From mboxrd@z Thu Jan 1 00:00:00 1970 From: haojian.zhuang@linaro.org (Haojian Zhuang) Date: Tue, 29 Jul 2014 20:15:45 +0800 Subject: [PATCH v15 07/12] ARM: dts: append hip04 dts In-Reply-To: <20140729121300.GH9245@arm.com> References: <1406555876-11989-1-git-send-email-haojian.zhuang@linaro.org> <1406555876-11989-8-git-send-email-haojian.zhuang@linaro.org> <20140728180609.GE2576@leverpostej> <20140729035321.GA8430@quad.lixom.net> <20140729111213.GE9245@arm.com> <20140729121300.GH9245@arm.com> Message-ID: To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On 29 July 2014 20:13, Will Deacon wrote: > On Tue, Jul 29, 2014 at 12:32:09PM +0100, Haojian Zhuang wrote: >> On 29 July 2014 19:12, Will Deacon wrote: >> > On Tue, Jul 29, 2014 at 04:53:21AM +0100, Olof Johansson wrote: >> >> On Tue, Jul 29, 2014 at 10:44:40AM +0800, Haojian Zhuang wrote: >> >> > Because I didn't get these materials yet. All clocks are enabled in bootloader. >> >> >> >> What materials? Technical documentation for the CPU you're upstreaming? >> >> >> >> It seems like a very bad idea to upstream a DT for a CPU that you don't >> >> have documentation for. It seems appropriate to wait until you have >> >> documentation so you can make sure that the hardware you're describing >> >> is actually what is there, and not something made-up or misdescribed. >> > >> > I completely agree. Furthermore, given that this SoC implements a dodgy >> > custom GIC, I don't think it's unreasonable for us to ask for documentation >> > describing that too, otherwise it's going to make maintaining a complicated >> > irqchip driver that much worse. >> > >> I don't have enough documents on clocks. And all clocks are already >> enabled in UEFI. There're also no requirement on change the clock >> frequency & disable them. So I set all these clocks as fixed rate. >> >> It's different from GIC. If I don't have documents on GIC, I can't >> write code for HIP04 GIC. HiP04 GIC supports 16-cores. ARM GICv2 >> supports 8-cores. So some registers offsets and bit fields are >> different from GICv2. > > Sure, and I'm asking for those documents to be made publicly available > somewhere so that we can refer to them if we need to change this code in > future. > > Will I can send the copy to you since it's already public. But the main issue is that this document is written in Chinese. I hope it could be written in English. Regards Haojian