linux-kernel - Re: [PATCH 1/6] arm64: dts: qcom: sm8150: add apps

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <6d0aa709dbd00ceda4e27d5d49ecc9ff@codeaurora.org>
Date:   Fri, 05 Jun 2020 20:34:12 +0530
From:   Sai Prakash Ranjan <saiprakash.ranjan@...eaurora.org>
To:     Nicolas Dechesne <nicolas.dechesne@...aro.org>
Cc:     Jonathan Marek <jonathan@...ek.ca>,
        Bjorn Andersson <bjorn.andersson@...aro.org>,
        linux-arm-msm <linux-arm-msm@...r.kernel.org>,
        Andy Gross <agross@...nel.org>,
        Rob Herring <robh+dt@...nel.org>,
        "open list:OPEN FIRMWARE AND FLATTENED DEVICE TREE BINDINGS" 
        <devicetree@...r.kernel.org>,
        open list <linux-kernel@...r.kernel.org>,
        devicetree-owner@...r.kernel.org
Subject: Re: [PATCH 1/6] arm64: dts: qcom: sm8150: add apps_smmu node

On 2020-06-05 20:21, Nicolas Dechesne wrote:
> On Fri, Jun 5, 2020 at 4:39 PM Sai Prakash Ranjan
> <saiprakash.ranjan@...eaurora.org> wrote:
>> 
>> Hi Nico,
>> 
>> On 2020-06-05 20:01, Nicolas Dechesne wrote:
>> > On Fri, Jun 5, 2020 at 4:14 PM Sai Prakash Ranjan
>> > <saiprakash.ranjan@...eaurora.org> wrote:
>> >>
>> >> On 2020-06-05 19:40, Jonathan Marek wrote:
>> >> > On 6/5/20 10:03 AM, Sai Prakash Ranjan wrote:
>> >> >> On 2020-05-29 08:45, Bjorn Andersson wrote:
>> >> >>> On Thu 28 May 20:02 PDT 2020, Jonathan Marek wrote:
>> >> >>>
>> >> >>>>
>> >> >>>>
>> >> >>>> On 5/28/20 10:52 PM, Bjorn Andersson wrote:
>> >> >>>> > On Sat 23 May 19:38 PDT 2020, Jonathan Marek wrote:
>> >> >>>> >
>> >> >>>> > > Add the apps_smmu node for sm8150. Note that adding the iommus field for
>> >> >>>> > > UFS is required because initializing the iommu removes the bypass mapping
>> >> >>>> > > that created by the bootloader.
>> >> >>>> > >
>> >> >>>> >
>> >> >>>> > Unrelated to the patch itself; how do you disable the splash screen on
>> >> >>>> > 8150? "fastboot oem select-display-panel none" doesn't seem to work for
>> >> >>>> > me on the MTP - and hence this would prevent my device from booting.
>> >> >>>> >
>> >> >>>> > Thanks,
>> >> >>>> > Bjorn
>> >> >>>> >
>> >> >>>>
>> >> >>>> I don't have a MTP, but on HDK855, "fastboot oem
>> >> >>>> select-display-panel none"
>> >> >>>> combined with setting the physical switch to HDMI mode (which
>> >> >>>> switches off
>> >> >>>> the 1440x2560 panel) gets it to not setup the display at all (just
>> >> >>>> the
>> >> >>>> fastboot command isn't enough).
>> >> >>>>
>> >> >>>
>> >> >>> Okay, I don't think we have anything equivalent on the MTP, but good
>> >> >>> to
>> >> >>> know.
>> >> >>>
>> >> >>
>> >> >> Actually I tried out this in SM8150 MTP and it works fine for me,
>> >> >>
>> >> >> "fastboot set_active a; fastboot set_active b; fastboot set_active a;
>> >> >> fastboot oem select-display-panel none; fastboot reboot bootloader;
>> >> >> fastboot boot boot-sm8150.img"
>> >> >>
>> >> >> Also I need to switch slots everytime like above, otherwise I always
>> >> >> see some error
>> >> >> while loading the boot image.
>> >> >>
>> >> >
>> >> > What is the error? If it is "FAILED (remote: Failed to
>> >> > load/authenticate boot image: Load Error)" then flashing/erasing
>> >> > boot_a will make it go away ("fastboot erase boot_a") for the next 6
>> >> > or so "failed" boots.
>> >> >
>> >>
>> >> Yes this exact error.
>> >
>> > The bootloader maintains a 'boot status' in one of the partition
>> > attributes. After a certain amount of 'failed' boot , it will switch
>> > to the other boot partition. It's the same thing on RB3/DB845c. In our
>> > release for DB845c, we are patching the bootloader so that this
>> > behavior is bypassed. On typical 'product' there is a user space
>> > application that will come and set the partition attribute to indicate
>> > the boot was successful.
>> >
>> > For the record, this is the patch we use on 845c:
>> > https://git.linaro.org/landing-teams/working/qualcomm/abl.git/commit/?h=release/LE.UM.2.3.7-09200-sda845.0&id=e3dc60213234ed626161a568ba587ddac63c5158
>> >
>> > rebuilding EDK2/ABL requires access to signing tools.. so it might not
>> > be possible for everyone. but in case you can, it should be
>> > straightforward to reuse this patch.
>> >
>> 
>> Thank you for these details and the patch, it's very useful.
>> I do have access to ABL code and the signing tools and can build one.
> 
> Good. Then the next problem you will likely face is that building QCOM
> ABL is far from being straightforward. Why would it be? ;)
> That's the script we use to build it ourselves:
> https://git.linaro.org/ci/job/configs.git/tree/lt-qcom-bootloader/dragonboard845c/builders.sh#n61
> 
> It has a reference to sectools which we have (internally) access to,
> but you have it too, and you should be able to leverage most of the
> script.

Looks like a cool tool, will definitely try it out :) Also internally we 
have another
tool to build ABL(if you are aware of kdev then you will know what this 
is called, guess ;))
which takes care of cloning and building and signing all the things 
required
(although very weirdly it clones sectools everytime which should be 
fixed, I just comment
that part out when I build) and all it takes is one command "make" :)

Thanks,
Sai

-- 
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a 
member
of Code Aurora Forum, hosted by The Linux Foundation