Hello, This attached patch re-designed the previous v2 according to Robert's comment. All I had to do is putting entrypoint for ForeignScan/CustomScan at ExecShutdownNode, because it is already modified to call child node first, earlier than ExecShutdownGather which releases DSM segment.
Thanks, 2017-02-20 9:25 GMT+09:00 Kouhei Kaigai <[email protected]>: >> -----Original Message----- >> From: [email protected] >> [mailto:[email protected]] On Behalf Of Robert Haas >> Sent: Monday, February 20, 2017 2:20 AM >> To: Kaigai Kouhei(海外 浩平) <[email protected]> >> Cc: Claudio Freire <[email protected]>; Amit Kapila >> <[email protected]>; pgsql-hackers <[email protected]> >> Subject: Re: ParallelFinish-hook of FDW/CSP (Re: [HACKERS] Steps inside >> ExecEndGather) >> >> On Fri, Feb 17, 2017 at 12:46 PM, Kouhei Kaigai <[email protected]> >> wrote: >> > The attached patch is revised one. >> > >> > Invocation of Exec(Foreign|Custom)ParallelFinish was moved to >> > ExecParallelRetrieveInstrumentation() not to walk on the plan- state >> > tree twice. >> > One (hypothetical) downside is, FDW/CSP can retrieve its own run-time >> > statistics only when query is executed under EXPLAIN ANALYZE. >> > >> > This enhancement allows FDW/CSP to collect its specific run- time >> > statistics more than Instrumentation, then show them as output of >> > EXPLAIN. My expected examples are GPU's kernel execution time, DMA >> > transfer ratio and so on. These statistics will never appear in the >> > Instrumentation structure, however, can be a hot- point of performance >> > bottleneck if CustomScan works on background workers. >> >> Would gather_shutdown_children_first.patch from >> https://www.postgresql.org/message-id/CAFiTN-s5KuRuDrQCEpiHHzmVf7JTtbn >> [email protected] >> help with this problem also? Suppose we did that, and then also added an >> ExecShutdownCustom method. Then you'd definitely be able to get control >> before the DSM went away, either from ExecEndNode() or ExecShutdownNode(). >> > Ah, yes, I couldn't find any problem around the above approach. > ExecShutdownGather() can be called by either ExecShutdownNode() or > ExecEndGather(). This patch allows to have an entrypoint for CSP/FDW > prior to release of the DSM. > > Thanks, > ---- > PG-Strom Project / NEC OSS Promotion Center > KaiGai Kohei <[email protected]> > > > -- > Sent via pgsql-hackers mailing list ([email protected]) > To make changes to your subscription: > http://www.postgresql.org/mailpref/pgsql-hackers -- KaiGai Kohei <[email protected]>
parallel-finish-fdw_csp.v3.patch
Description: Binary data
-- Sent via pgsql-hackers mailing list ([email protected]) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-hackers
