From: Dmitry Torokhov <dmitry.torokhov@gmail.com>
To: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Cc: "Wolfram Sang" <wsa@the-dreams.de>,
"Pali Rohár" <pali.rohar@gmail.com>,
"Michał Kępień" <kernel@kempniu.pl>,
"Jean Delvare" <jdelvare@suse.com>,
"Steven Honeyman" <stevenhoneyman@gmail.com>,
Valdis.Kletnieks@vt.edu,
"Jochen Eisinger" <jochen@penguin-breeder.org>,
"Gabriele Mazzotta" <gabriele.mzt@gmail.com>,
"Andy Lutomirski" <luto@kernel.org>,
Mario_Limonciello@dell.com, "Alex Hung" <alex.hung@canonical.com>,
"Takashi Iwai" <tiwai@suse.de>,
linux-i2c@vger.kernel.org, linux-kernel@vger.kernel.org,
platform-driver-x86@vger.kernel.org
Subject: Re: [PATCH] i2c: i801: Register optional lis3lv02d i2c device on Dell machines
Date: Wed, 4 Jan 2017 13:55:29 -0800 [thread overview]
Message-ID: <20170104215529.GC40341@dtor-ws> (raw)
In-Reply-To: <20170104214906.GA15702@mail.corp.redhat.com>
On Wed, Jan 04, 2017 at 10:49:06PM +0100, Benjamin Tissoires wrote:
> On Jan 04 2017 or thereabouts, Dmitry Torokhov wrote:
> > On Wed, Jan 04, 2017 at 06:46:19PM +0100, Wolfram Sang wrote:
> > >
> > > > How about:
> > > > ---
> > > > From daa7571bbf337704332c0cfeec9b8fd5aeae596f Mon Sep 17 00:00:00 2001
> > > > From: Benjamin Tissoires <benjamin.tissoires@redhat.com>
> > > > Date: Wed, 4 Jan 2017 18:26:54 +0100
> > > > Subject: [PATCH] I2C: add the source of the IRQ in struct i2c_client
> > > >
> > > > With commit 4d5538f5882a ("i2c: use an IRQ to report Host Notify events,
> > > > not alert"), the IRQ provided in struct i2c_client might be assigned while
> > > > it has not been explicitly declared by either the platform information
> > > > or OF or ACPI.
> > > > Some drivers (lis3lv02d) rely on the fact that the IRQ gets assigned or
> > > > not to trigger a different behavior (exposing /dev/freefall in this case).
> > > >
> > > > Provide a way for others to know who set the IRQ and so they can behave
> > > > accordingly.
> > > >
> > > > Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
> > > > ---
> > > > drivers/i2c/i2c-core.c | 7 +++++++
> > > > include/linux/i2c.h | 11 +++++++++++
> > > > 2 files changed, 18 insertions(+)
> > > >
> > > > diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
> > > > index cf9e396..226c75d 100644
> > > > --- a/drivers/i2c/i2c-core.c
> > > > +++ b/drivers/i2c/i2c-core.c
> > > > @@ -935,8 +935,12 @@ static int i2c_device_probe(struct device *dev)
> > > > irq = of_irq_get_byname(dev->of_node, "irq");
> > > > if (irq == -EINVAL || irq == -ENODATA)
> > > > irq = of_irq_get(dev->of_node, 0);
> > > > + if (irq > 0)
> > > > + client->irq_source = I2C_IRQ_SOURCE_OF;
> > > > } else if (ACPI_COMPANION(dev)) {
> > > > irq = acpi_dev_gpio_irq_get(ACPI_COMPANION(dev), 0);
> > > > + if (irq > 0)
> > > > + client->irq_source = I2C_IRQ_SOURCE_ACPI;
> > > > }
> > > > if (irq == -EPROBE_DEFER)
> > > > return irq;
> > > > @@ -947,6 +951,8 @@ static int i2c_device_probe(struct device *dev)
> > > > if (irq < 0) {
> > > > dev_dbg(dev, "Using Host Notify IRQ\n");
> > > > irq = i2c_smbus_host_notify_to_irq(client);
> > > > + if (irq > 0)
> > > > + client->irq_source = I2C_IRQ_SOURCE_HOST_NOTIFY;
> > > > }
> > > > if (irq < 0)
> > > > irq = 0;
> > > > @@ -1317,6 +1323,7 @@ i2c_new_device(struct i2c_adapter *adap, struct i2c_board_info const *info)
> > > > client->flags = info->flags;
> > > > client->addr = info->addr;
> > > > client->irq = info->irq;
> > > > + client->irq_source = I2C_IRQ_SOURCE_PLATFORM;
> > > >
> > > > strlcpy(client->name, info->type, sizeof(client->name));
> > > >
> > > > diff --git a/include/linux/i2c.h b/include/linux/i2c.h
> > > > index b2109c5..7d0368d 100644
> > > > --- a/include/linux/i2c.h
> > > > +++ b/include/linux/i2c.h
> > > > @@ -213,6 +213,13 @@ struct i2c_driver {
> > > > };
> > > > #define to_i2c_driver(d) container_of(d, struct i2c_driver, driver)
> > > >
> > > > +enum i2c_irq_source {
> > > > + I2C_IRQ_SOURCE_PLATFORM,
> > > > + I2C_IRQ_SOURCE_OF,
> > > > + I2C_IRQ_SOURCE_ACPI,
> > > > + I2C_IRQ_SOURCE_HOST_NOTIFY,
> > > > +};
> > > > +
> > > > /**
> > > > * struct i2c_client - represent an I2C slave device
> > > > * @flags: I2C_CLIENT_TEN indicates the device uses a ten bit chip address;
> > > > @@ -227,6 +234,9 @@ struct i2c_driver {
> > > > * userspace_devices list
> > > > * @slave_cb: Callback when I2C slave mode of an adapter is used. The adapter
> > > > * calls it to pass on slave events to the slave driver.
> > > > + * @irq_source: Enum which provides the source of the IRQ. Useful to know
> > > > + * if the IRQ was issued from Host Notify or if it was provided by an other
> > > > + * component.
> > >
> > > I'd think some documentation somewhere makes sense why we need to
> > > distinguish this in some cases?
> >
> > I'd rather drivers be oblivious of the source of interrupt. If they need
> > to distinguish between them that means that our IRQ abstration failed.
> >
> > >
> > > > *
> > > > * An i2c_client identifies a single device (i.e. chip) connected to an
> > > > * i2c bus. The behaviour exposed to Linux is defined by the driver
> > > > @@ -245,6 +255,7 @@ struct i2c_client {
> > > > #if IS_ENABLED(CONFIG_I2C_SLAVE)
> > > > i2c_slave_cb_t slave_cb; /* callback for slave mode */
> > > > #endif
> > > > + enum i2c_irq_source irq_source; /* which component assigned the irq */
> > > > };
> > > > #define to_i2c_client(d) container_of(d, struct i2c_client, dev)
> > > >
> > > > Dmitry, Wolfram, Jean, would this be acceptable for you?
> > >
> > > Adding something to i2c_driver is not exactly cheap, but from what I
> > > glimpsed from this thread, this is one of the cleanest solution to this
> > > problem?
> > >
> >
> > As Benjamin said, it is really property of device [instance], not
> > driver. I.e. driver could handle both wired IRQ and HostNotify-based
> > scheme similarly, it is device (and board) that knows how stuff is
> > connected.
> >
> > Maybe we could do something like this (untested):
> >
> >
> > From e362a0277fd1bd6112f258664d8831d9bc6b78da Mon Sep 17 00:00:00 2001
> > From: Dmitry Torokhov <dmitry.torokhov@gmail.com>
> > Date: Wed, 4 Jan 2017 09:33:43 -0800
> > Subject: [PATCH] i2c: do not enable fall back to Host Notify by default
> >
> > Falling back unconditionally to HostNotify as primary client's interrupt
> > breaks some drivers which alter their functionality depending on whether
> > interrupt is present or not, so let's introduce a board flag telling I2C
> > core explicitly if we want wired interrupt or HostNotify-based one:
> > I2C_CLIENT_HOST_NOTIFY.
> >
> > For DT-based systems we introduce "host-notofy" property that we convert
>
> typo: s/host-notofy/host-notify/
>
> > to I2C_CLIENT_HOST_NOTIFY board flag.
> >
> > Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
> > ---
> > Documentation/devicetree/bindings/i2c/i2c.txt | 8 ++++++++
> > drivers/i2c/i2c-core.c | 17 ++++++++---------
> > include/linux/i2c.h | 1 +
> > 3 files changed, 17 insertions(+), 9 deletions(-)
> >
> > diff --git a/Documentation/devicetree/bindings/i2c/i2c.txt b/Documentation/devicetree/bindings/i2c/i2c.txt
> > index 5fa691e6f638..cee9d5055fa2 100644
> > --- a/Documentation/devicetree/bindings/i2c/i2c.txt
> > +++ b/Documentation/devicetree/bindings/i2c/i2c.txt
> > @@ -62,6 +62,9 @@ wants to support one of the below features, it should adapt the bindings below.
> > "irq" and "wakeup" names are recognized by I2C core, other names are
> > left to individual drivers.
> >
> > +- host-notify
> > + device uses SMBus host notify protocol instead of interrupt line.
> > +
> > - multi-master
> > states that there is another master active on this bus. The OS can use
> > this information to adapt power management to keep the arbitration awake
> > @@ -81,6 +84,11 @@ Binding may contain optional "interrupts" property, describing interrupts
> > used by the device. I2C core will assign "irq" interrupt (or the very first
> > interrupt if not using interrupt names) as primary interrupt for the slave.
> >
> > +Alternatively, devices supporting SMbus Host Notify, and connected to
> > +adapters that support this feature, may use "host-notify" property. I2C
> > +core will create a virtual interrupt for Host Notify and assign it as
> > +primary interrupt for the slave.
> > +
> > Also, if device is marked as a wakeup source, I2C core will set up "wakeup"
> > interrupt for the device. If "wakeup" interrupt name is not present in the
> > binding, then primary interrupt will be used as wakeup interrupt.
> > diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
> > index cf9e396d7702..250969fa7670 100644
> > --- a/drivers/i2c/i2c-core.c
> > +++ b/drivers/i2c/i2c-core.c
> > @@ -931,7 +931,10 @@ static int i2c_device_probe(struct device *dev)
> > if (!client->irq) {
> > int irq = -ENOENT;
> >
> > - if (dev->of_node) {
> > + if (client->flags & I2C_CLIENT_HOST_HOTIFY) {
>
> typo: s/I2C_CLIENT_HOST_HOTIFY/I2C_CLIENT_HOST_NOTIFY/
>
> With these fixed, the code is:
> Tested-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
>
> I tested both with and without the I2C_CLIENT_HOST_HOTIFY flag on the
> Thinkpad T450s, and everything is in order.
>
> Thanks Dmitry for the patch!
Thanks Benjamin. Let me submit the patch "officially" and CC Rob & DT
folks on binding change.
--
Dmitry
next prev parent reply other threads:[~2017-01-04 22:03 UTC|newest]
Thread overview: 62+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-12-27 12:52 [PATCH] i2c: i801: Register optional lis3lv02d i2c device on Dell machines Pali Rohár
2016-12-27 13:43 ` Wolfram Sang
2016-12-27 13:51 ` Pali Rohár
2016-12-27 22:15 ` Andy Shevchenko
2016-12-27 22:41 ` Valdis.Kletnieks
2016-12-28 7:55 ` Andy Shevchenko
2016-12-28 9:05 ` Pali Rohár
2016-12-28 14:03 ` Wolfram Sang
2016-12-29 4:37 ` Valdis.Kletnieks
2016-12-28 8:38 ` Pali Rohár
2016-12-28 14:02 ` Wolfram Sang
2017-01-04 9:45 ` Jean Delvare
2016-12-29 8:29 ` Michał Kępień
2016-12-29 9:00 ` Pali Rohár
2016-12-29 13:47 ` Michał Kępień
2016-12-29 14:17 ` Pali Rohár
2016-12-29 21:09 ` Michał Kępień
2016-12-29 21:28 ` Pali Rohár
2017-01-03 9:06 ` Benjamin Tissoires
2017-01-03 9:23 ` Pali Rohár
2017-01-03 18:38 ` Dmitry Torokhov
2017-01-03 18:50 ` Pali Rohár
2017-01-03 18:58 ` Mario.Limonciello
2017-01-03 19:48 ` Dmitry Torokhov
2017-01-03 20:05 ` Pali Rohár
2017-01-03 20:24 ` Dmitry Torokhov
2017-01-03 20:39 ` Pali Rohár
2017-01-03 20:59 ` Dmitry Torokhov
2017-01-04 8:18 ` Pali Rohár
2017-01-04 9:05 ` Benjamin Tissoires
2017-01-04 9:18 ` Pali Rohár
2017-01-04 10:13 ` Benjamin Tissoires
2017-01-04 10:21 ` Pali Rohár
2017-01-04 10:32 ` Benjamin Tissoires
2017-01-04 11:22 ` Pali Rohár
2017-01-04 12:00 ` Pali Rohár
2017-01-04 13:02 ` Benjamin Tissoires
2017-01-04 16:06 ` Pali Rohár
2017-01-04 17:39 ` Benjamin Tissoires
2017-01-04 17:46 ` Wolfram Sang
2017-01-04 17:54 ` Dmitry Torokhov
2017-01-04 21:49 ` Benjamin Tissoires
2017-01-04 21:55 ` Dmitry Torokhov [this message]
2017-01-04 21:55 ` Wolfram Sang
2017-01-04 22:00 ` Dmitry Torokhov
2017-01-04 22:03 ` Dmitry Torokhov
2017-01-05 2:20 ` [PATCH] i2c: do not enable fall back to Host Notify by default kbuild test robot
2017-01-05 2:21 ` kbuild test robot
2017-01-05 8:54 ` [PATCH] i2c: i801: Register optional lis3lv02d i2c device on Dell machines Pali Rohár
2017-01-05 9:26 ` Benjamin Tissoires
2017-01-04 18:50 ` Jean Delvare
2017-01-04 9:53 ` Andy Shevchenko
2017-01-04 10:25 ` Benjamin Tissoires
2017-01-04 11:35 ` Pali Rohár
2017-01-04 12:33 ` Benjamin Tissoires
2017-01-03 21:29 ` Benjamin Tissoires
2017-01-04 6:36 ` Dmitry Torokhov
2017-01-04 9:13 ` Benjamin Tissoires
2017-01-04 9:25 ` Pali Rohár
2017-01-03 20:20 ` Andy Shevchenko
2017-01-04 10:18 ` Jean Delvare
2017-01-07 12:49 ` Wolfram Sang
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=20170104215529.GC40341@dtor-ws \
--to=dmitry.torokhov@gmail.com \
--cc=Mario_Limonciello@dell.com \
--cc=Valdis.Kletnieks@vt.edu \
--cc=alex.hung@canonical.com \
--cc=benjamin.tissoires@redhat.com \
--cc=gabriele.mzt@gmail.com \
--cc=jdelvare@suse.com \
--cc=jochen@penguin-breeder.org \
--cc=kernel@kempniu.pl \
--cc=linux-i2c@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@kernel.org \
--cc=pali.rohar@gmail.com \
--cc=platform-driver-x86@vger.kernel.org \
--cc=stevenhoneyman@gmail.com \
--cc=tiwai@suse.de \
--cc=wsa@the-dreams.de \
/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).