Thread: pgsql: Re-enable the old code in xlog.c that tried to use

Log Message:
-----------
Re-enable the old code in xlog.c that tried to use posix_fadvise(), so that
we can get some buildfarm feedback about whether that function is still
problematic.  (Note that the planned async-preread patch will not really
prove anything one way or the other in buildfarm testing, since it will
be inactive with default GUC settings.)

Modified Files:
--------------
    pgsql:
        configure (r1.625 -> r1.626)
        (http://anoncvs.postgresql.org/cvsweb.cgi/pgsql/configure?r1=1.625&r2=1.626)
        configure.in (r1.584 -> r1.585)
        (http://anoncvs.postgresql.org/cvsweb.cgi/pgsql/configure.in?r1=1.584&r2=1.585)
    pgsql/src/backend/access/transam:
        xlog.c (r1.326 -> r1.327)
        (http://anoncvs.postgresql.org/cvsweb.cgi/pgsql/src/backend/access/transam/xlog.c?r1=1.326&r2=1.327)
    pgsql/src/include:
        pg_config.h.in (r1.137 -> r1.138)
        (http://anoncvs.postgresql.org/cvsweb.cgi/pgsql/src/include/pg_config.h.in?r1=1.137&r2=1.138)
        pg_config_manual.h (r1.35 -> r1.36)
        (http://anoncvs.postgresql.org/cvsweb.cgi/pgsql/src/include/pg_config_manual.h?r1=1.35&r2=1.36)

tgl@postgresql.org (Tom Lane) writes:

> Log Message:
> -----------
> Re-enable the old code in xlog.c that tried to use posix_fadvise(), so that
> we can get some buildfarm feedback about whether that function is still
> problematic.  (Note that the planned async-preread patch will not really
> prove anything one way or the other in buildfarm testing, since it will
> be inactive with default GUC settings.)

! #if defined(USE_POSIX_FADVISE) && defined(POSIX_FADV_DONTNEED)

I think if we're not putting the AC_CHECK_RUN check in then this should just
be HAVE_POSIX_FADVISE not USE_POSIX_FADVISE. The latter was the define I made
if the run check passed.

--
  Gregory Stark
  EnterpriseDB          http://www.enterprisedb.com
  Get trained by Bruce Momjian - ask me about EnterpriseDB's PostgreSQL training!

Gregory Stark <greg.stark@enterprisedb.com> writes:
> tgl@postgresql.org (Tom Lane) writes:
> ! #if defined(USE_POSIX_FADVISE) && defined(POSIX_FADV_DONTNEED)

> I think if we're not putting the AC_CHECK_RUN check in then this should just
> be HAVE_POSIX_FADVISE not USE_POSIX_FADVISE. The latter was the define I made
> if the run check passed.

No, checking the DECL part as well is critical (if we don't see a
declaration we risk ABI incompatibility).  And in any case I'm insisting
on being able to turn this off in pg_config_manual.h.

            regards, tom lane