Re: DSA_ALLOC_NO_OOM vs dsm_create ERROR leaving a half-initialized pgstats hash entry - Mailing list pgsql-hackers

From Grigorev Jurij
Subject Re: DSA_ALLOC_NO_OOM vs dsm_create ERROR leaving a half-initialized pgstats hash entry
Date
Msg-id ee29f27c0c9f4520b3cf2135f7dc7a2b@localhost.localdomain
Whole thread
In response to Re: DSA_ALLOC_NO_OOM vs dsm_create ERROR leaving a half-initialized pgstats hash entry  (Michael Paquier <michael@paquier.xyz>)
List pgsql-hackers
Hi Michael,

Thanks very much for taking the time to clean this up and commit it.  I
reviewed d9de60c -- restoring the existing refcount and generation
comments was the right call, and the shorter wording still captures the
allocation-before-insert invariant.  The final ordering and cleanup
paths look good to me.

Thanks,
  Yuriy


pgsql-hackers by date:

Previous
From: Chao Li
Date:
Subject: Re: eliminate xl_heap_visible to reduce WAL (and eventually set VM on-access)
Next
From: Amit Langote
Date:
Subject: RI fastpath misses checking EXECUTE on functions