From mboxrd@z Thu Jan 1 00:00:00 1970 From: Pali =?utf-8?B?Um9ow6Fy?= Subject: Re: [RESEND] [PATCH v2 1/2] arm: devtree: Set system_rev from DT revision Date: Thu, 25 Jun 2015 09:18:51 +0200 Message-ID: <20150625071851.GB2890@pali> References: <1430902142-17035-1-git-send-email-pali.rohar@gmail.com> <3577431.y872Foi4kj@wuerfel> <20150506103752.GG30910@pali> <3529567.PeFp1NNa1q@wuerfel> <20150506114417.GI30910@pali> <20150625050138.GL4156@atomide.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: <20150625050138.GL4156@atomide.com> Sender: linux-kernel-owner@vger.kernel.org To: Tony Lindgren Cc: Arnd Bergmann , linux-arm-kernel@lists.infradead.org, Rob Herring , Russell King , Will Deacon , Ivaylo Dimitrov , Sebastian Reichel , Pavel Machek , Andreas =?utf-8?Q?F=C3=A4rber?= , linux-omap@vger.kernel.org, "linux-kernel@vger.kernel.org" , devicetree@vger.kernel.org List-Id: devicetree@vger.kernel.org On Wednesday 24 June 2015 22:01:38 Tony Lindgren wrote: > * Pali Roh=C3=A1r [150506 04:45]: > > On Wednesday 06 May 2015 13:04:01 Arnd Bergmann wrote: > > > >=20 > > > > It needs to be done in this code, so "system_rev" variable is s= et > > > > properly... > > >=20 > > > What I mean is which code accesses this variable that early? > > >=20 > >=20 > > ATAG code is doing it at same early stage, so I added it to same ea= rly > > stage... >=20 > Yes we should do this early like the other atags. > =20 > > > > > Also, it seems strange to have a string property and then use= kstrtouint > > > > > to convert it into a number. I think it should either be spec= ified in a DT > > > > > binding to be a string and then have the kernel not assume th= at it is a number, > > > > > or we should define it to be binary. > > > > >=20 > > > > > Arnd > > > >=20 > > > > Variable "system_rev" is number and it always was. So chaning t= ype will > > > > break more parts. > > > >=20 > > > > And it is string DT property to be human readable. Some other d= evelopers > > > > suggested for v2 to change it to string (from number). > > >=20 > > > Both of them would be human readable, you just use something else= to > > > read them ;-) > > >=20 > > > If we have a string here, we should just change all uses of syste= m_rev > > > in the kernel accordingly, there are only a few of them: >=20 > Let's just keep it as a hex as it was. After all it's an existing > interface in /proc that user space programs may expect to be in > hex format already. >=20 > Pali, care to repost the whole set again right after -rc1 with > with rev property naming and documentation added? Just keep it > as hex and let's forget any string conversion. >=20 > Regards, >=20 > Tony Ok, but what do you mean to forget any string conversion? --=20 Pali Roh=C3=A1r pali.rohar@gmail.com