On Mon, Apr 29, 2024 at 12:35 PM Manuel Traut <manut@xxxxxxxxx> wrote:
>
> On Mon, Apr 29, 2024 at 12:08:45PM +0200, Jens Wiklander wrote:
> > On Mon, Apr 29, 2024 at 11:41 AM Manuel Traut <manut@xxxxxxxxx> wrote:
> > >
> > > On Fri, Apr 26, 2024 at 03:24:21PM +0200, Jens Wiklander wrote:
> > > > On Thu, Apr 25, 2024 at 10:43 AM Manuel Traut <manut@xxxxxxxxx> wrote:
> > > > >
> > > > > On Mon, Apr 22, 2024 at 11:19:35AM +0200, Jens Wiklander 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>
> > > > > > ---
> > > > > > drivers/mmc/core/block.c | 241 ++++++++++++++++++++++++++++++++++++++-
> > > > > > 1 file changed, 239 insertions(+), 2 deletions(-)
> > > > > >
> > > > > > diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c
> > > > > > index 32d49100dff5..a7f126fbc605 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);
> > > > > >
> > > > > > /*
> > > > > > @@ -163,6 +207,7 @@ struct mmc_rpmb_data {
> > > > > > int id;
> > > > > > unsigned int part_index;
> > > > > > struct mmc_blk_data *md;
> > > > > > + struct rpmb_dev *rdev;
> > > > > > struct list_head node;
> > > > > > };
> > > > > >
> > > > > > @@ -2672,7 +2717,6 @@ 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);
> > > > > >
> > > > > > return nonseekable_open(inode, filp);
> > > > > > }
> > > > > > @@ -2682,7 +2726,6 @@ 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;
> > > > > > @@ -2703,10 +2746,165 @@ 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_simple_remove(&mmc_rpmb_ida, rpmb->id);
> > > > > > kfree(rpmb);
> > > > > > }
> > > > > >
> > > > > > +static void free_idata(struct mmc_blk_ioc_data **idata, unsigned int cmd_count)
> > > > > > +{
> > > > > > + unsigned int n;
> > > > > > +
> > > > > > + for (n = 0; n < cmd_count; n++)
> > > > > > + kfree(idata[n]);
> > > > > > + kfree(idata);
> > > > > > +}
> > > > > > +
> > > > > > +static struct mmc_blk_ioc_data **alloc_idata(struct mmc_rpmb_data *rpmb,
> > > > > > + unsigned int cmd_count)
> > > > > > +{
> > > > > > + struct mmc_blk_ioc_data **idata;
> > > > > > + unsigned int n;
> > > > > > +
> > > > > > + idata = kcalloc(cmd_count, sizeof(*idata), GFP_KERNEL);
> > > > > > + if (!idata)
> > > > > > + return NULL;
> > > > > > + for (n = 0; n < cmd_count; n++) {
> > > > > > + idata[n] = kcalloc(1, sizeof(**idata), GFP_KERNEL);
> > > > > > + if (!idata[n]) {
> > > > > > + free_idata(idata, n);
> > > > > > + return NULL;
> > > > > > + }
> > > > > > + idata[n]->rpmb = rpmb;
> > > > > > + }
> > > > > > +
> > > > > > + return idata;
> > > > > > +}
> > > > > > +
> > > > > > +static void set_idata(struct mmc_blk_ioc_data *idata, u32 opcode,
> > > > > > + int write_flag, u8 *buf, unsigned int buf_bytes)
> > > > > > +{
> > > > > > + /*
> > > > > > + * The size of an RPMB frame must match what's expected by the
> > > > > > + * hardware.
> > > > > > + */
> > > > > > + BUILD_BUG_ON(sizeof(struct rpmb_frame) != 512);
> > > > > > +
> > > > > > + idata->ic.opcode = opcode;
> > > > > > + idata->ic.flags = MMC_RSP_R1 | MMC_CMD_ADTC;
> > > > > > + idata->ic.write_flag = write_flag;
> > > > > > + idata->ic.blksz = sizeof(struct rpmb_frame);
> > > > > > + idata->ic.blocks = buf_bytes / idata->ic.blksz;
> > > > > > + idata->buf = buf;
> > > > >
> > > > > I tested the series on a i.MX8MM with a eMMC connected via the imx-sdhci
> > > > > controller. Reading from RPMB does not work. It ends in timeouts due to
> > > > > no response from the SDHCI controller.
> > > > >
> > > > > If idata->buf is allocated here with kmalloc(buf_bytes, GFP_KERNEL) and
> > > > > the content of buf is copied to the new allocated area, transfers succeed.
> > > > >
> > > > > Is it possible that idata->buf is not DMA capable? Any other ideas?
> > > >
> > > > Thanks for testing. I don't know, the idata->buf is allocated using
> > > > alloc_pages_exact(nr_pages * PAGE_SIZE, GFP_KERNEL | __GFP_ZERO); in
> > > > optee_pool_op_alloc_helper().
> > >
> > > Is this really true for idata->buf or isnt the complete RPMB frame memory
> > > allocated like this and therefore idata->buf not page aligned?
> >
> > You're right.
> >
> > >
> > > For RPMB via tee-supplicant the idata->buf is allocated within memdup_user
> > > and therefore page aligned.
> >
> > Yes, that's a difference. Have you tested with page-aligned buffers to
> > see if it helps?
>
> Yes, this helps. I tested with the following patch, but probably it can also
> be solved during frame allocation in optee?
Great, thanks for confirming. Yes, we should fix that in the secure world.
Cheers,
Jens
>
>
> commit b84a56c15abdcd07f4dacf0b7f482802f8ce752b
> Author: Manuel Traut <manut@xxxxxxxxx>
> Date: Tue Apr 23 13:22:27 2024 +0200
>
> mmc: core: block: rpmb: Allocate page aligned memory
>
> "Random" position in optee shared memory cannot be used for blk
> IO on an eMMC with the i.MX8 SDHCI.
>
> This is for sure not the best possible solution, but works
> for the moment.
>
> diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c
> index 30da8fd03..f123a6c96 100644
> --- a/drivers/mmc/core/block.c
> +++ b/drivers/mmc/core/block.c
> @@ -2786,6 +2786,8 @@ static struct mmc_blk_ioc_data **alloc_idata(struct mmc_rpmb_data *rpmb,
> return idata;
> }
>
> +#define DYNALLOC 1
> +
> static void set_idata(struct mmc_blk_ioc_data *idata, u32 opcode,
> int write_flag, u8 *buf, unsigned int buf_bytes)
> {
> @@ -2800,10 +2802,23 @@ static void set_idata(struct mmc_blk_ioc_data *idata, u32 opcode,
> idata->ic.write_flag = write_flag;
> idata->ic.blksz = sizeof(struct rpmb_frame);
> idata->ic.blocks = buf_bytes / idata->ic.blksz;
> +#ifdef DYNALLOC
> + idata->buf = kmalloc(buf_bytes, GFP_KERNEL);
> + memcpy(idata->buf, buf, buf_bytes);
> +#else
> idata->buf = buf;
> +#endif
> idata->buf_bytes = buf_bytes;
> }
>
> +#ifdef DYNALLOC
> +static void free_idata_buf(struct mmc_blk_ioc_data *idata, u8 *buf, unsigned int buf_bytes)
> +{
> + memcpy(buf, idata->buf, buf_bytes);
> + kfree(idata->buf);
> +}
> +#endif
> +
> static int mmc_route_rpmb_frames(struct device *dev, u8 *req,
> unsigned int req_len, u8 *resp,
> unsigned int resp_len)
> @@ -2901,6 +2916,13 @@ static int mmc_route_rpmb_frames(struct device *dev, u8 *req,
> blk_execute_rq(rq, false);
> ret = req_to_mmc_queue_req(rq)->drv_op_result;
>
> +#ifdef DYNALLOC
> + free_idata_buf(idata[0], req, req_len);
> + free_idata_buf(idata[1], resp, resp_len);
> + if (write)
> + free_idata_buf(idata[2], resp, resp_len);
> +#endif
> +
> blk_mq_free_request(rq);
>
> out:
>
> > > > Alternatively, it's from the memory
> > > > range mapped using memremap() in optee_config_shm_memremap(), but
> > > > that's only if you don't have "dynamic shared memory is enabled" in
> > > > the boot log.
> > >
> > > "dynamic shared memory is enabled" is in the bootlog, ..
> >
> > Great.
>
> Thanks
> Manuel
