[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH 4/4] lib: make safe_copy_string_from_guest() validate input


  • To: Jan Beulich <jbeulich@xxxxxxxx>, "xen-devel@xxxxxxxxxxxxxxxxxxxx" <xen-devel@xxxxxxxxxxxxxxxxxxxx>
  • From: Teddy Astie <teddy.astie@xxxxxxxxxx>
  • Date: Tue, 16 Jun 2026 11:55:35 +0200
  • Authentication-results: eu.smtp.expurgate.cloud; dkim=pass header.s=selector1 header.d=vates.tech header.i="@vates.tech" header.h="From:Subject:Date:Message-ID:To:Cc:MIME-Version:Content-Type:In-Reply-To:References:Feedback-ID"
  • Autocrypt: addr=teddy.astie@xxxxxxxxxx; keydata= xsDNBGn5sK8BDACuzSrrTjpVf4ay06OYB6yY0J1PqKffihoNMtrQRZjAHxoAPC7LTBVHV/XO Zw5HJc+9R71z1JV+iYg6z3jPziGKzX8Fj3ZXlzJPmpf1PuETH3KdbvtJT4ny+OGntnJntUoR KRPhTirr6yNeBk/637O3CQXjtqFUPZnko8OI/o1yawIBhJJAWicutjkkUgd28Bh6HV9EIumH tCBgn5/1A/fpm9624MMgYLsA8qjC4XsoovQvFCaO8HEhvfzrrTZHjn/nPeB9SigxIxXW8YaT VqMdqul07o72m3eA2mf+LMu9a04FX/d4wbxBLtELm+1jIrbtyaFZEMOLv/haSiS/Lj3btJH/ EoucejoZ5SH49ksmVAmKOLktOaTQ8b2gEvP7iaKiIiszCCtOSRohr+2GvDsDeLvVZnlR3I+S PhHar7TPKjFz0G3DPNolyjXywNqOAMpomSPi8lSwjAFsxOtQbcck/qRGRSNk4DAmH70pA+89 MXfQXZ3qt1Q01B1+sU0I8xsAEQEAAc0kVGVkZHkgQXN0aWUgPHRlZGR5LmFzdGllQHZhdGVz LnRlY2g+wsENBBMBCAA3FiEEGAIew9LzHY3pdrqtZg+p0QLLz9AFAmn5sK8FCQWjmoACGwME CwkIBwUVCAkKCwUWAgMBAAAKCRBmD6nRAsvP0ID6DACGOktArFbLKHNzuyOVCskwfUZPla6Z pd3GZ8r61SrAKePIr2BnpgPkd0hV3bSRkRLIrgjzR2NRCzfp0x0HfuhcYfAYPR46XHTvjaJE v99sT/vGUG1BZguYDOScSEpgSNaNlYum3RKZbMuROxdK8G+YHccJY8PvWSq2K2yiae2KGiAv 1yjnZxug9/PtDfX8vQFUSg2w1ukRDf50wvDohN1zUQfFtofOP2xCRsDZiHAlQ0pF+aUjXQhP eP3IdpfWc8cyRLXF06Rk46YMYCytweGtGdHcqAfrVthl84129ZPN422k/voW0sm14gjYlGcT UwgnYlFRk2FLq0QeKEDcS0aj3o3EVAQCrayoGzi1pnlIKE3PRGUcUzjGVvzQ/po24gOjwba9 Egr/Wmu3MQlx/7A8zT5QBzF/n+RYdLNQ0Eu6YnUwf0Z1uieqNaon+olyIRFiLb/hCZHO6ekN f5vrm2clHUbQAYaPQebknujoKBo6ZLHg0WM1gZS01Gz+aUpKsUfOwM0EafmwsAEMAKiQiZa3 yQMmc/h3sDbfVHPSiBA4IMI/NAB7IotzPHq1GzCpsoVILAhF/INbWjxJ3DbVf+en3/FvdVZg 2S38xtnth0njNdlVKpyxm054phKjbdoFDwaknWolS4hrddTmetSG5/52AjtmPFtlXAk0NmLv fJnW3seXVQbgM7sW/MNXPP5UKDpkGnLhnvej+GU0s3109sJeXT5ImVdphFs9cvyZyBT9t1Pb Rowv58EgV0zE4hbAeVkULAbxFV5b/ExTjjGVHoX7CVhWxvCiTqCUoXZRkUE9C3FnkzEFRkKb Yu6NCfiHfEyB3Xyg9hfdrRgjMRq907zCof+nDtWxGz1MSEuvTj1g9GZ049Bennqzjc/Q+0ov XoK4jm+Py0FiUGUaA6yhexficjH+kCR/xDbVnWrMhSLB4AuTBT9HjfZI6gk3uYLhoT8Pig4/ eVtR2Q1wZIJsFToR6ofGuyECwFcs+PUXN7fmGRSiPXgjAr/zIUBdW0VWCE3OGPNqtRk2E5s6 IQARAQABwsD8BBgBCAAmFiEEGAIew9LzHY3pdrqtZg+p0QLLz9AFAmn5sLAFCQWjmoACGwwA CgkQZg+p0QLLz9DncQwAg76IehTemLIfrB8T9WIBZrI4kUV7G7a4rjiVoUiHYN5QwhnbZnsa JDlt+Ezoqy/510eo2bCSzvW5xXYPgyjcuOPwgQo1Qp764QxyX6rld2f2RcWkDuBHun55ZWXj by8o21ginPRwruBVYY5rVf3DV1iBu4NurUeHtyFk/dS0XTOQi2wVUb17sW/+ybCEokdVacZG zOqP/OmwHrF8ylXlXnhQq6e3r+J+T8fuoGJelm/CJiMwyP6cEWE8sxVqX/iqwjwUYkuOCpE+ lOWSvdNHgoEkWR0RXBPQjnGmLKbfTl/QDXLk6NP2/r9uxm2HL6Ei3QJKSEdrp+XZaVnk/Off O485NOTKwGOxyWb006cTMh53xPkAJFQu4Tvdj+odsHz88jqw5wfPG0BYWx0I/FspYj7N9kZR 8ULR9nX0LvpzJ/kB4NgHIUt8YtIL6ZSfM2dbF7fKzvx1UqFfvozJZwFzfEieJLXa4nlGgR6D x9fhaZEsniw8/bYgC3igkk5YJiOa
  • Cc: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>, Julien Grall <julien@xxxxxxx>, Stefano Stabellini <sstabellini@xxxxxxxxxx>, Anthony PERARD <anthony.perard@xxxxxxxxxx>, Michal Orzel <michal.orzel@xxxxxxx>, Roger Pau Monné <roger.pau@xxxxxxxxxx>, Daniel Smith <dpsmith@xxxxxxxxxxxxxxxxxxxx>, Oleksii Kurochko <oleksii.kurochko@xxxxxxxxx>
  • Delivery-date: Tue, 16 Jun 2026 09:55:49 +0000
  • Feedback-id: default:8631fc262581453bbf619ec5b2062170:Sweego
  • List-id: Xen developer discussion <xen-devel.lists.xenproject.org>

Le 16/06/2026 à 11:05, Jan Beulich a écrit :
... rather than papering over guest flaws: Strings passed ought to be nul-
terminated (yet sadly libxc hasn't been doing so thus far). This way we
also avoid order-1 allocations, seeing that all present callers pass
PAGE_SIZE for max_size.


I'm not sure to fully understand the commit message, is it more about possible PAGE_SIZE+1 allocations (hence 2 pages required) or something else ?

Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
---
I can't spot any caller side use of FLASK_DEVICETREE_LABEL, hence there's
no corresponding prereq patch.

--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -11,6 +11,8 @@ The format is based on [Keep a Changelog
     to obtain an automatically allocated domid.  The prior sentinel values (0
     since the start of Xen, and DOMID_INVALID since Xen 4.21) now no longer
     represent a wildcard input.
+ - XEN_DOMCTL_DEV_DT's, FLASK_[GS]ETBOOL's, and FLASK_DEVICETREE_LABEL's input
+   string sizes need to include the nul terminator.
   - On x86:
     - Enable pf-fixup option by default for PVH dom0.
     - The libxenguest bzImage loader now uses the system liblz4 library.
--- a/xen/lib/guest-strcpy.c
+++ b/xen/lib/guest-strcpy.c
@@ -3,8 +3,8 @@
  #include <xen/err.h>
/*
- * The function copies a string from the guest and adds a NUL to
- * make sure the string is correctly terminated.
+ * The function copies a string from the guest and checks there's a NUL
+ * terminating the string.
   */
  char *safe_copy_string_from_guest(XEN_GUEST_HANDLE(char) u_buf,
                                    size_t size, size_t max_size)
@@ -14,8 +14,7 @@ char *safe_copy_string_from_guest(XEN_GU
      if ( size > max_size )
          return ERR_PTR(-ENOBUFS);
- /* Add an extra +1 to append \0 */
-    tmp = xmalloc_array(char, size + 1);
+    tmp = xmalloc_array(char, size);
      if ( !tmp )
          return ERR_PTR(-ENOMEM);
@@ -24,7 +23,12 @@ char *safe_copy_string_from_guest(XEN_GU
          xfree(tmp);
          return ERR_PTR(-EFAULT);
      }
-    tmp[size] = '\0';
+
+    if ( !memchr(tmp, 0, size) )
+    {
+        xfree(tmp);
+        return ERR_PTR(-EMSGSIZE);
+    }

EMSGSIZE feels a bit odd as the issue here is that there is no NUL terminator rather than the message being too large.

What about EILSEQ as we encountered a malformed string ?

return tmp;
  }
--- a/xen/include/public/domctl.h
+++ b/xen/include/public/domctl.h
@@ -574,7 +574,7 @@ struct xen_domctl_assign_device {
              uint32_t machine_sbdf;   /* machine PCI ID of assigned device */
          } pci;
          struct {
-            uint32_t size; /* Length of the path */
+            uint32_t size; /* Length of the path, including nul terminator */
              XEN_GUEST_HANDLE_64(char) path; /* Path to the device tree node */
  #ifdef __XEN__
              struct dt_device_node *dev; /* Resolved device node of the above 
*/
--- a/xen/include/public/xsm/flask_op.h
+++ b/xen/include/public/xsm/flask_op.h
@@ -26,7 +26,8 @@ typedef struct xen_flask_setenforce xen_
  struct xen_flask_sid_context {
      /* IN/OUT: sid to convert to/from string */
      uint32_t sid;
-    /* IN: size of the context buffer
+    /*
+     * IN: size of the context buffer, including nul terminator
       * OUT: actual size of the output context string
       */
      uint32_t size;
@@ -86,8 +87,11 @@ struct xen_flask_boolean {
      uint8_t new_value;
      /* IN: commit new value instead of only setting pending [SET] */
      uint8_t commit;
-    /* IN: size of boolean name buffer [GET/SET]
-     * OUT: actual size of name [GET only] */
+    /*
+     * IN: size of boolean name buffer [GET/SET]; must cover nul terminator
+     *     if "name" (below) is an input
+     * OUT: actual size of name [GET only]
+     */
      uint32_t size;
      /* IN: if bool_id is -1, used to find boolean [GET/SET]
       * OUT: textual name of boolean [GET only]
@@ -150,7 +154,7 @@ typedef struct xen_flask_relabel xen_fla
  struct xen_flask_devicetree_label {
      /* IN */
      uint32_t sid;
-    uint32_t length;
+    uint32_t length; /* length of the path, including nul terminator */
      XEN_GUEST_HANDLE(char) path;
  };
  typedef struct xen_flask_devicetree_label xen_flask_devicetree_label_t;



The rest looks good to me.

Teddy

Attachment: OpenPGP_0x660FA9D102CBCFD0.asc
Description: OpenPGP public key

Attachment: OpenPGP_signature.asc
Description: OpenPGP digital signature


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.