| From: | Jeff Davis <pgsql(at)j-davis(dot)com> |
|---|---|
| To: | Robert Haas <robertmhaas(at)gmail(dot)com> |
| Cc: | "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: 9.5: Better memory accounting, towards memory-bounded HashAgg |
| Date: | 2014-08-10 20:50:16 |
| Message-ID: | 1407703816.6623.2.camel@jeff-desktop |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Fri, 2014-08-08 at 01:16 -0700, Jeff Davis wrote:
> Either way, it's better to be conservative. Attached is a version of the
> patch with opt-in memory usage tracking. Child contexts inherit the
> setting from their parent.
There was a problem with the previous patch: the parent was unlinked
before the delete_context method was called, which meant that the
parent's memory was not being decremented.
Attached is a fix. It would be simpler to just reorder the operations in
MemoryContextDelete, but there is a comment warning against that. So, I
pass the old parent as an argument to the delete_context method.
Regards,
Jeff Davis
| Attachment | Content-Type | Size |
|---|---|---|
| memory-accounting-v3.patch | text/x-patch | 12.8 KB |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Jeff Davis | 2014-08-10 21:26:50 | 9.5: Memory-bounded HashAgg |
| Previous Message | Claudio Freire | 2014-08-10 17:43:17 | Re: Minmax indexes |