druid/processing
Vishesh Garg a27598a487
Segregate advance and advanceUninterruptibly flow in postJoinCursor to allow for interrupts in advance (#15222)
Currently advance function in postJoinCursor calls advanceUninterruptibly which in turn keeps calling baseCursor.advanceUninterruptibly until the post join condition matches, without checking for interrupts. This causes the CPU to hit 100% without getting a chance for query to be cancelled.

With this change, the call flow of advance and advanceUninterruptibly is separated out so that they call baseCursor.advance and baseCursor.advanceUninterruptibly in them, respectively, giving a chance for interrupts in the former case between successive calls to baseCursor.advance.
2023-10-30 14:39:15 +05:30
..
src Segregate advance and advanceUninterruptibly flow in postJoinCursor to allow for interrupts in advance (#15222) 2023-10-30 14:39:15 +05:30
pom.xml Prepare master for Druid 29 (#15121) 2023-10-11 10:33:45 +05:30