| From: | Michael Paquier <michael(at)paquier(dot)xyz> |
|---|---|
| To: | Ashutosh Bapat <ashutosh(dot)bapat(dot)oss(at)gmail(dot)com> |
| Cc: | pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Error in StrategySyncStart() prologue |
| Date: | 2025-01-30 08:36:32 |
| Message-ID: | Z5s6EE4oXFpb99xQ@paquier.xyz |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Thu, Jan 30, 2025 at 11:04:19AM +0530, Ashutosh Bapat wrote:
> The prologue of StrategySyncStart() mentions BufferSync(), but the
> latter function does not call the first function. Instead
> StrategySyncStart() is called by BgBufferSync() which uses the buffer
> id provided by the first function to decide the buffer from where to
> start syncing as mentioned in the first function's prologue.
> BufferSync, on the other hand, scans all the buffers and does not use
> the output of StrategySyncStart(). I think the prologue of
> StrategySyncStart() intends to refer to BgBufferSync() and not
> BufferSync(). PFA patch fixing the prologue.
Indeed, your suggestion sounds right. StrategySyncStart() is only
used in BgBufferSync() for the background writer since 9cd00c457e6a,
and this commit seems to have missed the comment update.
--
Michael
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Alexander Pyhalov | 2025-01-30 08:50:35 | Re: SQLFunctionCache and generic plans |
| Previous Message | Michael Banck | 2025-01-30 08:34:01 | Re: why there is not VACUUM FULL CONCURRENTLY? |