From: Ben Dooks <ben-linux-elnMNo+KYs3YtjvyW6yDsg@public.gmane.org>
To: dmitry pervushin
<dpervushin-L1vi/lXTdtvkgf6YlCu6wwC/G2K4zDHf@public.gmane.org>
Cc: Ben Dooks <ben-linux-elnMNo+KYs3YtjvyW6yDsg@public.gmane.org>,
linux-i2c-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: [PATCH, RFC] Freescale STMP: i2c driver
Date: Mon, 8 Jun 2009 23:50:34 +0100 [thread overview]
Message-ID: <20090608225034.GA20446@fluff.org.uk> (raw)
In-Reply-To: <1244059155.4074.19.camel-GL0tbJR47UuzLY3athcO2A@public.gmane.org>
On Wed, Jun 03, 2009 at 11:59:15PM +0400, dmitry pervushin wrote:
> Hello Ben & i2c people,
>
> Here is the patch to support I2C bus on Freescale STMP378x platform; any
> comments are welcome
>
> Signed-off-by: Dmitrij Frasenyak <d.frasenyak-L1vi/lXTdtvkgf6YlCu6wwC/G2K4zDHf@public.gmane.org>
should really have a from Dmitrij Frasenyak line as he is the original
authour and you're just pushing it, IIRC.
> Signed-off-by: dmitry pervishin <dpervushin-L1vi/lXTdtvkgf6YlCu6wwC/G2K4zDHf@public.gmane.org>
>
> ---
> drivers/i2c/busses/Kconfig | 7
> drivers/i2c/busses/Makefile | 1
> drivers/i2c/busses/i2c-stmp378x.c | 440 ++++++++++++++++++++++++++++++++++++++
> 3 files changed, 448 insertions(+)
>
> Index: linux-2.6-arm/drivers/i2c/busses/Kconfig
> ===================================================================
> --- linux-2.6-arm.orig/drivers/i2c/busses/Kconfig
> +++ linux-2.6-arm/drivers/i2c/busses/Kconfig
> @@ -513,6 +513,13 @@ config I2C_SIMTEC
> This driver can also be built as a module. If so, the module
> will be called i2c-simtec.
>
> +config I2C_STMP378X
> + tristate "STMP378x I2C adapter"
> + depends on ARCH_STMP378X
> + help
> + Include support of I2C adapter on Freescale STMP378x board; to build
> + the driver as a module, say M here - module will be called i2c-stmp378x.
> +
> config I2C_VERSATILE
> tristate "ARM Versatile/Realview I2C bus support"
> depends on ARCH_VERSATILE || ARCH_REALVIEW
> Index: linux-2.6-arm/drivers/i2c/busses/Makefile
> ===================================================================
> --- linux-2.6-arm.orig/drivers/i2c/busses/Makefile
> +++ linux-2.6-arm/drivers/i2c/busses/Makefile
> @@ -48,6 +48,7 @@ obj-$(CONFIG_I2C_S6000) += i2c-s6000.o
> obj-$(CONFIG_I2C_SH7760) += i2c-sh7760.o
> obj-$(CONFIG_I2C_SH_MOBILE) += i2c-sh_mobile.o
> obj-$(CONFIG_I2C_SIMTEC) += i2c-simtec.o
> +obj-$(CONFIG_I2C_STMP378X) += i2c-stmp378x.o
> obj-$(CONFIG_I2C_VERSATILE) += i2c-versatile.o
>
> # External I2C/SMBus adapter drivers
> Index: linux-2.6-arm/drivers/i2c/busses/i2c-stmp378x.c
> ===================================================================
> --- /dev/null
> +++ linux-2.6-arm/drivers/i2c/busses/i2c-stmp378x.c
> @@ -0,0 +1,440 @@
> +/*
> + * Freescale STMP378X I2C bus driver
> + *
> + * Author: Dmitrij Frasenyak <sed-L1vi/lXTdtvkgf6YlCu6wwC/G2K4zDHf@public.gmane.org>
> + *
> + * Copyright 2008 Freescale Semiconductor, Inc. All Rights Reserved.
> + * Copyright 2008 Embedded Alley Solutions, Inc All Rights Reserved.
> + */
> +
> +/*
> + * The code contained herein is licensed under the GNU General Public
> + * License. You may obtain a copy of the GNU General Public License
> + * Version 2 or later at the following locations:
> + *
> + * http://www.opensource.org/licenses/gpl-license.html
> + * http://www.gnu.org/copyleft/gpl.html
> + */
> +#include <linux/module.h>
> +#include <linux/delay.h>
> +#include <linux/i2c.h>
> +#include <linux/err.h>
> +#include <linux/interrupt.h>
> +#include <linux/completion.h>
> +#include <linux/platform_device.h>
> +#include <linux/clk.h>
> +#include <linux/io.h>
> +#include <linux/dma-mapping.h>
> +
> +#include <mach/platform.h>
> +#include <mach/stmp3xxx.h>
> +#include <mach/dma.h>
> +#include <mach/regs-i2c.h>
> +#include <mach/regs-apbx.h>
> +
> +static const u32 I2C_READ = 1,
> + I2C_WRITE = 0;
do you really want to be defining things with a prefix of I2C, that
might end up clashing with the i2c core?
> +static const struct stmp3xxx_dma_command cmd_i2c_select = {
> + .cmd = BF(1, APBX_CHn_CMD_XFER_COUNT) |
> + BF(1, APBX_CHn_CMD_CMDWORDS) |
> + BM_APBX_CHn_CMD_WAIT4ENDCMD |
> + BM_APBX_CHn_CMD_CHAIN |
> + BF(BV_APBX_CHn_CMD_COMMAND__DMA_READ, APBX_CHn_CMD_COMMAND),
what is BF() ?
> + .pio_words = {
> + [0] = BM_I2C_CTRL0_RETAIN_CLOCK |
> + BM_I2C_CTRL0_PRE_SEND_START |
> + BM_I2C_CTRL0_MASTER_MODE |
> + BM_I2C_CTRL0_DIRECTION |
> + BF(1, I2C_CTRL0_XFER_COUNT),
> + },
> +};
> +
> +static const struct stmp3xxx_dma_command cmd_i2c_write = {
> + .cmd = BM_APBX_CHn_CMD_SEMAPHORE |
> + BF(1, APBX_CHn_CMD_CMDWORDS) |
> + BM_APBX_CHn_CMD_WAIT4ENDCMD |
> + BM_APBX_CHn_CMD_IRQONCMPLT |
> + BF(BV_APBX_CHn_CMD_COMMAND__DMA_READ, APBX_CHn_CMD_COMMAND),
> + .pio_words = {
> + [0] = BM_I2C_CTRL0_PRE_SEND_START |
> + BM_I2C_CTRL0_MASTER_MODE |
> + BM_I2C_CTRL0_DIRECTION,
> + },
> +};
> +
> +static const struct stmp3xxx_dma_command cmd_i2c_read = {
> + .cmd = BM_APBX_CHn_CMD_SEMAPHORE |
> + BF(1, APBX_CHn_CMD_CMDWORDS) |
> + BM_APBX_CHn_CMD_WAIT4ENDCMD |
> + BM_APBX_CHn_CMD_IRQONCMPLT |
> + BF(BV_APBX_CHn_CMD_COMMAND__DMA_WRITE, APBX_CHn_CMD_COMMAND),
> + .pio_words = {
> + [0] = BM_I2C_CTRL0_SEND_NAK_ON_LAST |
> + BM_I2C_CTRL0_MASTER_MODE,
> + },
> +};
> +
> +struct stmp378x_i2c_dev {
> + struct device *dev;
> + int irq;
> + struct completion cmd_complete;
> + u32 cmd_err;
> + struct i2c_adapter adapter;
> + void __iomem *io;
> + unsigned dma;
> +
> + dma_addr_t bufp;
> + unsigned char *bufv;
> +
> + struct stmp3xxx_dma_descriptor i2c_dma_read[2],
> + i2c_dma_write;
> +};
would be nice to see some kerneldoc descriptions for the fields in
this structure.
> +static int stmp378x_i2c_init(struct stmp378x_i2c_dev *dev)
> +{
> + int err;
> +
> + err = stmp3xxx_dma_request(dev->dma, dev->dev, dev_name(dev->dev));
> + if (err)
> + goto out;
> + stmp3xxx_reset_block(dev->io, true);
> +
> + dev->bufv = dma_alloc_coherent(dev->dev, PAGE_SIZE,
> + &dev->bufp, GFP_KERNEL);
> + if (dma_mapping_error(dev->dev, dev->bufp))
> + goto out_dma;
hmm, doesn't dma_alloc_coherent() fail with an NULL buffer return?
> + err = stmp3xxx_request_pin_group(dev->dev->platform_data,
> + dev_name(dev->dev));
> + if (err)
> + goto out_free;
> +
> + err = stmp3xxx_dma_allocate_command(dev->dma, &dev->i2c_dma_read[0]);
> + if (err)
> + goto out_rd0;
> + err = stmp3xxx_dma_allocate_command(dev->dma, &dev->i2c_dma_read[1]);
> + if (err)
> + goto out_rd1;
> + err = stmp3xxx_dma_allocate_command(dev->dma, &dev->i2c_dma_write);
> + if (err)
> + goto out_wr;
> +
> + stmp3xxx_dma_reset_channel(dev->dma);
> + stmp3xxx_dma_clear_interrupt(dev->dma);
> + stmp3xxx_dma_enable_interrupt(dev->dma);
> +
> + return 0;
> +/*
> + stmp3xxx_dma_free_command(dev->dma, &dev->i2c_dma_write);
> +*/
> +out_wr:
> + stmp3xxx_dma_free_command(dev->dma, &dev->i2c_dma_read[1]);
> +out_rd1:
> + stmp3xxx_dma_free_command(dev->dma, &dev->i2c_dma_read[0]);
> +out_rd0:
> + stmp3xxx_release_pin_group(dev->dev->platform_data, dev_name(dev->dev));
> +out_free:
> + dma_free_coherent(dev->dev, PAGE_SIZE, dev->bufv, dev->bufp);
> +out_dma:
> + stmp3xxx_dma_release(dev->dma);
> +out:
> + return err;
> +}
> +
> +static void stmp378x_i2c_release(struct stmp378x_i2c_dev *dev)
> +{
> + stmp3xxx_dma_free_command(dev->dma, &dev->i2c_dma_write);
> + stmp3xxx_dma_free_command(dev->dma, &dev->i2c_dma_read[1]);
> + stmp3xxx_dma_free_command(dev->dma, &dev->i2c_dma_read[0]);
> + dma_free_coherent(dev->dev, PAGE_SIZE, dev->bufv, dev->bufp);
> + stmp3xxx_release_pin_group(dev->dev->platform_data, dev_name(dev->dev));
> + stmp3xxx_dma_release(dev->dma);
> +}
> +/*
> + * Low level master read/write transaction.
> + */
> +static int stmp378x_i2c_xfer_msg(struct i2c_adapter *adap,
> + struct i2c_msg *msg, int stop)
> +{
> + struct stmp378x_i2c_dev *dev = i2c_get_adapdata(adap);
> + int err;
> + struct stmp3xxx_dma_descriptor *c, *r;
> +
> + init_completion(&dev->cmd_complete);
> + dev->cmd_err = 0;
> +
> + dev_dbg(dev->dev, "start XFER\n");
> + dev_dbg(dev->dev, "addr: 0x%04x, len: %d, flags: 0x%x, stop: %d\n",
> + msg->addr, msg->len, msg->flags, stop);
> +
> + if ((msg->len == 0) || (msg->len > (PAGE_SIZE - 1)))
> + return -EINVAL;
> +
> + if (msg->flags & I2C_M_RD) {
> +
> + c = &dev->i2c_dma_read[0];
> + r = &dev->i2c_dma_read[1];
> +
> + /* first, setup the SELECT command */
> + memcpy(c->command, &cmd_i2c_select, sizeof(cmd_i2c_select));
> + c->command->buf_ptr = dev->bufp;
> + dev->bufv[0] = msg->addr | I2C_READ;
> +
> + /* then, setup the READ command */
> + memcpy(r->command, &cmd_i2c_read, sizeof(cmd_i2c_read));
> + r->command->cmd |= BF(msg->len, APBX_CHn_CMD_XFER_COUNT);
> + r->command->pio_words[0] |= BF(msg->len, I2C_CTRL0_XFER_COUNT);
> + if (stop)
> + r->command->pio_words[0] |= BM_I2C_CTRL0_POST_SEND_STOP;
> + /*
> + * yes, STMP controller can DMA transfer the data from any
> + * byte boundary
> + */
> + r->command->buf_ptr = dev->bufp + 1;
> +
> + /* and chain them together */
> + c->command->next = r->handle;
> +
> + } else {
> +
> + c = &dev->i2c_dma_write;
> + memcpy(c->command, &cmd_i2c_write, sizeof(cmd_i2c_write));
> + c->command->cmd |= BF(msg->len + 1, APBX_CHn_CMD_XFER_COUNT);
> + c->command->pio_words[0] |=
> + BF(msg->len + 1, I2C_CTRL0_XFER_COUNT);
> + if (stop)
> + c->command->pio_words[0] |=
> + BM_I2C_CTRL0_POST_SEND_STOP;
> + dev->bufv[0] = msg->addr | I2C_WRITE;
> + memcpy(dev->bufv + 1, msg->buf, msg->len);
> + }
> +
> + stmp3xxx_dma_go(dev->dma, c, 1);
> +
> + err = wait_for_completion_interruptible_timeout(&dev->cmd_complete,
> + msecs_to_jiffies(1000));
> +
> + if (err < 0) {
> + dev_dbg(dev->dev, "controler is timed out\n");
> + return -ETIMEDOUT;
> + }
> + if (!dev->cmd_err && (msg->flags & I2C_M_RD))
> + memcpy(msg->buf, dev->bufv + 1, msg->len);
> +
> + dev_dbg(dev->dev, "done with err=%d\n", dev->cmd_err);
> +
> + return dev->cmd_err;
> +}
> +
> +
> +static int
> +stmp378x_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg msgs[], int num)
> +{
> + int i;
> + int err;
> +
> + if (!msgs->len)
> + return -EINVAL;
> +
> + for (i = 0; i < num; i++) {
> + err = stmp378x_i2c_xfer_msg(adap, &msgs[i], (i == (num - 1)));
> + if (err)
> + return err;
> + }
> + return num;
> +}
> +
> +static u32
> +stmp378x_i2c_func(struct i2c_adapter *adap)
> +{
> + return I2C_FUNC_I2C | (I2C_FUNC_SMBUS_EMUL & ~I2C_FUNC_SMBUS_QUICK);
> +}
> +
> +#define I2C_IRQ_MASK 0x000000FF
> +static irqreturn_t
> +stmp378x_i2c_isr(int this_irq, void *dev_id)
> +{
> + struct stmp378x_i2c_dev *dev = dev_id;
> + u32 stat;
> + u32 done_mask = BM_I2C_CTRL1_DATA_ENGINE_CMPLT_IRQ |
> + BM_I2C_CTRL1_BUS_FREE_IRQ ;
> +
> + stat = readl(HW_I2C_CTRL1 + dev->io) & I2C_IRQ_MASK;
> + if (!stat)
> + return IRQ_NONE;
> +
> + if (stat & BM_I2C_CTRL1_NO_SLAVE_ACK_IRQ) {
> + dev->cmd_err = -EREMOTEIO;
> +
> + /*
> + * Stop DMA
> + * Clear NAK
> + */
> + stmp3xxx_setl(BM_I2C_CTRL1_CLR_GOT_A_NAK,
> + HW_I2C_CTRL1 + dev->io);
> + stmp3xxx_dma_reset_channel(dev->dma);
> + stmp3xxx_dma_clear_interrupt(dev->dma);
> +
> + complete(&dev->cmd_complete);
> +
> + goto done;
> + }
> +
> + /* Don't care about BM_I2C_CTRL1_OVERSIZE_XFER_TERM_IRQ */
> + if (stat & (BM_I2C_CTRL1_EARLY_TERM_IRQ |
> + BM_I2C_CTRL1_MASTER_LOSS_IRQ |
> + BM_I2C_CTRL1_SLAVE_STOP_IRQ |
> + BM_I2C_CTRL1_SLAVE_IRQ)) {
> + dev->cmd_err = -EIO;
> + complete(&dev->cmd_complete);
> + goto done;
> + }
> +
> + if ((stat & done_mask) == done_mask)
> + complete(&dev->cmd_complete);
> +
> +done:
> + stmp3xxx_clearl(stat, dev->io + HW_I2C_CTRL1);
> + return IRQ_HANDLED;
> +}
> +
> +static const struct i2c_algorithm stmp378x_i2c_algo = {
> + .master_xfer = stmp378x_i2c_xfer,
> + .functionality = stmp378x_i2c_func,
> +};
> +
> +
> +static int __devinit
> +stmp378x_i2c_probe(struct platform_device *pdev)
> +{
> + struct stmp378x_i2c_dev *dev;
> + struct i2c_adapter *adap;
> + struct resource *mem, *dma;
> + int err = 0;
> +
> + /* NOTE: driver uses the static register mapping */
> + dev = kzalloc(sizeof(struct stmp378x_i2c_dev), GFP_KERNEL);
> + if (!dev) {
> + dev_err(&pdev->dev, "no mem \n");
> + return -ENOMEM;
> + }
> +
> + dev->irq = platform_get_irq(pdev, 0); /* Error */
> + if (dev->irq < 0) {
> + dev_err(&pdev->dev, "no err_irq resource\n");
> + err = -ENODEV;
> + goto nores;
> + }
-ENODEV isn't what you want here, you'll end up losing errors in the
higher level.
> + mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> + if (!mem) {
> + dev_err(&pdev->dev, "no memory window\n");
> + err = -ENODEV;
> + goto nores;
> + }
> +
> + dma = platform_get_resource(pdev, IORESOURCE_DMA, 0);
> + if (!dma) {
> + dev_err(&pdev->dev, "no dma channel\n");
> + err = -ENODEV;
> + goto nores;
> + }
> + dev->dma = dma->start;
> +
> + dev->dev = &pdev->dev;
> + dev->io = ioremap(mem->start, resource_size(mem));
> + if (!dev->io) {
> + dev_err(&pdev->dev, "cannot ioremap\n");
> + err = -ENXIO;
> + goto nores;
> + }
> +
> + err = request_irq(dev->irq, stmp378x_i2c_isr, 0, pdev->name, dev);
> + if (err) {
> + dev_err(&pdev->dev, "Can't get IRQ\n");
> + goto no_err_irq;
> + }
> +
> + err = stmp378x_i2c_init(dev);
> + if (err) {
> + dev_err(&pdev->dev, "HW Init failed\n");
> + goto init_failed;
> + }
> +
> + stmp3xxx_setl(0xFF00, dev->io + HW_I2C_CTRL1);
> +
> + adap = &dev->adapter;
> + i2c_set_adapdata(adap, dev);
> + adap->owner = THIS_MODULE;
> + adap->class = I2C_CLASS_HWMON;
is this the correct class?
> + strncpy(adap->name, "378x I2C adapter", sizeof(adap->name));
> + adap->algo = &stmp378x_i2c_algo;
> + adap->dev.parent = &pdev->dev;
> +
> + adap->nr = pdev->id;
> + err = i2c_add_numbered_adapter(adap);
> + if (err) {
> + dev_err(&pdev->dev, "Failed to add adapter\n");
> + goto no_i2c_adapter;
> +
> + }
> +
> + return 0;
> +
> +no_i2c_adapter:
> + stmp378x_i2c_release(dev);
> +init_failed:
> + free_irq(dev->irq, dev);
> +no_err_irq:
> + iounmap(dev->io);
> +nores:
> + kfree(dev);
> + return err;
> +}
> +
> +static int __devexit
> +stmp378x_i2c_remove(struct platform_device *pdev)
> +{
> + struct stmp378x_i2c_dev *dev = platform_get_drvdata(pdev);
> + int res;
> +
> + res = i2c_del_adapter(&dev->adapter);
> + if (res)
> + return -EBUSY;
> +
> + stmp378x_i2c_release(dev);
> +
> + platform_set_drvdata(pdev, NULL);
> +
> + free_irq(dev->irq, dev);
> + iounmap(dev->io);
> +
> + kfree(dev);
> + return 0;
> +}
> +
> +static struct platform_driver stmp378x_i2c_driver = {
> + .probe = stmp378x_i2c_probe,
> + .remove = __devexit_p(stmp378x_i2c_remove),
> + .driver = {
> + .name = "i2c_stmp",
> + .owner = THIS_MODULE,
> + },
> +};
> +
> +static int __init stmp378x_i2c_init_driver(void)
> +{
> + return platform_driver_register(&stmp378x_i2c_driver);
> +}
> +module_init(stmp378x_i2c_init_driver);
> +
> +static void __exit stmp378x_i2c_exit_driver(void)
> +{
> + platform_driver_unregister(&stmp378x_i2c_driver);
> +}
> +module_exit(stmp378x_i2c_exit_driver);
> +
> +MODULE_AUTHOR("d.frasenyak-L1vi/lXTdtvkgf6YlCu6wwC/G2K4zDHf@public.gmane.org");
> +MODULE_DESCRIPTION("I2C for Freescale STMP378x");
> +MODULE_LICENSE("GPL");
no module alias for autoloading?
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-i2c" in
> the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
--
Ben (ben-elnMNo+KYs3YtjvyW6yDsg@public.gmane.org, http://www.fluff.org/)
'a smiley only costs 4 bytes'
next prev parent reply other threads:[~2009-06-08 22:50 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-06-03 19:59 [PATCH, RFC] Freescale STMP: i2c driver dmitry pervushin
[not found] ` <1244059155.4074.19.camel-GL0tbJR47UuzLY3athcO2A@public.gmane.org>
2009-06-08 22:50 ` Ben Dooks [this message]
[not found] ` <20090608225034.GA20446-elnMNo+KYs3pIgCt6eIbzw@public.gmane.org>
2009-06-09 7:46 ` Jean Delvare
2009-06-16 16:25 ` dmitry pervushin
[not found] ` <1245169539.32549.12.camel-GL0tbJR47UuzLY3athcO2A@public.gmane.org>
2009-06-22 10:51 ` Ben Dooks
[not found] ` <20090622105148.GC9006-elnMNo+KYs3pIgCt6eIbzw@public.gmane.org>
2009-06-22 13:41 ` dmitry pervushin
2009-06-15 8:19 ` Ben Dooks
[not found] ` <20090615081938.GA19878-elnMNo+KYs3pIgCt6eIbzw@public.gmane.org>
2009-06-16 16:31 ` dmitry pervushin
[not found] ` <1245169913.32549.19.camel-GL0tbJR47UuzLY3athcO2A@public.gmane.org>
2009-06-22 0:47 ` Ben Dooks
[not found] ` <20090622004741.GB9006-elnMNo+KYs3pIgCt6eIbzw@public.gmane.org>
2009-06-22 13:39 ` dmitry pervushin
2009-06-22 13:42 ` [PATCH, RFC] Freescale STMP: i2c driver, updated with correct signed-off lines dmitry pervushin
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=20090608225034.GA20446@fluff.org.uk \
--to=ben-linux-elnmno+kys3ytjvyw6ydsg@public.gmane.org \
--cc=dpervushin-L1vi/lXTdtvkgf6YlCu6wwC/G2K4zDHf@public.gmane.org \
--cc=linux-i2c-u79uwXL29TY76Z2rM5mHXA@public.gmane.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).