Re: Rework the way multixact truncations work - Mailing list pgsql-hackers

From Alvaro Herrera
Subject Re: Rework the way multixact truncations work
Date
Msg-id 20150923132909.GP295765@alvherre.pgsql
Whole thread Raw
In response to Re: Rework the way multixact truncations work  (Andres Freund <andres@anarazel.de>)
Responses Re: Rework the way multixact truncations work
List pgsql-hackers
> @@ -1210,8 +1211,14 @@ restart:;
>      (void) SlruScanDirectory(ctl, SlruScanDirCbDeleteCutoff, &cutoffPage);
>  }
>  
> -void
> -SlruDeleteSegment(SlruCtl ctl, char *filename)
> +/*
> + * Delete an individual SLRU segment, identified by the filename.
> + *
> + * NB: This does not touch the SLRU buffers themselves, callers have to ensure
> + * they either can't yet contain anything, or have already been cleaned out.
> + */
> +static void
> +SlruInternalDeleteSegment(SlruCtl ctl, char *filename)
>  {
>      char        path[MAXPGPATH];
>  
> @@ -1222,6 +1229,64 @@ SlruDeleteSegment(SlruCtl ctl, char *filename)
>  }
>  
>  /*
> + * Delete an individual SLRU segment, identified by the segment number.
> + */
> +void
> +SlruDeleteSegment(SlruCtl ctl, int segno)

Is it okay to change the ABI of SlruDeleteSegment?

-- 
Álvaro Herrera                http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services



pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: Calculage avg. width when operator = is missing
Next
From: Alvaro Herrera
Date:
Subject: Re: Calculage avg. width when operator = is missing