Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: [PATCH v5 3/3] gpiolib: Show correct direction from the beginning
To:     Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com>
CC:     Linus Walleij <linus.walleij@linaro.org>,
        Timur Tabi <timur@kernel.org>,
        Stephen Boyd <swboyd@chromium.org>,
        linux-gpio <linux-gpio@vger.kernel.org>,
        LKML <linux-kernel@vger.kernel.org>,
        Jeffrey Hugo <jhugo@codeaurora.org>,
        "linux-omap@vger.kernel.org" <linux-omap@vger.kernel.org>,
        Tony Lindgren <tony@atomide.com>,
        "Strashko, Grygorii" <grygorii.strashko@ti.com>
References: <20181005065300.22882-1-ricardo.ribalda@gmail.com>
 <20181005065300.22882-3-ricardo.ribalda@gmail.com>
 <9746eba6-63f5-feaf-e63f-7e0ea7638942@ti.com>
 <CAPybu_1PuRqOx-Fsmba4XCcRsJA0A9tdob1R=8N72-699cxomw@mail.gmail.com>
From:   Vignesh R <vigneshr@ti.com>
Message-ID: <8d180cc7-d7cb-66eb-001d-f2a5c371d179@ti.com>
Date:   Fri, 12 Oct 2018 11:29:58 +0530
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101
 Thunderbird/52.9.1
MIME-Version: 1.0
In-Reply-To: <CAPybu_1PuRqOx-Fsmba4XCcRsJA0A9tdob1R=8N72-699cxomw@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Hi,

On Thursday 11 October 2018 07:12 PM, Ricardo Ribalda Delgado wrote:
> Hi Vignesh
> 
> Ups, it does not look too good :S . Can you check if this change fixes it:
> 

Below diff works for me. I no longer see crash and gpio-hog seems to be
working. Thanks!

Regards
Vignesh

> diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c
> index e016b22658ff..bcd0ef49ce97 100644
> --- a/drivers/gpio/gpiolib.c
> +++ b/drivers/gpio/gpiolib.c
> @@ -1349,6 +1349,9 @@ int gpiochip_add_data_with_key(struct gpio_chip
> *chip, void *data,
> 
>         spin_unlock_irqrestore(&gpio_lock, flags);
> 
> +       for (i = 0; i < chip->ngpio; i++)
> +               gdev->descs[i].gdev = gdev;
> +
>  #ifdef CONFIG_PINCTRL
>         INIT_LIST_HEAD(&gdev->pin_ranges);
>  #endif
> @@ -1380,8 +1383,6 @@ int gpiochip_add_data_with_key(struct gpio_chip
> *chip, void *data,
>         for (i = 0; i < chip->ngpio; i++) {
>                 struct gpio_desc *desc = &gdev->descs[i];
> 
> -               desc->gdev = gdev;
> -
>                 if (chip->get_direction && gpiochip_line_is_valid(chip, i))
>                         desc->flags = !chip->get_direction(chip, i) ?
>                                         (1 << FLAG_IS_OUT) : 0;
> 
> 
> Thanks!
> On Thu, Oct 11, 2018 at 2:18 PM Vignesh R <vigneshr@ti.com> wrote:
>>
>> Hi,
>>
>> On Friday 05 October 2018 12:23 PM, Ricardo Ribalda Delgado wrote:
>>> Current code assumes that the direction is input if direction_input
>>> function is set.
>>> This might not be the case on GPIOs with programmable direction.
>>>
>>> Signed-off-by: Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com>
>>> Tested-by: Jeffrey Hugo <jhugo@codeaurora.org>
>>
>> This patch causes oops on TI's AM335x-ICEv2 board on next-20181011:
>>
>> [    0.563797] OMAP GPIO hardware version 0.1
>> [    0.577589] Unable to handle kernel NULL pointer dereference at virtual address 000002b8
>> [    0.586127] pgd = (ptrval)
>> [    0.588934] [000002b8] *pgd=00000000
>> [    0.592732] Internal error: Oops: 5 [#1] SMP ARM
>> [    0.597499] Modules linked in:
>> [    0.600668] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.19.0-rc7-next-20181011 #70
>> [    0.608466] Hardware name: Generic AM33XX (Flattened Device Tree)
>> [    0.614770] PC is at gpiod_hog+0x30/0x154
>> [    0.618913] LR is at of_gpiochip_add+0x2fc/0x4e4
>> [    0.623671] pc : [<c055213c>]    lr : [<c0553ff0>]    psr: 60000013
>> [    0.630130] sp : ce09bba0  ip : cdf08095  fp : 00000000
>> [    0.635516] r10: c0bfcf24  r9 : 00000000  r8 : 00000007
>> [    0.640902] r7 : cdf08088  r6 : 00000000  r5 : 00000000  r4 : ce191e40
>> [    0.647630] r3 : 00000000  r2 : 00000000  r1 : cdf08088  r0 : ce191e40
>> [    0.654361] Flags: nZCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment none
>> [    0.661718] Control: 10c5387d  Table: 80004019  DAC: 00000051
>> [    0.667642] Process swapper/0 (pid: 1, stack limit = 0x(ptrval))
>> [...]
>> [    0.973593] [<c055213c>] (gpiod_hog) from [<c0553ff0>] (of_gpiochip_add+0x2fc/0x4e4)
>> [    0.981588] [<c0553ff0>] (of_gpiochip_add) from [<c05528a0>] (gpiochip_add_data_with_key+
>> 0x5a0/0x990)
>> [    0.991102] [<c05528a0>] (gpiochip_add_data_with_key) from [<c0556d8c>] (omap_gpio_probe+
>> 0x37c/0x75c)
>> [    1.000613] [<c0556d8c>] (omap_gpio_probe) from [<c06049c0>] (platform_drv_probe+0x48/0x9
>> 8)
>> [    1.009237] [<c06049c0>] (platform_drv_probe) from [<c0602a04>] (really_probe+0x220/0x2d4
>> )
>> [    1.017764] [<c0602a04>] (really_probe) from [<c0602c18>] (driver_probe_device+0x5c/0x164
>> )
>> [    1.026293] [<c0602c18>] (driver_probe_device) from [<c0600d44>] (bus_for_each_drv+0x54/0
>> xb8)
>> [    1.035090] [<c0600d44>] (bus_for_each_drv) from [<c060276c>] (__device_attach+0xcc/0x13c
>> )
>> [    1.043615] [<c060276c>] (__device_attach) from [<c0601b88>] (bus_probe_device+0x88/0x90)
>> [    1.052051] [<c0601b88>] (bus_probe_device) from [<c05fea18>] (device_add+0x3d8/0x608)
>> [    1.060223] [<c05fea18>] (device_add) from [<c07343b8>] (of_platform_device_create_pdata+
>> 0x8c/0xc0)
>> [    1.069552] [<c07343b8>] (of_platform_device_create_pdata) from [<c07345c4>] (of_platform
>> _bus_create+0x190/0x228)
>> [    1.080134] [<c07345c4>] (of_platform_bus_create) from [<c0734610>] (of_platform_bus_crea
>> te+0x1dc/0x228)
>> [    1.089909] [<c0734610>] (of_platform_bus_create) from [<c073478c>] (of_platform_populate
>> +0x5c/0xac)
>> [    1.099333] [<c073478c>] (of_platform_populate) from [<c0d12614>] (pdata_quirks_init+0x6c
>> /0x90)
>> [    1.108306] [<c0d12614>] (pdata_quirks_init) from [<c0d12144>] (omap_generic_init+0xc/0x1
>> 8)
>> [    1.116933] [<c0d12144>] (omap_generic_init) from [<c0d03eb0>] (customize_machine+0x1c/0x
>> 30)
>> [    1.125645] [<c0d03eb0>] (customize_machine) from [<c01030e4>] (do_one_initcall+0x80/0x31
>> 0)
>> [    1.134265] [<c01030e4>] (do_one_initcall) from [<c0d01244>] (kernel_init_freeable+0x3c4/
>> 0x4ac)
>> [    1.143238] [<c0d01244>] (kernel_init_freeable) from [<c0900f28>] (kernel_init+0x8/0x114)
>> [    1.151673] [<c0900f28>] (kernel_init) from [<c01010b4>] (ret_from_fork+0x14/0x20)
>> [    1.159475] Exception stack(0xce09bfb0 to 0xce09bff8)
>> [    1.164686] bfa0:                                     00000000 00000000 00000000 00000000
>> [    1.173119] bfc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
>> [    1.181551] bfe0: 00000000 00000000 00000000 00000000 00000013 00000000
>> [    1.188378] Code: 0a000002 e3530000 01a09003 159392b4 (e59352b8)
>> [    1.194766] ---[ end trace d5c17cd400f50a22 ]---
>> [    1.199606] Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b
>> [    1.199606]
>> [    1.209052] ---[ end Kernel panic - not syncing: Attempted to kill init! exitcode=0x00000
>> 00b
>> [    1.209052]  ]---
>> [    3.172292] random: fast init done
>>
>> Full log: https://pastebin.ubuntu.com/p/jG8nN6CTBP/
>>
>> Reverting this patch from linux-next allows to boot to prompt.
>>
>> Regards
>> Vignesh
>>
>>> ---
>>>  drivers/gpio/gpiolib.c | 27 +++++++++++++--------------
>>>  1 file changed, 13 insertions(+), 14 deletions(-)
>>>
>>> diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c
>>> index 907019b67a58..e016b22658ff 100644
>>> --- a/drivers/gpio/gpiolib.c
>>> +++ b/drivers/gpio/gpiolib.c
>>> @@ -1349,20 +1349,6 @@ int gpiochip_add_data_with_key(struct gpio_chip *chip, void *data,
>>>
>>>       spin_unlock_irqrestore(&gpio_lock, flags);
>>>
>>> -     for (i = 0; i < chip->ngpio; i++) {
>>> -             struct gpio_desc *desc = &gdev->descs[i];
>>> -
>>> -             desc->gdev = gdev;
>>> -
>>> -             /* REVISIT: most hardware initializes GPIOs as inputs (often
>>> -              * with pullups enabled) so power usage is minimized. Linux
>>> -              * code should set the gpio direction first thing; but until
>>> -              * it does, and in case chip->get_direction is not set, we may
>>> -              * expose the wrong direction in sysfs.
>>> -              */
>>> -             desc->flags = !chip->direction_input ? (1 << FLAG_IS_OUT) : 0;
>>> -     }
>>> -
>>>  #ifdef CONFIG_PINCTRL
>>>       INIT_LIST_HEAD(&gdev->pin_ranges);
>>>  #endif
>>> @@ -1391,6 +1377,19 @@ int gpiochip_add_data_with_key(struct gpio_chip *chip, void *data,
>>>       if (status)
>>>               goto err_remove_chip;
>>>
>>> +     for (i = 0; i < chip->ngpio; i++) {
>>> +             struct gpio_desc *desc = &gdev->descs[i];
>>> +
>>> +             desc->gdev = gdev;
>>> +
>>> +             if (chip->get_direction && gpiochip_line_is_valid(chip, i))
>>> +                     desc->flags = !chip->get_direction(chip, i) ?
>>> +                                     (1 << FLAG_IS_OUT) : 0;
>>> +             else
>>> +                     desc->flags = !chip->direction_input ?
>>> +                                     (1 << FLAG_IS_OUT) : 0;
>>> +     }
>>> +
>>>       acpi_gpiochip_add(chip);
>>>
>>>       machine_gpiochip_add(chip);
>>>
>>
> 
>