Re: [HACKERS] Make LWLockAcquireCommon() inline?
On 2014-08-30 14:22:40 +0200, Andres Freund wrote: > Hi, > > when profiling optimized builds (linux, gcc 4.9) it's currently > LWLockAcquireCommon() showing up, not it's callers. Instruction level > profiles show that the tests for valptr show up in profiles to some > extent. Since most callers don't need the valptr logic it seems prudent > to mark the function inline which will then eliminate the superflous > branches. > > Arguments against? Done. Greetings, Andres Freund -- Andres Freund http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Training & Services -- Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-hackers
[HACKERS] Make LWLockAcquireCommon() inline?
Hi, when profiling optimized builds (linux, gcc 4.9) it's currently LWLockAcquireCommon() showing up, not it's callers. Instruction level profiles show that the tests for valptr show up in profiles to some extent. Since most callers don't need the valptr logic it seems prudent to mark the function inline which will then eliminate the superflous branches. Arguments against? Greetings, Andres Freund -- Andres Freund http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Training & Services -- Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-hackers
