This is an automated cherry-pick of #57641

What problem does this PR solve?

Issue Number: close #57546

Problem Summary:

Partial worker gets input by calling getChildInput function. In this function, we call Consume function to track memory occupied by input chunk. When sql's memory usage is high, there will be possible to lead to panic in Consume function. According to the synchronization rule, each time we get a chunk from getChildInput we need to call Done function for variable inflightChunkSync. However, when panic happens Done function is not called and the counter in inflightChunkSync will never be minused to 0 which leads to hang.

Solution: Each time waked up in channel in getChildInput function, we will set a variable named needDone. When panic happens in getChildInput function, we will catch the panic in this function and check if variable needDone is set to true then decide if we need to call Done function for inflightChunkSync, and rethrow the panic at last.

What changed and how does it work?

Check List

Tests

• Unit test
• Integration test
• Manual test (add detailed scripts or steps below)
• No need to test

  • I checked and no code files have been changed.

Side effects

• Performance regression: Consumes more CPU
• Performance regression: Consumes more Memory
• Breaking backward compatibility

Documentation

• Affects user behaviors
• Contains syntax changes
• Contains variable changes
• Contains experimental features
• Changes MySQL compatibility

Release note

Please refer to Release Notes Language Style Guide to write a quality release note.

fix issue that tidb may hang in hash agg when exceeding memory limit leads to panic