linux-kernel - Re: [PATCH v4 2/2] mfd: cros_ec: Add cros-usbpd-notify subdevice

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]

Date:   Tue, 7 Jan 2020 09:34:19 -0800
From:   Prashant Malani <pmalani@...omium.org>
To:     Enric Balletbo i Serra <enric.balletbo@...labora.com>
Cc:     Guenter Roeck <groeck@...omium.org>,
        Benson Leung <bleung@...omium.org>,
        Lee Jones <lee.jones@...aro.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v4 2/2] mfd: cros_ec: Add cros-usbpd-notify subdevice

Hi Enric,

On Tue, Jan 7, 2020 at 8:51 AM Enric Balletbo i Serra
<enric.balletbo@...labora.com> wrote:
>
> Hi Prashant,
>
> On 3/1/20 0:52, Prashant Malani wrote:
> > Hi Enric,
> >
> >>> Sorry to not catch this before, but a worry arose. Is non-ACPI platforms or
> >>> non-X86 platforms or on OF platforms?
> >>>
> >>> ARM64 for example has the CONFIG_ACPI symbol set to yes, with the below
> >>> condition condition will not work on Kevin for example and IIUC this is not what
> >>> we want, I think we want IS_ENABLED(CONFIG_OF)?
> >> Thanks for noting this. I will check with a kevin, and with the
> >> internal build flags to verify whether there are ARM64 which have the
> >> GOOG0003 PD notification device.
> >> I'll update this thread with my findings.
> > AFAICT from the Chrome OS kernel build step .config output, kevin
> > doesn't have CONFIG_ACPI enabled (it is marked as "# CONFIG_ACPI is
> > not set"), and it doesn't look like there are Chrome OS ARM devices
> > that use ACPI (I believe it's only used on Chrome OS x86-based
> > devices). So perhaps it is not a concern?
> >
>
> The problem is, although it is not used in your configs, it can be selected, and
> fwiw some defconfigs in mainline have it enabled, i.e the arm64 defconfig.
>
> I think you're testing the patch on x86 but I suspect we want also the notifier
> on some arm64 platforms (like kevin) right? In such case I won't get the
> notifier because CONFIG_OF and CONFIG_ACPI is enabled on my defconfig.
I tested the patchset on an arm64 platform and confirmed that
CONFIG_ACPI is enabled, but I suppose that is not a sufficient test
given that, as you mentioned, upstream configs like arm64 defconfig
have it enabled.
>
> My guess is that the logic should be if IS_ENABLED(CONFIG_OF) call
> cros_ec_check_features, otherwise ACPI will do the magic instead of
> (!IS_ENABLED(CONFIG_ACPI))
Noted. Thanks! I have a pending query on Patch 1/2. Once that is
resolved, I will update both patches.
>
> Best,
>  Enric
>
> >>
> >> Best,
> >>
> >>>
> >>> Thanks,
> >>>  Enric
> >>>
> >>>> +      */
> >>>> +     if (!IS_ENABLED(CONFIG_ACPI)) {
> >>>> +             if (cros_ec_check_features(ec, EC_FEATURE_USB_PD)) {
> >>>> +                     retval = mfd_add_hotplug_devices(ec->dev,
> >>>> +                                     cros_usbpd_notify_cells,
> >>>> +                                     ARRAY_SIZE(cros_usbpd_notify_cells));
> >>>> +                     if (retval)
> >>>> +                             dev_err(ec->dev,
> >>>> +                                     "failed to add PD notify devices: %d\n",
> >>>> +                                     retval);
> >>>> +             }
> >>>> +     }
> >>>> +
> >>>>       /*
> >>>>        * The following subdevices cannot be detected by sending the
> >>>>        * EC_FEATURE_GET_CMD to the Embedded Controller device.
> >>>>