Remove support for 'userlocks'?

On Fri, Jun 3, 2011 at 1:01 PM, Bruce Momjian <bruce@momjian.us> wrote:

According to our documentation, 'userlocks' were removed in PG 8.2:

       http://developer.postgresql.org/pgdocs/postgres/runtime-config-developer.html

       trace_userlocks (boolean)

       If on, emit information about user lock usage. Output is the same as
       for trace_locks, only for user locks.

       User locks were removed as of PostgreSQL version 8.2. This option
       currently has no effect.

       This parameter is only available if the LOCK_DEBUG macro was defined
       when PostgreSQL was compiled.

Should we remove this parameter and the supporting code since pre-8.2 is
now end-of-life?

hm, shouldn't it be replaced with trace_advisory_locks?

merlin

On 6/3/11 11:01 AM, Bruce Momjian wrote:

According to our documentation, 'userlocks' were removed in PG 8.2:

http://developer.postgresql.org/pgdocs/postgres/runtime-config-developer.html

I take it this doesn't trace advisory locks, and trace_locks does?

If so, then by all means remove the code. I highly doubt anyone is
using that option anyway ...

--
Josh Berkus
PostgreSQL Experts Inc.
http://pgexperts.com

Josh Berkus wrote:

On 6/3/11 11:01 AM, Bruce Momjian wrote:

According to our documentation, 'userlocks' were removed in PG 8.2:

http://developer.postgresql.org/pgdocs/postgres/runtime-config-developer.html

I take it this doesn't trace advisory locks, and trace_locks does?

If so, then by all means remove the code. I highly doubt anyone is
using that option anyway ...

OK, that is something I will do for 9.2.

--
Bruce Momjian <bruce@momjian.us> http://momjian.us
EnterpriseDB http://enterprisedb.com

+ It's impossible for everything to be true. +

Josh Berkus wrote:

On 6/3/11 11:01 AM, Bruce Momjian wrote:

According to our documentation, 'userlocks' were removed in PG 8.2:

http://developer.postgresql.org/pgdocs/postgres/runtime-config-developer.html

I take it this doesn't trace advisory locks, and trace_locks does?

If so, then by all means remove the code. I highly doubt anyone is
using that option anyway ...

The attached, applied patch removes all "traces" of trace_userlocks.

--
Bruce Momjian <bruce@momjian.us> http://momjian.us
EnterpriseDB http://enterprisedb.com

+ It's impossible for everything to be true. +