public inbox for stable@vger.kernel.org
 help / color / mirror / Atom feed
From: Ben Hutchings <ben@decadent.org.uk>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: akpm@linux-foundation.org,
	"Linus Torvalds" <torvalds@linux-foundation.org>,
	"Doug Ledford" <dledford@redhat.com>,
	"Jason Gunthorpe" <jgunthorpe@obsidianresearch.com>,
	"Jann Horn" <jann@thejh.net>
Subject: [PATCH 3.2 31/46] IB/security: Restrict use of the write() interface
Date: Sun, 12 Jun 2016 22:34:42 +0100	[thread overview]
Message-ID: <lsq.1465767282.268074732@decadent.org.uk> (raw)
In-Reply-To: <lsq.1465767281.501580564@decadent.org.uk>

3.2.81-rc1 review patch.  If anyone has any objections, please let me know.

------------------

From: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>

commit e6bd18f57aad1a2d1ef40e646d03ed0f2515c9e3 upstream.

The drivers/infiniband stack uses write() as a replacement for
bi-directional ioctl().  This is not safe. There are ways to
trigger write calls that result in the return structure that
is normally written to user space being shunted off to user
specified kernel memory instead.

For the immediate repair, detect and deny suspicious accesses to
the write API.

For long term, update the user space libraries and the kernel API
to something that doesn't present the same security vulnerabilities
(likely a structured ioctl() interface).

The impacted uAPI interfaces are generally only available if
hardware from drivers/infiniband is installed in the system.

Reported-by: Jann Horn <jann@thejh.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
[ Expanded check to all known write() entry points ]
Signed-off-by: Doug Ledford <dledford@redhat.com>
[bwh: Backported to 3.2:
 - Drop changes to hfi1
 - include/rdma/ib.h didn't exist, so create it with the usual header guard
   and include it in drivers/infiniband/core/ucma.c
 - ipath_write() has the same problem, so add the same restriction there]
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
---
--- a/drivers/infiniband/core/ucm.c
+++ b/drivers/infiniband/core/ucm.c
@@ -48,6 +48,7 @@
 
 #include <asm/uaccess.h>
 
+#include <rdma/ib.h>
 #include <rdma/ib_cm.h>
 #include <rdma/ib_user_cm.h>
 #include <rdma/ib_marshall.h>
@@ -1116,6 +1117,9 @@ static ssize_t ib_ucm_write(struct file
 	struct ib_ucm_cmd_hdr hdr;
 	ssize_t result;
 
+	if (WARN_ON_ONCE(!ib_safe_file_access(filp)))
+		return -EACCES;
+
 	if (len < sizeof(hdr))
 		return -EINVAL;
 
--- a/drivers/infiniband/core/ucma.c
+++ b/drivers/infiniband/core/ucma.c
@@ -47,6 +47,7 @@
 #include <rdma/ib_marshall.h>
 #include <rdma/rdma_cm.h>
 #include <rdma/rdma_cm_ib.h>
+#include <rdma/ib.h>
 
 MODULE_AUTHOR("Sean Hefty");
 MODULE_DESCRIPTION("RDMA Userspace Connection Manager Access");
@@ -1268,6 +1269,9 @@ static ssize_t ucma_write(struct file *f
 	struct rdma_ucm_cmd_hdr hdr;
 	ssize_t ret;
 
+	if (WARN_ON_ONCE(!ib_safe_file_access(filp)))
+		return -EACCES;
+
 	if (len < sizeof(hdr))
 		return -EINVAL;
 
--- a/drivers/infiniband/core/uverbs_main.c
+++ b/drivers/infiniband/core/uverbs_main.c
@@ -48,6 +48,8 @@
 
 #include <asm/uaccess.h>
 
+#include <rdma/ib.h>
+
 #include "uverbs.h"
 
 MODULE_AUTHOR("Roland Dreier");
@@ -580,6 +582,9 @@ static ssize_t ib_uverbs_write(struct fi
 	struct ib_uverbs_file *file = filp->private_data;
 	struct ib_uverbs_cmd_hdr hdr;
 
+	if (WARN_ON_ONCE(!ib_safe_file_access(filp)))
+		return -EACCES;
+
 	if (count < sizeof hdr)
 		return -EINVAL;
 
--- a/drivers/infiniband/hw/qib/qib_file_ops.c
+++ b/drivers/infiniband/hw/qib/qib_file_ops.c
@@ -45,6 +45,8 @@
 #include <linux/delay.h>
 #include <linux/export.h>
 
+#include <rdma/ib.h>
+
 #include "qib.h"
 #include "qib_common.h"
 #include "qib_user_sdma.h"
@@ -1971,6 +1973,9 @@ static ssize_t qib_write(struct file *fp
 	ssize_t ret = 0;
 	void *dest;
 
+	if (WARN_ON_ONCE(!ib_safe_file_access(fp)))
+		return -EACCES;
+
 	if (count < sizeof(cmd.type)) {
 		ret = -EINVAL;
 		goto bail;
--- /dev/null
+++ b/include/rdma/ib.h
@@ -0,0 +1,21 @@
+#if !defined(_RDMA_IB_H)
+#define _RDMA_IB_H
+
+#include <linux/sched.h>
+
+/*
+ * The IB interfaces that use write() as bi-directional ioctl() are
+ * fundamentally unsafe, since there are lots of ways to trigger "write()"
+ * calls from various contexts with elevated privileges. That includes the
+ * traditional suid executable error message writes, but also various kernel
+ * interfaces that can write to file descriptors.
+ *
+ * This function provides protection for the legacy API by restricting the
+ * calling context.
+ */
+static inline bool ib_safe_file_access(struct file *filp)
+{
+	return filp->f_cred == current_cred() && segment_eq(get_fs(), USER_DS);
+}
+
+#endif /* _RDMA_IB_H */
--- a/drivers/infiniband/hw/ipath/ipath_file_ops.c
+++ b/drivers/infiniband/hw/ipath/ipath_file_ops.c
@@ -44,6 +44,8 @@
 #include <linux/cpu.h>
 #include <asm/pgtable.h>
 
+#include <rdma/ib.h>
+
 #include "ipath_kernel.h"
 #include "ipath_common.h"
 #include "ipath_user_sdma.h"
@@ -2239,6 +2241,9 @@ static ssize_t ipath_write(struct file *
 	ssize_t ret = 0;
 	void *dest;
 
+	if (WARN_ON_ONCE(!ib_safe_file_access(fp)))
+		return -EACCES;
+
 	if (count < sizeof(cmd.type)) {
 		ret = -EINVAL;
 		goto bail;


  parent reply	other threads:[~2016-06-12 21:35 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-06-12 21:34 [PATCH 3.2 00/46] 3.2.81-rc1 review Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 04/46] x86/microcode/amd: Do not overwrite final patch levels Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 11/46] libahci: save port map for forced port map Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 18/46] Input: pmic8xxx-pwrkey - fix algorithm for converting trigger delay Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 08/46] x86, sparse: Do not force removal of __user when calling copy_to/from_user_nocheck() Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 10/46] Input: gtco - fix crash on detecting device without endpoints Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 14/46] usb: xhci: applying XHCI_PME_STUCK_QUIRK to Intel BXT B0 host Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 06/46] crypto: gcm - Fix rfc4543 decryption crash Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 07/46] x86: Add 1/2/4/8 byte optimization to 64bit __copy_{from,to}_user_inatomic Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 17/46] usb: hcd: out of bounds access in for_each_companion Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 03/46] x86/microcode/amd: Extract current patch level read to a function Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 15/46] usb: xhci: fix wild pointers in xhci_mem_cleanup Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 01/46] Revert "ax25: add link layer header validation function" Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 09/46] ipmi: fix timeout calculation when bmc is disconnected Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 16/46] USB: uas: Add a new NO_REPORT_LUNS quirk Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 13/46] nl80211: check netlink protocol in socket release notification Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 12/46] kvm: x86: do not leak guest xcr0 into host interrupt handlers Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 02/46] Revert "net: validate variable length ll headers" Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 05/46] crypto: gcm - fix rfc4543 to handle async crypto correctly Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 42/46] get_rock_ridge_filename(): handle malformed NM entries Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 29/46] USB: serial: cp210x: add Straizona Focusers device ids Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 46/46] net: fix a kernel infoleak in x25 module Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 27/46] batman-adv: Fix broadcast/ogm queue limit on a removed interface Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 34/46] EDAC: i7core, sb_edac: Don't return NOTIFY_BAD from mce_decoder callback Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 32/46] thp: introduce hugepage_vma_check() Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 23/46] s390/hugetlb: add hugepages_supported define Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 45/46] net: fix infoleak in rtnetlink Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 28/46] USB: serial: cp210x: add ID for Link ECU Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 25/46] batman-adv: Check skb size before using encapsulated ETH+VLAN header Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 33/46] mm/huge_memory: replace VM_NO_THP VM_BUG_ON with actual VMA check Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 19/46] powerpc: scan_features() updates incorrect bits for REAL_LE Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 43/46] nf_conntrack: avoid kernel pointer value leak in slab name Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 22/46] mm: hugetlb: allow hugepages_supported to be architecture specific Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 41/46] parisc: fix a bug when syscall number of tracee is __NR_Linux_syscalls Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 21/46] hugetlb: ensure hugepage access is denied if hugepages are not supported Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 20/46] atl2: Disable unimplemented scatter/gather feature Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 30/46] drm/i915: Fix system resume if PCI device remained enabled Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 40/46] proc: prevent accessing /proc/<PID>/environ until it's ready Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 36/46] Minimal fix-up of bad hashing behavior of hash_64() Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 44/46] net: fix infoleak in llc Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 39/46] crypto: hash - Fix page length clamping in hash walk Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 37/46] drm/radeon: make sure vertical front porch is at least 1 Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 38/46] ACPICA: Dispatcher: Update thread ID for recursive method calls Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 24/46] x86/mm/xen: Suppress hugetlbfs in PV guests Ben Hutchings
2016-06-12 21:34 ` [PATCH 3.2 35/46] Make hash_64() use a 64-bit multiply when appropriate Ben Hutchings
2016-06-12 21:34 ` Ben Hutchings [this message]
2016-06-14 21:11   ` [PATCH 3.2 31/46] IB/security: Restrict use of the write() interface Sudip Mukherjee
2016-06-14 21:23     ` Ben Hutchings
2016-06-14 22:04       ` Sudip Mukherjee
2016-06-12 21:34 ` [PATCH 3.2 26/46] batman-adv: Reduce refcnt of removed router when updating route Ben Hutchings
2016-06-12 23:13 ` [PATCH 3.2 00/46] 3.2.81-rc1 review Guenter Roeck
2016-06-12 23:49   ` Ben Hutchings
2016-06-13 18:45 ` Ben Hutchings
2016-06-14 21:56 ` Sudip Mukherjee
2016-06-14 22:16   ` Ben Hutchings
2016-06-14 22:35     ` Sudip Mukherjee
2017-10-08 18:55       ` Ben Hutchings

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=lsq.1465767282.268074732@decadent.org.uk \
    --to=ben@decadent.org.uk \
    --cc=akpm@linux-foundation.org \
    --cc=dledford@redhat.com \
    --cc=jann@thejh.net \
    --cc=jgunthorpe@obsidianresearch.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=torvalds@linux-foundation.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