From: Heiko Schocher <hs@denx.de>
To: u-boot@lists.denx.de
Subject: [U-Boot] [PATCH 08/55] i2c: Add a mux for GPIO-based I2C bus arbitration
Date: Sun, 05 Jul 2015 08:43:48 +0200 [thread overview]
Message-ID: <5598D224.8070706@denx.de> (raw)
In-Reply-To: <1435882592-487-9-git-send-email-sjg@chromium.org>
Hello Simon,
Am 03.07.2015 um 02:15 schrieb Simon Glass:
> While I2C supports multi-master buses this is difficult to get right. This
What do you mean here? Where are the problems? You have an i2c mux, or?
> driver provides a scheme based on two 'claim' GPIOs, one driven by the AP
> and one driver by the EC. With these they can communicate and reliably
What is AP and EC ?
> share the bus. This scheme has minimal overhead and involves very little
> code. It is used on snow to permit the EC and the AP to share access to
> the main system PMIC and battery. The scheme can survive reboots by either
> side without difficulty.
>
> Signed-off-by: Simon Glass <sjg@chromium.org>
> ---
>
> drivers/i2c/muxes/Kconfig | 9 ++
> drivers/i2c/muxes/Makefile | 1 +
> drivers/i2c/muxes/i2c-arb-gpio-challenge.c | 147 +++++++++++++++++++++++++++++
Nice!
Could you add a readme, which explains this a little bit more?
> 3 files changed, 157 insertions(+)
> create mode 100644 drivers/i2c/muxes/i2c-arb-gpio-challenge.c
>
> diff --git a/drivers/i2c/muxes/Kconfig b/drivers/i2c/muxes/Kconfig
> index a05b32d..bd3e078 100644
> --- a/drivers/i2c/muxes/Kconfig
> +++ b/drivers/i2c/muxes/Kconfig
> @@ -6,3 +6,12 @@ config I2C_MUX
> one of several buses using some sort of control mechanism. The
> bus select is handled automatically when that bus is accessed,
> using a suitable I2C MUX driver.
> +
> +config I2C_ARB_GPIO_CHALLENGE
> + bool "GPIO-based I2C arbitration"
> + depends on I2C_MUX
> + help
> + If you say yes to this option, support will be included for an
> + I2C multimaster arbitration scheme using GPIOs and a challenge &
> + response mechanism where masters have to claim the bus by asserting
> + a GPIO.
> diff --git a/drivers/i2c/muxes/Makefile b/drivers/i2c/muxes/Makefile
> index 7583e3a..612cc27 100644
> --- a/drivers/i2c/muxes/Makefile
> +++ b/drivers/i2c/muxes/Makefile
> @@ -3,4 +3,5 @@
> #
> # SPDX-License-Identifier: GPL-2.0+
> #
> +obj-$(CONFIG_I2C_ARB_GPIO_CHALLENGE) += i2c-arb-gpio-challenge.o
> obj-$(CONFIG_I2C_MUX) += i2c-mux-uclass.o
> diff --git a/drivers/i2c/muxes/i2c-arb-gpio-challenge.c b/drivers/i2c/muxes/i2c-arb-gpio-challenge.c
> new file mode 100644
> index 0000000..3f072c7
> --- /dev/null
> +++ b/drivers/i2c/muxes/i2c-arb-gpio-challenge.c
> @@ -0,0 +1,147 @@
> +/*
> + * Copyright (c) 2015 Google, Inc
> + * Written by Simon Glass <sjg@chromium.org>
> + *
> + * SPDX-License-Identifier: GPL-2.0+
> + */
> +
> +#include <common.h>
> +#include <dm.h>
> +#include <errno.h>
> +#include <i2c.h>
> +#include <asm/gpio.h>
> +
> +DECLARE_GLOBAL_DATA_PTR;
> +
> +struct i2c_arbitrator_priv {
> + struct gpio_desc ap_claim;
> + struct gpio_desc ec_claim;
> + uint slew_delay_us;
> + uint wait_retry_ms;
> + uint wait_free_ms;
> +};
> +
> +int i2c_arbitrator_deselect(struct udevice *mux, struct udevice *bus,
> + uint channel)
> +{
> + struct i2c_arbitrator_priv *priv = dev_get_priv(mux);
> + int ret;
> +
> + debug("%s: %s\n", __func__, mux->name);
> + ret = dm_gpio_set_value(&priv->ap_claim, 0);
> + udelay(priv->slew_delay_us);
> +
> + return ret;
> +}
> +
> +int i2c_arbitrator_select(struct udevice *mux, struct udevice *bus,
> + uint channel)
> +{
> + struct i2c_arbitrator_priv *priv = dev_get_priv(mux);
> + unsigned start;
> + int ret;
> +
> + debug("%s: %s\n", __func__, mux->name);
> + /* Start a round of trying to claim the bus */
> + start = get_timer(0);
> + do {
> + unsigned start_retry;
> + int waiting = 0;
> +
> + /* Indicate that we want to claim the bus */
> + ret = dm_gpio_set_value(&priv->ap_claim, 1);
> + if (ret)
> + goto err;
> + udelay(priv->slew_delay_us);
> +
> + /* Wait for the EC to release it */
> + start_retry = get_timer(0);
> + while (get_timer(start_retry) < priv->wait_retry_ms) {
> + ret = dm_gpio_get_value(&priv->ec_claim);
> + if (ret < 0) {
> + goto err;
> + } else if (!ret) {
> + /* We got it, so return */
> + return 0;
> + }
> +
> + if (!waiting)
> + waiting = 1;
> + }
> +
> + /* It didn't release, so give up, wait, and try again */
> + ret = dm_gpio_set_value(&priv->ap_claim, 0);
> + if (ret)
> + goto err;
> +
> + mdelay(priv->wait_retry_ms);
> + } while (get_timer(start) < priv->wait_free_ms);
> +
> + /* Give up, release our claim */
> + printf("I2C: Could not claim bus, timeout %lu\n", get_timer(start));
> + ret = -ETIMEDOUT;
> + ret = 0;
> +err:
> + return ret;
> +}
> +
> +static int i2c_arbitrator_probe(struct udevice *dev)
> +{
> + struct i2c_arbitrator_priv *priv = dev_get_priv(dev);
> + const void *blob = gd->fdt_blob;
> + int node = dev->of_offset;
> + int ret;
> +
> + debug("%s: %s\n", __func__, dev->name);
> + priv->slew_delay_us = fdtdec_get_int(blob, node, "slew-delay-us", 0);
> + priv->wait_retry_ms = fdtdec_get_int(blob, node, "wait-retry-us", 0) /
> + 1000;
> + priv->wait_free_ms = fdtdec_get_int(blob, node, "wait-free-us", 0) /
> + 1000;
> + ret = gpio_request_by_name(dev, "our-claim-gpio", 0, &priv->ap_claim,
> + GPIOD_IS_OUT);
> + if (ret)
> + goto err;
> + ret = gpio_request_by_name(dev, "their-claim-gpios", 0, &priv->ec_claim,
> + GPIOD_IS_IN);
What is "our" and "their"?
bye,
Heiko
> + if (ret)
> + goto err_ec_gpio;
> +
> + return 0;
> +
> +err_ec_gpio:
> + dm_gpio_free(dev, &priv->ap_claim);
> +err:
> + debug("%s: ret=%d\n", __func__, ret);
> + return ret;
> +}
> +
> +static int i2c_arbitrator_remove(struct udevice *dev)
> +{
> + struct i2c_arbitrator_priv *priv = dev_get_priv(dev);
> +
> + dm_gpio_free(dev, &priv->ap_claim);
> + dm_gpio_free(dev, &priv->ec_claim);
> +
> + return 0;
> +}
> +
> +static const struct i2c_mux_ops i2c_arbitrator_ops = {
> + .select = i2c_arbitrator_select,
> + .deselect = i2c_arbitrator_deselect,
> +};
> +
> +static const struct udevice_id i2c_arbitrator_ids[] = {
> + { .compatible = "i2c-arb-gpio-challenge" },
> + { }
> +};
> +
> +U_BOOT_DRIVER(i2c_arbitrator) = {
> + .name = "i2c_arbitrator",
> + .id = UCLASS_I2C_MUX,
> + .of_match = i2c_arbitrator_ids,
> + .probe = i2c_arbitrator_probe,
> + .remove = i2c_arbitrator_remove,
> + .ops = &i2c_arbitrator_ops,
> + .priv_auto_alloc_size = sizeof(struct i2c_arbitrator_priv),
> +};
>
--
DENX Software Engineering GmbH, Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
next prev parent reply other threads:[~2015-07-05 6:43 UTC|newest]
Thread overview: 139+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-07-03 0:15 [U-Boot] [PATCH 00/55] dm: exynos: Driver model improvements leading to spring support Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 01/55] dm: core: Support finding a device by phandle Simon Glass
2015-07-27 23:28 ` Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 02/55] dm: i2c: Add a message debug function Simon Glass
2015-07-03 6:52 ` Heiko Schocher denx
2015-07-27 23:28 ` Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 03/55] dm: i2c: Correct comment nits in dm_i2c_reg_read/write() Simon Glass
2015-07-03 6:53 ` Heiko Schocher denx
2015-07-27 23:28 ` Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 04/55] dm: i2c: Move definitions to the top of the header file Simon Glass
2015-07-03 6:55 ` Heiko Schocher denx
2015-07-27 23:28 ` Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 05/55] dm: i2c: Add a function to transfer messages Simon Glass
2015-07-03 6:58 ` Heiko Schocher denx
2015-07-27 23:28 ` Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 06/55] dm: i2c: Add support for multiplexed I2C buses Simon Glass
2015-07-05 6:10 ` Heiko Schocher
2015-07-06 16:38 ` Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 07/55] exynos: i2c: Correct bug in pinmux selection Simon Glass
2015-07-05 6:15 ` Heiko Schocher
2015-07-06 15:27 ` Simon Glass
2015-07-07 1:36 ` Minkyu Kang
2015-07-08 2:38 ` Simon Glass
2015-07-10 1:43 ` Minkyu Kang
2015-07-10 1:46 ` Simon Glass
2015-07-10 8:56 ` Przemyslaw Marczak
2015-07-10 12:14 ` Minkyu Kang
2015-07-07 6:51 ` Przemyslaw Marczak
2015-07-03 0:15 ` [U-Boot] [PATCH 08/55] i2c: Add a mux for GPIO-based I2C bus arbitration Simon Glass
2015-07-05 6:43 ` Heiko Schocher [this message]
2015-07-06 16:38 ` Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 09/55] exynos: i2c: Fix code style with ReadWriteByte() Simon Glass
2015-07-05 6:45 ` Heiko Schocher
2015-07-27 23:29 ` Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 10/55] exynos: i2c: Tidy up the driver model code Simon Glass
2015-07-06 5:40 ` Heiko Schocher
2015-07-27 23:28 ` Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 11/55] exynos: dts: Sync up I2C ports with the kernel Simon Glass
2015-07-10 11:51 ` Przemyslaw Marczak
2015-07-27 23:29 ` Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 12/55] exynos: dts: Add PMIC and regulator definitions Simon Glass
2015-07-10 11:51 ` Przemyslaw Marczak
2015-07-27 23:29 ` Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 13/55] exynos: dts: Support EC tunnel and main TPS65090 regulator Simon Glass
2015-07-27 23:29 ` Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 14/55] dm: cros_ec: Convert the I2C tunnel code to use driver model Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 15/55] cros_ec: Support the LDO access method used by spring Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 16/55] exynos: serial: Refactor init code for debug UART Simon Glass
2015-07-27 23:29 ` Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 17/55] exynos: Add debug UART support for Samsung S5P serial Simon Glass
2015-07-27 23:29 ` Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 18/55] exynos: Enable the debug UART in SPL Simon Glass
2015-07-27 23:30 ` Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 19/55] dm: gpio: Add support for setting a GPIO's pull direction Simon Glass
2015-07-05 4:55 ` Masahiro Yamada
2015-07-06 16:39 ` Simon Glass
2015-07-06 17:20 ` Masahiro Yamada
2015-07-06 17:33 ` Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 20/55] dm: exynos: gpio: Support pull-up/down in GPIOs Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 21/55] dm: power: Add a new driver for the TPS65090 PMIC Simon Glass
2015-07-10 11:52 ` Przemyslaw Marczak
2015-07-27 23:30 ` Simon Glass
2015-07-03 0:15 ` [U-Boot] [PATCH 22/55] dm: power: Add support for TPS65090 FETs Simon Glass
2015-07-10 11:52 ` Przemyslaw Marczak
2015-07-27 23:30 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 23/55] dm: power: Add support for the S5M8767 PMIC Simon Glass
2015-07-10 11:52 ` Przemyslaw Marczak
2015-07-27 23:30 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 24/55] dm: power: Add support for S5M8767 regulators Simon Glass
2015-07-10 11:52 ` Przemyslaw Marczak
2015-07-27 23:30 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 25/55] dm: pmic: max77686: Correct a few nits Simon Glass
2015-07-10 11:53 ` Przemyslaw Marczak
2015-07-27 23:30 ` Simon Glass
2015-07-28 6:19 ` Przemyslaw Marczak
2015-08-03 14:02 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 26/55] dm: pmic: Correct the pmic_reg_write() implementation Simon Glass
2015-07-10 11:53 ` Przemyslaw Marczak
2015-07-27 23:30 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 27/55] dm: power: max77686: Correct BUCK register access Simon Glass
2015-07-10 11:53 ` Przemyslaw Marczak
2015-07-27 23:30 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 28/55] dm: pmic: max77686: Support all BUCK regulators Simon Glass
2015-07-10 11:53 ` Przemyslaw Marczak
2015-07-30 2:05 ` Simon Glass
2015-07-30 8:22 ` Przemyslaw Marczak
2015-08-03 14:05 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 29/55] dm: power: Don't return an error when regulators are not autoset Simon Glass
2015-07-10 11:54 ` Przemyslaw Marczak
2015-07-27 23:30 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 30/55] dm: pmic: Display the regulator limits on error Simon Glass
2015-07-10 11:54 ` Przemyslaw Marczak
2015-07-27 23:30 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 31/55] dm: video: Add support for video bridges Simon Glass
2015-07-27 23:30 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 32/55] dm: video: Add support for the Parade PS8622/625 bridge Simon Glass
2015-07-27 23:30 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 33/55] dm: video: Add support for the NXP PTN3460 bridge Simon Glass
2015-07-27 23:30 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 34/55] exynos: spi: Convert the timeout to debug() Simon Glass
2015-07-27 23:30 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 35/55] exynos: Correct return value in exynos_mmc_init() Simon Glass
2015-07-27 23:31 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 36/55] exynos: Add support for the DisplayPort hotplug detect Simon Glass
2015-07-27 23:31 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 37/55] exynos: video: Correct debug output Simon Glass
2015-07-27 23:31 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 38/55] exynos: Tidy up CPU frequency display Simon Glass
2015-07-27 23:31 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 39/55] dm: gpio: Check a GPIO is valid before using it Simon Glass
2015-07-27 23:31 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 40/55] dts: exynos: snow: Add memory layout description Simon Glass
2015-07-27 23:31 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 41/55] dts: exynos: pit: Add a new node for the parade video bridge driver Simon Glass
2015-07-27 23:31 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 42/55] dts: exynos: snow: Add a new node for the NXP " Simon Glass
2015-07-27 23:31 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 43/55] exynos: dts: Drop the old TPS65090 I2C node Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 44/55] exynos: Add common board code for exynos5 boards that use device tree Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 45/55] exynos: Enable new features for exynos5 boards Simon Glass
2015-07-10 11:55 ` Przemyslaw Marczak
2015-07-28 16:29 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 46/55] exynos: Remove unneeded device tree control #ifdefs Simon Glass
2015-07-27 23:31 ` Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 47/55] exynos: config: Move common options to the common headers and tidy up Simon Glass
2015-07-10 11:56 ` Przemyslaw Marczak
2015-07-03 0:16 ` [U-Boot] [PATCH 48/55] exynos: Drop old exynos5420-specific board code Simon Glass
2015-07-10 11:56 ` Przemyslaw Marczak
2015-07-03 0:16 ` [U-Boot] [PATCH 49/55] exynos: Drop old exynos5250-specific " Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 50/55] power: Remove old TPS65090 drivers Simon Glass
2015-07-10 11:56 ` Przemyslaw Marczak
2015-07-03 0:16 ` [U-Boot] [PATCH 51/55] cros_ec: Remove the old tunnel code Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 52/55] video: Remove the old parade driver Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 53/55] dts: Drop unused compatible ID for the NXP video bridge Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 54/55] exynos: video: Remove non-device-tree code Simon Glass
2015-07-03 0:16 ` [U-Boot] [PATCH 55/55] exynos: Add support for spring Simon Glass
2015-07-20 14:19 ` [U-Boot] [PATCH 00/55] dm: exynos: Driver model improvements leading to spring support Simon Glass
2015-07-22 4:52 ` Minkyu Kang
2015-07-23 2:08 ` Simon Glass
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=5598D224.8070706@denx.de \
--to=hs@denx.de \
--cc=u-boot@lists.denx.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