On Tue, Nov 8, 2022 at 11:59 AM Amit Kapila <amit.kapil...@gmail.com> wrote: > > On Mon, Nov 7, 2022 at 3:17 PM rajesh singarapu <rajesh.rs0...@gmail.com> > wrote: > > > > In both TransactionGroupUpdateXidStatus and ProcArrayGroupClearXid > > global MyProc is used. for consistency, replaced with a function local > > variable. > > > > In ProcArrayGroupClearXid(), currently, we always pass MyProc as proc, > so the change suggested by you will work but I think if in the future > someone calls it with a different proc, then the change suggested by > you won't work.
Well, yes. Do you have any thoughts around such future usages of ProcArrayGroupClearXid()? > The change in TransactionGroupUpdateXidStatus() looks > good but If we don't want to change ProcArrayGroupClearXid() then I am > not sure if there is much value in making the change in > TransactionGroupUpdateXidStatus(). AFICS, there are many places in the code that use proc == MyProc (20 instances) or proc != MyProc (6 instances) sorts of things. I think defining a macro, something like below, is better for readability. However, I'm concerned that we might have to use it in 26 places. #define IsPGPROCMine(proc) (proc != NULL && proc == MyProc) or just #define IsPGPROCMine(proc) (proc == MyProc) -- Bharath Rupireddy PostgreSQL Contributors Team RDS Open Source Databases Amazon Web Services: https://aws.amazon.com
