From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from f352.mail.ru ([217.69.140.248]) by merlin.infradead.org with esmtps (Exim 4.80.1 #2 (Red Hat Linux)) id 1UUJVp-00052i-0H for barebox@lists.infradead.org; Mon, 22 Apr 2013 16:22:41 +0000 From: =?UTF-8?B?QWxleGFuZGVyIFNoaXlhbg==?= Mime-Version: 1.0 Date: Mon, 22 Apr 2013 20:22:39 +0400 Message-ID: <1366647759.828662080@f352.mail.ru> In-Reply-To: <20130422161540.GC13393@game.jcrosoft.org> References: <1366439718-31495-1-git-send-email-shc_work@mail.ru> <1366640359.9607106@f239.mail.ru> <20130422161540.GC13393@game.jcrosoft.org> Reply-To: =?UTF-8?B?QWxleGFuZGVyIFNoaXlhbg==?= List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: "barebox" Errors-To: barebox-bounces+u.kleine-koenig=pengutronix.de@lists.infradead.org Subject: =?UTF-8?B?UmVbMl06IFtQQVRDSF0gY2xvY2tzb3VyY2U6IEFkZCBwb2ludGVyIHRvIGNs?= =?UTF-8?B?b2Nrc291cmNlIHN0cnVjdHVyZSBpbiByZWFkIGZ1bmN0aW9u?= To: =?UTF-8?B?SmVhbi1DaHJpc3RvcGhlIFBMQUdOSU9MLVZJTExBUkQ=?= Cc: barebox@lists.infradead.org > On 18:19 Mon 22 Apr , Alexander Shiyan wrote: > > > On 13:13 Mon 22 Apr , Alexander Shiyan wrote: > > > > > On 10:35 Sat 20 Apr , Alexander Shiyan wrote: > > > > > > > > > > > > This patch adds pointer to clocksource structure in read function. > > > > > > For avoid name collisions, clocksource names for all drivers was > > > > > > changed to be unique. > > > > > why? > > > > > > > > > > no need the clocksource is hot path don't touch it > > > > > and we do not need the struct pointer as we can have only 1 cs in barebox > > > > > > > > This change allows to add an additional field in struct clocksource then > > > > we can remove static variables from clocksource drivers. > > > why we have only one clocksource at a time > > > what you propose will require to use container_of > > > > > > so will slow down a lot barebox > > > > I hope to see any comments about this patch from Sascha. > > Of course, we have only one instance of clocksource, but generally > > direction of this patch is avoid of usage static variables in clocksource drivers. > > With path we can remove all usage. Is this not good for you? > if this result by increase the instruction to execute it's worth If parameter is not used by unit I think compiler should compile this as before. E.g. whithout an additioal code. --- _______________________________________________ barebox mailing list barebox@lists.infradead.org http://lists.infradead.org/mailman/listinfo/barebox