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
next prev parent 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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).