Linux-NVME Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Shivam Kumar <kumar.shivam43666@gmail.com>
To: hch@lst.de
Cc: mlombard@arkamax.eu, sagi@grimberg.me, kch@nvidia.com,
	linux-nvme@lists.infradead.org, kbusch@kernel.org,
	gregkh@linuxfoundation.org, security@kernel.org,
	Shivam Kumar <kumar.shivam43666@gmail.com>
Subject: [PATCH v3] nvmet-tcp: set and enforce a default MDTS for TCP transport
Date: Fri,  8 May 2026 16:39:12 -0400	[thread overview]
Message-ID: <20260508203912.320938-1-kumar.shivam43666@gmail.com> (raw)
In-Reply-To: <DI4KADNWR1XG.ILIR3Q697X77@arkamax.eu>

Unlike other fabrics transports, the TCP target does not set a default
Maximum Data Transfer Size. With the configfs MDTS entry defaulting to 0
(no limit), a remote attacker can send a CapsuleCmd with an arbitrarily
large SGL length, causing sgl_alloc() in nvmet_tcp_map_data() to attempt
an excessive kernel allocation that triggers the OOM killer.

Set a default MDTS of 9 (2 MiB) for TCP. Enforce the limit server-side
in nvmet_tcp_map_data() by rejecting commands whose SGL length exceeds
the configured MDTS, returning NVME_SC_INVALID_FIELD as required by the
NVMe specification. Admins can still adjust via the configfs mdts
attribute if needed.

Signed-off-by: Shivam Kumar <kumar.shivam43666@gmail.com>
---
 drivers/nvme/target/tcp.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/nvme/target/tcp.c b/drivers/nvme/target/tcp.c
index 164a564ba3b4..098cf877c358 100644
--- a/drivers/nvme/target/tcp.c
+++ b/drivers/nvme/target/tcp.c
@@ -25,6 +25,7 @@
 #define NVMET_TCP_DEF_INLINE_DATA_SIZE	(4 * PAGE_SIZE)
 #define NVMET_TCP_MAXH2CDATA		0x400000 /* 16M arbitrary limit */
 #define NVMET_TCP_BACKLOG 128
+#define NVMET_TCP_DEF_MDTS		9 /* 2 MiB (2^(12+9)) */
 
 static int param_store_val(const char *str, int *val, int min, int max)
 {
@@ -422,6 +423,13 @@ static int nvmet_tcp_map_data(struct nvmet_tcp_cmd *cmd)
 	if (!len)
 		return 0;
 
+	/* Enforce MDTS: abort commands exceeding the advertised limit */
+	if (cmd->req.port->mdts) {
+		u8 mdts = cmd->req.port->mdts;
+		if (mdts < 20 && len > (1U << (12 + mdts)))
+			return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
+	}
+
 	if (sgl->type == ((NVME_SGL_FMT_DATA_DESC << 4) |
 			  NVME_SGL_FMT_OFFSET)) {
 		if (!nvme_is_write(cmd->req.cmd))
@@ -2077,6 +2085,8 @@ static int nvmet_tcp_add_port(struct nvmet_port *nport)
 	INIT_WORK(&port->accept_work, nvmet_tcp_accept_work);
 	if (port->nport->inline_data_size < 0)
 		port->nport->inline_data_size = NVMET_TCP_DEF_INLINE_DATA_SIZE;
+	if (nport->mdts < 0)
+		nport->mdts = NVMET_TCP_DEF_MDTS;
 
 	ret = sock_create(port->addr.ss_family, SOCK_STREAM,
 				IPPROTO_TCP, &port->sock);
-- 
2.34.1



  parent reply	other threads:[~2026-05-08 20:39 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <2026031805-stretch-skid-ae5b@gregkh>
2026-03-19  1:26 ` [PATCH] nvmet-tcp: bound sgl->length check in nvmet_tcp_map_data() Shivam Kumar
2026-03-19  7:59   ` Maurizio Lombardi
2026-03-19 18:00     ` Shivam Kumar
2026-03-20  7:48       ` Christoph Hellwig
2026-04-05 19:46         ` Shivam Kumar
2026-04-07  6:30           ` Christoph Hellwig
2026-04-08  6:30             ` Maurizio Lombardi
2026-04-09  6:08               ` Christoph Hellwig
2026-04-26 23:57                 ` Shivam Kumar
2026-04-27  0:44                 ` [PATCH v2] nvmet-tcp: set a default MDTS of 2 MiB for TCP transport Shivam Kumar
2026-04-28  6:07                   ` Maurizio Lombardi
2026-04-29  1:05                     ` Shivam Kumar
2026-05-08 20:39                     ` Shivam Kumar [this message]
2026-05-10 20:42                       ` [PATCH v3] nvmet-tcp: set and enforce a default MDTS " Sagi Grimberg
2026-05-11  6:36                         ` Christoph Hellwig
2026-05-11  8:01                           ` Sagi Grimberg
2026-05-11  8:04                             ` Christoph Hellwig

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=20260508203912.320938-1-kumar.shivam43666@gmail.com \
    --to=kumar.shivam43666@gmail.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=hch@lst.de \
    --cc=kbusch@kernel.org \
    --cc=kch@nvidia.com \
    --cc=linux-nvme@lists.infradead.org \
    --cc=mlombard@arkamax.eu \
    --cc=sagi@grimberg.me \
    --cc=security@kernel.org \
    /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