From: Paul Barker <paul.barker@sancloud.com>
To: shiva.linuxworks@gmail.com, tudor.ambarus@microchip.com,
michael@walle.cc, p.yadav@ti.com, miquel.raynal@bootlin.com,
richard@nod.at, vigneshr@ti.com
Cc: linux-mtd@lists.infradead.org, linux-kernel@vger.kernel.org,
Shivamurthy Shastri <sshivamurthy@micron.com>
Subject: Re: [PATCH 3/4] mtd: add advanced protection and security ioctls
Date: Mon, 6 Dec 2021 11:13:02 +0000 [thread overview]
Message-ID: <ea43e2a8-2415-8c89-c399-998233327ece@sancloud.com> (raw)
In-Reply-To: <4f0177a1-0425-1de8-8e6f-836682d225b9@sancloud.com>
[-- Attachment #1.1.1: Type: text/plain, Size: 8705 bytes --]
Apologies for the double-reply, spotted another thing I'd like to
comment on...
On 06/12/2021 10:42, Paul Barker wrote:
> On 27/10/2021 11:33, shiva.linuxworks@gmail.com wrote:
>> From: Shivamurthy Shastri <sshivamurthy@micron.com>
>>
>> Added new ioctls for advanced protection and security features.
>> These features are currently supported by new Micron SPI NOR flashes.
>>
>> Signed-off-by: Shivamurthy Shastri <sshivamurthy@micron.com>
>> ---
>> drivers/mtd/mtdchar.c | 145 +++++++++++++++++++++++++++++++++++++
>> include/uapi/mtd/mtd-abi.h | 11 +++
>> 2 files changed, 156 insertions(+)
>>
>> diff --git a/drivers/mtd/mtdchar.c b/drivers/mtd/mtdchar.c
>> index 155e991d9d75..97b97b80276d 100644
>> --- a/drivers/mtd/mtdchar.c
>> +++ b/drivers/mtd/mtdchar.c
>> @@ -654,6 +654,16 @@ static int mtdchar_ioctl(struct file *file, u_int
>> cmd, u_long arg)
>> case MTDFILEMODE:
>> case BLKPG:
>> case BLKRRPART:
>> + case SECURE_PACKET_READ:
>> + case SECURE_PACKET_WRITE:
>> + case RD_VLOCK_BITS:
>> + case WR_VLOCK_BITS:
>> + case RD_NVLOCK_BITS:
>> + case WR_NVLOCK_BITS:
>> + case ER_NVLOCK_BITS:
>> + case RD_GLOBAL_FREEZE_BITS:
>> + case WR_GLOBAL_FREEZE_BITS:
>> + case RD_PASSWORD:
>> break;
>
> It looks like you've listed all of the ioctls as "safe" commands so the
> write permission bit is not checked. My understanding is that all ioctls
> which may modify the data in flash need moving to the "dangerous"
> commands section below this so that the write permission bit is checked.
>
>> /* "dangerous" commands */
>> @@ -1017,6 +1027,141 @@ static int mtdchar_ioctl(struct file *file,
>> u_int cmd, u_long arg)
>> ret = 0;
>> break;
>> }
>> + case SECURE_PACKET_READ:
>> + {
>> + struct mtd_oob_buf buf;
>> + u8 *oobbuf;
>> +
>> + if (copy_from_user(&buf, argp, sizeof(buf)))
>> + ret = -EFAULT;
>> +
>> + oobbuf = kmalloc(buf.length, GFP_KERNEL);
>> + ret = master->_secure_packet_read(master, buf.length, oobbuf);
After reviewing patch 2 in this series it's clear that these function
pointers are NULL if a driver does not implement the sec_ops functions.
So unless the implementation there is changed, each of these calls needs
to be wrapped in a test to see if the corresponding pointer is valid and
to return -EOPNOTSUPP if it is NULL.
>> + if (copy_to_user(buf.ptr, oobbuf, buf.length))
>> + ret = -EFAULT;
>> + break;
>> + }
>> +
>> + case SECURE_PACKET_WRITE:
>> + {
>> + struct mtd_oob_buf buf;
>> + u8 *oobbuf;
>> +
>> + if (copy_from_user(&buf, argp, sizeof(buf)))
>> + ret = -EFAULT;
>> +
>> + oobbuf = memdup_user(buf.ptr, buf.length);
>> + ret = master->_secure_packet_write(master, buf.length, oobbuf);
>> + break;
>> + }
>> +
>> + case RD_VLOCK_BITS:
>> + {
>> + struct mtd_oob_buf buf;
>> + u8 *oobbuf;
>> +
>> + if (copy_from_user(&buf, argp, sizeof(buf)))
>> + ret = -EFAULT;
>> +
>> + oobbuf = kmalloc(buf.length, GFP_KERNEL);
>> + ret = master->_read_vlock_bits(master, buf.start, buf.length,
>> + oobbuf);
>> + if (copy_to_user(buf.ptr, oobbuf, buf.length))
>> + ret = -EFAULT;
>> + break;
>> + }
>> +
>> + case WR_VLOCK_BITS:
>> + {
>> + struct mtd_oob_buf buf;
>> + u8 *oobbuf;
>> +
>> + if (copy_from_user(&buf, argp, sizeof(buf)))
>> + ret = -EFAULT;
>> +
>> + oobbuf = memdup_user(buf.ptr, buf.length);
>> + ret = master->_write_vlock_bits(master, buf. start, buf.length,
>> + oobbuf);
>> + break;
>> + }
>> +
>> + case RD_NVLOCK_BITS:
>> + {
>> + struct mtd_oob_buf buf;
>> + u8 *oobbuf;
>> +
>> + if (copy_from_user(&buf, argp, sizeof(buf)))
>> + ret = -EFAULT;
>> +
>> + oobbuf = kmalloc(buf.length, GFP_KERNEL);
>> + ret = master->_read_nvlock_bits(master, buf.start, buf.length,
>> + oobbuf);
>> + if (copy_to_user(buf.ptr, oobbuf, buf.length))
>> + ret = -EFAULT;
>> + break;
>> + }
>> +
>> + case WR_NVLOCK_BITS:
>> + {
>> + struct mtd_oob_buf buf;
>> +
>> + if (copy_from_user(&buf, argp, sizeof(buf)))
>> + ret = -EFAULT;
>> +
>> + ret = master->_write_nvlock_bits(master, buf.start);
>> + break;
>> + }
>> +
>> + case ER_NVLOCK_BITS:
>> + {
>> + ret = master->_erase_nvlock_bits(master);
>> + break;
>> + }
>> +
>> + case RD_GLOBAL_FREEZE_BITS:
>> + {
>> + struct mtd_oob_buf buf;
>> + u8 *oobbuf;
>> +
>> + if (copy_from_user(&buf, argp, sizeof(buf)))
>> + ret = -EFAULT;
>> +
>> + oobbuf = kmalloc(buf.length, GFP_KERNEL);
>> + ret = master->_read_global_freeze_bits(master, buf.length,
>> + oobbuf);
>> + if (copy_to_user(buf.ptr, oobbuf, buf.length))
>> + ret = -EFAULT;
>> + break;
>> + }
>> +
>> + case WR_GLOBAL_FREEZE_BITS:
>> + {
>> + struct mtd_oob_buf buf;
>> + u8 *oobbuf;
>> +
>> + if (copy_from_user(&buf, argp, sizeof(buf)))
>> + ret = -EFAULT;
>> +
>> + oobbuf = memdup_user(buf.ptr, buf.length);
>> + ret = master->_write_global_freeze_bits(master, buf.length,
>> + oobbuf);
>> + break;
>> + }
>> +
>> + case RD_PASSWORD:
>> + {
>> + struct mtd_oob_buf buf;
>> + u8 *oobbuf;
>> +
>> + if (copy_from_user(&buf, argp, sizeof(buf)))
>> + ret = -EFAULT;
>> +
>> + oobbuf = kmalloc(buf.length, GFP_KERNEL);
>> + ret = master->_read_password(master, buf.length, oobbuf);
>> + if (copy_to_user(buf.ptr, oobbuf, buf.length))
>> + ret = -EFAULT;
>> + break;
>> + }
>> }
>> return ret;
>> diff --git a/include/uapi/mtd/mtd-abi.h b/include/uapi/mtd/mtd-abi.h
>> index b869990c2db2..dbd7bf60d484 100644
>> --- a/include/uapi/mtd/mtd-abi.h
>> +++ b/include/uapi/mtd/mtd-abi.h
>> @@ -208,6 +208,17 @@ struct otp_info {
>> /* Erase a given range of user data (must be in mode
>> %MTD_FILE_MODE_OTP_USER) */
>> #define OTPERASE _IOW('M', 25, struct otp_info)
>> +#define SECURE_PACKET_READ _IOWR('M', 26, struct mtd_oob_buf)
>> +#define SECURE_PACKET_WRITE _IOWR('M', 27, struct mtd_oob_buf)
>> +#define RD_VLOCK_BITS _IOWR('M', 28, struct mtd_oob_buf)
>> +#define WR_VLOCK_BITS _IOWR('M', 29, struct mtd_oob_buf)
>> +#define RD_NVLOCK_BITS _IOWR('M', 30, struct mtd_oob_buf)
>> +#define WR_NVLOCK_BITS _IOWR('M', 31, struct mtd_oob_buf)
>> +#define ER_NVLOCK_BITS _IO('M', 32)
>> +#define RD_GLOBAL_FREEZE_BITS _IOWR('M', 33, struct mtd_oob_buf)
>> +#define WR_GLOBAL_FREEZE_BITS _IOWR('M', 34, struct mtd_oob_buf)
>> +#define RD_PASSWORD _IOWR('M', 35, struct mtd_oob_buf)
>> +
>
> All other ioctls defined in this header are preceeded by a comment which
> briefly explains what they do. I think this is needed for these new
> ioctls as well.
>
>> /*
>> * Obsolete legacy interface. Keep it in order not to break userspace
>> * interfaces
>>
>
> Thanks,
>
--
Paul Barker
Principal Software Engineer
SanCloud Ltd
e: paul.barker@sancloud.com
w: https://sancloud.co.uk/
[-- Attachment #1.1.2: OpenPGP public key --]
[-- Type: application/pgp-keys, Size: 7643 bytes --]
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 236 bytes --]
next prev parent reply other threads:[~2021-12-06 11:13 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-27 10:33 [PATCH 0/4] enabling Advanced protection and security features shiva.linuxworks
2021-10-27 10:33 ` [PATCH 1/4] mtd: spi-nor: micron-st: add advanced " shiva.linuxworks
2021-11-08 15:43 ` Michael Walle
2021-12-06 10:49 ` Paul Barker
2021-10-27 10:33 ` [PATCH 2/4] mtd: spi-nor: add advanced protection and security features support shiva.linuxworks
2021-10-27 21:00 ` kernel test robot
2021-10-27 23:01 ` kernel test robot
2021-10-28 4:43 ` kernel test robot
2021-12-06 11:03 ` Paul Barker
2021-10-27 10:33 ` [PATCH 3/4] mtd: add advanced protection and security ioctls shiva.linuxworks
2021-12-06 10:42 ` Paul Barker
2021-12-06 11:13 ` Paul Barker [this message]
2021-10-27 10:33 ` [PATCH 4/4] mtd: spi-nor: micron-st: add mt25qu128abb and mt25ql128abb shiva.linuxworks
2021-12-06 11:05 ` Paul Barker
2021-10-27 10:54 ` [PATCH 0/4] enabling Advanced protection and security features Richard Weinberger
2021-11-08 15:06 ` [EXT] " Shivamurthy Shastri (sshivamurthy)
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=ea43e2a8-2415-8c89-c399-998233327ece@sancloud.com \
--to=paul.barker@sancloud.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mtd@lists.infradead.org \
--cc=michael@walle.cc \
--cc=miquel.raynal@bootlin.com \
--cc=p.yadav@ti.com \
--cc=richard@nod.at \
--cc=shiva.linuxworks@gmail.com \
--cc=sshivamurthy@micron.com \
--cc=tudor.ambarus@microchip.com \
--cc=vigneshr@ti.com \
/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).