On Thu, May 18, 2017 at 8:27 PM, Bob Friesenhahn <bfrie...@simple.dallas.tx.us> wrote: > On Thu, 18 May 2017, Kim Gräsman wrote: > >> On Thu, May 18, 2017 at 6:10 PM, Bob Friesenhahn >> <bfrie...@simple.dallas.tx.us> wrote: >>> >>> Does anyone know why using 'PRAGMA cache_size=0' (or some other small >>> value) >>> to attempt to decrease memory usage (and it is reported as immediately >>> decreased in the shell by .stats) actually significantly increases heap >>> memory usage? >> >> >> It sounds like you have active operation with a larger cache size >> before issuing the PRAGMA, is that so? > > > The request is issued early on when the connection is first opened so no > actual queries have been issued at that time.
Then my (black-box) guess is that you're seeing the bump from allocating heap space for whatever structures the schema needs. > An earlier developer had tried the same thing almost 5 years ago (with a > much older sqlite) and noticed a 200k jump in heap usage. We're at 3.14.1, so I'm also speaking from an older timeframe. - Kim _______________________________________________ sqlite-users mailing list sqlite-users@mailinglists.sqlite.org http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
