From: Greg Kroah-Hartman Date: Thu, 28 Mar 2019 21:59:22 +0000 (+0100) Subject: 4.14-stable patches X-Git-Tag: v3.18.138~73 X-Git-Url: http://git.ipfire.org/?a=commitdiff_plain;h=cac80c347de881a1ae04e63f7e35d39f154d440c;p=thirdparty%2Fkernel%2Fstable-queue.git 4.14-stable patches added patches: dccp-do-not-use-ipv6-header-for-ipv4-flow.patch genetlink-fix-a-memory-leak-on-error-path.patch mac8390-fix-mmio-access-size-probe.patch misdn-hfcpci-test-both-vendor-device-id-for-digium-hfc4s.patch net-aquantia-fix-rx-checksum-offload-for-udp-tcp-over-ipv6.patch net-datagram-fix-unbounded-loop-in-__skb_try_recv_datagram.patch net-packet-set-__gfp_nowarn-upon-allocation-in-alloc_pg_vec.patch net-rose-fix-a-possible-stack-overflow.patch net-stmmac-fix-memory-corruption-with-large-mtus.patch net-sysfs-call-dev_hold-if-kobject_init_and_add-success.patch net-sysfs-fix-memory-leak-in-netdev_register_kobject.patch packets-always-register-packet-sk-in-the-same-order.patch rhashtable-still-do-rehash-when-we-get-eexist.patch sctp-get-sctphdr-by-offset-in-sctp_compute_cksum.patch tcp-do-not-use-ipv6-header-for-ipv4-flow.patch thunderx-eliminate-extra-calls-to-put_page-for-pages-held-for-recycling.patch thunderx-enable-page-recycling-for-non-xdp-case.patch tun-properly-test-for-iff_up.patch vxlan-don-t-call-gro_cells_destroy-before-device-is-unregistered.patch --- diff --git a/queue-4.14/dccp-do-not-use-ipv6-header-for-ipv4-flow.patch b/queue-4.14/dccp-do-not-use-ipv6-header-for-ipv4-flow.patch new file mode 100644 index 00000000000..754c6953174 --- /dev/null +++ b/queue-4.14/dccp-do-not-use-ipv6-header-for-ipv4-flow.patch @@ -0,0 +1,34 @@ +From foo@baz Thu Mar 28 21:57:57 CET 2019 +From: Eric Dumazet +Date: Tue, 19 Mar 2019 05:46:18 -0700 +Subject: dccp: do not use ipv6 header for ipv4 flow + +From: Eric Dumazet + +[ Upstream commit e0aa67709f89d08c8d8e5bdd9e0b649df61d0090 ] + +When a dual stack dccp listener accepts an ipv4 flow, +it should not attempt to use an ipv6 header or +inet6_iif() helper. + +Fixes: 3df80d9320bc ("[DCCP]: Introduce DCCPv6") +Signed-off-by: Eric Dumazet +Signed-off-by: David S. Miller +Signed-off-by: Greg Kroah-Hartman +--- + net/dccp/ipv6.c | 4 ++-- + 1 file changed, 2 insertions(+), 2 deletions(-) + +--- a/net/dccp/ipv6.c ++++ b/net/dccp/ipv6.c +@@ -433,8 +433,8 @@ static struct sock *dccp_v6_request_recv + newnp->ipv6_mc_list = NULL; + newnp->ipv6_ac_list = NULL; + newnp->ipv6_fl_list = NULL; +- newnp->mcast_oif = inet6_iif(skb); +- newnp->mcast_hops = ipv6_hdr(skb)->hop_limit; ++ newnp->mcast_oif = inet_iif(skb); ++ newnp->mcast_hops = ip_hdr(skb)->ttl; + + /* + * No need to charge this sock to the relevant IPv6 refcnt debug socks count diff --git a/queue-4.14/genetlink-fix-a-memory-leak-on-error-path.patch b/queue-4.14/genetlink-fix-a-memory-leak-on-error-path.patch new file mode 100644 index 00000000000..08cd7a5c7c6 --- /dev/null +++ b/queue-4.14/genetlink-fix-a-memory-leak-on-error-path.patch @@ -0,0 +1,42 @@ +From foo@baz Thu Mar 28 21:57:57 CET 2019 +From: YueHaibing +Date: Thu, 21 Mar 2019 15:02:50 +0800 +Subject: genetlink: Fix a memory leak on error path + +From: YueHaibing + +[ Upstream commit ceabee6c59943bdd5e1da1a6a20dc7ee5f8113a2 ] + +In genl_register_family(), when idr_alloc() fails, +we forget to free the memory we possibly allocate for +family->attrbuf. + +Reported-by: Hulk Robot +Fixes: 2ae0f17df1cd ("genetlink: use idr to track families") +Signed-off-by: YueHaibing +Reviewed-by: Kirill Tkhai +Signed-off-by: David S. Miller +Signed-off-by: Greg Kroah-Hartman +--- + net/netlink/genetlink.c | 3 ++- + 1 file changed, 2 insertions(+), 1 deletion(-) + +--- a/net/netlink/genetlink.c ++++ b/net/netlink/genetlink.c +@@ -365,7 +365,7 @@ int genl_register_family(struct genl_fam + start, end + 1, GFP_KERNEL); + if (family->id < 0) { + err = family->id; +- goto errout_locked; ++ goto errout_free; + } + + err = genl_validate_assign_mc_groups(family); +@@ -384,6 +384,7 @@ int genl_register_family(struct genl_fam + + errout_remove: + idr_remove(&genl_fam_idr, family->id); ++errout_free: + kfree(family->attrbuf); + errout_locked: + genl_unlock_all(); diff --git a/queue-4.14/mac8390-fix-mmio-access-size-probe.patch b/queue-4.14/mac8390-fix-mmio-access-size-probe.patch new file mode 100644 index 00000000000..d2a8e1e76a6 --- /dev/null +++ b/queue-4.14/mac8390-fix-mmio-access-size-probe.patch @@ -0,0 +1,74 @@ +From foo@baz Thu Mar 28 21:57:57 CET 2019 +From: Finn Thain +Date: Sat, 16 Mar 2019 14:21:19 +1100 +Subject: mac8390: Fix mmio access size probe + +From: Finn Thain + +[ Upstream commit bb9e5c5bcd76f4474eac3baf643d7a39f7bac7bb ] + +The bug that Stan reported is as follows. After a restart, a 16-bit NIC +may be incorrectly identified as a 32-bit NIC and stop working. + +mac8390 slot.E: Memory length resource not found, probing +mac8390 slot.E: Farallon EtherMac II-C (type farallon) +mac8390 slot.E: MAC 00:00:c5:30:c2:99, IRQ 61, 32 KB shared memory at 0xfeed0000, 32-bit access. + +The bug never arises after a cold start and only intermittently after a +warm start. (I didn't investigate why the bug is intermittent.) + +It turns out that memcpy_toio() is deprecated and memcmp_withio() also +has issues. Replacing these calls with mmio accessors fixes the problem. + +Reported-and-tested-by: Stan Johnson +Fixes: 2964db0f5904 ("m68k: Mac DP8390 update") +Signed-off-by: Finn Thain +Signed-off-by: David S. Miller +Signed-off-by: Greg Kroah-Hartman +--- + drivers/net/ethernet/8390/mac8390.c | 19 ++++++++++++------- + 1 file changed, 12 insertions(+), 7 deletions(-) + +--- a/drivers/net/ethernet/8390/mac8390.c ++++ b/drivers/net/ethernet/8390/mac8390.c +@@ -156,8 +156,6 @@ static void dayna_block_output(struct ne + #define memcpy_fromio(a, b, c) memcpy((a), (void *)(b), (c)) + #define memcpy_toio(a, b, c) memcpy((void *)(a), (b), (c)) + +-#define memcmp_withio(a, b, c) memcmp((a), (void *)(b), (c)) +- + /* Slow Sane (16-bit chunk memory read/write) Cabletron uses this */ + static void slow_sane_get_8390_hdr(struct net_device *dev, + struct e8390_pkt_hdr *hdr, int ring_page); +@@ -237,19 +235,26 @@ static enum mac8390_type __init mac8390_ + + static enum mac8390_access __init mac8390_testio(volatile unsigned long membase) + { +- unsigned long outdata = 0xA5A0B5B0; +- unsigned long indata = 0x00000000; ++ u32 outdata = 0xA5A0B5B0; ++ u32 indata = 0; ++ + /* Try writing 32 bits */ +- memcpy_toio(membase, &outdata, 4); +- /* Now compare them */ +- if (memcmp_withio(&outdata, membase, 4) == 0) ++ nubus_writel(outdata, membase); ++ /* Now read it back */ ++ indata = nubus_readl(membase); ++ if (outdata == indata) + return ACCESS_32; ++ ++ outdata = 0xC5C0D5D0; ++ indata = 0; ++ + /* Write 16 bit output */ + word_memcpy_tocard(membase, &outdata, 4); + /* Now read it back */ + word_memcpy_fromcard(&indata, membase, 4); + if (outdata == indata) + return ACCESS_16; ++ + return ACCESS_UNKNOWN; + } + diff --git a/queue-4.14/misdn-hfcpci-test-both-vendor-device-id-for-digium-hfc4s.patch b/queue-4.14/misdn-hfcpci-test-both-vendor-device-id-for-digium-hfc4s.patch new file mode 100644 index 00000000000..81c644be61a --- /dev/null +++ b/queue-4.14/misdn-hfcpci-test-both-vendor-device-id-for-digium-hfc4s.patch @@ -0,0 +1,36 @@ +From foo@baz Thu Mar 28 21:57:57 CET 2019 +From: Bjorn Helgaas +Date: Mon, 18 Mar 2019 08:51:06 -0500 +Subject: mISDN: hfcpci: Test both vendor & device ID for Digium HFC4S + +From: Bjorn Helgaas + +[ Upstream commit fae846e2b7124d4b076ef17791c73addf3b26350 ] + +The device ID alone does not uniquely identify a device. Test both the +vendor and device ID to make sure we don't mistakenly think some other +vendor's 0xB410 device is a Digium HFC4S. Also, instead of the bare hex +ID, use the same constant (PCI_DEVICE_ID_DIGIUM_HFC4S) used in the device +ID table. + +No functional change intended. + +Signed-off-by: Bjorn Helgaas +Signed-off-by: David S. Miller +Signed-off-by: Greg Kroah-Hartman +--- + drivers/isdn/hardware/mISDN/hfcmulti.c | 3 ++- + 1 file changed, 2 insertions(+), 1 deletion(-) + +--- a/drivers/isdn/hardware/mISDN/hfcmulti.c ++++ b/drivers/isdn/hardware/mISDN/hfcmulti.c +@@ -4367,7 +4367,8 @@ setup_pci(struct hfc_multi *hc, struct p + if (m->clock2) + test_and_set_bit(HFC_CHIP_CLOCK2, &hc->chip); + +- if (ent->device == 0xB410) { ++ if (ent->vendor == PCI_VENDOR_ID_DIGIUM && ++ ent->device == PCI_DEVICE_ID_DIGIUM_HFC4S) { + test_and_set_bit(HFC_CHIP_B410P, &hc->chip); + test_and_set_bit(HFC_CHIP_PCM_MASTER, &hc->chip); + test_and_clear_bit(HFC_CHIP_PCM_SLAVE, &hc->chip); diff --git a/queue-4.14/net-aquantia-fix-rx-checksum-offload-for-udp-tcp-over-ipv6.patch b/queue-4.14/net-aquantia-fix-rx-checksum-offload-for-udp-tcp-over-ipv6.patch new file mode 100644 index 00000000000..c200126e8bf --- /dev/null +++ b/queue-4.14/net-aquantia-fix-rx-checksum-offload-for-udp-tcp-over-ipv6.patch @@ -0,0 +1,41 @@ +From foo@baz Thu Mar 28 21:57:57 CET 2019 +From: Dmitry Bogdanov +Date: Sat, 16 Mar 2019 08:28:18 +0000 +Subject: net: aquantia: fix rx checksum offload for UDP/TCP over IPv6 + +From: Dmitry Bogdanov + +[ Upstream commit a7faaa0c5dc7d091cc9f72b870d7edcdd6f43f12 ] + +TCP/UDP checksum validity was propagated to skb +only if IP checksum is valid. +But for IPv6 there is no validity as there is no checksum in IPv6. +This patch propagates TCP/UDP checksum validity regardless of IP checksum. + +Fixes: 018423e90bee ("net: ethernet: aquantia: Add ring support code") +Signed-off-by: Igor Russkikh +Signed-off-by: Nikita Danilov +Signed-off-by: Dmitry Bogdanov +Signed-off-by: David S. Miller +Signed-off-by: Greg Kroah-Hartman +--- + drivers/net/ethernet/aquantia/atlantic/aq_ring.c | 5 +++-- + 1 file changed, 3 insertions(+), 2 deletions(-) + +--- a/drivers/net/ethernet/aquantia/atlantic/aq_ring.c ++++ b/drivers/net/ethernet/aquantia/atlantic/aq_ring.c +@@ -270,11 +270,12 @@ int aq_ring_rx_clean(struct aq_ring_s *s + } else { + if (buff->is_ip_cso) { + __skb_incr_checksum_unnecessary(skb); +- if (buff->is_udp_cso || buff->is_tcp_cso) +- __skb_incr_checksum_unnecessary(skb); + } else { + skb->ip_summed = CHECKSUM_NONE; + } ++ ++ if (buff->is_udp_cso || buff->is_tcp_cso) ++ __skb_incr_checksum_unnecessary(skb); + } + + skb_set_hash(skb, buff->rss_hash, diff --git a/queue-4.14/net-datagram-fix-unbounded-loop-in-__skb_try_recv_datagram.patch b/queue-4.14/net-datagram-fix-unbounded-loop-in-__skb_try_recv_datagram.patch new file mode 100644 index 00000000000..615740333dc --- /dev/null +++ b/queue-4.14/net-datagram-fix-unbounded-loop-in-__skb_try_recv_datagram.patch @@ -0,0 +1,37 @@ +From foo@baz Thu Mar 28 21:57:57 CET 2019 +From: Paolo Abeni +Date: Mon, 25 Mar 2019 14:18:06 +0100 +Subject: net: datagram: fix unbounded loop in __skb_try_recv_datagram() + +From: Paolo Abeni + +[ Upstream commit 0b91bce1ebfc797ff3de60c8f4a1e6219a8a3187 ] + +Christoph reported a stall while peeking datagram with an offset when +busy polling is enabled. __skb_try_recv_datagram() uses as the loop +termination condition 'queue empty'. When peeking, the socket +queue can be not empty, even when no additional packets are received. + +Address the issue explicitly checking for receive queue changes, +as currently done by __skb_wait_for_more_packets(). + +Fixes: 2b5cd0dfa384 ("net: Change return type of sk_busy_loop from bool to void") +Reported-and-tested-by: Christoph Paasch +Signed-off-by: Paolo Abeni +Signed-off-by: David S. Miller +Signed-off-by: Greg Kroah-Hartman +--- + net/core/datagram.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +--- a/net/core/datagram.c ++++ b/net/core/datagram.c +@@ -281,7 +281,7 @@ struct sk_buff *__skb_try_recv_datagram( + break; + + sk_busy_loop(sk, flags & MSG_DONTWAIT); +- } while (!skb_queue_empty(&sk->sk_receive_queue)); ++ } while (sk->sk_receive_queue.prev != *last); + + error = -EAGAIN; + diff --git a/queue-4.14/net-packet-set-__gfp_nowarn-upon-allocation-in-alloc_pg_vec.patch b/queue-4.14/net-packet-set-__gfp_nowarn-upon-allocation-in-alloc_pg_vec.patch new file mode 100644 index 00000000000..3de212a5c80 --- /dev/null +++ b/queue-4.14/net-packet-set-__gfp_nowarn-upon-allocation-in-alloc_pg_vec.patch @@ -0,0 +1,81 @@ +From foo@baz Thu Mar 28 21:57:57 CET 2019 +From: Christoph Paasch +Date: Mon, 18 Mar 2019 23:14:52 -0700 +Subject: net/packet: Set __GFP_NOWARN upon allocation in alloc_pg_vec + +From: Christoph Paasch + +[ Upstream commit 398f0132c14754fcd03c1c4f8e7176d001ce8ea1 ] + +Since commit fc62814d690c ("net/packet: fix 4gb buffer limit due to overflow check") +one can now allocate packet ring buffers >= UINT_MAX. However, syzkaller +found that that triggers a warning: + +[ 21.100000] WARNING: CPU: 2 PID: 2075 at mm/page_alloc.c:4584 __alloc_pages_nod0 +[ 21.101490] Modules linked in: +[ 21.101921] CPU: 2 PID: 2075 Comm: syz-executor.0 Not tainted 5.0.0 #146 +[ 21.102784] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 0.5.1 01/01/2011 +[ 21.103887] RIP: 0010:__alloc_pages_nodemask+0x2a0/0x630 +[ 21.104640] Code: fe ff ff 65 48 8b 04 25 c0 de 01 00 48 05 90 0f 00 00 41 bd 01 00 00 00 48 89 44 24 48 e9 9c fe 3 +[ 21.107121] RSP: 0018:ffff88805e1cf920 EFLAGS: 00010246 +[ 21.107819] RAX: 0000000000000000 RBX: ffffffff85a488a0 RCX: 0000000000000000 +[ 21.108753] RDX: 0000000000000000 RSI: dffffc0000000000 RDI: 0000000000000000 +[ 21.109699] RBP: 1ffff1100bc39f28 R08: ffffed100bcefb67 R09: ffffed100bcefb67 +[ 21.110646] R10: 0000000000000001 R11: ffffed100bcefb66 R12: 000000000000000d +[ 21.111623] R13: 0000000000000000 R14: ffff88805e77d888 R15: 000000000000000d +[ 21.112552] FS: 00007f7c7de05700(0000) GS:ffff88806d100000(0000) knlGS:0000000000000000 +[ 21.113612] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +[ 21.114405] CR2: 000000000065c000 CR3: 000000005e58e006 CR4: 00000000001606e0 +[ 21.115367] Call Trace: +[ 21.115705] ? __alloc_pages_slowpath+0x21c0/0x21c0 +[ 21.116362] alloc_pages_current+0xac/0x1e0 +[ 21.116923] kmalloc_order+0x18/0x70 +[ 21.117393] kmalloc_order_trace+0x18/0x110 +[ 21.117949] packet_set_ring+0x9d5/0x1770 +[ 21.118524] ? packet_rcv_spkt+0x440/0x440 +[ 21.119094] ? lock_downgrade+0x620/0x620 +[ 21.119646] ? __might_fault+0x177/0x1b0 +[ 21.120177] packet_setsockopt+0x981/0x2940 +[ 21.120753] ? __fget+0x2fb/0x4b0 +[ 21.121209] ? packet_release+0xab0/0xab0 +[ 21.121740] ? sock_has_perm+0x1cd/0x260 +[ 21.122297] ? selinux_secmark_relabel_packet+0xd0/0xd0 +[ 21.123013] ? __fget+0x324/0x4b0 +[ 21.123451] ? selinux_netlbl_socket_setsockopt+0x101/0x320 +[ 21.124186] ? selinux_netlbl_sock_rcv_skb+0x3a0/0x3a0 +[ 21.124908] ? __lock_acquire+0x529/0x3200 +[ 21.125453] ? selinux_socket_setsockopt+0x5d/0x70 +[ 21.126075] ? __sys_setsockopt+0x131/0x210 +[ 21.126533] ? packet_release+0xab0/0xab0 +[ 21.127004] __sys_setsockopt+0x131/0x210 +[ 21.127449] ? kernel_accept+0x2f0/0x2f0 +[ 21.127911] ? ret_from_fork+0x8/0x50 +[ 21.128313] ? do_raw_spin_lock+0x11b/0x280 +[ 21.128800] __x64_sys_setsockopt+0xba/0x150 +[ 21.129271] ? lockdep_hardirqs_on+0x37f/0x560 +[ 21.129769] do_syscall_64+0x9f/0x450 +[ 21.130182] entry_SYSCALL_64_after_hwframe+0x49/0xbe + +We should allocate with __GFP_NOWARN to handle this. + +Cc: Kal Conley +Cc: Andrey Konovalov +Fixes: fc62814d690c ("net/packet: fix 4gb buffer limit due to overflow check") +Signed-off-by: Christoph Paasch +Signed-off-by: David S. Miller +Signed-off-by: Greg Kroah-Hartman +--- + net/packet/af_packet.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +--- a/net/packet/af_packet.c ++++ b/net/packet/af_packet.c +@@ -4232,7 +4232,7 @@ static struct pgv *alloc_pg_vec(struct t + struct pgv *pg_vec; + int i; + +- pg_vec = kcalloc(block_nr, sizeof(struct pgv), GFP_KERNEL); ++ pg_vec = kcalloc(block_nr, sizeof(struct pgv), GFP_KERNEL | __GFP_NOWARN); + if (unlikely(!pg_vec)) + goto out; + diff --git a/queue-4.14/net-rose-fix-a-possible-stack-overflow.patch b/queue-4.14/net-rose-fix-a-possible-stack-overflow.patch new file mode 100644 index 00000000000..279d1001634 --- /dev/null +++ b/queue-4.14/net-rose-fix-a-possible-stack-overflow.patch @@ -0,0 +1,131 @@ +From foo@baz Thu Mar 28 21:57:57 CET 2019 +From: Eric Dumazet +Date: Fri, 15 Mar 2019 10:41:14 -0700 +Subject: net: rose: fix a possible stack overflow + +From: Eric Dumazet + +[ Upstream commit e5dcc0c3223c45c94100f05f28d8ef814db3d82c ] + +rose_write_internal() uses a temp buffer of 100 bytes, but a manual +inspection showed that given arbitrary input, rose_create_facilities() +can fill up to 110 bytes. + +Lets use a tailroom of 256 bytes for peace of mind, and remove +the bounce buffer : we can simply allocate a big enough skb +and adjust its length as needed. + +syzbot report : + +BUG: KASAN: stack-out-of-bounds in memcpy include/linux/string.h:352 [inline] +BUG: KASAN: stack-out-of-bounds in rose_create_facilities net/rose/rose_subr.c:521 [inline] +BUG: KASAN: stack-out-of-bounds in rose_write_internal+0x597/0x15d0 net/rose/rose_subr.c:116 +Write of size 7 at addr ffff88808b1ffbef by task syz-executor.0/24854 + +CPU: 0 PID: 24854 Comm: syz-executor.0 Not tainted 5.0.0+ #97 +Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 +Call Trace: + __dump_stack lib/dump_stack.c:77 [inline] + dump_stack+0x172/0x1f0 lib/dump_stack.c:113 + print_address_description.cold+0x7c/0x20d mm/kasan/report.c:187 + kasan_report.cold+0x1b/0x40 mm/kasan/report.c:317 + check_memory_region_inline mm/kasan/generic.c:185 [inline] + check_memory_region+0x123/0x190 mm/kasan/generic.c:191 + memcpy+0x38/0x50 mm/kasan/common.c:131 + memcpy include/linux/string.h:352 [inline] + rose_create_facilities net/rose/rose_subr.c:521 [inline] + rose_write_internal+0x597/0x15d0 net/rose/rose_subr.c:116 + rose_connect+0x7cb/0x1510 net/rose/af_rose.c:826 + __sys_connect+0x266/0x330 net/socket.c:1685 + __do_sys_connect net/socket.c:1696 [inline] + __se_sys_connect net/socket.c:1693 [inline] + __x64_sys_connect+0x73/0xb0 net/socket.c:1693 + do_syscall_64+0x103/0x610 arch/x86/entry/common.c:290 + entry_SYSCALL_64_after_hwframe+0x49/0xbe +RIP: 0033:0x458079 +Code: ad b8 fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 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 0f 83 7b b8 fb ff c3 66 2e 0f 1f 84 00 00 00 00 +RSP: 002b:00007f47b8d9dc78 EFLAGS: 00000246 ORIG_RAX: 000000000000002a +RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 0000000000458079 +RDX: 000000000000001c RSI: 0000000020000040 RDI: 0000000000000004 +RBP: 000000000073bf00 R08: 0000000000000000 R09: 0000000000000000 +R10: 0000000000000000 R11: 0000000000000246 R12: 00007f47b8d9e6d4 +R13: 00000000004be4a4 R14: 00000000004ceca8 R15: 00000000ffffffff + +The buggy address belongs to the page: +page:ffffea00022c7fc0 count:0 mapcount:0 mapping:0000000000000000 index:0x0 +flags: 0x1fffc0000000000() +raw: 01fffc0000000000 0000000000000000 ffffffff022c0101 0000000000000000 +raw: 0000000000000000 0000000000000000 00000000ffffffff 0000000000000000 +page dumped because: kasan: bad access detected + +Memory state around the buggy address: + ffff88808b1ffa80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 + ffff88808b1ffb00: 00 00 00 00 00 00 00 00 f1 f1 f1 f1 00 00 00 03 +>ffff88808b1ffb80: f2 f2 00 00 00 00 00 00 00 00 00 00 00 00 04 f3 + ^ + ffff88808b1ffc00: f3 f3 f3 f3 00 00 00 00 00 00 00 00 00 00 00 00 + ffff88808b1ffc80: 00 00 00 00 00 00 00 f1 f1 f1 f1 f1 f1 01 f2 01 + +Signed-off-by: Eric Dumazet +Reported-by: syzbot +Signed-off-by: David S. Miller +Signed-off-by: Greg Kroah-Hartman +--- + net/rose/rose_subr.c | 21 ++++++++++++--------- + 1 file changed, 12 insertions(+), 9 deletions(-) + +--- a/net/rose/rose_subr.c ++++ b/net/rose/rose_subr.c +@@ -105,16 +105,17 @@ void rose_write_internal(struct sock *sk + struct sk_buff *skb; + unsigned char *dptr; + unsigned char lci1, lci2; +- char buffer[100]; +- int len, faclen = 0; ++ int maxfaclen = 0; ++ int len, faclen; ++ int reserve; + +- len = AX25_BPQ_HEADER_LEN + AX25_MAX_HEADER_LEN + ROSE_MIN_LEN + 1; ++ reserve = AX25_BPQ_HEADER_LEN + AX25_MAX_HEADER_LEN + 1; ++ len = ROSE_MIN_LEN; + + switch (frametype) { + case ROSE_CALL_REQUEST: + len += 1 + ROSE_ADDR_LEN + ROSE_ADDR_LEN; +- faclen = rose_create_facilities(buffer, rose); +- len += faclen; ++ maxfaclen = 256; + break; + case ROSE_CALL_ACCEPTED: + case ROSE_CLEAR_REQUEST: +@@ -123,15 +124,16 @@ void rose_write_internal(struct sock *sk + break; + } + +- if ((skb = alloc_skb(len, GFP_ATOMIC)) == NULL) ++ skb = alloc_skb(reserve + len + maxfaclen, GFP_ATOMIC); ++ if (!skb) + return; + + /* + * Space for AX.25 header and PID. + */ +- skb_reserve(skb, AX25_BPQ_HEADER_LEN + AX25_MAX_HEADER_LEN + 1); ++ skb_reserve(skb, reserve); + +- dptr = skb_put(skb, skb_tailroom(skb)); ++ dptr = skb_put(skb, len); + + lci1 = (rose->lci >> 8) & 0x0F; + lci2 = (rose->lci >> 0) & 0xFF; +@@ -146,7 +148,8 @@ void rose_write_internal(struct sock *sk + dptr += ROSE_ADDR_LEN; + memcpy(dptr, &rose->source_addr, ROSE_ADDR_LEN); + dptr += ROSE_ADDR_LEN; +- memcpy(dptr, buffer, faclen); ++ faclen = rose_create_facilities(dptr, rose); ++ skb_put(skb, faclen); + dptr += faclen; + break; + diff --git a/queue-4.14/net-stmmac-fix-memory-corruption-with-large-mtus.patch b/queue-4.14/net-stmmac-fix-memory-corruption-with-large-mtus.patch new file mode 100644 index 00000000000..8ac1de7cce7 --- /dev/null +++ b/queue-4.14/net-stmmac-fix-memory-corruption-with-large-mtus.patch @@ -0,0 +1,64 @@ +From foo@baz Thu Mar 28 21:57:57 CET 2019 +From: Aaro Koskinen +Date: Mon, 18 Mar 2019 23:36:08 +0200 +Subject: net: stmmac: fix memory corruption with large MTUs + +From: Aaro Koskinen + +[ Upstream commit 223a960c01227e4dbcb6f9fa06b47d73bda21274 ] + +When using 16K DMA buffers and ring mode, the DES3 refill is not working +correctly as the function is using a bogus pointer for checking the +private data. As a result stale pointers will remain in the RX descriptor +ring, so DMA will now likely overwrite/corrupt some already freed memory. + +As simple reproducer, just receive some UDP traffic: + + # ifconfig eth0 down; ifconfig eth0 mtu 9000; ifconfig eth0 up + # iperf3 -c 192.168.253.40 -u -b 0 -R + +If you didn't crash by now check the RX descriptors to find non-contiguous +RX buffers: + + cat /sys/kernel/debug/stmmaceth/eth0/descriptors_status + [...] + 1 [0x2be5020]: 0xa3220321 0x9ffc1ffc 0x72d70082 0x130e207e + ^^^^^^^^^^^^^^^^^^^^^ + 2 [0x2be5040]: 0xa3220321 0x9ffc1ffc 0x72998082 0x1311a07e + ^^^^^^^^^^^^^^^^^^^^^ + +A simple ping test will now report bad data: + + # ping -s 8200 192.168.253.40 + PING 192.168.253.40 (192.168.253.40) 8200(8228) bytes of data. + 8208 bytes from 192.168.253.40: icmp_seq=1 ttl=64 time=1.00 ms + wrong data byte #8144 should be 0xd0 but was 0x88 + +Fix the wrong pointer. Also we must refill DES3 only if the DMA buffer +size is 16K. + +Fixes: 54139cf3bb33 ("net: stmmac: adding multiple buffers for rx") +Signed-off-by: Aaro Koskinen +Acked-by: Jose Abreu +Signed-off-by: David S. Miller +Signed-off-by: Greg Kroah-Hartman +--- + drivers/net/ethernet/stmicro/stmmac/ring_mode.c | 5 +++-- + 1 file changed, 3 insertions(+), 2 deletions(-) + +--- a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c ++++ b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c +@@ -114,10 +114,11 @@ static unsigned int stmmac_is_jumbo_frm( + + static void stmmac_refill_desc3(void *priv_ptr, struct dma_desc *p) + { +- struct stmmac_priv *priv = (struct stmmac_priv *)priv_ptr; ++ struct stmmac_rx_queue *rx_q = priv_ptr; ++ struct stmmac_priv *priv = rx_q->priv_data; + + /* Fill DES3 in case of RING mode */ +- if (priv->dma_buf_sz >= BUF_SIZE_8KiB) ++ if (priv->dma_buf_sz == BUF_SIZE_16KiB) + p->des3 = cpu_to_le32(le32_to_cpu(p->des2) + BUF_SIZE_8KiB); + } + diff --git a/queue-4.14/net-sysfs-call-dev_hold-if-kobject_init_and_add-success.patch b/queue-4.14/net-sysfs-call-dev_hold-if-kobject_init_and_add-success.patch new file mode 100644 index 00000000000..d8e880d4542 --- /dev/null +++ b/queue-4.14/net-sysfs-call-dev_hold-if-kobject_init_and_add-success.patch @@ -0,0 +1,62 @@ +From foo@baz Thu Mar 28 21:57:57 CET 2019 +From: YueHaibing +Date: Tue, 19 Mar 2019 10:16:53 +0800 +Subject: net-sysfs: call dev_hold if kobject_init_and_add success + +From: YueHaibing + +[ Upstream commit a3e23f719f5c4a38ffb3d30c8d7632a4ed8ccd9e ] + +In netdev_queue_add_kobject and rx_queue_add_kobject, +if sysfs_create_group failed, kobject_put will call +netdev_queue_release to decrease dev refcont, however +dev_hold has not be called. So we will see this while +unregistering dev: + +unregister_netdevice: waiting for bcsh0 to become free. Usage count = -1 + +Reported-by: Hulk Robot +Fixes: d0d668371679 ("net: don't decrement kobj reference count on init failure") +Signed-off-by: YueHaibing +Signed-off-by: David S. Miller +Signed-off-by: Greg Kroah-Hartman +--- + net/core/net-sysfs.c | 6 ++++-- + 1 file changed, 4 insertions(+), 2 deletions(-) + +--- a/net/core/net-sysfs.c ++++ b/net/core/net-sysfs.c +@@ -917,6 +917,8 @@ static int rx_queue_add_kobject(struct n + if (error) + return error; + ++ dev_hold(queue->dev); ++ + if (dev->sysfs_rx_queue_group) { + error = sysfs_create_group(kobj, dev->sysfs_rx_queue_group); + if (error) { +@@ -926,7 +928,6 @@ static int rx_queue_add_kobject(struct n + } + + kobject_uevent(kobj, KOBJ_ADD); +- dev_hold(queue->dev); + + return error; + } +@@ -1327,6 +1328,8 @@ static int netdev_queue_add_kobject(stru + if (error) + return error; + ++ dev_hold(queue->dev); ++ + #ifdef CONFIG_BQL + error = sysfs_create_group(kobj, &dql_group); + if (error) { +@@ -1336,7 +1339,6 @@ static int netdev_queue_add_kobject(stru + #endif + + kobject_uevent(kobj, KOBJ_ADD); +- dev_hold(queue->dev); + + return 0; + } diff --git a/queue-4.14/net-sysfs-fix-memory-leak-in-netdev_register_kobject.patch b/queue-4.14/net-sysfs-fix-memory-leak-in-netdev_register_kobject.patch new file mode 100644 index 00000000000..53805128d3a --- /dev/null +++ b/queue-4.14/net-sysfs-fix-memory-leak-in-netdev_register_kobject.patch @@ -0,0 +1,86 @@ +From foo@baz Thu Mar 28 21:57:57 CET 2019 +From: Wang Hai +Date: Wed, 20 Mar 2019 14:25:05 -0400 +Subject: net-sysfs: Fix memory leak in netdev_register_kobject + +From: Wang Hai + +[ Upstream commit 6b70fc94afd165342876e53fc4b2f7d085009945 ] + +When registering struct net_device, it will call + register_netdevice -> + netdev_register_kobject -> + device_initialize(dev); + dev_set_name(dev, "%s", ndev->name) + device_add(dev) + register_queue_kobjects(ndev) + +In netdev_register_kobject(), if device_add(dev) or +register_queue_kobjects(ndev) failed. Register_netdevice() +will return error, causing netdev_freemem(ndev) to be +called to free net_device, however put_device(&dev->dev)->..-> +kobject_cleanup() won't be called, resulting in a memory leak. + +syzkaller report this: +BUG: memory leak +unreferenced object 0xffff8881f4fad168 (size 8): +comm "syz-executor.0", pid 3575, jiffies 4294778002 (age 20.134s) +hex dump (first 8 bytes): + 77 70 61 6e 30 00 ff ff wpan0... +backtrace: + [<000000006d2d91d7>] kstrdup_const+0x3d/0x50 mm/util.c:73 + [<00000000ba9ff953>] kvasprintf_const+0x112/0x170 lib/kasprintf.c:48 + [<000000005555ec09>] kobject_set_name_vargs+0x55/0x130 lib/kobject.c:281 + [<0000000098d28ec3>] dev_set_name+0xbb/0xf0 drivers/base/core.c:1915 + [<00000000b7553017>] netdev_register_kobject+0xc0/0x410 net/core/net-sysfs.c:1727 + [<00000000c826a797>] register_netdevice+0xa51/0xeb0 net/core/dev.c:8711 + [<00000000857bfcfd>] cfg802154_update_iface_num.isra.2+0x13/0x90 [ieee802154] + [<000000003126e453>] ieee802154_llsec_fill_key_id+0x1d5/0x570 [ieee802154] + [<00000000e4b3df51>] 0xffffffffc1500e0e + [<00000000b4319776>] platform_drv_probe+0xc6/0x180 drivers/base/platform.c:614 + [<0000000037669347>] really_probe+0x491/0x7c0 drivers/base/dd.c:509 + [<000000008fed8862>] driver_probe_device+0xdc/0x240 drivers/base/dd.c:671 + [<00000000baf52041>] device_driver_attach+0xf2/0x130 drivers/base/dd.c:945 + [<00000000c7cc8dec>] __driver_attach+0x10e/0x210 drivers/base/dd.c:1022 + [<0000000057a757c2>] bus_for_each_dev+0x154/0x1e0 drivers/base/bus.c:304 + [<000000005f5ae04b>] bus_add_driver+0x427/0x5e0 drivers/base/bus.c:645 + +Reported-by: Hulk Robot +Fixes: 1fa5ae857bb1 ("driver core: get rid of struct device's bus_id string array") +Signed-off-by: Wang Hai +Reviewed-by: Andy Shevchenko +Reviewed-by: Stephen Hemminger +Signed-off-by: David S. Miller +Signed-off-by: Greg Kroah-Hartman +--- + net/core/net-sysfs.c | 14 +++++++++----- + 1 file changed, 9 insertions(+), 5 deletions(-) + +--- a/net/core/net-sysfs.c ++++ b/net/core/net-sysfs.c +@@ -1598,16 +1598,20 @@ int netdev_register_kobject(struct net_d + + error = device_add(dev); + if (error) +- return error; ++ goto error_put_device; + + error = register_queue_kobjects(ndev); +- if (error) { +- device_del(dev); +- return error; +- } ++ if (error) ++ goto error_device_del; + + pm_runtime_set_memalloc_noio(dev, true); + ++ return 0; ++ ++error_device_del: ++ device_del(dev); ++error_put_device: ++ put_device(dev); + return error; + } + diff --git a/queue-4.14/packets-always-register-packet-sk-in-the-same-order.patch b/queue-4.14/packets-always-register-packet-sk-in-the-same-order.patch new file mode 100644 index 00000000000..c2fe15312d1 --- /dev/null +++ b/queue-4.14/packets-always-register-packet-sk-in-the-same-order.patch @@ -0,0 +1,71 @@ +From foo@baz Thu Mar 28 21:57:57 CET 2019 +From: Maxime Chevallier +Date: Sat, 16 Mar 2019 14:41:30 +0100 +Subject: packets: Always register packet sk in the same order + +From: Maxime Chevallier + +[ Upstream commit a4dc6a49156b1f8d6e17251ffda17c9e6a5db78a ] + +When using fanouts with AF_PACKET, the demux functions such as +fanout_demux_cpu will return an index in the fanout socket array, which +corresponds to the selected socket. + +The ordering of this array depends on the order the sockets were added +to a given fanout group, so for FANOUT_CPU this means sockets are bound +to cpus in the order they are configured, which is OK. + +However, when stopping then restarting the interface these sockets are +bound to, the sockets are reassigned to the fanout group in the reverse +order, due to the fact that they were inserted at the head of the +interface's AF_PACKET socket list. + +This means that traffic that was directed to the first socket in the +fanout group is now directed to the last one after an interface restart. + +In the case of FANOUT_CPU, traffic from CPU0 will be directed to the +socket that used to receive traffic from the last CPU after an interface +restart. + +This commit introduces a helper to add a socket at the tail of a list, +then uses it to register AF_PACKET sockets. + +Note that this changes the order in which sockets are listed in /proc and +with sock_diag. + +Fixes: dc99f600698d ("packet: Add fanout support") +Signed-off-by: Maxime Chevallier +Acked-by: Willem de Bruijn +Signed-off-by: David S. Miller +Signed-off-by: Greg Kroah-Hartman +--- + include/net/sock.h | 6 ++++++ + net/packet/af_packet.c | 2 +- + 2 files changed, 7 insertions(+), 1 deletion(-) + +--- a/include/net/sock.h ++++ b/include/net/sock.h +@@ -682,6 +682,12 @@ static inline void sk_add_node_rcu(struc + hlist_add_head_rcu(&sk->sk_node, list); + } + ++static inline void sk_add_node_tail_rcu(struct sock *sk, struct hlist_head *list) ++{ ++ sock_hold(sk); ++ hlist_add_tail_rcu(&sk->sk_node, list); ++} ++ + static inline void __sk_nulls_add_node_rcu(struct sock *sk, struct hlist_nulls_head *list) + { + hlist_nulls_add_head_rcu(&sk->sk_nulls_node, list); +--- a/net/packet/af_packet.c ++++ b/net/packet/af_packet.c +@@ -3281,7 +3281,7 @@ static int packet_create(struct net *net + } + + mutex_lock(&net->packet.sklist_lock); +- sk_add_node_rcu(sk, &net->packet.sklist); ++ sk_add_node_tail_rcu(sk, &net->packet.sklist); + mutex_unlock(&net->packet.sklist_lock); + + preempt_disable(); diff --git a/queue-4.14/rhashtable-still-do-rehash-when-we-get-eexist.patch b/queue-4.14/rhashtable-still-do-rehash-when-we-get-eexist.patch new file mode 100644 index 00000000000..e136262d20d --- /dev/null +++ b/queue-4.14/rhashtable-still-do-rehash-when-we-get-eexist.patch @@ -0,0 +1,48 @@ +From foo@baz Thu Mar 28 21:57:57 CET 2019 +From: Herbert Xu +Date: Thu, 21 Mar 2019 09:39:52 +0800 +Subject: rhashtable: Still do rehash when we get EEXIST + +From: Herbert Xu + +[ Upstream commit 408f13ef358aa5ad56dc6230c2c7deb92cf462b1 ] + +As it stands if a shrink is delayed because of an outstanding +rehash, we will go into a rescheduling loop without ever doing +the rehash. + +This patch fixes this by still carrying out the rehash and then +rescheduling so that we can shrink after the completion of the +rehash should it still be necessary. + +The return value of EEXIST captures this case and other cases +(e.g., another thread expanded/rehashed the table at the same +time) where we should still proceed with the rehash. + +Fixes: da20420f83ea ("rhashtable: Add nested tables") +Reported-by: Josh Elsasser +Signed-off-by: Herbert Xu +Tested-by: Josh Elsasser +Signed-off-by: David S. Miller +Signed-off-by: Greg Kroah-Hartman +--- + lib/rhashtable.c | 8 ++++++-- + 1 file changed, 6 insertions(+), 2 deletions(-) + +--- a/lib/rhashtable.c ++++ b/lib/rhashtable.c +@@ -459,8 +459,12 @@ static void rht_deferred_worker(struct w + else if (tbl->nest) + err = rhashtable_rehash_alloc(ht, tbl, tbl->size); + +- if (!err) +- err = rhashtable_rehash_table(ht); ++ if (!err || err == -EEXIST) { ++ int nerr; ++ ++ nerr = rhashtable_rehash_table(ht); ++ err = err ?: nerr; ++ } + + mutex_unlock(&ht->mutex); + diff --git a/queue-4.14/sctp-get-sctphdr-by-offset-in-sctp_compute_cksum.patch b/queue-4.14/sctp-get-sctphdr-by-offset-in-sctp_compute_cksum.patch new file mode 100644 index 00000000000..7a60846f6aa --- /dev/null +++ b/queue-4.14/sctp-get-sctphdr-by-offset-in-sctp_compute_cksum.patch @@ -0,0 +1,40 @@ +From foo@baz Thu Mar 28 21:57:57 CET 2019 +From: Xin Long +Date: Mon, 18 Mar 2019 19:47:00 +0800 +Subject: sctp: get sctphdr by offset in sctp_compute_cksum + +From: Xin Long + +[ Upstream commit 273160ffc6b993c7c91627f5a84799c66dfe4dee ] + +sctp_hdr(skb) only works when skb->transport_header is set properly. + +But in Netfilter, skb->transport_header for ipv6 is not guaranteed +to be right value for sctphdr. It would cause to fail to check the +checksum for sctp packets. + +So fix it by using offset, which is always right in all places. + +v1->v2: + - Fix the changelog. + +Fixes: e6d8b64b34aa ("net: sctp: fix and consolidate SCTP checksumming code") +Reported-by: Li Shuang +Signed-off-by: Xin Long +Signed-off-by: David S. Miller +Signed-off-by: Greg Kroah-Hartman +--- + include/net/sctp/checksum.h | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +--- a/include/net/sctp/checksum.h ++++ b/include/net/sctp/checksum.h +@@ -60,7 +60,7 @@ static inline __wsum sctp_csum_combine(_ + static inline __le32 sctp_compute_cksum(const struct sk_buff *skb, + unsigned int offset) + { +- struct sctphdr *sh = sctp_hdr(skb); ++ struct sctphdr *sh = (struct sctphdr *)(skb->data + offset); + __le32 ret, old = sh->checksum; + const struct skb_checksum_ops ops = { + .update = sctp_csum_update, diff --git a/queue-4.14/series b/queue-4.14/series index 5c51886f32b..9a998c2677d 100644 --- a/queue-4.14/series +++ b/queue-4.14/series @@ -2,3 +2,22 @@ bluetooth-check-l2cap-option-sizes-returned-from-l2cap_get_conf_opt.patch bluetooth-verify-that-l2cap_get_conf_opt-provides-large-enough-buffer.patch video-fbdev-set-pixclock-0-in-goldfishfb.patch stmmac-copy-unicast-mac-address-to-mac-registers.patch +dccp-do-not-use-ipv6-header-for-ipv4-flow.patch +genetlink-fix-a-memory-leak-on-error-path.patch +misdn-hfcpci-test-both-vendor-device-id-for-digium-hfc4s.patch +net-datagram-fix-unbounded-loop-in-__skb_try_recv_datagram.patch +net-packet-set-__gfp_nowarn-upon-allocation-in-alloc_pg_vec.patch +net-rose-fix-a-possible-stack-overflow.patch +net-stmmac-fix-memory-corruption-with-large-mtus.patch +net-sysfs-call-dev_hold-if-kobject_init_and_add-success.patch +net-sysfs-fix-memory-leak-in-netdev_register_kobject.patch +packets-always-register-packet-sk-in-the-same-order.patch +rhashtable-still-do-rehash-when-we-get-eexist.patch +tcp-do-not-use-ipv6-header-for-ipv4-flow.patch +thunderx-enable-page-recycling-for-non-xdp-case.patch +thunderx-eliminate-extra-calls-to-put_page-for-pages-held-for-recycling.patch +vxlan-don-t-call-gro_cells_destroy-before-device-is-unregistered.patch +sctp-get-sctphdr-by-offset-in-sctp_compute_cksum.patch +net-aquantia-fix-rx-checksum-offload-for-udp-tcp-over-ipv6.patch +mac8390-fix-mmio-access-size-probe.patch +tun-properly-test-for-iff_up.patch diff --git a/queue-4.14/tcp-do-not-use-ipv6-header-for-ipv4-flow.patch b/queue-4.14/tcp-do-not-use-ipv6-header-for-ipv4-flow.patch new file mode 100644 index 00000000000..87b9b7d447c --- /dev/null +++ b/queue-4.14/tcp-do-not-use-ipv6-header-for-ipv4-flow.patch @@ -0,0 +1,40 @@ +From foo@baz Thu Mar 28 21:57:57 CET 2019 +From: Eric Dumazet +Date: Tue, 19 Mar 2019 05:45:35 -0700 +Subject: tcp: do not use ipv6 header for ipv4 flow + +From: Eric Dumazet + +[ Upstream commit 89e4130939a20304f4059ab72179da81f5347528 ] + +When a dual stack tcp listener accepts an ipv4 flow, +it should not attempt to use an ipv6 header or tcp_v6_iif() helper. + +Fixes: 1397ed35f22d ("ipv6: add flowinfo for tcp6 pkt_options for all cases") +Fixes: df3687ffc665 ("ipv6: add the IPV6_FL_F_REFLECT flag to IPV6_FL_A_GET") +Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") +Signed-off-by: Eric Dumazet +Signed-off-by: David S. Miller +Signed-off-by: Greg Kroah-Hartman +--- + net/ipv6/tcp_ipv6.c | 8 ++++---- + 1 file changed, 4 insertions(+), 4 deletions(-) + +--- a/net/ipv6/tcp_ipv6.c ++++ b/net/ipv6/tcp_ipv6.c +@@ -1083,11 +1083,11 @@ static struct sock *tcp_v6_syn_recv_sock + newnp->ipv6_fl_list = NULL; + newnp->pktoptions = NULL; + newnp->opt = NULL; +- newnp->mcast_oif = tcp_v6_iif(skb); +- newnp->mcast_hops = ipv6_hdr(skb)->hop_limit; +- newnp->rcv_flowinfo = ip6_flowinfo(ipv6_hdr(skb)); ++ newnp->mcast_oif = inet_iif(skb); ++ newnp->mcast_hops = ip_hdr(skb)->ttl; ++ newnp->rcv_flowinfo = 0; + if (np->repflow) +- newnp->flow_label = ip6_flowlabel(ipv6_hdr(skb)); ++ newnp->flow_label = 0; + + /* + * No need to charge this sock to the relevant IPv6 refcnt debug socks count diff --git a/queue-4.14/thunderx-eliminate-extra-calls-to-put_page-for-pages-held-for-recycling.patch b/queue-4.14/thunderx-eliminate-extra-calls-to-put_page-for-pages-held-for-recycling.patch new file mode 100644 index 00000000000..05ce70f688a --- /dev/null +++ b/queue-4.14/thunderx-eliminate-extra-calls-to-put_page-for-pages-held-for-recycling.patch @@ -0,0 +1,58 @@ +From foo@baz Thu Mar 28 21:57:57 CET 2019 +From: Dean Nelson +Date: Tue, 26 Mar 2019 11:53:26 -0400 +Subject: thunderx: eliminate extra calls to put_page() for pages held for recycling + +From: Dean Nelson + +[ Upstream commit cd35ef91490ad8049dd180bb060aff7ee192eda9 ] + +For the non-XDP case, commit 773225388dae15e72790 ("net: thunderx: Optimize +page recycling for XDP") added code to nicvf_free_rbdr() that, when releasing +the additional receive buffer page reference held for recycling, repeatedly +calls put_page() until the page's _refcount goes to zero. Which results in +the page being freed. + +This is not okay if the page's _refcount was greater than 1 (in the non-XDP +case), because nicvf_free_rbdr() should not be subtracting more than what +nicvf_alloc_page() had previously added to the page's _refcount, which was +only 1 (in the non-XDP case). + +This can arise if a received packet is still being processed and the receive +buffer (i.e., skb->head) has not yet been freed via skb_free_head() when +nicvf_free_rbdr() is spinning through the aforementioned put_page() loop. + +If this should occur, when the received packet finishes processing and +skb_free_head() is called, various problems can ensue. Exactly what, depends on +whether the page has already been reallocated or not, anything from "BUG: Bad +page state ... ", to "Unable to handle kernel NULL pointer dereference ..." or +"Unable to handle kernel paging request...". + +So this patch changes nicvf_free_rbdr() to only call put_page() once for pages +held for recycling (in the non-XDP case). + +Fixes: 773225388dae ("net: thunderx: Optimize page recycling for XDP") +Signed-off-by: Dean Nelson +Signed-off-by: David S. Miller +Signed-off-by: Greg Kroah-Hartman +--- + drivers/net/ethernet/cavium/thunder/nicvf_queues.c | 7 +++---- + 1 file changed, 3 insertions(+), 4 deletions(-) + +--- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c ++++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c +@@ -364,11 +364,10 @@ static void nicvf_free_rbdr(struct nicvf + while (head < rbdr->pgcnt) { + pgcache = &rbdr->pgcache[head]; + if (pgcache->page && page_ref_count(pgcache->page) != 0) { +- if (!rbdr->is_xdp) { +- put_page(pgcache->page); +- continue; ++ if (rbdr->is_xdp) { ++ page_ref_sub(pgcache->page, ++ pgcache->ref_count - 1); + } +- page_ref_sub(pgcache->page, pgcache->ref_count - 1); + put_page(pgcache->page); + } + head++; diff --git a/queue-4.14/thunderx-enable-page-recycling-for-non-xdp-case.patch b/queue-4.14/thunderx-enable-page-recycling-for-non-xdp-case.patch new file mode 100644 index 00000000000..b01af9a5a0a --- /dev/null +++ b/queue-4.14/thunderx-enable-page-recycling-for-non-xdp-case.patch @@ -0,0 +1,59 @@ +From foo@baz Thu Mar 28 21:57:57 CET 2019 +From: Dean Nelson +Date: Tue, 26 Mar 2019 11:53:19 -0400 +Subject: thunderx: enable page recycling for non-XDP case + +From: Dean Nelson + +[ Upstream commit b3e208069477588c06f4d5d986164b435bb06e6d ] + +Commit 773225388dae15e72790 ("net: thunderx: Optimize page recycling for XDP") +added code to nicvf_alloc_page() that inadvertently disables receive buffer +page recycling for the non-XDP case by always NULL'ng the page pointer. + +This patch corrects two if-conditionals to allow for the recycling of non-XDP +mode pages by only setting the page pointer to NULL when the page is not ready +for recycling. + +Fixes: 773225388dae ("net: thunderx: Optimize page recycling for XDP") +Signed-off-by: Dean Nelson +Signed-off-by: David S. Miller +Signed-off-by: Greg Kroah-Hartman +--- + drivers/net/ethernet/cavium/thunder/nicvf_queues.c | 23 ++++++++++----------- + 1 file changed, 11 insertions(+), 12 deletions(-) + +--- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c ++++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c +@@ -105,20 +105,19 @@ static inline struct pgcache *nicvf_allo + /* Check if page can be recycled */ + if (page) { + ref_count = page_ref_count(page); +- /* Check if this page has been used once i.e 'put_page' +- * called after packet transmission i.e internal ref_count +- * and page's ref_count are equal i.e page can be recycled. ++ /* This page can be recycled if internal ref_count and page's ++ * ref_count are equal, indicating that the page has been used ++ * once for packet transmission. For non-XDP mode, internal ++ * ref_count is always '1'. + */ +- if (rbdr->is_xdp && (ref_count == pgcache->ref_count)) +- pgcache->ref_count--; +- else +- page = NULL; +- +- /* In non-XDP mode, page's ref_count needs to be '1' for it +- * to be recycled. +- */ +- if (!rbdr->is_xdp && (ref_count != 1)) ++ if (rbdr->is_xdp) { ++ if (ref_count == pgcache->ref_count) ++ pgcache->ref_count--; ++ else ++ page = NULL; ++ } else if (ref_count != 1) { + page = NULL; ++ } + } + + if (!page) { diff --git a/queue-4.14/tun-properly-test-for-iff_up.patch b/queue-4.14/tun-properly-test-for-iff_up.patch new file mode 100644 index 00000000000..f6a1310331d --- /dev/null +++ b/queue-4.14/tun-properly-test-for-iff_up.patch @@ -0,0 +1,82 @@ +From foo@baz Thu Mar 28 21:57:57 CET 2019 +From: Eric Dumazet +Date: Thu, 14 Mar 2019 20:19:47 -0700 +Subject: tun: properly test for IFF_UP + +From: Eric Dumazet + +[ Upstream commit 4477138fa0ae4e1b699786ef0600863ea6e6c61c ] + +Same reasons than the ones explained in commit 4179cb5a4c92 +("vxlan: test dev->flags & IFF_UP before calling netif_rx()") + +netif_rx_ni() or napi_gro_frags() must be called under a strict contract. + +At device dismantle phase, core networking clears IFF_UP +and flush_all_backlogs() is called after rcu grace period +to make sure no incoming packet might be in a cpu backlog +and still referencing the device. + +A similar protocol is used for gro layer. + +Most drivers call netif_rx() from their interrupt handler, +and since the interrupts are disabled at device dismantle, +netif_rx() does not have to check dev->flags & IFF_UP + +Virtual drivers do not have this guarantee, and must +therefore make the check themselves. + +Fixes: 1bd4978a88ac ("tun: honor IFF_UP in tun_get_user()") +Signed-off-by: Eric Dumazet +Reported-by: syzbot +Signed-off-by: David S. Miller +Signed-off-by: Greg Kroah-Hartman +--- + drivers/net/tun.c | 15 +++++++++++---- + 1 file changed, 11 insertions(+), 4 deletions(-) + +--- a/drivers/net/tun.c ++++ b/drivers/net/tun.c +@@ -1403,9 +1403,6 @@ static ssize_t tun_get_user(struct tun_s + u32 rxhash; + int skb_xdp = 1; + +- if (!(tun->dev->flags & IFF_UP)) +- return -EIO; +- + if (!(tun->flags & IFF_NO_PI)) { + if (len < sizeof(pi)) + return -EINVAL; +@@ -1493,9 +1490,11 @@ static ssize_t tun_get_user(struct tun_s + err = skb_copy_datagram_from_iter(skb, 0, from, len); + + if (err) { ++ err = -EFAULT; ++drop: + this_cpu_inc(tun->pcpu_stats->rx_dropped); + kfree_skb(skb); +- return -EFAULT; ++ return err; + } + } + +@@ -1566,11 +1565,19 @@ static ssize_t tun_get_user(struct tun_s + } + + rxhash = __skb_get_hash_symmetric(skb); ++ ++ rcu_read_lock(); ++ if (unlikely(!(tun->dev->flags & IFF_UP))) { ++ err = -EIO; ++ goto drop; ++ } ++ + #ifndef CONFIG_4KSTACKS + tun_rx_batched(tun, tfile, skb, more); + #else + netif_rx_ni(skb); + #endif ++ rcu_read_unlock(); + + stats = get_cpu_ptr(tun->pcpu_stats); + u64_stats_update_begin(&stats->syncp); diff --git a/queue-4.14/vxlan-don-t-call-gro_cells_destroy-before-device-is-unregistered.patch b/queue-4.14/vxlan-don-t-call-gro_cells_destroy-before-device-is-unregistered.patch new file mode 100644 index 00000000000..65ae956a3cc --- /dev/null +++ b/queue-4.14/vxlan-don-t-call-gro_cells_destroy-before-device-is-unregistered.patch @@ -0,0 +1,47 @@ +From foo@baz Thu Mar 28 21:57:57 CET 2019 +From: Zhiqiang Liu +Date: Sat, 16 Mar 2019 17:02:54 +0800 +Subject: vxlan: Don't call gro_cells_destroy() before device is unregistered + +From: Zhiqiang Liu + +[ Upstream commit cc4807bb609230d8959fd732b0bf3bd4c2de8eac ] + +Commit ad6c9986bcb62 ("vxlan: Fix GRO cells race condition between +receive and link delete") fixed a race condition for the typical case a vxlan +device is dismantled from the current netns. But if a netns is dismantled, +vxlan_destroy_tunnels() is called to schedule a unregister_netdevice_queue() +of all the vxlan tunnels that are related to this netns. + +In vxlan_destroy_tunnels(), gro_cells_destroy() is called and finished before +unregister_netdevice_queue(). This means that the gro_cells_destroy() call is +done too soon, for the same reasons explained in above commit. + +So we need to fully respect the RCU rules, and thus must remove the +gro_cells_destroy() call or risk use after-free. + +Fixes: 58ce31cca1ff ("vxlan: GRO support at tunnel layer") +Signed-off-by: Suanming.Mou +Suggested-by: Eric Dumazet +Reviewed-by: Stefano Brivio +Reviewed-by: Zhiqiang Liu +Signed-off-by: David S. Miller +Signed-off-by: Greg Kroah-Hartman +--- + drivers/net/vxlan.c | 4 +--- + 1 file changed, 1 insertion(+), 3 deletions(-) + +--- a/drivers/net/vxlan.c ++++ b/drivers/net/vxlan.c +@@ -3793,10 +3793,8 @@ static void __net_exit vxlan_exit_net(st + /* If vxlan->dev is in the same netns, it has already been added + * to the list by the previous loop. + */ +- if (!net_eq(dev_net(vxlan->dev), net)) { +- gro_cells_destroy(&vxlan->gro_cells); ++ if (!net_eq(dev_net(vxlan->dev), net)) + unregister_netdevice_queue(vxlan->dev, &list); +- } + } + + unregister_netdevice_many(&list);