Linux block layer
 help / color / mirror / Atom feed
From: Ming Lei <tom.leiming@gmail.com>
To: Jens Axboe <axboe@kernel.dk>, linux-block@vger.kernel.org
Cc: Caleb Sander Mateos <csander@purestorage.com>,
	Uday Shankar <ushankar@purestorage.com>,
	Ming Lei <tom.leiming@gmail.com>
Subject: [PATCH v2] ublk: validate physical_bs_shift, io_min_shift and io_opt_shift
Date: Wed,  6 May 2026 16:22:38 +0800	[thread overview]
Message-ID: <20260506082238.22363-1-tom.leiming@gmail.com> (raw)

ublk_validate_params() checks logical_bs_shift is within
[9, PAGE_SHIFT] but has no upper bound for physical_bs_shift,
io_min_shift, or io_opt_shift. A malicious userspace can set any
of these to a large value (e.g., 44), causing undefined behavior
from `1 << shift` in ublk_ctrl_start_dev() since the result is
stored in 32-bit unsigned int.

Cap all three at ilog2(SZ_256M) (28). 256M is big enough to cover
all practical block sizes, and originates from the maximum physical
block size possible in NVMe (lba_size * (1 + npwg), where npwg is
16-bit).

Also zero out ub->params with memset() when copy_from_user() fails
or ublk_validate_params() returns error, so that no stale or partial
params survive for a subsequent START_DEV to consume.

Fixes: 71f28f3136af ("ublk_drv: add io_uring based userspace block driver")
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
---
V2:
	- zero out ub->params in case of partial copy or validation failure

 drivers/block/ublk_drv.c | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

diff --git a/drivers/block/ublk_drv.c b/drivers/block/ublk_drv.c
index d10460d29e4a..57ec900f0ce0 100644
--- a/drivers/block/ublk_drv.c
+++ b/drivers/block/ublk_drv.c
@@ -900,6 +900,20 @@ static int ublk_validate_params(const struct ublk_device *ub)
 		if (p->logical_bs_shift > PAGE_SHIFT || p->logical_bs_shift < 9)
 			return -EINVAL;
 
+		/*
+		 * 256M is a reasonable upper bound for physical block size,
+		 * io_min and io_opt; it aligns with the maximum physical
+		 * block size possible in NVMe.
+		 */
+		if (p->physical_bs_shift > ilog2(SZ_256M))
+			return -EINVAL;
+
+		if (p->io_min_shift > ilog2(SZ_256M))
+			return -EINVAL;
+
+		if (p->io_opt_shift > ilog2(SZ_256M))
+			return -EINVAL;
+
 		if (p->logical_bs_shift > p->physical_bs_shift)
 			return -EINVAL;
 
@@ -4992,13 +5006,15 @@ static int ublk_ctrl_set_params(struct ublk_device *ub,
 		 */
 		ret = -EACCES;
 	} else if (copy_from_user(&ub->params, argp, ph.len)) {
+		/* zero out partial copy so no stale params survive */
+		memset(&ub->params, 0, sizeof(ub->params));
 		ret = -EFAULT;
 	} else {
 		/* clear all we don't support yet */
 		ub->params.types &= UBLK_PARAM_TYPE_ALL;
 		ret = ublk_validate_params(ub);
 		if (ret)
-			ub->params.types = 0;
+			memset(&ub->params, 0, sizeof(ub->params));
 	}
 	mutex_unlock(&ub->mutex);
 
-- 
2.53.0


             reply	other threads:[~2026-05-06  8:22 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-05-06  8:22 Ming Lei [this message]
2026-05-06 10:43 ` [PATCH v2] ublk: validate physical_bs_shift, io_min_shift and io_opt_shift Jens Axboe

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=20260506082238.22363-1-tom.leiming@gmail.com \
    --to=tom.leiming@gmail.com \
    --cc=axboe@kernel.dk \
    --cc=csander@purestorage.com \
    --cc=linux-block@vger.kernel.org \
    --cc=ushankar@purestorage.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