On Wed, 14 Aug 2024 at 17:36, Jens Wiklander <jens.wiklander@xxxxxxxxxx> wrote:
>
> Register eMMC RPMB partition with the RPMB subsystem and provide
> an implementation for the RPMB access operations abstracting
> the actual multi step process.
>
> Add a callback to extract the needed device information at registration
> to avoid accessing the struct mmc_card at a later stage as we're not
> holding a reference counter for this struct.
>
> Taking the needed reference to md->disk in mmc_blk_alloc_rpmb_part()
> instead of in mmc_rpmb_chrdev_open(). This is needed by the
> route_frames() function pointer in struct rpmb_ops.
>
> Signed-off-by: Tomas Winkler <tomas.winkler@xxxxxxxxx>
> Signed-off-by: Alexander Usyskin <alexander.usyskin@xxxxxxxxx>
> Signed-off-by: Jens Wiklander <jens.wiklander@xxxxxxxxxx>
> Tested-by: Manuel Traut <manut@xxxxxxxxx>
> Reviewed-by: Linus Walleij <linus.walleij@xxxxxxxxxx>
First of all, sorry for the long delay! Besides only one comment, see
more below, this looks good to me.
> ---
> drivers/mmc/core/block.c | 242 ++++++++++++++++++++++++++++++++++++++-
> 1 file changed, 240 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c
> index 2c9963248fcb..cc7318089cf2 100644
> --- a/drivers/mmc/core/block.c
> +++ b/drivers/mmc/core/block.c
> @@ -33,6 +33,7 @@
> #include <linux/cdev.h>
> #include <linux/mutex.h>
> #include <linux/scatterlist.h>
> +#include <linux/string.h>
> #include <linux/string_helpers.h>
> #include <linux/delay.h>
> #include <linux/capability.h>
> @@ -40,6 +41,7 @@
> #include <linux/pm_runtime.h>
> #include <linux/idr.h>
> #include <linux/debugfs.h>
> +#include <linux/rpmb.h>
>
> #include <linux/mmc/ioctl.h>
> #include <linux/mmc/card.h>
> @@ -76,6 +78,48 @@ MODULE_ALIAS("mmc:block");
> #define MMC_EXTRACT_INDEX_FROM_ARG(x) ((x & 0x00FF0000) >> 16)
> #define MMC_EXTRACT_VALUE_FROM_ARG(x) ((x & 0x0000FF00) >> 8)
>
> +/**
> + * struct rpmb_frame - rpmb frame as defined by eMMC 5.1 (JESD84-B51)
> + *
> + * @stuff : stuff bytes
> + * @key_mac : The authentication key or the message authentication
> + * code (MAC) depending on the request/response type.
> + * The MAC will be delivered in the last (or the only)
> + * block of data.
> + * @data : Data to be written or read by signed access.
> + * @nonce : Random number generated by the host for the requests
> + * and copied to the response by the RPMB engine.
> + * @write_counter: Counter value for the total amount of the successful
> + * authenticated data write requests made by the host.
> + * @addr : Address of the data to be programmed to or read
> + * from the RPMB. Address is the serial number of
> + * the accessed block (half sector 256B).
> + * @block_count : Number of blocks (half sectors, 256B) requested to be
> + * read/programmed.
> + * @result : Includes information about the status of the write counter
> + * (valid, expired) and result of the access made to the RPMB.
> + * @req_resp : Defines the type of request and response to/from the memory.
> + *
> + * The stuff bytes and big-endian properties are modeled to fit to the spec.
> + */
> +struct rpmb_frame {
> + u8 stuff[196];
> + u8 key_mac[32];
> + u8 data[256];
> + u8 nonce[16];
> + __be32 write_counter;
> + __be16 addr;
> + __be16 block_count;
> + __be16 result;
> + __be16 req_resp;
> +} __packed;
> +
> +#define RPMB_PROGRAM_KEY 0x1 /* Program RPMB Authentication Key */
> +#define RPMB_GET_WRITE_COUNTER 0x2 /* Read RPMB write counter */
> +#define RPMB_WRITE_DATA 0x3 /* Write data to RPMB partition */
> +#define RPMB_READ_DATA 0x4 /* Read data from RPMB partition */
> +#define RPMB_RESULT_READ 0x5 /* Read result request (Internal) */
> +
> static DEFINE_MUTEX(block_mutex);
>
> /*
> @@ -155,6 +199,7 @@ static const struct bus_type mmc_rpmb_bus_type = {
> * @id: unique device ID number
> * @part_index: partition index (0 on first)
> * @md: parent MMC block device
> + * @rdev: registered RPMB device
> * @node: list item, so we can put this device on a list
> */
> struct mmc_rpmb_data {
> @@ -163,6 +208,7 @@ struct mmc_rpmb_data {
> int id;
> unsigned int part_index;
> struct mmc_blk_data *md;
> + struct rpmb_dev *rdev;
> struct list_head node;
> };
>
> @@ -2670,7 +2716,6 @@ static int mmc_rpmb_chrdev_open(struct inode *inode, struct file *filp)
>
> get_device(&rpmb->dev);
> filp->private_data = rpmb;
> - mmc_blk_get(rpmb->md->disk);
If I understand correctly, the call to mmc_blk_get() is here to
reference count the usage of the md->disk. So when a user decides to
open the chrdev, we are keeping a reference to the corresponding data
around, until the corresponding file descriptor is closed and
mmc_rpmb_chrdev_release() is called. We do the similar thing if the
regular mmc block device is opened.
If we change to increase/decrease the reference count from
mmc_blk_alloc_rpmb_part()/mmc_blk_rpmb_device_release() instead - are
you certain that we are still taking care of the above scenario
correctly?
I guess a test that opens the chrdev and then unbinding the mmc host
driver should tell us. :-)
>
> return nonseekable_open(inode, filp);
> }
> @@ -2680,7 +2725,6 @@ static int mmc_rpmb_chrdev_release(struct inode *inode, struct file *filp)
> struct mmc_rpmb_data *rpmb = container_of(inode->i_cdev,
> struct mmc_rpmb_data, chrdev);
>
> - mmc_blk_put(rpmb->md);
> put_device(&rpmb->dev);
>
> return 0;
> @@ -2701,10 +2745,165 @@ static void mmc_blk_rpmb_device_release(struct device *dev)
> {
> struct mmc_rpmb_data *rpmb = dev_get_drvdata(dev);
>
> + rpmb_dev_unregister(rpmb->rdev);
> + mmc_blk_put(rpmb->md);
> ida_free(&mmc_rpmb_ida, rpmb->id);
> kfree(rpmb);
> }
>
[...]
Kind regards
Uffe
