All of lore.kernel.org
 help / color / mirror / Atom feed
From: Kent Gibson <warthog618@gmail.com>
To: Bartosz Golaszewski <brgl@bgdev.pl>
Cc: Bartosz Golaszewski <bgolaszewski@baylibre.com>,
	"open list:GPIO SUBSYSTEM" <linux-gpio@vger.kernel.org>
Subject: Re: strace decoding for GPIO uAPI
Date: Tue, 29 Sep 2020 23:04:30 +0800	[thread overview]
Message-ID: <20200929150430.GA551060@sol> (raw)
In-Reply-To: <CAMRc=MeUOOq9Zw+fvab1+rwKf066j9GqEMJ4mCHqVjMfwdo-Gw@mail.gmail.com>

On Tue, Sep 29, 2020 at 03:20:22PM +0200, Bartosz Golaszewski wrote:
> On Mon, Sep 28, 2020 at 3:42 PM Kent Gibson <warthog618@gmail.com> wrote:
> >
> > On Tue, Sep 22, 2020 at 11:50:51AM +0200, Bartosz Golaszewski wrote:
> > > On Tue, Sep 22, 2020 at 9:41 AM Arnd Bergmann <arnd@arndb.de> wrote:
> > > >
> > > > On Tue, Sep 22, 2020 at 4:34 AM Kent Gibson <warthog618@gmail.com> wrote:
> > > > > +/**
> > > >
> > > > Having different-sized members in the union makes it hard for
> > > > something like strace to print the contents. How about just making
> > > > them all __aligned_u64 even when 32 bits are sufficient?
> > > >
> > >
> > > Ah yes, adding support for GPIO ioctl()'s to strace has been on my
> > > TODO list for 3 years now. :(
> > >
> >
> > I'm looking at doing this as it would be useful to have.
> >
> > Just starting with the v1 uAPI, as v2 isn't merged yet.
> >
> > I think I've got it basically done, but I thought I'd run it past you
> > before sending a patch to the strace list.
> >
> > Successful calls currently look like this:
> >
> > ioctl(3, GPIO_GET_CHIPINFO_IOCTL, {name="gpiochip1", label="gpio-mockup-B", lines=8}) = 0
> >
> > ioctl(3, GPIO_GET_LINEHANDLE_IOCTL, {lines=8, lineoffsets=[3, 2, 1, 0, 4, 5, 6, 7], flags=GPIOHANDLE_REQUEST_INPUT, default_values=[0, 0, 0, 0, 0, 0, 0, 0], consumer_label="", fd=7}) = 0
> >
> > ioctl(0, GPIO_GET_LINEEVENT_IOCTL, {lineoffset=2, handleflags=GPIOHANDLE_REQUEST_INPUT, eventflags=GPIOEVENT_REQUEST_BOTH_EDGES, consumer_label="high to low", fd=3}) = 0
> >
> > ioctl(3, GPIO_GET_LINEINFO_IOCTL, {line_offset=1, flags=GPIOLINE_FLAG_KERNEL|GPIOLINE_FLAG_IS_OUT, name="gpio-mockup-A-1", consumer="output atv-lo to as-is atv-hi"}) = 0
> >
> > ioctl(7, GPIOHANDLE_GET_LINE_VALUES_IOCTL, {values=[1, 1, 0, 1, 1, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...]}) = 0
> >
> > ioctl(7, GPIOHANDLE_SET_LINE_VALUES_IOCTL, {values=[1, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...]}) = 0
> >
> > ioctl(7, GPIOHANDLE_SET_CONFIG_IOCTL, {flags=GPIOHANDLE_REQUEST_ACTIVE_LOW, default_values=[0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...]}) = 0
> >
> > ioctl(3, GPIO_GET_LINEINFO_WATCH_IOCTL, {line_offset=3, flags=0, name="gpio-mockup-A-3", consumer=""}) = 0
> >
> > ioctl(3, GPIO_GET_LINEINFO_UNWATCH_IOCTL, {offset=3}) = 0
> >
> > For unsuccessful calls it only prints the input fields, and drops the
> > output fields. e.g.
> >
> > ioctl(3, GPIO_GET_LINEINFO_WATCH_IOCTL, {line_offset=3}) = -1 EPERM (Operation not permitted)
> >
> > Does that work for you?
> >
> 
> Kent,
> 
> This is perfect! Thanks for doing this! For the error cases: I'm not
> sure how strace handles this for other ioctls() - maybe we could print
> something like 'flags=N/A'? Unless what you did is the standard way,
> then leave it.
> 

I think that is pretty standard, but I will double check.

Just debugging the uAPI v2 decoding now, though I wont submit that
until the uAPI is in an -rc.  It may be useful for debugging in the
meantime, so I'll push a copy to my github fork[1] shortly - the v1
decoding is there already.

I've noticed a few things I want to tidy up in the v1 decoding
before I submit that though.  Should have it ready to submit in the
next few days.

Cheers,
Kent.

[1] https://github.com/warthog618/strace/tree/gpio

  reply	other threads:[~2020-09-29 15:04 UTC|newest]

Thread overview: 77+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-09-22  2:31 [PATCH v9 00/20] gpio: cdev: add uAPI v2 Kent Gibson
2020-09-22  2:31 ` [PATCH v9 01/20] gpiolib: cdev: gpio_desc_to_lineinfo() should set info offset Kent Gibson
2020-09-22  2:31 ` [PATCH v9 02/20] gpiolib: cdev: replace strncpy() with strscpy() Kent Gibson
2020-09-22  2:31 ` [PATCH v9 03/20] gpio: uapi: define GPIO_MAX_NAME_SIZE for array sizes Kent Gibson
2020-09-22  2:31 ` [PATCH v9 04/20] gpio: uapi: define uAPI v2 Kent Gibson
2020-09-22  7:41   ` Arnd Bergmann
2020-09-22  9:50     ` Bartosz Golaszewski
2020-09-22 10:07       ` Kent Gibson
2020-09-28 13:42       ` strace decoding for GPIO uAPI Kent Gibson
2020-09-29 13:20         ` Bartosz Golaszewski
2020-09-29 15:04           ` Kent Gibson [this message]
2020-09-29 16:19           ` Andy Shevchenko
2020-09-23 10:04   ` [PATCH v9 04/20] gpio: uapi: define uAPI v2 Andy Shevchenko
2020-09-23 10:30     ` Kent Gibson
2020-09-23 11:16       ` Andy Shevchenko
2020-09-23 12:18         ` Arnd Bergmann
2020-09-23 15:15           ` Andy Shevchenko
2020-09-22  2:31 ` [PATCH v9 05/20] gpiolib: make cdev a build option Kent Gibson
2020-09-22  2:31 ` [PATCH v9 06/20] gpiolib: add build option for CDEV v1 ABI Kent Gibson
2020-09-22  2:31 ` [PATCH v9 07/20] gpiolib: cdev: support GPIO_V2_GET_LINE_IOCTL and GPIO_V2_LINE_GET_VALUES_IOCTL Kent Gibson
2020-09-23 11:11   ` Andy Shevchenko
2020-09-24  8:09     ` Kent Gibson
2020-09-25 10:06       ` Andy Shevchenko
2020-09-26  9:16         ` Kent Gibson
2020-09-27  9:00           ` Andy Shevchenko
2020-09-27 12:39             ` Kent Gibson
2020-09-24 14:29     ` Bartosz Golaszewski
2020-09-22  2:31 ` [PATCH v9 08/20] gpiolib: cdev: support GPIO_V2_GET_LINEINFO_IOCTL and GPIO_V2_GET_LINEINFO_WATCH_IOCTL Kent Gibson
2020-09-23 15:41   ` Andy Shevchenko
2020-09-24  2:39     ` Kent Gibson
2020-09-24  8:39       ` Andy Shevchenko
2020-09-24  9:48         ` Kent Gibson
2020-09-25 10:12           ` Andy Shevchenko
2020-09-25 11:56             ` Kent Gibson
2020-09-25 14:43               ` Andy Shevchenko
2020-09-25  5:32     ` Kent Gibson
2020-09-22  2:31 ` [PATCH v9 09/20] gpiolib: cdev: support edge detection for uAPI v2 Kent Gibson
2020-09-23 15:47   ` Andy Shevchenko
2020-09-24  3:07     ` Kent Gibson
2020-09-25  9:35       ` Andy Shevchenko
2020-09-25 12:26         ` Kent Gibson
2020-09-25 14:48           ` Andy Shevchenko
2020-09-22  2:31 ` [PATCH v9 10/20] gpiolib: cdev: support GPIO_V2_LINE_SET_CONFIG_IOCTL Kent Gibson
2020-09-23 16:14   ` Andy Shevchenko
2020-09-23 16:15     ` Andy Shevchenko
2020-09-24  3:24       ` Kent Gibson
2020-09-24  8:26         ` Andy Shevchenko
2020-09-24  9:26           ` Kent Gibson
2020-09-25  9:49             ` Andy Shevchenko
2020-09-22  2:31 ` [PATCH v9 11/20] gpiolib: cdev: support GPIO_V2_LINE_SET_VALUES_IOCTL Kent Gibson
2020-09-23 16:18   ` Andy Shevchenko
2020-09-24  7:32     ` Kent Gibson
2020-09-24  8:21       ` Andy Shevchenko
2020-09-24  9:08         ` Kent Gibson
2020-09-24 12:46       ` Kent Gibson
2020-09-25  9:57         ` Andy Shevchenko
2020-09-25 12:16           ` Kent Gibson
2020-09-22  2:31 ` [PATCH v9 12/20] gpiolib: cdev: support setting debounce Kent Gibson
2020-09-23 16:27   ` Andy Shevchenko
2020-09-24  7:48     ` Kent Gibson
2020-09-25  9:43       ` Andy Shevchenko
2020-09-22  2:31 ` [PATCH v9 13/20] gpio: uapi: document uAPI v1 as deprecated Kent Gibson
2020-09-22  7:49   ` Arnd Bergmann
2020-09-22  8:11     ` Andy Shevchenko
2020-09-22  9:05     ` Kent Gibson
2020-09-22  2:31 ` [PATCH v9 14/20] tools: gpio: port lsgpio to v2 uAPI Kent Gibson
2020-09-22  2:31 ` [PATCH v9 15/20] tools: gpio: port gpio-watch " Kent Gibson
2020-09-22  2:31 ` [PATCH v9 16/20] tools: gpio: rename nlines to num_lines Kent Gibson
2020-09-22  2:31 ` [PATCH v9 17/20] tools: gpio: port gpio-hammer to v2 uAPI Kent Gibson
2020-09-23 16:30   ` Andy Shevchenko
2020-09-24  7:50     ` Kent Gibson
2020-09-22  2:31 ` [PATCH v9 18/20] tools: gpio: port gpio-event-mon " Kent Gibson
2020-09-22  2:31 ` [PATCH v9 19/20] tools: gpio: add multi-line monitoring to gpio-event-mon Kent Gibson
2020-09-22  2:31 ` [PATCH v9 20/20] tools: gpio: add debounce support " Kent Gibson
2020-09-23 16:35 ` [PATCH v9 00/20] gpio: cdev: add uAPI v2 Andy Shevchenko
2020-09-24  8:00   ` Kent Gibson
2020-09-25  9:44     ` Andy Shevchenko

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200929150430.GA551060@sol \
    --to=warthog618@gmail.com \
    --cc=bgolaszewski@baylibre.com \
    --cc=brgl@bgdev.pl \
    --cc=linux-gpio@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.