.03% |
Covered Lines: | 7231 |
Relevant Lines: | 7505 |
Currently (buffer-manager runtime) this just works because lifetime extension makes sure that the final reference is dropped on TDAG pruning. The IDAG runtime detects the attempt to destroy a buffer from the wrong thread (scheduler thread, which prunes the IDAG) and aborts, but for some reason the most recent benchmark run timed out instead of aborting. Not sure why that was, but it worked on my machine.
It is illegal to value-capture buffers etc into host task lambdas, and actually breaks with the new runtime. Two instances of this pattern, both in
system_benchmarks.cc
, were missed in #244.