[BUG] Skipped initialization of some xl_xact_parsed_prepare fields - Mailing list pgsql-hackers

From Daniil Davydov
Subject [BUG] Skipped initialization of some xl_xact_parsed_prepare fields
Date
Msg-id CAJDiXgh-6Epb2XiJe4uL0zF-cf0_s_7Lw1TfEHDMLzYjEmfGOw@mail.gmail.com
Whole thread Raw
List pgsql-hackers
Hi,
I noticed that inside ParsePrepareRecord function we are missing
initialization of `nstats` and `nabortstats` fields of
xl_xact_parsed_prepare structure:
*** (current code in master)
memset(parsed, 0, sizeof(*parsed));

parsed->xact_time = xlrec->prepared_at;
parsed->origin_lsn = xlrec->origin_lsn;
parsed->origin_timestamp = xlrec->origin_timestamp;
parsed->twophase_xid = xlrec->xid;
parsed->dbId = xlrec->database;
parsed->nsubxacts = xlrec->nsubxacts;
parsed->nrels = xlrec->ncommitrels;
parsed->nabortrels = xlrec->nabortrels;
parsed->nmsgs = xlrec->ninvalmsgs;
***

Thus, they will always be 0 after calling the ParsePrepareRecord
function, but `stats` and `abortstats` pointers are set correctly:
*** (current code in master)
parsed->stats = (xl_xact_stats_item *) bufptr;
bufptr += MAXALIGN(xlrec->ncommitstats * sizeof(xl_xact_stats_item));

parsed->abortstats = (xl_xact_stats_item *) bufptr;
bufptr += MAXALIGN(xlrec->nabortstats * sizeof(xl_xact_stats_item));
***

If we look (for example) on parsing of a commit record, we could see
that both `nstats` and `stats` fields are set inside the
ParseCommitRecord function.
For now, zeroed number of stats lead to invalid output of
`xact_desc_prepare`, because it relies on values of `nstats` and
`nabortstats` fields:
*** (current code in master)
xact_desc_stats(buf, "commit ", parsed.nstats, parsed.stats);
xact_desc_stats(buf, "abort ", parsed.nabortstats, parsed.abortstats);
***

I suppose to add initialization of `nstats` and `nabortstats` to
ParsePrepareRecord (see attached patch).

--
Best regards,
Daniil Davydov

Attachment

pgsql-hackers by date:

Previous
From: wenhui qiu
Date:
Subject: Re: Should we optimize the `ORDER BY random() LIMIT x` case?
Next
From: Richard Guo
Date:
Subject: Re: Assert failure in base_yyparse