From: Wolfram Sang <wsa+renesas@sang-engineering.com>
To: linux-i2c@vger.kernel.org
Cc: linux-renesas-soc@vger.kernel.org,
Wolfram Sang <wsa+renesas@sang-engineering.com>
Subject: [RFC PATCH] i2c: gpio: fault-injector: add 'lose_arbitration' injector
Date: Mon, 21 Jan 2019 15:28:39 +0100 [thread overview]
Message-ID: <20190121142839.19011-1-wsa+renesas@sang-engineering.com> (raw)
Here is a fault injector simulating 'arbitration lost' from multi-master
setups. Read the docs for its usage.
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
---
This is the most reliable result I came up with so far for simulating lost
arbitration (after playing a lot with falling SDA as trigger first, but SCL
seems the way to go). Works fine with the i2c-sh_mobile driver on a Renesas
Lager board. I am not super-happy with the interrupt latency causing some bits
to be non-disturbed, but for now, I don't see a way around it except for
busy-polling which I think is too excessive. RFC for now because someone still
might have a better idea :)
Needs my previous fault-injector cleanup patches, a branch for consuming is here:
git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux.git renesas/iic-arbitration-lost
Now let's see how to fix the sh_mobile driver...
Documentation/i2c/gpio-fault-injection | 26 +++++++++++++++
drivers/i2c/busses/i2c-gpio.c | 61 ++++++++++++++++++++++++++++++++++
2 files changed, 87 insertions(+)
diff --git a/Documentation/i2c/gpio-fault-injection b/Documentation/i2c/gpio-fault-injection
index 1a44e3edc0c4..b6f36ffe55e1 100644
--- a/Documentation/i2c/gpio-fault-injection
+++ b/Documentation/i2c/gpio-fault-injection
@@ -83,3 +83,29 @@ This is why bus recovery (up to 9 clock pulses) must either check SDA or send
additional STOP conditions to ensure the bus has been released. Otherwise
random data will be written to a device!
+Lost arbitration
+================
+
+Here, we want to simulate the condition where the master under tests loses the
+bus arbitration against another master in a multi-master setup.
+
+"lose_arbitration"
+------------------
+
+This file is write only and you need to write the number of desired lost
+arbitrations in a row. The calling process will then sleep and interfere with
+transfers from the master under test when they appear until that number is
+reached. The process is interruptible, though.
+
+Arbitration lost is achieved by waiting for SCL going down by the master under
+test and then pulling SDA low for some time. So, the I2C address sent out
+should be corrupted and that should be detected properly. That means that the
+address sent out should have a lot of '1' bits to be able to detect corruption.
+There doesn't need to be a device at this address because arbitration lost
+should be detected beforehand. Also note, that SCL going down is monitored
+using interrupts, so the interrupt latency might cause the first bits to be not
+corrupted. A good starting script for using this fault injector:
+
+# echo 1 > lose_arbitration &
+# i2cget -y <bus_to_test> 0x3f
+
diff --git a/drivers/i2c/busses/i2c-gpio.c b/drivers/i2c/busses/i2c-gpio.c
index ca04fa25a141..c630172b4787 100644
--- a/drivers/i2c/busses/i2c-gpio.c
+++ b/drivers/i2c/busses/i2c-gpio.c
@@ -7,12 +7,14 @@
* it under the terms of the GNU General Public License version 2 as
* published by the Free Software Foundation.
*/
+#include <linux/completion.h>
#include <linux/debugfs.h>
#include <linux/delay.h>
#include <linux/gpio/consumer.h>
#include <linux/i2c-algo-bit.h>
#include <linux/i2c.h>
#include <linux/init.h>
+#include <linux/interrupt.h>
#include <linux/module.h>
#include <linux/of.h>
#include <linux/platform_data/i2c-gpio.h>
@@ -27,6 +29,7 @@ struct i2c_gpio_private_data {
struct i2c_gpio_platform_data pdata;
#ifdef CONFIG_I2C_GPIO_FAULT_INJECTOR
struct dentry *debug_dir;
+ struct completion irq_happened;
#endif
};
@@ -162,6 +165,59 @@ static int fops_incomplete_write_byte_set(void *data, u64 addr)
}
DEFINE_DEBUGFS_ATTRIBUTE(fops_incomplete_write_byte, NULL, fops_incomplete_write_byte_set, "%llu\n");
+static irqreturn_t lose_arbitration_irq(int irq, void *dev_id)
+{
+ struct i2c_gpio_private_data *priv = dev_id;
+
+ setsda(&priv->bit_data, 0);
+ udelay(200);
+ setsda(&priv->bit_data, 1);
+
+ complete(&priv->irq_happened);
+ return IRQ_HANDLED;
+}
+
+static int fops_lose_arbitration_set(void *data, u64 num_faults)
+{
+ struct i2c_gpio_private_data *priv = data;
+ int irq = gpiod_to_irq(priv->scl);
+ int ret, i;
+
+ i2c_lock_bus(&priv->adap, I2C_LOCK_ROOT_ADAPTER);
+
+ /*
+ * Interrupt on falling SCL. This ensures that the master under test has
+ * really started the transfer. Interrupt on falling SDA did only
+ * exercise 'bus busy' detection on some HW but not 'arbitration lost'.
+ * Note that the interrupt latency may cause the first bits to be
+ * transmitted correctly.
+ */
+ ret = gpiod_direction_input(priv->scl);
+ if (ret)
+ goto unlock;
+
+ ret = request_irq(irq, lose_arbitration_irq, IRQF_TRIGGER_FALLING,
+ "i2c-gpio-fi", priv);
+ if (ret)
+ goto output;
+
+ for (i = 0; i < num_faults; i++) {
+ ret = wait_for_completion_interruptible(&priv->irq_happened);
+ if (ret)
+ break;
+ reinit_completion(&priv->irq_happened);
+ }
+
+ free_irq(irq, priv);
+ output:
+ ret = gpiod_direction_output(priv->scl, 1);
+ unlock:
+ i2c_unlock_bus(&priv->adap, I2C_LOCK_ROOT_ADAPTER);
+
+ return ret;
+}
+DEFINE_DEBUGFS_ATTRIBUTE(fops_lose_arbitration, NULL, fops_lose_arbitration_set, "%llu\n");
+
static void i2c_gpio_fault_injector_init(struct platform_device *pdev)
{
struct i2c_gpio_private_data *priv = platform_get_drvdata(pdev);
@@ -181,10 +237,15 @@ static void i2c_gpio_fault_injector_init(struct platform_device *pdev)
if (!priv->debug_dir)
return;
+ init_completion(&priv->irq_happened);
+
debugfs_create_file_unsafe("incomplete_address_phase", 0200, priv->debug_dir,
priv, &fops_incomplete_addr_phase);
debugfs_create_file_unsafe("incomplete_write_byte", 0200, priv->debug_dir,
priv, &fops_incomplete_write_byte);
+ if (priv->bit_data.getscl)
+ debugfs_create_file_unsafe("lose_arbitration", 0200, priv->debug_dir,
+ priv, &fops_lose_arbitration);
debugfs_create_file_unsafe("scl", 0600, priv->debug_dir, priv, &fops_scl);
debugfs_create_file_unsafe("sda", 0600, priv->debug_dir, priv, &fops_sda);
}
--
2.11.0
next reply other threads:[~2019-01-21 14:28 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-21 14:28 Wolfram Sang [this message]
2019-02-05 13:31 ` [RFC PATCH] i2c: gpio: fault-injector: add 'lose_arbitration' injector Wolfram Sang
2019-02-11 10:30 ` Geert Uytterhoeven
2019-02-11 20:59 ` Wolfram Sang
2019-02-12 8:31 ` Geert Uytterhoeven
2019-02-16 13:04 ` 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=20190121142839.19011-1-wsa+renesas@sang-engineering.com \
--to=wsa+renesas@sang-engineering.com \
--cc=linux-i2c@vger.kernel.org \
--cc=linux-renesas-soc@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).