From cbaaf2ee5c8898dec03f25795d0675110b667994 Mon Sep 17 00:00:00 2001 From: Frederick Lawler Date: Mon, 9 Jan 2023 10:39:06 -0600 Subject: [PATCH 1/3] net: sched: disallow noqueue for qdisc classes stable inclusion from stable-5.10.163 commit 9f7bc28a6b8afc2274e25650511555e93f45470f category: bugfix issue: I6BIHC CVE: CVE-2022-47929 Signed-off-by: gaochao --------------------------------------- commit 96398560f26aa07e8f2969d73c8197e6a6d10407 upstream. While experimenting with applying noqueue to a classful queue discipline, we discovered a NULL pointer dereference in the __dev_queue_xmit() path that generates a kernel OOPS: # dev=enp0s5 # tc qdisc replace dev $dev root handle 1: htb default 1 # tc class add dev $dev parent 1: classid 1:1 htb rate 10mbit # tc qdisc add dev $dev parent 1:1 handle 10: noqueue # ping -I $dev -w 1 -c 1 1.1.1.1 [ 2.172856] BUG: kernel NULL pointer dereference, address: 0000000000000000 [ 2.173217] #PF: supervisor instruction fetch in kernel mode ... [ 2.178451] Call Trace: [ 2.178577] [ 2.178686] htb_enqueue+0x1c8/0x370 [ 2.178880] dev_qdisc_enqueue+0x15/0x90 [ 2.179093] __dev_queue_xmit+0x798/0xd00 [ 2.179305] ? _raw_write_lock_bh+0xe/0x30 [ 2.179522] ? __local_bh_enable_ip+0x32/0x70 [ 2.179759] ? ___neigh_create+0x610/0x840 [ 2.179968] ? eth_header+0x21/0xc0 [ 2.180144] ip_finish_output2+0x15e/0x4f0 [ 2.180348] ? dst_output+0x30/0x30 [ 2.180525] ip_push_pending_frames+0x9d/0xb0 [ 2.180739] raw_sendmsg+0x601/0xcb0 [ 2.180916] ? _raw_spin_trylock+0xe/0x50 [ 2.181112] ? _raw_spin_unlock_irqrestore+0x16/0x30 [ 2.181354] ? get_page_from_freelist+0xcd6/0xdf0 [ 2.181594] ? sock_sendmsg+0x56/0x60 [ 2.181781] sock_sendmsg+0x56/0x60 [ 2.181958] __sys_sendto+0xf7/0x160 [ 2.182139] ? handle_mm_fault+0x6e/0x1d0 [ 2.182366] ? do_user_addr_fault+0x1e1/0x660 [ 2.182627] __x64_sys_sendto+0x1b/0x30 [ 2.182881] do_syscall_64+0x38/0x90 [ 2.183085] entry_SYSCALL_64_after_hwframe+0x63/0xcd ... [ 2.187402] Previously in commit d66d6c3152e8 ("net: sched: register noqueue qdisc"), NULL was set for the noqueue discipline on noqueue init so that __dev_queue_xmit() falls through for the noqueue case. This also sets a bypass of the enqueue NULL check in the register_qdisc() function for the struct noqueue_disc_ops. Classful queue disciplines make it past the NULL check in __dev_queue_xmit() because the discipline is set to htb (in this case), and then in the call to __dev_xmit_skb(), it calls into htb_enqueue() which grabs a leaf node for a class and then calls qdisc_enqueue() by passing in a queue discipline which assumes ->enqueue() is not set to NULL. Fix this by not allowing classes to be assigned to the noqueue discipline. Linux TC Notes states that classes cannot be set to the noqueue discipline. [1] Let's enforce that here. Links: 1. https://linux-tc-notes.sourceforge.net/tc/doc/sch_noqueue.txt Fixes: d66d6c3152e8 ("net: sched: register noqueue qdisc") Cc: stable@vger.kernel.org Signed-off-by: Frederick Lawler Reviewed-by: Jakub Sitnicki Link: https://lore.kernel.org/r/20230109163906.706000-1-fred@cloudflare.com Signed-off-by: Jakub Kicinski Signed-off-by: Greg Kroah-Hartman --- net/sched/sch_api.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 6e18aa417782..01bec5e746d3 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -1113,6 +1113,11 @@ static int qdisc_graft(struct net_device *dev, struct Qdisc *parent, return -ENOENT; } + if (new && new->ops == &noqueue_qdisc_ops) { + NL_SET_ERR_MSG(extack, "Cannot assign noqueue to a class"); + return -EINVAL; + } + err = cops->graft(parent, cl, new, &old, extack); if (err) return err; -- Gitee From dc57b66eedb0b929291d7b76d4589ca8a947c5e1 Mon Sep 17 00:00:00 2001 From: Zheyu Ma Date: Fri, 29 Jul 2022 19:02:16 +0800 Subject: [PATCH 2/3] i2c: ismt: Fix an out-of-bounds bug in ismt_access() mainline inclusion from mainline-6.2-rc1 commit 39244cc754829bf707dccd12e2ce37510f5b1f8d category: bugfix issue: I5O7R3 CVE: CVE-2022-2873 Signed-off-by: gaochao --------------------------------------- When the driver does not check the data from the user, the variable 'data->block[0]' may be very large to cause an out-of-bounds bug. The following log can reveal it: [ 33.995542] i2c i2c-1: ioctl, cmd=0x720, arg=0x7ffcb3dc3a20 [ 33.995978] ismt_smbus 0000:00:05.0: I2C_SMBUS_BLOCK_DATA: WRITE [ 33.996475] ================================================================== [ 33.996995] BUG: KASAN: out-of-bounds in ismt_access.cold+0x374/0x214b [ 33.997473] Read of size 18446744073709551615 at addr ffff88810efcfdb1 by task ismt_poc/485 [ 33.999450] Call Trace: [ 34.001849] memcpy+0x20/0x60 [ 34.002077] ismt_access.cold+0x374/0x214b [ 34.003382] __i2c_smbus_xfer+0x44f/0xfb0 [ 34.004007] i2c_smbus_xfer+0x10a/0x390 [ 34.004291] i2cdev_ioctl_smbus+0x2c8/0x710 [ 34.005196] i2cdev_ioctl+0x5ec/0x74c Fix this bug by checking the size of 'data->block[0]' first. Fixes: 13f35ac14cd0 ("i2c: Adding support for Intel iSMT SMBus 2.0 host controller") Signed-off-by: Zheyu Ma Signed-off-by: Wolfram Sang --- drivers/i2c/busses/i2c-ismt.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/i2c/busses/i2c-ismt.c b/drivers/i2c/busses/i2c-ismt.c index 3d2d92640651..cec2b2ae7684 100644 --- a/drivers/i2c/busses/i2c-ismt.c +++ b/drivers/i2c/busses/i2c-ismt.c @@ -507,6 +507,9 @@ static int ismt_access(struct i2c_adapter *adap, u16 addr, if (read_write == I2C_SMBUS_WRITE) { /* Block Write */ dev_dbg(dev, "I2C_SMBUS_BLOCK_DATA: WRITE\n"); + if (data->block[0] < 1 || data->block[0] > I2C_SMBUS_BLOCK_MAX) + return -EINVAL; + dma_size = data->block[0] + 1; dma_direction = DMA_TO_DEVICE; desc->wr_len_cmd = dma_size; -- Gitee From 70aa23b1a958f72b4aaac7261b0cdd1faa3ccac7 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Tue, 10 Jan 2023 08:59:06 +0800 Subject: [PATCH 3/3] ipv6: raw: Deduct extension header length in rawv6_push_pending_frames stable inclusion from stable-5.10.164 commit 6c9e2c11c33c35563d34d12b343d43b5c12200b5 category: bugfix issue: I6BQ51 CVE: CVE-2023-0394 Signed-off-by: gaochao --------------------------------------- commit cb3e9864cdbe35ff6378966660edbcbac955fe17 upstream. The total cork length created by ip6_append_data includes extension headers, so we must exclude them when comparing them against the IPV6_CHECKSUM offset which does not include extension headers. Reported-by: Kyle Zeng Fixes: 357b40a18b04 ("[IPV6]: IPV6_CHECKSUM socket option can corrupt kernel memory") Signed-off-by: Herbert Xu Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- net/ipv6/raw.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 31eb54e92b3f..110254f44a46 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -539,6 +539,7 @@ static int rawv6_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, static int rawv6_push_pending_frames(struct sock *sk, struct flowi6 *fl6, struct raw6_sock *rp) { + struct ipv6_txoptions *opt; struct sk_buff *skb; int err = 0; int offset; @@ -556,6 +557,9 @@ static int rawv6_push_pending_frames(struct sock *sk, struct flowi6 *fl6, offset = rp->offset; total_len = inet_sk(sk)->cork.base.length; + opt = inet6_sk(sk)->cork.opt; + total_len -= opt ? opt->opt_flen : 0; + if (offset >= total_len - 1) { err = -EINVAL; ip6_flush_pending_frames(sk); -- Gitee