From: Mike Pattrick mkp@redhat.com

[ Upstream commit c21ab2afa2c64896a7f0e3cbc6845ec63dcfad2e ]

Currently queue_userspace_packet will call kfree_skb for all frames, whether or not an error occurred. This can result in a single dropped frame being reported as multiple drops in dropwatch. This functions caller may also call kfree_skb in case of an error. This patch will consume the skbs instead and allow caller's to use kfree_skb.

Signed-off-by: Mike Pattrick mkp@redhat.com Link: https://bugzilla.redhat.com/show_bug.cgi?id=2109957 Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- net/openvswitch/datapath.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index 42616aca0ce4..8319628ab428 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -553,8 +553,9 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, out: if (err) skb_tx_error(skb); - kfree_skb(user_skb); - kfree_skb(nskb); + consume_skb(user_skb); + consume_skb(nskb); + return err; }

From: Tetsuo Handa penguin-kernel@I-love.SAKURA.ne.jp

[ Upstream commit b383e8abed41cc6ff1a3b34de75df9397fa4878c ]

syzbot is reporting uninit value at ath9k_htc_rx_msg() [1], for ioctl(USB_RAW_IOCTL_EP_WRITE) can call ath9k_hif_usb_rx_stream() with pkt_len = 0 but ath9k_hif_usb_rx_stream() uses __dev_alloc_skb(pkt_len + 32, GFP_ATOMIC) based on an assumption that pkt_len is valid. As a result, ath9k_hif_usb_rx_stream() allocates skb with uninitialized memory and ath9k_htc_rx_msg() is reading from uninitialized memory.

Since bytes accessed by ath9k_htc_rx_msg() is not known until ath9k_htc_rx_msg() is called, it would be difficult to check minimal valid pkt_len at "if (pkt_len > 2 * MAX_RX_BUF_SIZE) {" line in ath9k_hif_usb_rx_stream().

We have two choices. One is to workaround by adding __GFP_ZERO so that ath9k_htc_rx_msg() sees 0 if pkt_len is invalid. The other is to let ath9k_htc_rx_msg() validate pkt_len before accessing. This patch chose the latter.

Note that I'm not sure threshold condition is correct, for I can't find details on possible packet length used by this protocol.

Link: https://syzkaller.appspot.com/bug?extid=2ca247c2d60c7023de7f [1] Reported-by: syzbot syzbot+2ca247c2d60c7023de7f@syzkaller.appspotmail.com Signed-off-by: Tetsuo Handa penguin-kernel@I-love.SAKURA.ne.jp Acked-by: Toke Høiland-Jørgensen toke@toke.dk Signed-off-by: Kalle Valo quic_kvalo@quicinc.com Link: https://lore.kernel.org/r/7acfa1be-4b5c-b2ce-de43-95b0593fb3e5@I-love.SAKURA... Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/wireless/ath/ath9k/htc_hst.c | 43 +++++++++++++++--------- 1 file changed, 28 insertions(+), 15 deletions(-)

diff --git a/drivers/net/wireless/ath/ath9k/htc_hst.c b/drivers/net/wireless/ath/ath9k/htc_hst.c index e37de14bc502..6d69cf69fd86 100644 --- a/drivers/net/wireless/ath/ath9k/htc_hst.c +++ b/drivers/net/wireless/ath/ath9k/htc_hst.c @@ -367,33 +367,27 @@ void ath9k_htc_txcompletion_cb(struct htc_target *htc_handle, }

static void ath9k_htc_fw_panic_report(struct htc_target *htc_handle, - struct sk_buff *skb) + struct sk_buff *skb, u32 len) { uint32_t *pattern = (uint32_t *)skb->data;

- switch (*pattern) { - case 0x33221199: - { + if (*pattern == 0x33221199 && len >= sizeof(struct htc_panic_bad_vaddr)) { struct htc_panic_bad_vaddr *htc_panic; htc_panic = (struct htc_panic_bad_vaddr *) skb->data; dev_err(htc_handle->dev, "ath: firmware panic! " "exccause: 0x%08x; pc: 0x%08x; badvaddr: 0x%08x.\n", htc_panic->exccause, htc_panic->pc, htc_panic->badvaddr); - break; - } - case 0x33221299: - { + return; + } + if (*pattern == 0x33221299) { struct htc_panic_bad_epid *htc_panic; htc_panic = (struct htc_panic_bad_epid *) skb->data; dev_err(htc_handle->dev, "ath: firmware panic! " "bad epid: 0x%08x\n", htc_panic->epid); - break; - } - default: - dev_err(htc_handle->dev, "ath: unknown panic pattern!\n"); - break; + return; } + dev_err(htc_handle->dev, "ath: unknown panic pattern!\n"); }

/* @@ -414,16 +408,26 @@ void ath9k_htc_rx_msg(struct htc_target *htc_handle, if (!htc_handle || !skb) return;

+ /* A valid message requires len >= 8. + * + * sizeof(struct htc_frame_hdr) == 8 + * sizeof(struct htc_ready_msg) == 8 + * sizeof(struct htc_panic_bad_vaddr) == 16 + * sizeof(struct htc_panic_bad_epid) == 8 + */ + if (unlikely(len < sizeof(struct htc_frame_hdr))) + goto invalid; htc_hdr = (struct htc_frame_hdr *) skb->data; epid = htc_hdr->endpoint_id;

if (epid == 0x99) { - ath9k_htc_fw_panic_report(htc_handle, skb); + ath9k_htc_fw_panic_report(htc_handle, skb, len); kfree_skb(skb); return; }

if (epid < 0 || epid >= ENDPOINT_MAX) { +invalid: if (pipe_id != USB_REG_IN_PIPE) dev_kfree_skb_any(skb); else @@ -435,21 +439,30 @@ void ath9k_htc_rx_msg(struct htc_target *htc_handle,

/* Handle trailer */ if (htc_hdr->flags & HTC_FLAGS_RECV_TRAILER) { - if (be32_to_cpu(*(__be32 *) skb->data) == 0x00C60000) + if (be32_to_cpu(*(__be32 *) skb->data) == 0x00C60000) { /* Move past the Watchdog pattern */ htc_hdr = (struct htc_frame_hdr *)(skb->data + 4); + len -= 4; + } }

/* Get the message ID */ + if (unlikely(len < sizeof(struct htc_frame_hdr) + sizeof(__be16))) + goto invalid; msg_id = (__be16 *) ((void *) htc_hdr + sizeof(struct htc_frame_hdr));

/* Now process HTC messages */ switch (be16_to_cpu(*msg_id)) { case HTC_MSG_READY_ID: + if (unlikely(len < sizeof(struct htc_ready_msg))) + goto invalid; htc_process_target_rdy(htc_handle, htc_hdr); break; case HTC_MSG_CONNECT_SERVICE_RESPONSE_ID: + if (unlikely(len < sizeof(struct htc_frame_hdr) + + sizeof(struct htc_conn_svc_rspmsg))) + goto invalid; htc_process_conn_rsp(htc_handle, htc_hdr); break; default:

From: GUO Zihua guozihua@huawei.com

[ Upstream commit 0dbaf0fa62329d9fe452d9041a707a33f6274f1f ]

Since Linux now supports CFI, it will be a good idea to fix mismatched return type for implementation of hooks. Otherwise this might get cought out by CFI and cause a panic.

eth_xmit() would return either NETDEV_TX_BUSY or NETDEV_TX_OK, so change the return type to netdev_tx_t directly.

Signed-off-by: GUO Zihua guozihua@huawei.com Link: https://lore.kernel.org/r/20220902081612.60405-1-guozihua@huawei.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/xscale/ixp4xx_eth.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/xscale/ixp4xx_eth.c b/drivers/net/ethernet/xscale/ixp4xx_eth.c index aee55c03def0..d2aed667c0da 100644 --- a/drivers/net/ethernet/xscale/ixp4xx_eth.c +++ b/drivers/net/ethernet/xscale/ixp4xx_eth.c @@ -835,7 +835,7 @@ static void eth_txdone_irq(void *unused) } }

-static int eth_xmit(struct sk_buff *skb, struct net_device *dev) +static netdev_tx_t eth_xmit(struct sk_buff *skb, struct net_device *dev) { struct port *port = netdev_priv(dev); unsigned int txreadyq = port->plat->txreadyq;

From: Sergei Antonov saproj@gmail.com

[ Upstream commit 9df696b3b3a4c96c3219eb87c7bf03fb50e490b8 ]

Sparse found a number of endianness-related issues of these kinds:

.../ftmac100.c:192:32: warning: restricted __le32 degrades to integer

.../ftmac100.c:208:23: warning: incorrect type in assignment (different base types) .../ftmac100.c:208:23: expected unsigned int rxdes0 .../ftmac100.c:208:23: got restricted __le32 [usertype]

.../ftmac100.c:249:23: warning: invalid assignment: &= .../ftmac100.c:249:23: left side has type unsigned int .../ftmac100.c:249:23: right side has type restricted __le32

.../ftmac100.c:527:16: warning: cast to restricted __le32

Change type of some fields from 'unsigned int' to '__le32' to fix it.

Signed-off-by: Sergei Antonov saproj@gmail.com Reviewed-by: Andrew Lunn andrew@lunn.ch Link: https://lore.kernel.org/r/20220902113749.1408562-1-saproj@gmail.com Signed-off-by: Paolo Abeni pabeni@redhat.com Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/faraday/ftmac100.h | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/faraday/ftmac100.h b/drivers/net/ethernet/faraday/ftmac100.h index 46a0c47b1ee1..0731d65e856c 100644 --- a/drivers/net/ethernet/faraday/ftmac100.h +++ b/drivers/net/ethernet/faraday/ftmac100.h @@ -135,9 +135,9 @@ * Transmit descriptor, aligned to 16 bytes */ struct ftmac100_txdes { - unsigned int txdes0; - unsigned int txdes1; - unsigned int txdes2; /* TXBUF_BADR */ + __le32 txdes0; + __le32 txdes1; + __le32 txdes2; /* TXBUF_BADR */ unsigned int txdes3; /* not used by HW */ } __attribute__ ((aligned(16)));

@@ -156,9 +156,9 @@ struct ftmac100_txdes { * Receive descriptor, aligned to 16 bytes */ struct ftmac100_rxdes { - unsigned int rxdes0; - unsigned int rxdes1; - unsigned int rxdes2; /* RXBUF_BADR */ + __le32 rxdes0; + __le32 rxdes1; + __le32 rxdes2; /* RXBUF_BADR */ unsigned int rxdes3; /* not used by HW */ } __attribute__ ((aligned(16)));

From: Tetsuo Handa penguin-kernel@I-love.SAKURA.ne.jp

[ Upstream commit 2d2cb3066f2c90cd8ca540b36ba7a55e7f2406e0 ]

syzbot is reporting cancel_delayed_work() without INIT_DELAYED_WORK() at l2cap_chan_del() [1], for CONF_NOT_COMPLETE flag (which meant to prevent l2cap_chan_del() from calling cancel_delayed_work()) is cleared by timer which fires before l2cap_chan_del() is called by closing file descriptor created by socket(AF_BLUETOOTH, SOCK_STREAM, BTPROTO_L2CAP).

l2cap_bredr_sig_cmd(L2CAP_CONF_REQ) and l2cap_bredr_sig_cmd(L2CAP_CONF_RSP) are calling l2cap_ertm_init(chan), and they call l2cap_chan_ready() (which clears CONF_NOT_COMPLETE flag) only when l2cap_ertm_init(chan) succeeded.

l2cap_sock_init() does not call l2cap_ertm_init(chan), and it instead sets CONF_NOT_COMPLETE flag by calling l2cap_chan_set_defaults(). However, when connect() is requested, "command 0x0409 tx timeout" happens after 2 seconds from connect() request, and CONF_NOT_COMPLETE flag is cleared after 4 seconds from connect() request, for l2cap_conn_start() from l2cap_info_timeout() callback scheduled by

schedule_delayed_work(&conn->info_timer, L2CAP_INFO_TIMEOUT);

in l2cap_connect() is calling l2cap_chan_ready().

Fix this problem by initializing delayed works used by L2CAP_MODE_ERTM mode as soon as l2cap_chan_create() allocates a channel, like I did in commit be8597239379f0f5 ("Bluetooth: initialize skb_queue_head at l2cap_chan_create()").

Link: https://syzkaller.appspot.com/bug?extid=83672956c7aa6af698b3 [1] Reported-by: syzbot syzbot+83672956c7aa6af698b3@syzkaller.appspotmail.com Signed-off-by: Tetsuo Handa penguin-kernel@I-love.SAKURA.ne.jp Signed-off-by: Luiz Augusto von Dentz luiz.von.dentz@intel.com Signed-off-by: Sasha Levin sashal@kernel.org --- net/bluetooth/l2cap_core.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index e45a12378bd1..75a76353525c 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -63,6 +63,9 @@ static void l2cap_send_disconn_req(struct l2cap_chan *chan, int err);

static void l2cap_tx(struct l2cap_chan *chan, struct l2cap_ctrl *control, struct sk_buff_head *skbs, u8 event); +static void l2cap_retrans_timeout(struct work_struct *work); +static void l2cap_monitor_timeout(struct work_struct *work); +static void l2cap_ack_timeout(struct work_struct *work);

static inline u8 bdaddr_type(u8 link_type, u8 bdaddr_type) { @@ -470,6 +473,9 @@ struct l2cap_chan *l2cap_chan_create(void) write_unlock(&chan_list_lock);

INIT_DELAYED_WORK(&chan->chan_timer, l2cap_chan_timeout); + INIT_DELAYED_WORK(&chan->retrans_timer, l2cap_retrans_timeout); + INIT_DELAYED_WORK(&chan->monitor_timer, l2cap_monitor_timeout); + INIT_DELAYED_WORK(&chan->ack_timer, l2cap_ack_timeout);

chan->state = BT_OPEN;

@@ -3154,10 +3160,6 @@ int l2cap_ertm_init(struct l2cap_chan *chan) chan->rx_state = L2CAP_RX_STATE_RECV; chan->tx_state = L2CAP_TX_STATE_XMIT;

- INIT_DELAYED_WORK(&chan->retrans_timer, l2cap_retrans_timeout); - INIT_DELAYED_WORK(&chan->monitor_timer, l2cap_monitor_timeout); - INIT_DELAYED_WORK(&chan->ack_timer, l2cap_ack_timeout); - skb_queue_head_init(&chan->srej_q);

err = l2cap_seq_list_init(&chan->srej_list, chan->tx_win);

From: Nathan Huckleberry nhuck@google.com

[ Upstream commit 5972ca946098487c5155fe13654743f9010f5ed5 ]

The ndo_start_xmit field in net_device_ops is expected to be of type netdev_tx_t (*ndo_start_xmit)(struct sk_buff *skb, struct net_device *dev).

The mismatched return type breaks forward edge kCFI since the underlying function definition does not match the function hook definition.

The return type of emac_dev_xmit should be changed from int to netdev_tx_t.

Reported-by: Dan Carpenter error27@gmail.com Link: https://github.com/ClangBuiltLinux/linux/issues/1703 Cc: llvm@lists.linux.dev Signed-off-by: Nathan Huckleberry nhuck@google.com Reviewed-by: Nathan Chancellor nathan@kernel.org Link: https://lore.kernel.org/r/20220912195023.810319-1-nhuck@google.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/ti/davinci_emac.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/ti/davinci_emac.c b/drivers/net/ethernet/ti/davinci_emac.c index da536385075a..f226e8f84835 100644 --- a/drivers/net/ethernet/ti/davinci_emac.c +++ b/drivers/net/ethernet/ti/davinci_emac.c @@ -955,7 +955,7 @@ static void emac_tx_handler(void *token, int len, int status) * * Returns success(NETDEV_TX_OK) or error code (typically out of desc's) */ -static int emac_dev_xmit(struct sk_buff *skb, struct net_device *ndev) +static netdev_tx_t emac_dev_xmit(struct sk_buff *skb, struct net_device *ndev) { struct device *emac_dev = &ndev->dev; int ret_code;

From: Luiz Augusto von Dentz luiz.von.dentz@intel.com

[ Upstream commit 448a496f760664d3e2e79466aa1787e6abc922b5 ]

device_add shall not be called multiple times as stated in its documentation:

'Do not call this routine or device_register() more than once for any device structure'

Syzkaller reports a bug as follows [1]: ------------[ cut here ]------------ kernel BUG at lib/list_debug.c:33! invalid opcode: 0000 [#1] PREEMPT SMP KASAN [...] Call Trace: <TASK> __list_add include/linux/list.h:69 [inline] list_add_tail include/linux/list.h:102 [inline] kobj_kset_join lib/kobject.c:164 [inline] kobject_add_internal+0x18f/0x8f0 lib/kobject.c:214 kobject_add_varg lib/kobject.c:358 [inline] kobject_add+0x150/0x1c0 lib/kobject.c:410 device_add+0x368/0x1e90 drivers/base/core.c:3452 hci_conn_add_sysfs+0x9b/0x1b0 net/bluetooth/hci_sysfs.c:53 hci_le_cis_estabilished_evt+0x57c/0xae0 net/bluetooth/hci_event.c:6799 hci_le_meta_evt+0x2b8/0x510 net/bluetooth/hci_event.c:7110 hci_event_func net/bluetooth/hci_event.c:7440 [inline] hci_event_packet+0x63d/0xfd0 net/bluetooth/hci_event.c:7495 hci_rx_work+0xae7/0x1230 net/bluetooth/hci_core.c:4007 process_one_work+0x991/0x1610 kernel/workqueue.c:2289 worker_thread+0x665/0x1080 kernel/workqueue.c:2436 kthread+0x2e4/0x3a0 kernel/kthread.c:376 ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:306 </TASK>

Link: https://syzkaller.appspot.com/bug?id=da3246e2d33afdb92d66bc166a0934c5b146404... Signed-off-by: Luiz Augusto von Dentz luiz.von.dentz@intel.com Tested-by: Hawkins Jiawei yin31149@gmail.com Signed-off-by: Sasha Levin sashal@kernel.org --- net/bluetooth/hci_sysfs.c | 3 +++ 1 file changed, 3 insertions(+)

diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c index b568f7c21b30..0c9249339790 100644 --- a/net/bluetooth/hci_sysfs.c +++ b/net/bluetooth/hci_sysfs.c @@ -48,6 +48,9 @@ void hci_conn_add_sysfs(struct hci_conn *conn)

BT_DBG("conn %p", conn);

+ if (device_is_registered(&conn->dev)) + return; + dev_set_name(&conn->dev, "%s:%d", hdev->name, conn->handle);

if (device_add(&conn->dev) < 0) {

From: Daniel Golle daniel@makrotopia.org

[ Upstream commit d3aad83d05aec0cfd7670cf0028f2ad4b81de92e ]

The function rt2800_iq_calibrate is intended for Rt5592 only. Don't call it for MT7620 which has it's own calibration functions.

Reported-by: Serge Vasilugin vasilugin@yandex.ru Signed-off-by: Daniel Golle daniel@makrotopia.org Signed-off-by: Kalle Valo kvalo@kernel.org Link: https://lore.kernel.org/r/31a1c34ddbd296b82f38c18c9ae7339059215fdc.166344515... Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/wireless/ralink/rt2x00/rt2800lib.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ralink/rt2x00/rt2800lib.c b/drivers/net/wireless/ralink/rt2x00/rt2800lib.c index d2c289446c00..1309c136f7f3 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2800lib.c +++ b/drivers/net/wireless/ralink/rt2x00/rt2800lib.c @@ -3835,7 +3835,8 @@ static void rt2800_config_channel(struct rt2x00_dev *rt2x00dev, reg += 2 * rt2x00dev->lna_gain; rt2800_bbp_write_with_rx_chain(rt2x00dev, 66, reg);

- rt2800_iq_calibrate(rt2x00dev, rf->channel); + if (rt2x00_rt(rt2x00dev, RT5592)) + rt2800_iq_calibrate(rt2x00dev, rf->channel); }

bbp = rt2800_bbp_read(rt2x00dev, 4);

From: Daniel Golle daniel@makrotopia.org

[ Upstream commit cbde6ed406a51092d9e8a2df058f5f8490f27443 ]

Instead of using the default value 33 (pci), set US_CYC_CNT init based on Programming guide: If available, set chipset bus clock with fallback to cpu clock/3.

Reported-by: Serge Vasilugin vasilugin@yandex.ru Signed-off-by: Daniel Golle daniel@makrotopia.org Acked-by: Stanislaw Gruszka stf_xl@wp.pl Signed-off-by: Kalle Valo kvalo@kernel.org Link: https://lore.kernel.org/r/3e275d259f476f597dab91a9c395015ef3fe3284.166344515... Signed-off-by: Sasha Levin sashal@kernel.org --- .../net/wireless/ralink/rt2x00/rt2800lib.c | 21 +++++++++++++++++++ 1 file changed, 21 insertions(+)

diff --git a/drivers/net/wireless/ralink/rt2x00/rt2800lib.c b/drivers/net/wireless/ralink/rt2x00/rt2800lib.c index 0c90bf0540b9..57fa472b5c4e 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2800lib.c +++ b/drivers/net/wireless/ralink/rt2x00/rt2800lib.c @@ -5567,6 +5567,27 @@ static int rt2800_init_registers(struct rt2x00_dev *rt2x00dev) reg = rt2800_register_read(rt2x00dev, US_CYC_CNT); rt2x00_set_field32(&reg, US_CYC_CNT_CLOCK_CYCLE, 125); rt2800_register_write(rt2x00dev, US_CYC_CNT, reg); + } else if (rt2x00_is_soc(rt2x00dev)) { + struct clk *clk = clk_get_sys("bus", NULL); + int rate; + + if (IS_ERR(clk)) { + clk = clk_get_sys("cpu", NULL); + + if (IS_ERR(clk)) { + rate = 125; + } else { + rate = clk_get_rate(clk) / 3000000; + clk_put(clk); + } + } else { + rate = clk_get_rate(clk) / 1000000; + clk_put(clk); + } + + reg = rt2800_register_read(rt2x00dev, US_CYC_CNT); + rt2x00_set_field32(&reg, US_CYC_CNT_CLOCK_CYCLE, rate); + rt2800_register_write(rt2x00dev, US_CYC_CNT, reg); }

reg = rt2800_register_read(rt2x00dev, HT_FBK_CFG0);

From: Liu Jian liujian56@huawei.com

[ Upstream commit 3f8ef65af927db247418d4e1db49164d7a158fc5 ]

Fixes the below NULL pointer dereference:

[...] [ 14.471200] Call Trace: [ 14.471562] <TASK> [ 14.471882] lock_acquire+0x245/0x2e0 [ 14.472416] ? remove_wait_queue+0x12/0x50 [ 14.473014] ? _raw_spin_lock_irqsave+0x17/0x50 [ 14.473681] _raw_spin_lock_irqsave+0x3d/0x50 [ 14.474318] ? remove_wait_queue+0x12/0x50 [ 14.474907] remove_wait_queue+0x12/0x50 [ 14.475480] sk_stream_wait_memory+0x20d/0x340 [ 14.476127] ? do_wait_intr_irq+0x80/0x80 [ 14.476704] do_tcp_sendpages+0x287/0x600 [ 14.477283] tcp_bpf_push+0xab/0x260 [ 14.477817] tcp_bpf_sendmsg_redir+0x297/0x500 [ 14.478461] ? __local_bh_enable_ip+0x77/0xe0 [ 14.479096] tcp_bpf_send_verdict+0x105/0x470 [ 14.479729] tcp_bpf_sendmsg+0x318/0x4f0 [ 14.480311] sock_sendmsg+0x2d/0x40 [ 14.480822] ____sys_sendmsg+0x1b4/0x1c0 [ 14.481390] ? copy_msghdr_from_user+0x62/0x80 [ 14.482048] ___sys_sendmsg+0x78/0xb0 [ 14.482580] ? vmf_insert_pfn_prot+0x91/0x150 [ 14.483215] ? __do_fault+0x2a/0x1a0 [ 14.483738] ? do_fault+0x15e/0x5d0 [ 14.484246] ? __handle_mm_fault+0x56b/0x1040 [ 14.484874] ? lock_is_held_type+0xdf/0x130 [ 14.485474] ? find_held_lock+0x2d/0x90 [ 14.486046] ? __sys_sendmsg+0x41/0x70 [ 14.486587] __sys_sendmsg+0x41/0x70 [ 14.487105] ? intel_pmu_drain_pebs_core+0x350/0x350 [ 14.487822] do_syscall_64+0x34/0x80 [ 14.488345] entry_SYSCALL_64_after_hwframe+0x63/0xcd [...]

The test scenario has the following flow:

thread1 thread2 ----------- --------------- tcp_bpf_sendmsg tcp_bpf_send_verdict tcp_bpf_sendmsg_redir sock_close tcp_bpf_push_locked __sock_release tcp_bpf_push //inet_release do_tcp_sendpages sock->ops->release sk_stream_wait_memory // tcp_close sk_wait_event sk->sk_prot->close release_sock(__sk); *** lock_sock(sk); __tcp_close sock_orphan(sk) sk->sk_wq = NULL release_sock **** lock_sock(__sk); remove_wait_queue(sk_sleep(sk), &wait); sk_sleep(sk) //NULL pointer dereference &rcu_dereference_raw(sk->sk_wq)->wait

While waiting for memory in thread1, the socket is released with its wait queue because thread2 has closed it. This caused by tcp_bpf_send_verdict didn't increase the f_count of psock->sk_redir->sk_socket->file in thread1.

We should check if SOCK_DEAD flag is set on wakeup in sk_stream_wait_memory before accessing the wait queue.

Suggested-by: Jakub Sitnicki jakub@cloudflare.com Signed-off-by: Liu Jian liujian56@huawei.com Signed-off-by: Daniel Borkmann daniel@iogearbox.net Acked-by: John Fastabend john.fastabend@gmail.com Cc: Eric Dumazet edumazet@google.com Link: https://lore.kernel.org/bpf/20220823133755.314697-2-liujian56@huawei.com Signed-off-by: Sasha Levin sashal@kernel.org --- net/core/stream.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/core/stream.c b/net/core/stream.c index cbe52b169070..e5c6c9e5e0aa 100644 --- a/net/core/stream.c +++ b/net/core/stream.c @@ -159,7 +159,8 @@ int sk_stream_wait_memory(struct sock *sk, long *timeo_p) *timeo_p = current_timeo; } out: - remove_wait_queue(sk_sleep(sk), &wait); + if (!sock_flag(sk, SOCK_DEAD)) + remove_wait_queue(sk_sleep(sk), &wait); return err;

do_error:

From: Andrew Gaul gaul@gaul.org

[ Upstream commit 93e2be344a7db169b7119de21ac1bf253b8c6907 ]

My system shows almost 10 million of these messages over a 24-hour period which pollutes my logs.

Signed-off-by: Andrew Gaul gaul@google.com Link: https://lore.kernel.org/r/20221002034128.2026653-1-gaul@google.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/usb/r8152.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index a5a4fef09b93..1ed358d0da84 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -1382,7 +1382,9 @@ static void intr_callback(struct urb *urb) "Stop submitting intr, status %d\n", status); return; case -EOVERFLOW: - netif_info(tp, intr, tp->netdev, "intr status -EOVERFLOW\n"); + if (net_ratelimit()) + netif_info(tp, intr, tp->netdev, + "intr status -EOVERFLOW\n"); goto resubmit; /* -EPIPE: should clear the halt */ default: