Content-Type: multipart/mixed; boundary="===============4314844856481051638==" MIME-Version: 1.0 From: Sergey Senozhatsky Subject: Re: [Powertop] [PATCH] abstract_cpu: do not call total_pstate_time() in validate() Date: Fri, 14 Sep 2012 13:16:13 +0300 Message-ID: <20120914101613.GA3053@swordfish.minsk.epam.com> In-Reply-To: 1347616856.6783.11.camel@mattotaupa To: powertop@lists.01.org List-ID: --===============4314844856481051638== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable On (09/14/12 12:00), Paul Menzel wrote: > = > Dear Sergey, > = > Am Donnerstag, den 13.09.2012, 15:41 +0300 schrieb Sergey Senozhatsky: > > abstract_cpu: do not call total_pstate_time() in validate() > = > copying the commit summary into the commit message body is not useful. > = > Maybe also describe why this was a problem. > while I agree, this commit was not fixing any real problem (perhaps, meager performance impact), but just cleaned up abstract_cpu part after f77ea5fe2ad46b3c5de46b8447d23bd12958d2c8 commit: - if (my_time !=3D children[i]->total_pstate_time()) - printf("My (%i) time %llu is not the same a= s child (%i) time %llu\n", - first_cpu, - (unsigned long long)my_time, - children[i]->number, - (unsigned long long)children[i]->to= tal_pstate_time()); for such trivial commits I usually cut corners since commit does what the s= ummary says. may be I should be more accurate. -ss --===============4314844856481051638==--