Re: pgsql: Allow vacuum command to process indexes in parallel.

On Tue, 31 Mar 2020 at 17:28, Amit Kapila <amit.kapila16@gmail.com> wrote:
>
> On Mon, Mar 30, 2020 at 8:30 AM Mahendra Singh Thalor
> <mahi6run@gmail.com> wrote:
> >
> > > > > 2077          if ((shared_balance >= VacuumCostLimit) &&
> > > > > >>>     CID ...:  Incorrect expression  (UNINTENDED_INTEGER_DIVISION)
> > > > > >>>     Dividing integer expressions "VacuumCostLimit" and "nworkers", and then converting the integer
quotientto type "double". Any remainder, or fractional part of the quotient, is ignored.
 
> > > > > 2078                  (VacuumCostBalanceLocal > 0.5 * (VacuumCostLimit / nworkers)))
> > > > > 2079          {
> > > > > 2080                  /* Compute sleep time based on the local cost balance */
> > > > > 2081                  msec = VacuumCostDelay * VacuumCostBalanceLocal / VacuumCostLimit;
> > > > > 2082                  pg_atomic_sub_fetch_u32(VacuumSharedCostBalance, VacuumCostBalanceLocal);
> > > > > 2083                  VacuumCostBalanceLocal = 0;
> > > >
> > > > Which seems like a fair enough complaint?
> > > >
> > >
> > > I'll look into it.
> > >
> >
> > Hi,
> > Attaching patch to fix this but I don't have coverity setup so I
> > haven't verified fix.
> >
>
> - (VacuumCostBalanceLocal > 0.5 * (VacuumCostLimit / nworkers)))
> + (VacuumCostBalanceLocal > (int) (0.5 * ((double) VacuumCostLimit /
> nworkers))))
>   {
>
> I think typecasting to double should be enough to fix this coverity error.

I also think same so attaching updated patch.

-- 
Thanks and Regards
Mahendra Singh Thalor
EnterpriseDB: http://www.enterprisedb.com