开始学linux驱动，三个月熟悉，立帖为证！每日汇报进展

发表于 2010-1-19 12:33

出差归来，继续

发表于 2010-1-19 15:05

支持，linux学习更快入门可以看看海同linux课程

发表于 2010-1-19 16:10

原来出差了

发表于 2010-1-19 21:46

今天帮同事整了个USB转串口的工装，USB转串口芯片用的是ft232，同事弄了好多天，都无法下手，然后俺这边直接在驱动里面选择上该芯片，写上测试应用程序就OK了，几个同事大呼linux真强大。

发表于 2010-1-19 22:01

出差时一直在思索sysfs文件系统。
好像/sys目录下只能查看属性，设置属性之类的，我刚开始一直想通过open来打开/sys目录下的某个设备，但是一直没有成功，貌似理解有问题。
写一个设备的驱动查看/sys目录下面，/device下面有该设备，/driver也有该设备驱动，这个时候我想通过open来打开这个设备，也不知道从何下手？iceut兄弟帮我解答下。
打算继续看这一块

发表于 2010-1-19 22:02

明天linux计划：争取搞定spi dataflash驱动

发表于 2010-1-19 22:03

支持一下

发表于 2010-1-19 23:58

路过！！！！！！！！

发表于 2010-1-20 15:30

一起学习,加油:)

发表于 2010-1-20 17:32

过年了买块ARM开发板来学习
好像挺带劲的操作系统于我还是很神奇的未知世界！！

发表于 2010-1-20 20:39

今天试了下spi接口的data flash驱动，linux内核里面已经自带了芯片at45db161驱动，编译进去，启动系统，不能用。
提示检测不到flash。
然后看了下文件board-sam9260ek.c和板子的电路，发现at45db161使用的是片选0，而board-sam9260ek.c使用的是片选1，把sam9260ek.c里面的at45db161更改成片选0后，成功。

发表于 2010-1-20 20:54

手上的板子使用的SPI接口的SD卡，SPI0,片选1。也就是说用的和data flash芯片 at45db161同一路SPI，片选不一样而已。
在驱动里面选择（*）MMC/SD card support
                           （*）MMC block device driver
                           <*> MMC/SD over SPI
然后在文件board-sam9260ek.c里面的ek_spi_devices做如下改动
static struct spi_board_info ek_spi_devices[] = {

#if !defined(CONFIG_MMC_AT91)
{ // DataFlash chip
.modalias = "mtd_dataflash",
// .chip_select = 1,
      .chip_select = 0,    //zhh
.max_speed_hz = 15 * 1000 * 1000,
.bus_num = 0,
},
#if defined(CONFIG_MTD_AT91_DATAFLASH_CARD)
{ // DataFlash card
.modalias = "mtd_dataflash",
.chip_select = 0,
.max_speed_hz = 15 * 1000 * 1000,
.bus_num = 0,
},
#endif
#endif
#if defined(CONFIG_MMC_SPI)
{ /* DataFlash card */
.modalias = "mmc_spi",
.chip_select = 1,
.max_speed_hz = 15 * 1000 * 1000,
.bus_num = 0,
},
#endif

#if defined(CONFIG_SND_AT73C213) || defined(CONFIG_SND_AT73C213_MODULE)
{ /* AT73C213 DAC */
.modalias = "at73c213",
.chip_select = 0,
.max_speed_hz = 10 * 1000 * 1000,
.bus_num = 1,
.mode = SPI_MODE_1,
.platform_data = &at73c213_data,
},
#endif
};
   也就是增加了mmc_spi的初始化部分。
编译完毕，
启动系统，
提示
can't share SPI bus

为了简单，我直接把SPI其他部分屏蔽掉了，重新编译，启动，提示如下：
mmc0: error -84 whilst initialising SD card

一路跟踪
mmc_sd_init_card() ----->mmc_send_cid()----->mmc_send_cxd_data()----->
static int
mmc_send_cxd_data(struct mmc_card *card, struct mmc_host *host,
u32 opcode, void *buf, unsigned len)
{
struct mmc_request mrq;
struct mmc_command cmd;
struct mmc_data data;
struct scatterlist sg;
void *data_buf;

/* dma onto stack is unsafe/nonportable, but callers to this
   * routine normally provide temporary on-stack buffers ...
   */

data_buf = kmalloc(len, GFP_KERNEL);
if (data_buf == NULL)
return -ENOMEM;

memset(&mrq, 0, sizeof(struct mmc_request));
memset(&cmd, 0, sizeof(struct mmc_command));
memset(&data, 0, sizeof(struct mmc_data));

mrq.cmd = &cmd;
mrq.data = &data;

cmd.opcode = opcode;
cmd.arg = 0;

/* NOTE HACK:  the MMC_RSP_SPI_R1 is always correct here, but we
   * rely on callers to never use this with "native" calls for reading
   * CSD or CID.  Native versions of those commands use the R2 type,
   * not R1 plus a data block.
   */
cmd.flags = MMC_RSP_SPI_R1 | MMC_RSP_R1 | MMC_CMD_ADTC;

data.blksz = len;
data.blocks = 1;
data.flags = MMC_DATA_READ;
data.sg = &sg;
data.sg_len = 1;

sg_init_one(&sg, data_buf, len);

if (card)
mmc_set_data_timeout(&data, card);

mmc_wait_for_req(host, &mrq);

memcpy(buf, data_buf, len);
kfree(data_buf);

if (cmd.error)
{
      return cmd.error;
}
if (data.error)
{
      printk("hello mmc_send_cxd_data data.error is %d\n",data.error);
      return data.error;
}

return 0;
}
就是上面的  (data.error)，从这里跳出去的。

我估计我得把SD卡协议看看才能解决这个问题

发表于 2010-1-21 10:04

三个月？哎，spi这种三个星期足够了。

发表于 2010-1-21 20:29

把SPI模式的SD卡协议简单的看了下，知道SD卡的初始化过程，读写数据是怎么回事了。

然后跟踪SD卡的初始化，发现SD卡复位，进入IDLE模式都没问题，而问题是出在获取CID上，程序如下：
static int mmc_sd_init_card(struct mmc_host *host, u32 ocr,
struct mmc_card *oldcard)
{
            省略
----------------

err = mmc_send_if_cond(host, ocr);
if (!err)
ocr |= 1 << 30;

err = mmc_send_app_op_cond(host, ocr, NULL);
if (err)
goto err;

/*
   * For SPI, enable CRC as appropriate.
   */
if (mmc_host_is_spi(host)) {
//err = mmc_spi_set_crc(host, use_spi_crc);
err = mmc_spi_set_crc(host, 0);  //zhh
if (err)
goto err;
}

/*
   * Fetch CID from card.
   */
if (mmc_host_is_spi(host))
{
      err = mmc_send_cid(host, cid);
}
else
err = mmc_all_send_cid(host, cid);
if (err)
goto err;
-----------------------------
省略
}
于是猜想会不会是spi加了ECC的原因，网上有位高手说SPI模式的SD卡加ECC是自讨苦吃，于是我把ECC去掉了，程序上面红色部分。

然后就成功了。

查看dev是179 0
然后mknod
试了下，OK。

发表于 2010-1-21 20:31

接下来打算解决data flash与SD卡共享SPI的问题

发表于 2010-1-21 21:32

理论上来说，data flash与SD卡共用一组SPI，CS不同应该是可以的，但是看了下SPI sd卡的源码，如下：
/* We can use the bus safely iff nobody else will interfere with us.
   * Most commands consist of one SPI message to issue a command, then
   * several more to collect its response, then possibly more for data
   * transfer.  Clocking access to other devices during that period will
   * corrupt the command execution.
   *
   * Until we have software primitives which guarantee non-interference,
   * we'll aim for a hardware-level guarantee.
   *
   * REVISIT we can't guarantee another device won't be added later...
   */
if (spi->master->num_chipselect > 1) {
struct count_children cc;

cc.n = 0;
cc.bus = spi->dev.bus;
status = device_for_each_child(spi->dev.parent, &cc,
maybe_count_child);
if (status < 0) {
dev_err(&spi->dev, "can't share SPI bus\n");
return status;
}

dev_warn(&spi->dev, "ASSUMING SPI bus stays unshared!\n");
}
觉得共享似乎不太可能了，用过的兄弟指点下吧。

发表于 2010-1-21 21:38

这段代码果然有些问题，如下
Subject: [PATCH 2/2] mmc_spi: lock the SPI bus when accessing the card - msg#06885
List: linux-kernel
Date: Prev Next Index Thread: Prev Next Index
From: Yi Li <yi.li@xxxxxxxxxx>

The MMC/SPI spec does not play well with typical SPI design -- it often
needs to send out a command in one message, read a response, then do some
other arbitrary step. Since we can't let another SPI client use the bus
during this time, use the new SPI lock/unlock functions to provide the
required exclusivity.

Signed-off-by: Yi Li <yi.li@xxxxxxxxxx>
Signed-off-by: Mike Frysinger <vapier@xxxxxxxxxx>
---
drivers/mmc/host/mmc_spi.c | 29 ++---------------------------
1 files changed, 2 insertions(+), 27 deletions(-)

diff --git a/drivers/mmc/host/mmc_spi.c b/drivers/mmc/host/mmc_spi.c
index a461017..a96e058 100644
--- a/drivers/mmc/host/mmc_spi.c
+++ b/drivers/mmc/host/mmc_spi.c
@@ -1084,6 +1084,7 @@ static void mmc_spi_request(struct mmc_host *mmc, struct
mmc_request *mrq)
#endif

/* issue command; then optionally data and stop */
+ spi_lock_bus(host->spi);
status = mmc_spi_command_send(host, mrq, mrq->cmd, mrq->data != NULL);
if (status == 0 && mrq->data) {
mmc_spi_data_do(host, mrq->cmd, mrq->data, mrq->data->blksz);
@@ -1092,7 +1093,7 @@ static void mmc_spi_request(struct mmc_host *mmc, struct
mmc_request *mrq)
else
mmc_cs_off(host);
}
-
+ spi_unlock_bus(host->spi);
mmc_request_done(host->mmc, mrq);
}

@@ -1337,32 +1338,6 @@ static int mmc_spi_probe(struct spi_device *spi)
return status;
}

- /* We can use the bus safely iff nobody else will interfere with us.
- * Most commands consist of one SPI message to issue a command, then
- * several more to collect its response, then possibly more for data
- * transfer. Clocking access to other devices during that period will
- * corrupt the command execution.
- *
- * Until we have software primitives which guarantee non-interference,
- * we'll aim for a hardware-level guarantee.
- *
- * REVISIT we can't guarantee another device won't be added later...
- */
- if (spi->master->num_chipselect > 1) {
- struct count_children cc;
-
- cc.n = 0;
- cc.bus = spi->dev.bus;
- status = device_for_each_child(spi->dev.parent, &cc,
- maybe_count_child);
- if (status < 0) {
- dev_err(&spi->dev, "can't share SPI bus\n");
- return status;
- }
-
- dev_warn(&spi->dev, "ASSUMING SPI bus stays unshared!\n");
- }
-
/* We need a supply of ones to transmit. This is the only time
* the CPU touches these, so cache coherency isn't a concern.
*
--
1.6.5.rc1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

发表于 2010-1-22 08:57

佩服楼主的**不懈，加油。

发表于 2010-1-23 12:28

不错。很入门了。

发表于 2010-1-24 11:48

谢谢icecut兄弟的鼓励。linux里面全部都有了，我好像什么都没做，至于入门，就真谈不上呢。

[嵌入式linux] 开始学linux驱动，三个月熟悉，立帖为证！每日汇报进展

浏览过的版块

精华达人奖章

湍急之河流

常驻人口

技术奇才奖章

技术领袖奖章

荣誉元老奖章

坚毅之洋流

十世金身

技术新星奖章

社区建设奖章

欢快之小溪

以坛为家

技术高手奖章

甘甜之泉水

缘定三生