On 3/19/21 3:57 AM, Tom Lane wrote: > Tomas Vondra <tomas.von...@enterprisedb.com> writes: >> So crake and florican seem to be happy now. Not sure about lapwing yet. >> But interestingly enough, prion and curculio got unhappy. They worked >> fine with the older test, but now it fails with the "no such file or >> directory" message. I wonder what makes them different from the other >> x86_64 machines ... > > I'm confused about why the new query would use a temp file at all. > Maybe they aren't using the same plan? >
Because generate_series() stashes the results into a tuplestore, and the test sets work_mem to just 64kB. Maybe I'm missing some detail, though. The plan is extremely simple - just Function Scan feeding data into an Insert, not sure what other plan could be used. regards -- Tomas Vondra EnterpriseDB: http://www.enterprisedb.com The Enterprise PostgreSQL Company
