public inbox for stable@vger.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Eric Dumazet <edumazet@google.com>,
	syzbot <syzkaller@googlegroups.com>,
	David Ahern <dsahern@kernel.org>,
	Jakub Kicinski <kuba@kernel.org>, Sasha Levin <sashal@kernel.org>
Subject: [PATCH 5.10 26/86] ipv6: make ip6_rt_gc_expire an atomic_t
Date: Tue, 26 Apr 2022 10:20:54 +0200	[thread overview]
Message-ID: <20220426081741.964618835@linuxfoundation.org> (raw)
In-Reply-To: <20220426081741.202366502@linuxfoundation.org>

From: Eric Dumazet <edumazet@google.com>

[ Upstream commit 9cb7c013420f98fa6fd12fc6a5dc055170c108db ]

Reads and Writes to ip6_rt_gc_expire always have been racy,
as syzbot reported lately [1]

There is a possible risk of under-flow, leading
to unexpected high value passed to fib6_run_gc(),
although I have not observed this in the field.

Hosts hitting ip6_dst_gc() very hard are under pretty bad
state anyway.

[1]
BUG: KCSAN: data-race in ip6_dst_gc / ip6_dst_gc

read-write to 0xffff888102110744 of 4 bytes by task 13165 on cpu 1:
 ip6_dst_gc+0x1f3/0x220 net/ipv6/route.c:3311
 dst_alloc+0x9b/0x160 net/core/dst.c:86
 ip6_dst_alloc net/ipv6/route.c:344 [inline]
 icmp6_dst_alloc+0xb2/0x360 net/ipv6/route.c:3261
 mld_sendpack+0x2b9/0x580 net/ipv6/mcast.c:1807
 mld_send_cr net/ipv6/mcast.c:2119 [inline]
 mld_ifc_work+0x576/0x800 net/ipv6/mcast.c:2651
 process_one_work+0x3d3/0x720 kernel/workqueue.c:2289
 worker_thread+0x618/0xa70 kernel/workqueue.c:2436
 kthread+0x1a9/0x1e0 kernel/kthread.c:376
 ret_from_fork+0x1f/0x30

read-write to 0xffff888102110744 of 4 bytes by task 11607 on cpu 0:
 ip6_dst_gc+0x1f3/0x220 net/ipv6/route.c:3311
 dst_alloc+0x9b/0x160 net/core/dst.c:86
 ip6_dst_alloc net/ipv6/route.c:344 [inline]
 icmp6_dst_alloc+0xb2/0x360 net/ipv6/route.c:3261
 mld_sendpack+0x2b9/0x580 net/ipv6/mcast.c:1807
 mld_send_cr net/ipv6/mcast.c:2119 [inline]
 mld_ifc_work+0x576/0x800 net/ipv6/mcast.c:2651
 process_one_work+0x3d3/0x720 kernel/workqueue.c:2289
 worker_thread+0x618/0xa70 kernel/workqueue.c:2436
 kthread+0x1a9/0x1e0 kernel/kthread.c:376
 ret_from_fork+0x1f/0x30

value changed: 0x00000bb3 -> 0x00000ba9

Reported by Kernel Concurrency Sanitizer on:
CPU: 0 PID: 11607 Comm: kworker/0:21 Not tainted 5.18.0-rc1-syzkaller-00037-g42e7a03d3bad-dirty #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Workqueue: mld mld_ifc_work

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: syzbot <syzkaller@googlegroups.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20220413181333.649424-1-eric.dumazet@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 include/net/netns/ipv6.h |  4 ++--
 net/ipv6/route.c         | 11 ++++++-----
 2 files changed, 8 insertions(+), 7 deletions(-)

diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index 1c0fbe3abf24..f179996c6184 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -78,8 +78,8 @@ struct netns_ipv6 {
 	struct dst_ops		ip6_dst_ops;
 	rwlock_t		fib6_walker_lock;
 	spinlock_t		fib6_gc_lock;
-	unsigned int		 ip6_rt_gc_expire;
-	unsigned long		 ip6_rt_last_gc;
+	atomic_t		ip6_rt_gc_expire;
+	unsigned long		ip6_rt_last_gc;
 	unsigned char		flowlabel_has_excl;
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
 	bool			fib6_has_custom_rules;
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 776b1b58c5dc..6ace9f0ac22f 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -3192,6 +3192,7 @@ static int ip6_dst_gc(struct dst_ops *ops)
 	int rt_elasticity = net->ipv6.sysctl.ip6_rt_gc_elasticity;
 	int rt_gc_timeout = net->ipv6.sysctl.ip6_rt_gc_timeout;
 	unsigned long rt_last_gc = net->ipv6.ip6_rt_last_gc;
+	unsigned int val;
 	int entries;
 
 	entries = dst_entries_get_fast(ops);
@@ -3202,13 +3203,13 @@ static int ip6_dst_gc(struct dst_ops *ops)
 	    entries <= rt_max_size)
 		goto out;
 
-	net->ipv6.ip6_rt_gc_expire++;
-	fib6_run_gc(net->ipv6.ip6_rt_gc_expire, net, true);
+	fib6_run_gc(atomic_inc_return(&net->ipv6.ip6_rt_gc_expire), net, true);
 	entries = dst_entries_get_slow(ops);
 	if (entries < ops->gc_thresh)
-		net->ipv6.ip6_rt_gc_expire = rt_gc_timeout>>1;
+		atomic_set(&net->ipv6.ip6_rt_gc_expire, rt_gc_timeout >> 1);
 out:
-	net->ipv6.ip6_rt_gc_expire -= net->ipv6.ip6_rt_gc_expire>>rt_elasticity;
+	val = atomic_read(&net->ipv6.ip6_rt_gc_expire);
+	atomic_set(&net->ipv6.ip6_rt_gc_expire, val - (val >> rt_elasticity));
 	return entries > rt_max_size;
 }
 
@@ -6363,7 +6364,7 @@ static int __net_init ip6_route_net_init(struct net *net)
 	net->ipv6.sysctl.ip6_rt_min_advmss = IPV6_MIN_MTU - 20 - 40;
 	net->ipv6.sysctl.skip_notify_on_dev_down = 0;
 
-	net->ipv6.ip6_rt_gc_expire = 30*HZ;
+	atomic_set(&net->ipv6.ip6_rt_gc_expire, 30*HZ);
 
 	ret = 0;
 out:
-- 
2.35.1




  parent reply	other threads:[~2022-04-26  8:38 UTC|newest]

Thread overview: 97+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-04-26  8:20 [PATCH 5.10 00/86] 5.10.113-rc1 review Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 01/86] etherdevice: Adjust ether_addr* prototypes to silence -Wstringop-overead Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 02/86] mm: page_alloc: fix building error on -Werror=array-compare Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 03/86] tracing: Dump stacktrace trigger to the corresponding instance Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 04/86] perf tools: Fix segfault accessing sample_id xyarray Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 05/86] gfs2: assign rgrp glock before compute_bitstructs Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 06/86] net/sched: cls_u32: fix netns refcount changes in u32_change() Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 07/86] ALSA: usb-audio: Clear MIDI port active flag after draining Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 08/86] ALSA: hda/realtek: Add quirk for Clevo NP70PNP Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 09/86] dm: fix mempool NULL pointer race when completing IO Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 10/86] ASoC: atmel: Remove system clock tree configuration for at91sam9g20ek Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 11/86] ASoC: msm8916-wcd-digital: Check failure for devm_snd_soc_register_component Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 12/86] ASoC: codecs: wcd934x: do not switch off SIDO Buck when codec is in use Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 13/86] dmaengine: imx-sdma: Fix error checking in sdma_event_remap Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 14/86] dmaengine: mediatek:Fix PM usage reference leak of mtk_uart_apdma_alloc_chan_resources Greg Kroah-Hartman
2022-04-27 20:28   ` Pavel Machek
2022-04-27 20:32     ` Pavel Machek
2022-04-26  8:20 ` [PATCH 5.10 15/86] spi: spi-mtk-nor: initialize spi controller after resume Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 16/86] esp: limit skb_page_frag_refill use to a single page Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 17/86] igc: Fix infinite loop in release_swfw_sync Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 18/86] igc: Fix BUG: scheduling while atomic Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 19/86] rxrpc: Restore removed timer deletion Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 20/86] net/smc: Fix sock leak when release after smc_shutdown() Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 21/86] net/packet: fix packet_sock xmit return value checking Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 22/86] ip6_gre: Avoid updating tunnel->tun_hlen in __gre6_xmit() Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 23/86] ip6_gre: Fix skb_under_panic " Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 24/86] net/sched: cls_u32: fix possible leak in u32_init_knode() Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 25/86] l3mdev: l3mdev_master_upper_ifindex_by_index_rcu should be using netdev_master_upper_dev_get_rcu Greg Kroah-Hartman
2022-04-26  8:20 ` Greg Kroah-Hartman [this message]
2022-04-26  8:20 ` [PATCH 5.10 27/86] netlink: reset network and mac headers in netlink_dump() Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 28/86] net: stmmac: Use readl_poll_timeout_atomic() in atomic state Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 29/86] dmaengine: idxd: add RO check for wq max_batch_size write Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 30/86] dmaengine: idxd: add RO check for wq max_transfer_size write Greg Kroah-Hartman
2022-04-26  8:20 ` [PATCH 5.10 31/86] selftests: mlxsw: vxlan_flooding: Prevent flooding of unwanted packets Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 32/86] arm64/mm: Remove [PUD|PMD]_TABLE_BIT from [pud|pmd]_bad() Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 33/86] arm64: mm: fix p?d_leaf() Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 34/86] ARM: vexpress/spc: Avoid negative array index when !SMP Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 35/86] reset: tegra-bpmp: Restore Handle errors in BPMP response Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 36/86] platform/x86: samsung-laptop: Fix an unsigned comparison which can never be negative Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 37/86] ALSA: usb-audio: Fix undefined behavior due to shift overflowing the constant Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 38/86] arm64: dts: imx: Fix imx8*-var-som touchscreen property sizes Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 39/86] vxlan: fix error return code in vxlan_fdb_append Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 40/86] cifs: Check the IOCB_DIRECT flag, not O_DIRECT Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 41/86] net: atlantic: Avoid out-of-bounds indexing Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 42/86] mt76: Fix undefined behavior due to shift overflowing the constant Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 43/86] brcmfmac: sdio: " Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 44/86] dpaa_eth: Fix missing of_node_put in dpaa_get_ts_info() Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 45/86] drm/msm/mdp5: check the return of kzalloc() Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 46/86] net: macb: Restart tx only if queue pointer is lagging Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 47/86] scsi: qedi: Fix failed disconnect handling Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 48/86] stat: fix inconsistency between struct stat and struct compat_stat Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 49/86] nvme: add a quirk to disable namespace identifiers Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 50/86] nvme-pci: disable namespace identifiers for Qemu controllers Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 51/86] EDAC/synopsys: Read the error count from the correct register Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 52/86] mm, hugetlb: allow for "high" userspace addresses Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 53/86] oom_kill.c: futex: delay the OOM reaper to allow time for proper futex cleanup Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 54/86] mm/mmu_notifier.c: fix race in mmu_interval_notifier_remove() Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 55/86] ata: pata_marvell: Check the bmdma_addr beforing reading Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 56/86] dma: at_xdmac: fix a missing check on list iterator Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 57/86] net: atlantic: invert deep par in pm functions, preventing null derefs Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 58/86] xtensa: patch_text: Fixup last cpu should be master Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 59/86] xtensa: fix a7 clobbering in coprocessor context load/store Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 60/86] openvswitch: fix OOB access in reserve_sfa_size() Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 61/86] gpio: Request interrupts after IRQ is initialized Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 62/86] ASoC: soc-dapm: fix two incorrect uses of list iterator Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 63/86] e1000e: Fix possible overflow in LTR decoding Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 64/86] ARC: entry: fix syscall_trace_exit argument Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 65/86] arm_pmu: Validate single/group leader events Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 66/86] sched/pelt: Fix attach_entity_load_avg() corner case Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 67/86] perf/core: Fix perf_mmap fail when CONFIG_PERF_USE_VMALLOC enabled Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 68/86] drm/panel/raspberrypi-touchscreen: Avoid NULL deref if not initialised Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 69/86] drm/panel/raspberrypi-touchscreen: Initialise the bridge in prepare Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 70/86] KVM: PPC: Fix TCE handling for VFIO Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 71/86] drm/vc4: Use pm_runtime_resume_and_get to fix pm_runtime_get_sync() usage Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 72/86] powerpc/perf: Fix power9 event alternatives Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 73/86] perf report: Set PERF_SAMPLE_DATA_SRC bit for Arm SPE event Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 74/86] ext4: fix fallocate to use file_modified to update permissions consistently Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 75/86] ext4: fix symlink file size not match to file content Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 76/86] ext4: fix use-after-free in ext4_search_dir Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 77/86] ext4: limit length to bitmap_maxbytes - blocksize in punch_hole Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 78/86] ext4, doc: fix incorrect h_reserved size Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 79/86] ext4: fix overhead calculation to account for the reserved gdt blocks Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 80/86] ext4: force overhead calculation if the s_overhead_cluster makes no sense Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 81/86] can: isotp: stop timeout monitoring when no first frame was sent Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 82/86] jbd2: fix a potential race while discarding reserved buffers after an abort Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 83/86] spi: atmel-quadspi: Fix the buswidth adjustment between spi-mem and controller Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 84/86] staging: ion: Prevent incorrect reference counting behavour Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 85/86] block/compat_ioctl: fix range check in BLKGETSIZE Greg Kroah-Hartman
2022-04-26  8:21 ` [PATCH 5.10 86/86] Revert "net: micrel: fix KS8851_MLL Kconfig" Greg Kroah-Hartman
2022-04-26 16:20 ` [PATCH 5.10 00/86] 5.10.113-rc1 review Jon Hunter
2022-04-26 17:12 ` Florian Fainelli
2022-04-26 19:58 ` Pavel Machek
2022-04-26 20:12 ` Guenter Roeck
2022-04-26 20:14 ` Shuah Khan
2022-04-27  1:54 ` Slade Watkins
2022-04-27  8:08 ` Naresh Kamboju
2022-04-27 11:09 ` Sudip Mukherjee

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=20220426081741.964618835@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=dsahern@kernel.org \
    --cc=edumazet@google.com \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=sashal@kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=syzkaller@googlegroups.com \
    /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