public inbox for linux-usb@vger.kernel.org
 help / color / mirror / Atom feed
From: Oliver Neukum <oneukum@suse.com>
To: linux-usb@vger.kernel.org
Cc: Oliver Neukum <oneukum@suse.com>
Subject: [RFC 5/6] usb: usb_parse_endpoint must not count duplicated endpoints
Date: Thu, 11 Apr 2024 14:43:03 +0200	[thread overview]
Message-ID: <20240411124722.17343-6-oneukum@suse.com> (raw)
In-Reply-To: <20240411124722.17343-1-oneukum@suse.com>

When an interface is parsed the number of endpoints claimed to exist
is compared to the number of endpoint descriptors actually found.
Duplicated endpoints are not parsed in usb_parse_endpoint but
usb_parse_interface counts them. That makes no sense.
To correct this usb_parse_endpoint needs to return feedback
about the validity of parsed endpoints.

Signed-off-by: Oliver Neukum <oneukum@suse.com>
---
 drivers/usb/core/config.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/drivers/usb/core/config.c b/drivers/usb/core/config.c
index 055910fc6b19..50acc9021247 100644
--- a/drivers/usb/core/config.c
+++ b/drivers/usb/core/config.c
@@ -254,7 +254,7 @@ static bool config_endpoint_is_duplicate(struct usb_host_config *config,
 static int usb_parse_endpoint(struct device *ddev, int cfgno,
 		struct usb_host_config *config, int inum, int asnum,
 		struct usb_host_interface *ifp, int num_ep,
-		unsigned char *buffer, int size)
+		unsigned char *buffer, int size, bool *valid)
 {
 	struct usb_device *udev = to_usb_device(ddev);
 	unsigned char *buffer0 = buffer;
@@ -270,6 +270,7 @@ static int usb_parse_endpoint(struct device *ddev, int cfgno,
 
 	buffer += d->bLength;
 	size -= d->bLength;
+	*valid = false;
 
 	if (d->bDescriptorType != USB_DT_ENDPOINT)
 		goto skip_to_next_endpoint_or_interface_descriptor;
@@ -313,6 +314,7 @@ static int usb_parse_endpoint(struct device *ddev, int cfgno,
 		}
 	}
 
+	*valid = true;
 	endpoint = &ifp->endpoint[ifp->desc.bNumEndpoints];
 	++ifp->desc.bNumEndpoints;
 
@@ -581,14 +583,17 @@ static int usb_parse_interface(struct device *ddev, int cfgno,
 	/* Parse all the endpoint descriptors */
 	n = 0;
 	while (size >= sizeof(struct usb_descriptor_header)) { /* minimum length to get bDescriptorType */
+		bool valid;
+
 		if (((struct usb_descriptor_header *) buffer)->bDescriptorType
 		     == USB_DT_INTERFACE)
 			break;
 		retval = usb_parse_endpoint(ddev, cfgno, config, inum, asnum,
-				alt, num_ep, buffer, size);
+				alt, num_ep, buffer, size, &valid);
 		if (retval < 0)
 			return retval;
-		++n;
+		if (valid)
+			++n;
 
 		buffer += retval;
 		size -= retval;
-- 
2.44.0


  parent reply	other threads:[~2024-04-11 12:47 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-11 12:42 Hardening the parser during enumerations Oliver Neukum
2024-04-11 12:42 ` [RFC 1/6] usb: usb_parse_endpoint ignore reserved bits Oliver Neukum
2024-04-11 14:11   ` Greg KH
2024-04-11 14:27     ` Oliver Neukum
2024-04-11 14:58       ` Greg KH
2024-04-11 15:35   ` Alan Stern
2024-04-11 15:40     ` Oliver Neukum
2024-04-11 12:43 ` [RFC 2/6] usb: avoid overrunning a buffer in usb_parse_interface Oliver Neukum
2024-04-11 15:39   ` Alan Stern
2024-04-11 17:36     ` Alan Stern
2024-04-11 12:43 ` [RFC 3/6] usb: usb_parse_endpoint needs to guard against short descriptors Oliver Neukum
2024-04-11 15:57   ` Alan Stern
2024-04-11 12:43 ` [RFC 4/6] usb: usb_parse_endpoint guard against an incromprehensible preamble Oliver Neukum
2024-04-11 16:00   ` Alan Stern
2024-04-11 12:43 ` Oliver Neukum [this message]
2024-04-11 16:04   ` [RFC 5/6] usb: usb_parse_endpoint must not count duplicated endpoints Alan Stern
2024-04-11 12:43 ` [RFC 6/6] usb: config: find_next_descriptor can overflow buffer Oliver Neukum
2024-04-11 16:16   ` Alan Stern
2024-04-11 14:09 ` Hardening the parser during enumerations Greg KH
2024-04-11 15:37   ` Oliver Neukum
2024-04-12  7:54     ` Greg KH

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=20240411124722.17343-6-oneukum@suse.com \
    --to=oneukum@suse.com \
    --cc=linux-usb@vger.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