[5.10,06/86] net/sched: cls_u32: fix netns refcount changes in u32_change()

From: Eric Dumazet <edumazet@google.com>

From: Eric Dumazet <edumazet@google.com>

commit 3db09e762dc79584a69c10d74a6b98f89a9979f8 upstream.

We are now able to detect extra put_net() at the moment
they happen, instead of much later in correct code paths.

u32_init_knode() / tcf_exts_init() populates the ->exts.net
pointer, but as mentioned in tcf_exts_init(),
the refcount on netns has not been elevated yet.

The refcount is taken only once tcf_exts_get_net()
is called.

So the two u32_destroy_key() calls from u32_change()
are attempting to release an invalid reference on the netns.

syzbot report:

refcount_t: decrement hit 0; leaking memory.
WARNING: CPU: 0 PID: 21708 at lib/refcount.c:31 refcount_warn_saturate+0xbf/0x1e0 lib/refcount.c:31
Modules linked in:
CPU: 0 PID: 21708 Comm: syz-executor.5 Not tainted 5.18.0-rc2-next-20220412-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:refcount_warn_saturate+0xbf/0x1e0 lib/refcount.c:31
Code: 1d 14 b6 b2 09 31 ff 89 de e8 6d e9 89 fd 84 db 75 e0 e8 84 e5 89 fd 48 c7 c7 40 aa 26 8a c6 05 f4 b5 b2 09 01 e8 e5 81 2e 05 <0f> 0b eb c4 e8 68 e5 89 fd 0f b6 1d e3 b5 b2 09 31 ff 89 de e8 38
RSP: 0018:ffffc900051af1b0 EFLAGS: 00010286
RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
RDX: 0000000000040000 RSI: ffffffff8160a0c8 RDI: fffff52000a35e28
RBP: 0000000000000004 R08: 0000000000000000 R09: 0000000000000000
R10: ffffffff81604a9e R11: 0000000000000000 R12: 1ffff92000a35e3b
R13: 00000000ffffffef R14: ffff8880211a0194 R15: ffff8880577d0a00
FS:  00007f25d183e700(0000) GS:ffff8880b9c00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f19c859c028 CR3: 0000000051009000 CR4: 00000000003506f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <TASK>
 __refcount_dec include/linux/refcount.h:344 [inline]
 refcount_dec include/linux/refcount.h:359 [inline]
 ref_tracker_free+0x535/0x6b0 lib/ref_tracker.c:118
 netns_tracker_free include/net/net_namespace.h:327 [inline]
 put_net_track include/net/net_namespace.h:341 [inline]
 tcf_exts_put_net include/net/pkt_cls.h:255 [inline]
 u32_destroy_key.isra.0+0xa7/0x2b0 net/sched/cls_u32.c:394
 u32_change+0xe01/0x3140 net/sched/cls_u32.c:909
 tc_new_tfilter+0x98d/0x2200 net/sched/cls_api.c:2148
 rtnetlink_rcv_msg+0x80d/0xb80 net/core/rtnetlink.c:6016
 netlink_rcv_skb+0x153/0x420 net/netlink/af_netlink.c:2495
 netlink_unicast_kernel net/netlink/af_netlink.c:1319 [inline]
 netlink_unicast+0x543/0x7f0 net/netlink/af_netlink.c:1345
 netlink_sendmsg+0x904/0xe00 net/netlink/af_netlink.c:1921
 sock_sendmsg_nosec net/socket.c:705 [inline]
 sock_sendmsg+0xcf/0x120 net/socket.c:725
 ____sys_sendmsg+0x6e2/0x800 net/socket.c:2413
 ___sys_sendmsg+0xf3/0x170 net/socket.c:2467
 __sys_sendmsg+0xe5/0x1b0 net/socket.c:2496
 do_syscall_x64 arch/x86/entry/common.c:50 [inline]
 do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
 entry_SYSCALL_64_after_hwframe+0x44/0xae
RIP: 0033:0x7f25d0689049
Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f25d183e168 EFLAGS: 00000246 ORIG_RAX: 000000000000002e
RAX: ffffffffffffffda RBX: 00007f25d079c030 RCX: 00007f25d0689049
RDX: 0000000000000000 RSI: 0000000020000340 RDI: 0000000000000005
RBP: 00007f25d06e308d R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 00007ffd0b752e3f R14: 00007f25d183e300 R15: 0000000000022000
 </TASK>

Fixes: 35c55fc156d8 ("cls_u32: use tcf_exts_get_net() before call_rcu()")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: syzbot <syzkaller@googlegroups.com>
Cc: Cong Wang <xiyou.wangcong@gmail.com>
Cc: Jiri Pirko <jiri@resnulli.us>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 net/sched/cls_u32.c |   16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

Message ID	20220426081741.391271584@linuxfoundation.org
State	Superseded
Headers	show Return-Path: <stable-owner@kernel.org> 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>, Cong Wang <xiyou.wangcong@gmail.com>, Jiri Pirko <jiri@resnulli.us>, Jamal Hadi Salim <jhs@mojatatu.com>, Jakub Kicinski <kuba@kernel.org> Subject: [PATCH 5.10 06/86] net/sched: cls_u32: fix netns refcount changes in u32_change() Date: Tue, 26 Apr 2022 10:20:34 +0200 Message-Id: <20220426081741.391271584@linuxfoundation.org> In-Reply-To: <20220426081741.202366502@linuxfoundation.org> References: <20220426081741.202366502@linuxfoundation.org> User-Agent: quilt/0.66 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Precedence: bulk
Series	None \| expand [5.10,02/86] mm: page_alloc: fix building error on -Werror=array-compare [5.10,03/86] tracing: Dump stacktrace trigger to the corresponding instance [5.10,04/86] perf tools: Fix segfault accessing sample_id xyarray [5.10,05/86] gfs2: assign rgrp glock before compute_bitstructs [5.10,06/86] net/sched: cls_u32: fix netns refcount changes in u32_change() [5.10,07/86] ALSA: usb-audio: Clear MIDI port active flag after draining [5.10,08/86] ALSA: hda/realtek: Add quirk for Clevo NP70PNP [5.10,09/86] dm: fix mempool NULL pointer race when completing IO [5.10,10/86] ASoC: atmel: Remove system clock tree configuration for at91sam9g20ek [5.10,11/86] ASoC: msm8916-wcd-digital: Check failure for devm_snd_soc_register_component [5.10,12/86] ASoC: codecs: wcd934x: do not switch off SIDO Buck when codec is in use [5.10,13/86] dmaengine: imx-sdma: Fix error checking in sdma_event_remap [5.10,14/86] dmaengine: mediatek:Fix PM usage reference leak of mtk_uart_apdma_alloc_chan_resources [5.10,15/86] spi: spi-mtk-nor: initialize spi controller after resume [5.10,16/86] esp: limit skb_page_frag_refill use to a single page [5.10,17/86] igc: Fix infinite loop in release_swfw_sync [5.10,18/86] igc: Fix BUG: scheduling while atomic [5.10,19/86] rxrpc: Restore removed timer deletion [5.10,20/86] net/smc: Fix sock leak when release after smc_shutdown() [5.10,21/86] net/packet: fix packet_sock xmit return value checking [5.10,22/86] ip6_gre: Avoid updating tunnel->tun_hlen in __gre6_xmit() [5.10,23/86] ip6_gre: Fix skb_under_panic in __gre6_xmit() [5.10,24/86] net/sched: cls_u32: fix possible leak in u32_init_knode() [5.10,25/86] l3mdev: l3mdev_master_upper_ifindex_by_index_rcu should be using netdev_master_upper... [5.10,26/86] ipv6: make ip6_rt_gc_expire an atomic_t [5.10,27/86] netlink: reset network and mac headers in netlink_dump() [5.10,28/86] net: stmmac: Use readl_poll_timeout_atomic() in atomic state [5.10,29/86] dmaengine: idxd: add RO check for wq max_batch_size write [5.10,30/86] dmaengine: idxd: add RO check for wq max_transfer_size write [5.10,31/86] selftests: mlxsw: vxlan_flooding: Prevent flooding of unwanted packets [5.10,32/86] arm64/mm: Remove [PUD\|PMD]_TABLE_BIT from [pud\|pmd]_bad() [5.10,33/86] arm64: mm: fix p?d_leaf() [5.10,34/86] ARM: vexpress/spc: Avoid negative array index when !SMP [5.10,35/86] reset: tegra-bpmp: Restore Handle errors in BPMP response [5.10,36/86] platform/x86: samsung-laptop: Fix an unsigned comparison which can never be negative [5.10,37/86] ALSA: usb-audio: Fix undefined behavior due to shift overflowing the constant [5.10,38/86] arm64: dts: imx: Fix imx8*-var-som touchscreen property sizes [5.10,39/86] vxlan: fix error return code in vxlan_fdb_append [5.10,40/86] cifs: Check the IOCB_DIRECT flag, not O_DIRECT [5.10,41/86] net: atlantic: Avoid out-of-bounds indexing [5.10,42/86] mt76: Fix undefined behavior due to shift overflowing the constant [5.10,43/86] brcmfmac: sdio: Fix undefined behavior due to shift overflowing the constant [5.10,44/86] dpaa_eth: Fix missing of_node_put in dpaa_get_ts_info() [5.10,45/86] drm/msm/mdp5: check the return of kzalloc() [5.10,46/86] net: macb: Restart tx only if queue pointer is lagging [5.10,47/86] scsi: qedi: Fix failed disconnect handling [5.10,48/86] stat: fix inconsistency between struct stat and struct compat_stat [5.10,49/86] nvme: add a quirk to disable namespace identifiers [5.10,50/86] nvme-pci: disable namespace identifiers for Qemu controllers [5.10,51/86] EDAC/synopsys: Read the error count from the correct register [5.10,52/86] mm, hugetlb: allow for "high" userspace addresses [5.10,53/86] oom_kill.c: futex: delay the OOM reaper to allow time for proper futex cleanup [5.10,54/86] mm/mmu_notifier.c: fix race in mmu_interval_notifier_remove() [5.10,55/86] ata: pata_marvell: Check the bmdma_addr beforing reading [5.10,56/86] dma: at_xdmac: fix a missing check on list iterator [5.10,57/86] net: atlantic: invert deep par in pm functions, preventing null derefs [5.10,58/86] xtensa: patch_text: Fixup last cpu should be master [5.10,59/86] xtensa: fix a7 clobbering in coprocessor context load/store [5.10,60/86] openvswitch: fix OOB access in reserve_sfa_size() [5.10,61/86] gpio: Request interrupts after IRQ is initialized [5.10,62/86] ASoC: soc-dapm: fix two incorrect uses of list iterator [5.10,63/86] e1000e: Fix possible overflow in LTR decoding [5.10,64/86] ARC: entry: fix syscall_trace_exit argument [5.10,65/86] arm_pmu: Validate single/group leader events [5.10,66/86] sched/pelt: Fix attach_entity_load_avg() corner case [5.10,67/86] perf/core: Fix perf_mmap fail when CONFIG_PERF_USE_VMALLOC enabled [5.10,68/86] drm/panel/raspberrypi-touchscreen: Avoid NULL deref if not initialised [5.10,69/86] drm/panel/raspberrypi-touchscreen: Initialise the bridge in prepare [5.10,70/86] KVM: PPC: Fix TCE handling for VFIO [5.10,71/86] drm/vc4: Use pm_runtime_resume_and_get to fix pm_runtime_get_sync() usage [5.10,72/86] powerpc/perf: Fix power9 event alternatives [5.10,73/86] perf report: Set PERF_SAMPLE_DATA_SRC bit for Arm SPE event [5.10,74/86] ext4: fix fallocate to use file_modified to update permissions consistently [5.10,75/86] ext4: fix symlink file size not match to file content [5.10,76/86] ext4: fix use-after-free in ext4_search_dir [5.10,77/86] ext4: limit length to bitmap_maxbytes - blocksize in punch_hole [5.10,78/86] ext4, doc: fix incorrect h_reserved size [5.10,79/86] ext4: fix overhead calculation to account for the reserved gdt blocks [5.10,80/86] ext4: force overhead calculation if the s_overhead_cluster makes no sense [5.10,81/86] can: isotp: stop timeout monitoring when no first frame was sent [5.10,82/86] jbd2: fix a potential race while discarding reserved buffers after an abort [5.10,83/86] spi: atmel-quadspi: Fix the buswidth adjustment between spi-mem and controller [5.10,84/86] staging: ion: Prevent incorrect reference counting behavour [5.10,85/86] block/compat_ioctl: fix range check in BLKGETSIZE [5.10,86/86] Revert "net: micrel: fix KS8851_MLL Kconfig"

[5.10,06/86] net/sched: cls_u32: fix netns refcount changes in u32_change()

Commit Message

Patch