Re: Question for coverage report - Mailing list pgsql-hackers

From Tom Lane
Subject Re: Question for coverage report
Date
Msg-id 795389.1761148306@sss.pgh.pa.us
Whole thread Raw
In response to Re: Question for coverage report  (Jacob Champion <jacob.champion@enterprisedb.com>)
List pgsql-hackers
Jacob Champion <jacob.champion@enterprisedb.com> writes:
> (I don't know the answer to this question, but I will note that clang
> (15.0.7) does not seem to make this mistake on my machine, and reports
> a call count of zero for the `return` on line 1495. Looking at the
> disassembly, it seems to add more instrumentation points than what Tom
> showed for gcc.)

Interesting.  I also realized, after re-reading the snippet I showed,
that gcc is treating the code leading up to a CALL instruction as a
separate basic block from the code following the CALL.  So that begs
the question of which count is shown for the function call's line
at the source-code level.  It'd only differ when the function throws
an error, presumably.

            regards, tom lane



pgsql-hackers by date:

Previous
From: Álvaro Herrera
Date:
Subject: Re: LISTEN/NOTIFY bug: VACUUM sets frozenxid past a xid in async queue
Next
From: Tom Lane
Date:
Subject: Re: Confine vacuum skip logic to lazy_scan_skip