From mboxrd@z Thu Jan 1 00:00:00 1970 Content-Type: multipart/mixed; boundary="===============4112993796020157462==" MIME-Version: 1.0 From: Magnus Fromreide Subject: Re: [Powertop] [PATCH] Get rid of ncurses' TRUE and FALSE Date: Mon, 23 Jul 2012 09:31:33 +0200 Message-ID: <1343028693.3103.4.camel@sara> In-Reply-To: 500CF009.908@samsung.com To: powertop@lists.01.org List-ID: --===============4112993796020157462== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable On Mon, 2012-07-23 at 10:32 +0400, Igor Zhbanov wrote: > Hi Chris! > Thank you for your answer. > = > Chris Ferron wrote: > > On 07/20/2012 12:06 AM, Igor Zhbanov wrote: > >> > >> Linaro has made only few commits after forking from mainline. > >> > >> They are: > >> > >> 1) adding __NR_perf_event_open define > > Please insure this is fresh code authored by you and detailed if you = > > submit, and I will be happy to take it. > Yes, this is just a system call number copied from unistd.h for ARM = > platform. > Without it PowerTOP will not know what system call number = > __NR_perf_event_open has on ARM. Why can't it take __NR_perf_event_open from the libc headers? That is, why can't the whole ifdef under /* some people have stale headers */ go? Powertop is rather demanding with regards to the environment anyhow and magic constants are never fun. /MF --===============4112993796020157462==--