On 25.04.2013 07:40, Olivier Cochard-Labbé wrote: > On Wed, Apr 24, 2013 at 1:46 PM, Sami Halabi <sodynet1_at_gmail.com> wrote: >> 3. there some point of improved performance (without fw) that went down >> again somewhere before Clang got prod. > > Found it ! > > It's commit 242402: "Rework the known mutexes..." Again one has to be really careful drawing any firm conclusions from this as it was measured on a Pentium4 and UP kernel (GENERIC would add WITNESS and INVARIANT overhead as well). The Pentium4 is about the worst micro-architecture when it comes to locks and easily regresses. At the same time modern Intel Core i[3-7] and AMD64 may actually improve with these changes. Unless more recent micro-archs have been shown to exhibit the same regression we can't claim this change was bad (other than for Pentium4). -- Andre > ministat -s 242401.forwarding 242402.forwarding > x 242401.forwarding > + 242402.forwarding > +---------------------------------------------------------------------------------------+ > | + > | > |+ + + + > x xx x x| > | > |____A____| | > | |_____A_M___| > | > +---------------------------------------------------------------------------------------+ > N Min Max Median Avg Stddev > x 5 417527 420242 418902 419074 1049.7974 > + 5 402211 404828 404096 403689 1237.6696 > Difference at 95.0% confidence > -15385 +/- 1673.69 > -3.67119% +/- 0.399377% > (Student's t, pooled s = 1147.58) > _______________________________________________ > freebsd-current_at_freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-current > To unsubscribe, send any mail to "freebsd-current-unsubscribe_at_freebsd.org" > >Received on Thu Apr 25 2013 - 07:28:30 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:40:37 UTC