| From: | Michael Paquier <michael(dot)paquier(at)gmail(dot)com> |
|---|---|
| To: | Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> |
| Cc: | Jim Nasby <Jim(dot)Nasby(at)bluetreble(dot)com>, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Pg Hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Odd behavior with PG_TRY |
| Date: | 2017-01-06 11:43:14 |
| Message-ID: | CAB7nPqS8mVqFd9mw84UbaofaiXrfd0pgRfeZrhbL8DMdHJayTA@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Thu, Jan 5, 2017 at 7:10 PM, Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> wrote:
> Memory contexts used in catch block also doesn't seem to be marked as
> volatile, you might want to try by marking them as volatile. Also, it
> might worth trying it on some other system to see if you are by any
> chance hitting the problem similar to what Tom has faced.
If a variable is modified within PG_TRY and then referenced in
PG_CATCH it needs to be marked as volatile to be strictly in
conformance with POSIX. This also ensures that any compiler does not
do any stupid optimizations with those variables in the way they are
referenced and used.
--
Michael
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Ashutosh Bapat | 2017-01-06 12:25:01 | Re: postgres_fdw : altering foreign table not invalidating prepare statement execution plan. |
| Previous Message | Ashutosh Bapat | 2017-01-06 11:34:09 | Re: Parallel Append implementation |