pgsql: Don't use a non-locked pre-test of the spinlock on x86_64 - Mailing list pgsql-committers

From tgl@svr1.postgresql.org (Tom Lane)
Subject pgsql: Don't use a non-locked pre-test of the spinlock on x86_64
Date
Msg-id 20051011200131.64BCAD7342@svr1.postgresql.org
Whole thread Raw
List pgsql-committers
Log Message:
-----------
Don't use a non-locked pre-test of the spinlock on x86_64 machines.
The pre-test has been shown to be a big loss on Opterons and at best a
wash on EM64T.

Modified Files:
--------------
    pgsql/src/include/storage:
        s_lock.h (r1.140 -> r1.141)
        (http://developer.postgresql.org/cvsweb.cgi/pgsql/src/include/storage/s_lock.h.diff?r1=1.140&r2=1.141)

pgsql-committers by date:

Previous
From: mha@pgfoundry.org (User Mha)
Date:
Subject: pginstaller - web: Add mention of multiprocess architecture Add
Next
From: tgl@svr1.postgresql.org (Tom Lane)
Date:
Subject: pgsql: Add infrastructure for making spins_per_delay variable depending