Re: [HACKERS] GSoC 2017: weekly progress reports (week 8) - Mailing list pgsql-hackers

Project: Explicitly support predicate locks in index AMs besides b-tree

Hi,

During this week, I worked on predicate locking in spgist index. I think, for spgist index, predicate lock only on leaf pages will be enough as spgist searches can determine if there is a match or not only at leaf level.

I have done following things in this week.

1) read the source code of spgist index to understand  the access method

2) found appropriate places to insert calls to existing functions

3) created tests (to verify serialization failures and to demonstrate the feature of reduced false positives) for 'point' and 'box' data types.

    

    link to code and tests: https://github.com/shubhambaraiss/postgres/commit/d9ae709c93ff038478ada411c621faeabe6813cb

I will attach the patch shortly.

Regards,

Shubham

Sent with Mailtrack