mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Greg KH <gregkh@suse.de>
To: linux-kernel@vger.kernel.org, stable@kernel.org
Cc: stable-review@kernel.org, torvalds@linux-foundation.org,
	akpm@linux-foundation.org, alan@lxorguk.ukuu.org.uk,
	Eric Dumazet <eric.dumazet@gmail.com>,
	Johannes Berg <johannes.berg@intel.com>,
	"David S. Miller" <davem@davemloft.net>
Subject: [45/59] netlink: fix compat recvmsg
Date: Tue, 24 Aug 2010 15:24:57 -0700	[thread overview]
Message-ID: <20100824222526.329149319@clark.site> (raw)
In-Reply-To: <20100824224625.GA5449@kroah.com>

2.6.32-stable review patch.  If anyone has any objections, please let us know.

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

From: Johannes Berg <johannes.berg@intel.com>

commit 68d6ac6d2740b6a55f3ae92a4e0be6d881904b32 upstream.

Since
commit 1dacc76d0014a034b8aca14237c127d7c19d7726
Author: Johannes Berg <johannes@sipsolutions.net>
Date:   Wed Jul 1 11:26:02 2009 +0000

    net/compat/wext: send different messages to compat tasks

we had a race condition when setting and then
restoring frag_list. Eric attempted to fix it,
but the fix created even worse problems.

However, the original motivation I had when I
added the code that turned out to be racy is
no longer clear to me, since we only copy up
to skb->len to userspace, which doesn't include
the frag_list length. As a result, not doing
any frag_list clearing and restoring avoids
the race condition, while not introducing any
other problems.

Additionally, while preparing this patch I found
that since none of the remaining netlink code is
really aware of the frag_list, we need to use the
original skb's information for packet information
and credentials. This fixes, for example, the
group information received by compat tasks.

Cc: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 net/netlink/af_netlink.c |   46 ++++++++++++++++------------------------------
 1 file changed, 16 insertions(+), 30 deletions(-)

--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -1363,7 +1363,7 @@ static int netlink_recvmsg(struct kiocb
 	struct netlink_sock *nlk = nlk_sk(sk);
 	int noblock = flags&MSG_DONTWAIT;
 	size_t copied;
-	struct sk_buff *skb, *frag __maybe_unused = NULL;
+	struct sk_buff *skb, *data_skb;
 	int err;
 
 	if (flags&MSG_OOB)
@@ -1375,45 +1375,35 @@ static int netlink_recvmsg(struct kiocb
 	if (skb == NULL)
 		goto out;
 
+	data_skb = skb;
+
 #ifdef CONFIG_COMPAT_NETLINK_MESSAGES
 	if (unlikely(skb_shinfo(skb)->frag_list)) {
-		bool need_compat = !!(flags & MSG_CMSG_COMPAT);
-
 		/*
-		 * If this skb has a frag_list, then here that means that
-		 * we will have to use the frag_list skb for compat tasks
-		 * and the regular skb for non-compat tasks.
+		 * If this skb has a frag_list, then here that means that we
+		 * will have to use the frag_list skb's data for compat tasks
+		 * and the regular skb's data for normal (non-compat) tasks.
 		 *
-		 * The skb might (and likely will) be cloned, so we can't
-		 * just reset frag_list and go on with things -- we need to
-		 * keep that. For the compat case that's easy -- simply get
-		 * a reference to the compat skb and free the regular one
-		 * including the frag. For the non-compat case, we need to
-		 * avoid sending the frag to the user -- so assign NULL but
-		 * restore it below before freeing the skb.
+		 * If we need to send the compat skb, assign it to the
+		 * 'data_skb' variable so that it will be used below for data
+		 * copying. We keep 'skb' for everything else, including
+		 * freeing both later.
 		 */
-		if (need_compat) {
-			struct sk_buff *compskb = skb_shinfo(skb)->frag_list;
-			skb_get(compskb);
-			kfree_skb(skb);
-			skb = compskb;
-		} else {
-			frag = skb_shinfo(skb)->frag_list;
-			skb_shinfo(skb)->frag_list = NULL;
-		}
+		if (flags & MSG_CMSG_COMPAT)
+			data_skb = skb_shinfo(skb)->frag_list;
 	}
 #endif
 
 	msg->msg_namelen = 0;
 
-	copied = skb->len;
+	copied = data_skb->len;
 	if (len < copied) {
 		msg->msg_flags |= MSG_TRUNC;
 		copied = len;
 	}
 
-	skb_reset_transport_header(skb);
-	err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied);
+	skb_reset_transport_header(data_skb);
+	err = skb_copy_datagram_iovec(data_skb, 0, msg->msg_iov, copied);
 
 	if (msg->msg_name) {
 		struct sockaddr_nl *addr = (struct sockaddr_nl *)msg->msg_name;
@@ -1433,11 +1423,7 @@ static int netlink_recvmsg(struct kiocb
 	}
 	siocb->scm->creds = *NETLINK_CREDS(skb);
 	if (flags & MSG_TRUNC)
-		copied = skb->len;
-
-#ifdef CONFIG_COMPAT_NETLINK_MESSAGES
-	skb_shinfo(skb)->frag_list = frag;
-#endif
+		copied = data_skb->len;
 
 	skb_free_datagram(sk, skb);
 



  parent reply	other threads:[~2010-08-24 22:57 UTC|newest]

Thread overview: 62+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-08-24 22:46 [00/59] 2.6.32.21-stable review Greg KH
2010-08-24 22:24 ` [01/59] memstick: fix hangs on unexpected device removal in mspro_blk Greg KH
2010-08-24 22:24 ` [02/59] ASoC: Fix inverted mute controls for WM8580 Greg KH
2010-08-24 22:24 ` [03/59] ASoC: Remove DSP mode support for WM8776 Greg KH
2010-08-24 22:24 ` [04/59] ALSA: riptide - Fix detection / load of firmware files Greg KH
2010-08-24 22:24 ` [05/59] ALSA: emu10k1 - delay the PCM interrupts (add pcm_irq_delay parameter) Greg KH
2010-08-24 22:24 ` [06/59] ALSA: hda - Fix missing stream for second ADC on Realtek ALC260 HDA codec Greg KH
2010-08-24 22:24 ` [07/59] ocfs2: do not overwrite error codes in ocfs2_init_acl Greg KH
2010-08-24 22:24 ` [08/59] ocfs2/dlm: fix a dead lock Greg KH
2010-08-24 22:24 ` [09/59] ocfs2 fix o2dlm dlm run purgelist (rev 3) Greg KH
2010-08-24 22:24 ` [10/59] ocfs2: Count more refcount records in file system fragmentation Greg KH
2010-08-24 22:24 ` [11/59] ocfs2/dlm: avoid incorrect bit set in refmap on recovery master Greg KH
2010-08-24 22:24 ` [12/59] ocfs2/dlm: remove potential deadlock -V3 Greg KH
2010-08-24 22:24 ` [13/59] x86, hotplug: Serialize CPU hotplug to avoid bringup concurrency issues Greg KH
2010-08-24 22:24 ` [14/59] x86, apic: Fix apic=debug boot crash Greg KH
2010-08-24 22:24 ` [15/59] Fix the nested PR lock calling issue in ACL Greg KH
2010-08-24 22:24 ` [16/59] hwmon: (pc87360) Fix device resource declaration Greg KH
2010-08-24 22:24 ` [17/59] ARM: Tighten check for allowable CPSR values Greg KH
2010-08-24 22:24 ` [18/59] nfs: Add "lookupcache" to displayed mount options Greg KH
2010-08-24 22:24 ` [19/59] ath5k: disable ASPM L0s for all cards Greg KH
2010-08-24 22:24 ` [20/59] pxa3xx: fix ns2cycle equation Greg KH
2010-08-24 22:24 ` [21/59] drm/i915/edp: Flush the write before waiting for PLLs Greg KH
2010-08-24 22:24 ` [22/59] dm mpath: fix NULL pointer dereference when path parameters missing Greg KH
2010-08-24 22:24 ` [23/59] dm ioctl: release _hash_lock between devices in remove_all Greg KH
2010-08-24 22:24 ` [24/59] mm: make the vma list be doubly linked Greg KH
2010-08-24 22:24 ` [25/59] mm: make the mlock() stack guard page checks stricter Greg KH
2010-08-24 22:24 ` [26/59] mm: make stack guard page logic use vm_prev pointer Greg KH
2010-08-24 22:24 ` [27/59] drm/i915: fix hibernation since i915 self-reclaim fixes Greg KH
2010-08-24 22:24 ` [28/59] drm/i915: add reclaimable to i915 self-reclaimable page allocations Greg KH
2010-08-24 22:24 ` [29/59] slab: fix object alignment Greg KH
2010-08-24 22:24 ` [30/59] sunxvr500: Ignore secondary output PCI devices Greg KH
2010-08-24 22:24 ` [31/59] sparc64: Add missing ID to parport probing code Greg KH
2010-08-24 22:24 ` [32/59] sparc64: Fix rwsem constant bug leading to hangs Greg KH
2010-08-24 22:24 ` [33/59] sparc64: Fix atomic64_t routine return values Greg KH
2010-08-24 22:24 ` [34/59] net: Fix a memmove bug in dev_gro_receive() Greg KH
2010-08-24 22:24 ` [35/59] can: add limit for nframes and clean up signed/unsigned variables Greg KH
2010-08-24 22:24 ` [36/59] isdn: fix information leak Greg KH
2010-08-24 22:24 ` [37/59] act_nat: the checksum of ICMP doesnt have pseudo header Greg KH
2010-08-24 22:24 ` [38/59] vmscan: raise the bar to PAGEOUT_IO_SYNC stalls Greg KH
2010-08-24 22:24 ` [39/59] pcmcia: avoid buffer overflow in pcmcia_setup_isa_irq Greg KH
2010-08-24 22:24 ` [40/59] ext4: consolidate in_range() definitions Greg KH
2010-08-24 22:24 ` [41/59] Oprofile: Change CPUIDS from decimal to hex, and add some comments Greg KH
2010-08-24 22:24 ` [42/59] oprofile: add support for Intel processor model 30 Greg KH
2010-08-24 22:24 ` [43/59] fixes for using make 3.82 Greg KH
2010-08-24 22:24 ` [44/59] ALSA: intel8x0: Mute External Amplifier by default for ThinkPad X31 Greg KH
2010-08-24 22:24 ` Greg KH [this message]
2010-08-24 22:24 ` [46/59] drm/radeon/kms: fix typo in radeon_compute_pll_gain Greg KH
2010-08-24 22:24 ` [47/59] drm: stop information leak of old kernel stack Greg KH
2010-08-24 22:25 ` [48/59] powerpc: Fix typo in uImage target Greg KH
2010-08-24 22:25 ` [49/59] powerpc: Initialise paca->kstack before early_setup_secondary Greg KH
2010-08-26  7:10   ` Matt Evans
2010-08-26 23:25     ` Greg KH
2010-08-24 22:25 ` [50/59] USB: option: add Celot CT-650 Greg KH
2010-08-24 22:25 ` [51/59] USB: add device IDs for igotu to navman Greg KH
2010-08-24 22:25 ` [52/59] USB: pl2303: New vendor and product id Greg KH
2010-08-24 22:25 ` [53/59] USB: CP210x Fix Break On/Off Greg KH
2010-08-24 22:25 ` [54/59] USB: ftdi_sio: fix endianess of max packet size Greg KH
2010-08-24 22:25 ` [55/59] USB: io_ti: check firmware version before updating Greg KH
2010-08-24 22:25 ` [56/59] USB: xhci: Remove buggy assignment in next_trb() Greg KH
2010-08-24 22:25 ` [57/59] USB: ftdi_sio: Add ID for Ionics PlugComputer Greg KH
2010-08-24 22:25 ` [58/59] USB: ftdi_sio: add product ID for Lenz LI-USB Greg KH
2010-08-24 22:25 ` [59/59] x86, apic: ack all pending irqs when crashed/on kexec 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=20100824222526.329149319@clark.site \
    --to=gregkh@suse.de \
    --cc=akpm@linux-foundation.org \
    --cc=alan@lxorguk.ukuu.org.uk \
    --cc=davem@davemloft.net \
    --cc=eric.dumazet@gmail.com \
    --cc=johannes.berg@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable-review@kernel.org \
    --cc=stable@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

all inboxes | Powered by JetHome®