On Thu, Dec 12, 2019 at 8:44 PM Robert Haas <robertmh...@gmail.com> wrote: > > On Tue, Dec 10, 2019 at 4:32 PM Andres Freund <and...@anarazel.de> wrote: > > and then, within an if(): > > > > /* > > * We don't try to do group update optimization if a > > process has > > * overflowed the subxids array in its PGPROC, since in > > that case we > > * don't have a complete list of XIDs for it. > > */ > > Assert(THRESHOLD_SUBTRANS_CLOG_OPT <= > > PGPROC_MAX_CACHED_SUBXIDS); > > > > Even if these weren't redundant, it can't make sense to test such a > > static condition only within an if? Is it possible this was actually > > intended to test something different? > > Based on the comment, I imagine it might've been intended to read > Assert(nsubxids <= PGPROC_MAX_CACHED_SUBXIDS). >
Do you think we need such an Assert after having StaticAssert for (THRESHOLD_SUBTRANS_CLOG_OPT <= PGPROC_MAX_CACHED_SUBXIDS) and then an if statement containing (nsubxids <= THRESHOLD_SUBTRANS_CLOG_OPT) just before this Assert? Sure, we can keep this for extra safety, but I don't see the need for it. -- With Regards, Amit Kapila. EnterpriseDB: http://www.enterprisedb.com
