From mboxrd@z Thu Jan 1 00:00:00 1970 From: xuwei5@hisilicon.com (Wei Xu) Date: Thu, 25 Jun 2015 08:57:35 +0100 Subject: [GIT PULL]Hisilicon hip04 soc and D01 board DT updates for v4.2 In-Reply-To: References: <554CC781.9010705@hisilicon.com> <5554E035.6020004@hisilicon.com> <2091000.vbEIJY2sm3@wuerfel> Message-ID: <558BB46F.8030406@hisilicon.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On 6/25/2015 1:24 AM, Kevin Hilman wrote: > On Fri, May 15, 2015 at 2:28 AM, Arnd Bergmann wrote: >> On Thursday 14 May 2015 18:49:41 Wei Xu wrote: >>> On 5/14/2015 2:55 PM, Tyler Baker wrote: >>>> On 14 May 2015 at 06:36, Tyler Baker wrote: >>>>> On 14 May 2015 at 01:49, Wei Xu wrote: >>>>>> On 5/13/2015 8:20 PM, Tyler Baker wrote: >>>>>>> On 12 May 2015 at 08:09, Wei Xu wrote: >>>>>>> When this was pulled into arm-soc, the kernelci.org bot reported >>>>>>> hip04-d01 boot failures[1]. Before bisecting, I decided to wait a day >>>>>>> and double check the status of -next in case there were needed >>>>>>> dependencies from other trees. However, now there are the same >>>>>>> hip04-d01 boot failures in next-20150513[2]. My robot has bisected[3] >>>>>>> these boot failures down to... >>>>>>> >>>>>>> 828fcfe3c84ade68ecc3f1ba78baebc1654977f3 is the first bad commit >>>>>>> commit 828fcfe3c84ade68ecc3f1ba78baebc1654977f3 >>>>>>> Author: Zhou Wang >>>>>>> Date: Wed Mar 25 14:57:45 2015 +0800 >>>>>>> >>>>>>> ARM: dts: hip04: add GPIO pieces >>>>>>> >>>>>>> Hisilicon Soc hip04 has four GPIO controllers, each one has 32 >>>>>>> GPIOs and can be configured to be an interrupt controller.The GPIO >>>>>>> controllers are compatible with the snps,dw-apb-gpio driver. >>>>>>> This patch add the corresponding device tree nodes. >>>>>>> >>>>>>> Acked-by: Linus Walleij >>>>>>> Signed-off-by: Zhou Wang >>>>>>> Signed-off-by: Wei Xu >>>>>> >>>>>> >>>>>> It is a little strange. >>>>>> I checked to this commit "28837bc3e732610ebc7c88ce205dbe43245b1cb8" >>>>>> at "git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc.git". >>>>>> and compiled with "hisi_defconfig". >>>>>> >>>>>> It is booting OK and the log is as below. >>>>>> And I have also uploaded the binary at "https://github.com/hisilicon/boards/tree/master/D01/for-tyler". >>>>>> Could you please confirm it again? >>>>> >>>>> It appears we are using different image formats to boot. The link >>>>> above shows .kernel, .monitor, .text...can I split the kernel//dtb out >>>>> from .kernel? >>>>> >>>>> I am using zImage and hip04-d01.dtb directly[1] with UEFI to TFTP >>>>> boot. Could you try this method? >>> >>>> For good measure, I tried booting next-20150514 on the D01 hardware in >>>> the Linaro lab to confirm the same failure I see locally. It fails to >>>> boot as well[1]. >>> >>> Thanks! >>> I have tried your binary files and found it is stuck in the UEFI. >>> I will continue to investigate this issue. >>> >>> Hi Arnd, >>> >>> Could you please revert that branch? >>> After I resolve this issue I will send another pull request. >>> Sorry to bother you! >>> Thanks! >>> >> >> I have applied the patch below on the for-next branch now, to revert everything >> from this pull request. I have not changed the contents of the next/dt branch >> for now, and am still waiting for a bug fix. Once you know what caused the >> problem, please send a fix that applies on top of the next/dt branch in arm-soc, >> and I'll reactivate it. Hi Kevin, > > FYI... I'm not aware of any fixes for this, so I'll be removing this > branch from the next/dt branch of arm-soc before sending to Linus. > OK. We are still investigating this issue. Once we resolved it, I will send out a new patch. Thanks! Best Regards, Wei > Kevin > > . >