linux-2.6-microblaze.git
7 months agollc: call sock_orphan() at release time
Eric Dumazet [Fri, 26 Jan 2024 16:55:32 +0000 (16:55 +0000)]
llc: call sock_orphan() at release time

syzbot reported an interesting trace [1] caused by a stale sk->sk_wq
pointer in a closed llc socket.

In commit ff7b11aa481f ("net: socket: set sock->sk to NULL after
calling proto_ops::release()") Eric Biggers hinted that some protocols
are missing a sock_orphan(), we need to perform a full audit.

In net-next, I plan to clear sock->sk from sock_orphan() and
amend Eric patch to add a warning.

[1]
 BUG: KASAN: slab-use-after-free in list_empty include/linux/list.h:373 [inline]
 BUG: KASAN: slab-use-after-free in waitqueue_active include/linux/wait.h:127 [inline]
 BUG: KASAN: slab-use-after-free in sock_def_write_space_wfree net/core/sock.c:3384 [inline]
 BUG: KASAN: slab-use-after-free in sock_wfree+0x9a8/0x9d0 net/core/sock.c:2468
Read of size 8 at addr ffff88802f4fc880 by task ksoftirqd/1/27

CPU: 1 PID: 27 Comm: ksoftirqd/1 Not tainted 6.8.0-rc1-syzkaller-00049-g6098d87eaf31 #0
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.2-debian-1.16.2-1 04/01/2014
Call Trace:
 <TASK>
  __dump_stack lib/dump_stack.c:88 [inline]
  dump_stack_lvl+0xd9/0x1b0 lib/dump_stack.c:106
  print_address_description mm/kasan/report.c:377 [inline]
  print_report+0xc4/0x620 mm/kasan/report.c:488
  kasan_report+0xda/0x110 mm/kasan/report.c:601
  list_empty include/linux/list.h:373 [inline]
  waitqueue_active include/linux/wait.h:127 [inline]
  sock_def_write_space_wfree net/core/sock.c:3384 [inline]
  sock_wfree+0x9a8/0x9d0 net/core/sock.c:2468
  skb_release_head_state+0xa3/0x2b0 net/core/skbuff.c:1080
  skb_release_all net/core/skbuff.c:1092 [inline]
  napi_consume_skb+0x119/0x2b0 net/core/skbuff.c:1404
  e1000_unmap_and_free_tx_resource+0x144/0x200 drivers/net/ethernet/intel/e1000/e1000_main.c:1970
  e1000_clean_tx_irq drivers/net/ethernet/intel/e1000/e1000_main.c:3860 [inline]
  e1000_clean+0x4a1/0x26e0 drivers/net/ethernet/intel/e1000/e1000_main.c:3801
  __napi_poll.constprop.0+0xb4/0x540 net/core/dev.c:6576
  napi_poll net/core/dev.c:6645 [inline]
  net_rx_action+0x956/0xe90 net/core/dev.c:6778
  __do_softirq+0x21a/0x8de kernel/softirq.c:553
  run_ksoftirqd kernel/softirq.c:921 [inline]
  run_ksoftirqd+0x31/0x60 kernel/softirq.c:913
  smpboot_thread_fn+0x660/0xa10 kernel/smpboot.c:164
  kthread+0x2c6/0x3a0 kernel/kthread.c:388
  ret_from_fork+0x45/0x80 arch/x86/kernel/process.c:147
  ret_from_fork_asm+0x11/0x20 arch/x86/entry/entry_64.S:242
 </TASK>

Allocated by task 5167:
  kasan_save_stack+0x33/0x50 mm/kasan/common.c:47
  kasan_save_track+0x14/0x30 mm/kasan/common.c:68
  unpoison_slab_object mm/kasan/common.c:314 [inline]
  __kasan_slab_alloc+0x81/0x90 mm/kasan/common.c:340
  kasan_slab_alloc include/linux/kasan.h:201 [inline]
  slab_post_alloc_hook mm/slub.c:3813 [inline]
  slab_alloc_node mm/slub.c:3860 [inline]
  kmem_cache_alloc_lru+0x142/0x6f0 mm/slub.c:3879
  alloc_inode_sb include/linux/fs.h:3019 [inline]
  sock_alloc_inode+0x25/0x1c0 net/socket.c:308
  alloc_inode+0x5d/0x220 fs/inode.c:260
  new_inode_pseudo+0x16/0x80 fs/inode.c:1005
  sock_alloc+0x40/0x270 net/socket.c:634
  __sock_create+0xbc/0x800 net/socket.c:1535
  sock_create net/socket.c:1622 [inline]
  __sys_socket_create net/socket.c:1659 [inline]
  __sys_socket+0x14c/0x260 net/socket.c:1706
  __do_sys_socket net/socket.c:1720 [inline]
  __se_sys_socket net/socket.c:1718 [inline]
  __x64_sys_socket+0x72/0xb0 net/socket.c:1718
  do_syscall_x64 arch/x86/entry/common.c:52 [inline]
  do_syscall_64+0xd3/0x250 arch/x86/entry/common.c:83
 entry_SYSCALL_64_after_hwframe+0x63/0x6b

Freed by task 0:
  kasan_save_stack+0x33/0x50 mm/kasan/common.c:47
  kasan_save_track+0x14/0x30 mm/kasan/common.c:68
  kasan_save_free_info+0x3f/0x60 mm/kasan/generic.c:640
  poison_slab_object mm/kasan/common.c:241 [inline]
  __kasan_slab_free+0x121/0x1b0 mm/kasan/common.c:257
  kasan_slab_free include/linux/kasan.h:184 [inline]
  slab_free_hook mm/slub.c:2121 [inline]
  slab_free mm/slub.c:4299 [inline]
  kmem_cache_free+0x129/0x350 mm/slub.c:4363
  i_callback+0x43/0x70 fs/inode.c:249
  rcu_do_batch kernel/rcu/tree.c:2158 [inline]
  rcu_core+0x819/0x1680 kernel/rcu/tree.c:2433
  __do_softirq+0x21a/0x8de kernel/softirq.c:553

Last potentially related work creation:
  kasan_save_stack+0x33/0x50 mm/kasan/common.c:47
  __kasan_record_aux_stack+0xba/0x100 mm/kasan/generic.c:586
  __call_rcu_common.constprop.0+0x9a/0x7b0 kernel/rcu/tree.c:2683
  destroy_inode+0x129/0x1b0 fs/inode.c:315
  iput_final fs/inode.c:1739 [inline]
  iput.part.0+0x560/0x7b0 fs/inode.c:1765
  iput+0x5c/0x80 fs/inode.c:1755
  dentry_unlink_inode+0x292/0x430 fs/dcache.c:400
  __dentry_kill+0x1ca/0x5f0 fs/dcache.c:603
  dput.part.0+0x4ac/0x9a0 fs/dcache.c:845
  dput+0x1f/0x30 fs/dcache.c:835
  __fput+0x3b9/0xb70 fs/file_table.c:384
  task_work_run+0x14d/0x240 kernel/task_work.c:180
  exit_task_work include/linux/task_work.h:38 [inline]
  do_exit+0xa8a/0x2ad0 kernel/exit.c:871
  do_group_exit+0xd4/0x2a0 kernel/exit.c:1020
  __do_sys_exit_group kernel/exit.c:1031 [inline]
  __se_sys_exit_group kernel/exit.c:1029 [inline]
  __x64_sys_exit_group+0x3e/0x50 kernel/exit.c:1029
  do_syscall_x64 arch/x86/entry/common.c:52 [inline]
  do_syscall_64+0xd3/0x250 arch/x86/entry/common.c:83
 entry_SYSCALL_64_after_hwframe+0x63/0x6b

The buggy address belongs to the object at ffff88802f4fc800
 which belongs to the cache sock_inode_cache of size 1408
The buggy address is located 128 bytes inside of
 freed 1408-byte region [ffff88802f4fc800ffff88802f4fcd80)

The buggy address belongs to the physical page:
page:ffffea0000bd3e00 refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x2f4f8
head:ffffea0000bd3e00 order:3 entire_mapcount:0 nr_pages_mapped:0 pincount:0
anon flags: 0xfff00000000840(slab|head|node=0|zone=1|lastcpupid=0x7ff)
page_type: 0xffffffff()
raw: 00fff00000000840 ffff888013b06b40 0000000000000000 0000000000000001
raw: 0000000000000000 0000000080150015 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected
page_owner tracks the page as allocated
page last allocated via order 3, migratetype Reclaimable, gfp_mask 0xd20d0(__GFP_IO|__GFP_FS|__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP|__GFP_NOMEMALLOC|__GFP_RECLAIMABLE), pid 4956, tgid 4956 (sshd), ts 31423924727, free_ts 0
  set_page_owner include/linux/page_owner.h:31 [inline]
  post_alloc_hook+0x2d0/0x350 mm/page_alloc.c:1533
  prep_new_page mm/page_alloc.c:1540 [inline]
  get_page_from_freelist+0xa28/0x3780 mm/page_alloc.c:3311
  __alloc_pages+0x22f/0x2440 mm/page_alloc.c:4567
  __alloc_pages_node include/linux/gfp.h:238 [inline]
  alloc_pages_node include/linux/gfp.h:261 [inline]
  alloc_slab_page mm/slub.c:2190 [inline]
  allocate_slab mm/slub.c:2354 [inline]
  new_slab+0xcc/0x3a0 mm/slub.c:2407
  ___slab_alloc+0x4af/0x19a0 mm/slub.c:3540
  __slab_alloc.constprop.0+0x56/0xa0 mm/slub.c:3625
  __slab_alloc_node mm/slub.c:3678 [inline]
  slab_alloc_node mm/slub.c:3850 [inline]
  kmem_cache_alloc_lru+0x379/0x6f0 mm/slub.c:3879
  alloc_inode_sb include/linux/fs.h:3019 [inline]
  sock_alloc_inode+0x25/0x1c0 net/socket.c:308
  alloc_inode+0x5d/0x220 fs/inode.c:260
  new_inode_pseudo+0x16/0x80 fs/inode.c:1005
  sock_alloc+0x40/0x270 net/socket.c:634
  __sock_create+0xbc/0x800 net/socket.c:1535
  sock_create net/socket.c:1622 [inline]
  __sys_socket_create net/socket.c:1659 [inline]
  __sys_socket+0x14c/0x260 net/socket.c:1706
  __do_sys_socket net/socket.c:1720 [inline]
  __se_sys_socket net/socket.c:1718 [inline]
  __x64_sys_socket+0x72/0xb0 net/socket.c:1718
  do_syscall_x64 arch/x86/entry/common.c:52 [inline]
  do_syscall_64+0xd3/0x250 arch/x86/entry/common.c:83
 entry_SYSCALL_64_after_hwframe+0x63/0x6b
page_owner free stack trace missing

Memory state around the buggy address:
 ffff88802f4fc780: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
 ffff88802f4fc800: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>ffff88802f4fc880: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
                   ^
 ffff88802f4fc900: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff88802f4fc980: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb

Fixes: 43815482370c ("net: sock_def_readable() and friends RCU conversion")
Reported-and-tested-by: syzbot+32b89eaa102b372ff76d@syzkaller.appspotmail.com
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Eric Biggers <ebiggers@google.com>
Cc: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://lore.kernel.org/r/20240126165532.3396702-1-edumazet@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
7 months agoMerge branch 'net-stmmac-dwmac-imx-time-based-scheduling-support'
Paolo Abeni [Tue, 30 Jan 2024 11:58:13 +0000 (12:58 +0100)]
Merge branch 'net-stmmac-dwmac-imx-time-based-scheduling-support'

Esben Haabendal says:

====================
net: stmmac: dwmac-imx: Time Based Scheduling support

This small patch series allows using TBS support of the i.MX Ethernet QOS
controller for etf qdisc offload.
It achieves this in a similar manner that it is done in dwmac-intel.c,
dwmac-mediatek.c and stmmac_pci.c.

Changes since v1:

- Simplified for loop by starting at index 1.
- Fixed problem with indentation.
====================

Link: https://lore.kernel.org/r/cover.1706256158.git.esben@geanix.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
7 months agonet: stmmac: dwmac-imx: set TSO/TBS TX queues default settings
Esben Haabendal [Fri, 26 Jan 2024 09:10:42 +0000 (10:10 +0100)]
net: stmmac: dwmac-imx: set TSO/TBS TX queues default settings

TSO and TBS cannot coexist. For now we set i.MX Ethernet QOS controller to
use the first TX queue with TSO and the rest for TBS.

TX queues with TBS can support etf qdisc hw offload.

Signed-off-by: Esben Haabendal <esben@geanix.com>
Reviewed-by: Kurt Kanzenbach <kurt@linutronix.de>
Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
7 months agonet: stmmac: do not clear TBS enable bit on link up/down
Esben Haabendal [Fri, 26 Jan 2024 09:10:41 +0000 (10:10 +0100)]
net: stmmac: do not clear TBS enable bit on link up/down

With the dma conf being reallocated on each call to stmmac_open(), any
information in there is lost, unless we specifically handle it.

The STMMAC_TBS_EN bit is set when adding an etf qdisc, and the etf qdisc
therefore would stop working when link was set down and then back up.

Fixes: ba39b344e924 ("net: ethernet: stmicro: stmmac: generate stmmac dma conf before open")
Cc: stable@vger.kernel.org
Signed-off-by: Esben Haabendal <esben@geanix.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
7 months agoipv6: Ensure natural alignment of const ipv6 loopback and router addresses
Helge Deller [Fri, 26 Jan 2024 08:32:20 +0000 (09:32 +0100)]
ipv6: Ensure natural alignment of const ipv6 loopback and router addresses

On a parisc64 kernel I sometimes notice this kernel warning:
Kernel unaligned access to 0x40ff8814 at ndisc_send_skb+0xc0/0x4d8

The address 0x40ff8814 points to the in6addr_linklocal_allrouters
variable and the warning simply means that some ipv6 function tries to
read a 64-bit word directly from the not-64-bit aligned
in6addr_linklocal_allrouters variable.

Unaligned accesses are non-critical as the architecture or exception
handlers usually will fix it up at runtime. Nevertheless it may trigger
a performance penality for some architectures. For details read the
"unaligned-memory-access" kernel documentation.

The patch below ensures that the ipv6 loopback and router addresses will
always be naturally aligned. This prevents the unaligned accesses for
all architectures.

Signed-off-by: Helge Deller <deller@gmx.de>
Fixes: 034dfc5df99eb ("ipv6: export in6addr_loopback to modules")
Acked-by: Paolo Abeni <pabeni@redhat.com>
Link: https://lore.kernel.org/r/ZbNuFM1bFqoH-UoY@p100
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
7 months agoselftests: net: add missing config for nftables-backed iptables
Jakub Kicinski [Fri, 26 Jan 2024 20:13:08 +0000 (12:13 -0800)]
selftests: net: add missing config for nftables-backed iptables

Modern OSes use iptables implementation with nf_tables as a backend,
e.g.:

$ iptables -V
iptables v1.8.8 (nf_tables)

Pablo points out that we need CONFIG_NFT_COMPAT to make that work,
otherwise we see a lot of:

  Warning: Extension DNAT revision 0 not supported, missing kernel module?

with DNAT being just an example here, other modules we need
include udp, TTL, length etc.

Link: https://lore.kernel.org/r/20240126201308.2903602-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agonet: dsa: qca8k: fix illegal usage of GPIO
Michal Vokáč [Fri, 26 Jan 2024 10:49:35 +0000 (11:49 +0100)]
net: dsa: qca8k: fix illegal usage of GPIO

When working with GPIO, its direction must be set either when the GPIO is
requested by gpiod_get*() or later on by one of the gpiod_direction_*()
functions. Neither of this is done here which results in undefined
behavior on some systems.

As the reset GPIO is used right after it is requested here, it makes sense
to configure it as GPIOD_OUT_HIGH right away. With that, the following
gpiod_set_value_cansleep(1) becomes redundant and can be safely
removed.

Fixes: a653f2f538f9 ("net: dsa: qca8k: introduce reset via gpio feature")
Signed-off-by: Michal Vokáč <michal.vokac@ysoft.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/1706266175-3408-1-git-send-email-michal.vokac@ysoft.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agotcp: add sanity checks to rx zerocopy
Eric Dumazet [Thu, 25 Jan 2024 10:33:17 +0000 (10:33 +0000)]
tcp: add sanity checks to rx zerocopy

TCP rx zerocopy intent is to map pages initially allocated
from NIC drivers, not pages owned by a fs.

This patch adds to can_map_frag() these additional checks:

- Page must not be a compound one.
- page->mapping must be NULL.

This fixes the panic reported by ZhangPeng.

syzbot was able to loopback packets built with sendfile(),
mapping pages owned by an ext4 file to TCP rx zerocopy.

r3 = socket$inet_tcp(0x2, 0x1, 0x0)
mmap(&(0x7f0000ff9000/0x4000)=nil, 0x4000, 0x0, 0x12, r3, 0x0)
r4 = socket$inet_tcp(0x2, 0x1, 0x0)
bind$inet(r4, &(0x7f0000000000)={0x2, 0x4e24, @multicast1}, 0x10)
connect$inet(r4, &(0x7f00000006c0)={0x2, 0x4e24, @empty}, 0x10)
r5 = openat$dir(0xffffffffffffff9c, &(0x7f00000000c0)='./file0\x00',
    0x181e42, 0x0)
fallocate(r5, 0x0, 0x0, 0x85b8)
sendfile(r4, r5, 0x0, 0x8ba0)
getsockopt$inet_tcp_TCP_ZEROCOPY_RECEIVE(r4, 0x6, 0x23,
    &(0x7f00000001c0)={&(0x7f0000ffb000/0x3000)=nil, 0x3000, 0x0, 0x0, 0x0,
    0x0, 0x0, 0x0, 0x0}, &(0x7f0000000440)=0x40)
r6 = openat$dir(0xffffffffffffff9c, &(0x7f00000000c0)='./file0\x00',
    0x181e42, 0x0)

Fixes: 93ab6cc69162 ("tcp: implement mmap() for zero copy receive")
Link: https://lore.kernel.org/netdev/5106a58e-04da-372a-b836-9d3d0bd2507b@huawei.com/T/
Reported-and-bisected-by: ZhangPeng <zhangpeng362@huawei.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Arjun Roy <arjunroy@google.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: linux-mm@vger.kernel.org
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: linux-fsdevel@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>
7 months agonfc: nci: free rx_data_reassembly skb on NCI device cleanup
Fedor Pchelkin [Thu, 25 Jan 2024 09:53:09 +0000 (12:53 +0300)]
nfc: nci: free rx_data_reassembly skb on NCI device cleanup

rx_data_reassembly skb is stored during NCI data exchange for processing
fragmented packets. It is dropped only when the last fragment is processed
or when an NTF packet with NCI_OP_RF_DEACTIVATE_NTF opcode is received.
However, the NCI device may be deallocated before that which leads to skb
leak.

As by design the rx_data_reassembly skb is bound to the NCI device and
nothing prevents the device to be freed before the skb is processed in
some way and cleaned, free it on the NCI device cleanup.

Found by Linux Verification Center (linuxtesting.org) with Syzkaller.

Fixes: 6a2968aaf50c ("NFC: basic NCI protocol implementation")
Cc: stable@vger.kernel.org
Reported-by: syzbot+6b7c68d9c21e4ee4251b@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/lkml/000000000000f43987060043da7b@google.com/
Signed-off-by: Fedor Pchelkin <pchelkin@ispras.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
7 months agonet: hsr: remove WARN_ONCE() in send_hsr_supervision_frame()
Nikita Zhandarovich [Wed, 24 Jan 2024 10:21:47 +0000 (02:21 -0800)]
net: hsr: remove WARN_ONCE() in send_hsr_supervision_frame()

Syzkaller reported [1] hitting a warning after failing to allocate
resources for skb in hsr_init_skb(). Since a WARN_ONCE() call will
not help much in this case, it might be prudent to switch to
netdev_warn_once(). At the very least it will suppress syzkaller
reports such as [1].

Just in case, use netdev_warn_once() in send_prp_supervision_frame()
for similar reasons.

[1]
HSR: Could not send supervision frame
WARNING: CPU: 1 PID: 85 at net/hsr/hsr_device.c:294 send_hsr_supervision_frame+0x60a/0x810 net/hsr/hsr_device.c:294
RIP: 0010:send_hsr_supervision_frame+0x60a/0x810 net/hsr/hsr_device.c:294
...
Call Trace:
 <IRQ>
 hsr_announce+0x114/0x370 net/hsr/hsr_device.c:382
 call_timer_fn+0x193/0x590 kernel/time/timer.c:1700
 expire_timers kernel/time/timer.c:1751 [inline]
 __run_timers+0x764/0xb20 kernel/time/timer.c:2022
 run_timer_softirq+0x58/0xd0 kernel/time/timer.c:2035
 __do_softirq+0x21a/0x8de kernel/softirq.c:553
 invoke_softirq kernel/softirq.c:427 [inline]
 __irq_exit_rcu kernel/softirq.c:632 [inline]
 irq_exit_rcu+0xb7/0x120 kernel/softirq.c:644
 sysvec_apic_timer_interrupt+0x95/0xb0 arch/x86/kernel/apic/apic.c:1076
 </IRQ>
 <TASK>
 asm_sysvec_apic_timer_interrupt+0x1a/0x20 arch/x86/include/asm/idtentry.h:649
...

This issue is also found in older kernels (at least up to 5.10).

Cc: stable@vger.kernel.org
Reported-by: syzbot+3ae0a3f42c84074b7c8e@syzkaller.appspotmail.com
Fixes: 121c33b07b31 ("net: hsr: introduce common code for skb initialization")
Signed-off-by: Nikita Zhandarovich <n.zhandarovich@fintech.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
7 months agonet: lan966x: Fix port configuration when using SGMII interface
Horatiu Vultur [Wed, 24 Jan 2024 10:17:58 +0000 (11:17 +0100)]
net: lan966x: Fix port configuration when using SGMII interface

In case the interface between the MAC and the PHY is SGMII, then the bit
GIGA_MODE on the MAC side needs to be set regardless of the speed at
which it is running.

Fixes: d28d6d2e37d1 ("net: lan966x: add port module support")
Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Reviewed-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
7 months agoMAINTAINERS: Add connector headers to NETWORKING DRIVERS
Simon Horman [Wed, 24 Jan 2024 09:16:22 +0000 (09:16 +0000)]
MAINTAINERS: Add connector headers to NETWORKING DRIVERS

Commit 46cf789b68b2 ("connector: Move maintainence under networking
drivers umbrella.") moved the connector maintenance but did not include
the connector header files.

It seems that it has always been implied that these headers were
maintained along with the rest of the connector code, both before and
after the cited commit. Make this explicit.

Signed-off-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
7 months agoipmr: fix kernel panic when forwarding mcast packets
Nicolas Dichtel [Thu, 25 Jan 2024 14:18:47 +0000 (15:18 +0100)]
ipmr: fix kernel panic when forwarding mcast packets

The stacktrace was:
[   86.305548] BUG: kernel NULL pointer dereference, address: 0000000000000092
[   86.306815] #PF: supervisor read access in kernel mode
[   86.307717] #PF: error_code(0x0000) - not-present page
[   86.308624] PGD 0 P4D 0
[   86.309091] Oops: 0000 [#1] PREEMPT SMP NOPTI
[   86.309883] CPU: 2 PID: 3139 Comm: pimd Tainted: G     U             6.8.0-6wind-knet #1
[   86.311027] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.11.1-0-g0551a4be2c-prebuilt.qemu-project.org 04/01/2014
[   86.312728] RIP: 0010:ip_mr_forward (/build/work/knet/net/ipv4/ipmr.c:1985)
[ 86.313399] Code: f9 1f 0f 87 85 03 00 00 48 8d 04 5b 48 8d 04 83 49 8d 44 c5 00 48 8b 40 70 48 39 c2 0f 84 d9 00 00 00 49 8b 46 58 48 83 e0 fe <80> b8 92 00 00 00 00 0f 84 55 ff ff ff 49 83 47 38 01 45 85 e4 0f
[   86.316565] RSP: 0018:ffffad21c0583ae0 EFLAGS: 00010246
[   86.317497] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
[   86.318596] RDX: ffff9559cb46c000 RSI: 0000000000000000 RDI: 0000000000000000
[   86.319627] RBP: ffffad21c0583b30 R08: 0000000000000000 R09: 0000000000000000
[   86.320650] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000001
[   86.321672] R13: ffff9559c093a000 R14: ffff9559cc00b800 R15: ffff9559c09c1d80
[   86.322873] FS:  00007f85db661980(0000) GS:ffff955a79d00000(0000) knlGS:0000000000000000
[   86.324291] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   86.325314] CR2: 0000000000000092 CR3: 000000002f13a000 CR4: 0000000000350ef0
[   86.326589] Call Trace:
[   86.327036]  <TASK>
[   86.327434] ? show_regs (/build/work/knet/arch/x86/kernel/dumpstack.c:479)
[   86.328049] ? __die (/build/work/knet/arch/x86/kernel/dumpstack.c:421 /build/work/knet/arch/x86/kernel/dumpstack.c:434)
[   86.328508] ? page_fault_oops (/build/work/knet/arch/x86/mm/fault.c:707)
[   86.329107] ? do_user_addr_fault (/build/work/knet/arch/x86/mm/fault.c:1264)
[   86.329756] ? srso_return_thunk (/build/work/knet/arch/x86/lib/retpoline.S:223)
[   86.330350] ? __irq_work_queue_local (/build/work/knet/kernel/irq_work.c:111 (discriminator 1))
[   86.331013] ? exc_page_fault (/build/work/knet/./arch/x86/include/asm/paravirt.h:693 /build/work/knet/arch/x86/mm/fault.c:1515 /build/work/knet/arch/x86/mm/fault.c:1563)
[   86.331702] ? asm_exc_page_fault (/build/work/knet/./arch/x86/include/asm/idtentry.h:570)
[   86.332468] ? ip_mr_forward (/build/work/knet/net/ipv4/ipmr.c:1985)
[   86.333183] ? srso_return_thunk (/build/work/knet/arch/x86/lib/retpoline.S:223)
[   86.333920] ipmr_mfc_add (/build/work/knet/./include/linux/rcupdate.h:782 /build/work/knet/net/ipv4/ipmr.c:1009 /build/work/knet/net/ipv4/ipmr.c:1273)
[   86.334583] ? __pfx_ipmr_hash_cmp (/build/work/knet/net/ipv4/ipmr.c:363)
[   86.335357] ip_mroute_setsockopt (/build/work/knet/net/ipv4/ipmr.c:1470)
[   86.336135] ? srso_return_thunk (/build/work/knet/arch/x86/lib/retpoline.S:223)
[   86.336854] ? ip_mroute_setsockopt (/build/work/knet/net/ipv4/ipmr.c:1470)
[   86.337679] do_ip_setsockopt (/build/work/knet/net/ipv4/ip_sockglue.c:944)
[   86.338408] ? __pfx_unix_stream_read_actor (/build/work/knet/net/unix/af_unix.c:2862)
[   86.339232] ? srso_return_thunk (/build/work/knet/arch/x86/lib/retpoline.S:223)
[   86.339809] ? aa_sk_perm (/build/work/knet/security/apparmor/include/cred.h:153 /build/work/knet/security/apparmor/net.c:181)
[   86.340342] ip_setsockopt (/build/work/knet/net/ipv4/ip_sockglue.c:1415)
[   86.340859] raw_setsockopt (/build/work/knet/net/ipv4/raw.c:836)
[   86.341408] ? security_socket_setsockopt (/build/work/knet/security/security.c:4561 (discriminator 13))
[   86.342116] sock_common_setsockopt (/build/work/knet/net/core/sock.c:3716)
[   86.342747] do_sock_setsockopt (/build/work/knet/net/socket.c:2313)
[   86.343363] __sys_setsockopt (/build/work/knet/./include/linux/file.h:32 /build/work/knet/net/socket.c:2336)
[   86.344020] __x64_sys_setsockopt (/build/work/knet/net/socket.c:2340)
[   86.344766] do_syscall_64 (/build/work/knet/arch/x86/entry/common.c:52 /build/work/knet/arch/x86/entry/common.c:83)
[   86.345433] ? srso_return_thunk (/build/work/knet/arch/x86/lib/retpoline.S:223)
[   86.346161] ? syscall_exit_work (/build/work/knet/./include/linux/audit.h:357 /build/work/knet/kernel/entry/common.c:160)
[   86.346938] ? srso_return_thunk (/build/work/knet/arch/x86/lib/retpoline.S:223)
[   86.347657] ? syscall_exit_to_user_mode (/build/work/knet/kernel/entry/common.c:215)
[   86.348538] ? srso_return_thunk (/build/work/knet/arch/x86/lib/retpoline.S:223)
[   86.349262] ? do_syscall_64 (/build/work/knet/./arch/x86/include/asm/cpufeature.h:171 /build/work/knet/arch/x86/entry/common.c:98)
[   86.349971] entry_SYSCALL_64_after_hwframe (/build/work/knet/arch/x86/entry/entry_64.S:129)

The original packet in ipmr_cache_report() may be queued and then forwarded
with ip_mr_forward(). This last function has the assumption that the skb
dst is set.

After the below commit, the skb dst is dropped by ipv4_pktinfo_prepare(),
which causes the oops.

Fixes: bb7403655b3c ("ipmr: support IP_PKTINFO on cache report IGMP msg")
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20240125141847.1931933-1-nicolas.dichtel@6wind.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agonet: dsa: mt7530: fix 10M/100M speed on MT7988 switch
Daniel Golle [Wed, 24 Jan 2024 05:17:25 +0000 (05:17 +0000)]
net: dsa: mt7530: fix 10M/100M speed on MT7988 switch

Setup PMCR port register for actual speed and duplex on internally
connected PHYs of the MT7988 built-in switch. This fixes links with
speeds other than 1000M.

Fixes: 110c18bfed41 ("net: dsa: mt7530: introduce driver for MT7988 built-in switch")
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Acked-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Link: https://lore.kernel.org/r/a5b04dfa8256d8302f402545a51ac4c626fdba25.1706071272.git.daniel@makrotopia.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agoip6_tunnel: make sure to pull inner header in __ip6_tnl_rcv()
Eric Dumazet [Thu, 25 Jan 2024 17:05:57 +0000 (17:05 +0000)]
ip6_tunnel: make sure to pull inner header in __ip6_tnl_rcv()

syzbot found __ip6_tnl_rcv() could access unitiliazed data [1].

Call pskb_inet_may_pull() to fix this, and initialize ipv6h
variable after this call as it can change skb->head.

[1]
 BUG: KMSAN: uninit-value in __INET_ECN_decapsulate include/net/inet_ecn.h:253 [inline]
 BUG: KMSAN: uninit-value in INET_ECN_decapsulate include/net/inet_ecn.h:275 [inline]
 BUG: KMSAN: uninit-value in IP6_ECN_decapsulate+0x7df/0x1e50 include/net/inet_ecn.h:321
  __INET_ECN_decapsulate include/net/inet_ecn.h:253 [inline]
  INET_ECN_decapsulate include/net/inet_ecn.h:275 [inline]
  IP6_ECN_decapsulate+0x7df/0x1e50 include/net/inet_ecn.h:321
  ip6ip6_dscp_ecn_decapsulate+0x178/0x1b0 net/ipv6/ip6_tunnel.c:727
  __ip6_tnl_rcv+0xd4e/0x1590 net/ipv6/ip6_tunnel.c:845
  ip6_tnl_rcv+0xce/0x100 net/ipv6/ip6_tunnel.c:888
 gre_rcv+0x143f/0x1870
  ip6_protocol_deliver_rcu+0xda6/0x2a60 net/ipv6/ip6_input.c:438
  ip6_input_finish net/ipv6/ip6_input.c:483 [inline]
  NF_HOOK include/linux/netfilter.h:314 [inline]
  ip6_input+0x15d/0x430 net/ipv6/ip6_input.c:492
  ip6_mc_input+0xa7e/0xc80 net/ipv6/ip6_input.c:586
  dst_input include/net/dst.h:461 [inline]
  ip6_rcv_finish+0x5db/0x870 net/ipv6/ip6_input.c:79
  NF_HOOK include/linux/netfilter.h:314 [inline]
  ipv6_rcv+0xda/0x390 net/ipv6/ip6_input.c:310
  __netif_receive_skb_one_core net/core/dev.c:5532 [inline]
  __netif_receive_skb+0x1a6/0x5a0 net/core/dev.c:5646
  netif_receive_skb_internal net/core/dev.c:5732 [inline]
  netif_receive_skb+0x58/0x660 net/core/dev.c:5791
  tun_rx_batched+0x3ee/0x980 drivers/net/tun.c:1555
  tun_get_user+0x53af/0x66d0 drivers/net/tun.c:2002
  tun_chr_write_iter+0x3af/0x5d0 drivers/net/tun.c:2048
  call_write_iter include/linux/fs.h:2084 [inline]
  new_sync_write fs/read_write.c:497 [inline]
  vfs_write+0x786/0x1200 fs/read_write.c:590
  ksys_write+0x20f/0x4c0 fs/read_write.c:643
  __do_sys_write fs/read_write.c:655 [inline]
  __se_sys_write fs/read_write.c:652 [inline]
  __x64_sys_write+0x93/0xd0 fs/read_write.c:652
  do_syscall_x64 arch/x86/entry/common.c:52 [inline]
  do_syscall_64+0x6d/0x140 arch/x86/entry/common.c:83
 entry_SYSCALL_64_after_hwframe+0x63/0x6b

Uninit was created at:
  slab_post_alloc_hook+0x129/0xa70 mm/slab.h:768
  slab_alloc_node mm/slub.c:3478 [inline]
  kmem_cache_alloc_node+0x5e9/0xb10 mm/slub.c:3523
  kmalloc_reserve+0x13d/0x4a0 net/core/skbuff.c:560
  __alloc_skb+0x318/0x740 net/core/skbuff.c:651
  alloc_skb include/linux/skbuff.h:1286 [inline]
  alloc_skb_with_frags+0xc8/0xbd0 net/core/skbuff.c:6334
  sock_alloc_send_pskb+0xa80/0xbf0 net/core/sock.c:2787
  tun_alloc_skb drivers/net/tun.c:1531 [inline]
  tun_get_user+0x1e8a/0x66d0 drivers/net/tun.c:1846
  tun_chr_write_iter+0x3af/0x5d0 drivers/net/tun.c:2048
  call_write_iter include/linux/fs.h:2084 [inline]
  new_sync_write fs/read_write.c:497 [inline]
  vfs_write+0x786/0x1200 fs/read_write.c:590
  ksys_write+0x20f/0x4c0 fs/read_write.c:643
  __do_sys_write fs/read_write.c:655 [inline]
  __se_sys_write fs/read_write.c:652 [inline]
  __x64_sys_write+0x93/0xd0 fs/read_write.c:652
  do_syscall_x64 arch/x86/entry/common.c:52 [inline]
  do_syscall_64+0x6d/0x140 arch/x86/entry/common.c:83
 entry_SYSCALL_64_after_hwframe+0x63/0x6b

CPU: 0 PID: 5034 Comm: syz-executor331 Not tainted 6.7.0-syzkaller-00562-g9f8413c4a66f #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 11/17/2023

Fixes: 0d3c703a9d17 ("ipv6: Cleanup IPv6 tunnel receive path")
Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240125170557.2663942-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agoselftests: net: give more time for GRO aggregation
Paolo Abeni [Thu, 25 Jan 2024 18:09:06 +0000 (19:09 +0100)]
selftests: net: give more time for GRO aggregation

The gro.sh test-case relay on the gro_flush_timeout to ensure
that all the segments belonging to any given batch are properly
aggregated.

The other end, the sender is a user-space program transmitting
each packet with a separate write syscall. A busy host and/or
stracing the sender program can make the relevant segments reach
the GRO engine after the flush timeout triggers.

Give the GRO flush timeout more slack, to avoid sporadic self-tests
failures.

Fixes: 9af771d2ec04 ("selftests/net: allow GRO coalesce test on veth")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Tested-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/bffec2beab3a5672dd13ecabe4fad81d2155b367.1706206101.git.pabeni@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agoselftests: net: add missing required classifier
Paolo Abeni [Thu, 25 Jan 2024 08:22:50 +0000 (09:22 +0100)]
selftests: net: add missing required classifier

the udpgro_fraglist self-test uses the BPF classifiers, but the
current net self-test configuration does not include it, causing
CI failures:

 # selftests: net: udpgro_frglist.sh
 # ipv6
 # tcp - over veth touching data
 # -l 4 -6 -D 2001:db8::1 -t rx -4 -t
 # Error: TC classifier not found.
 # We have an error talking to the kernel
 # Error: TC classifier not found.
 # We have an error talking to the kernel

Add the missing knob.

Fixes: edae34a3ed92 ("selftests net: add UDP GRO fraglist + bpf self-tests")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Maciej Żenczykowski <maze@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/7c3643763b331e9a400e1874fe089193c99a1c3f.1706170897.git.pabeni@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agobnxt_en: Make PTP timestamp HWRM more silent
Breno Leitao [Thu, 25 Jan 2024 13:41:03 +0000 (05:41 -0800)]
bnxt_en: Make PTP timestamp HWRM more silent

commit 056bce63c469 ("bnxt_en: Make PTP TX timestamp HWRM query silent")
changed a netdev_err() to netdev_WARN_ONCE().

netdev_WARN_ONCE() is it generates a kernel WARNING, which is bad, for
the following reasons:

 * You do not a kernel warning if the firmware queries are late
 * In busy networks, timestamp query failures fairly regularly
 * A WARNING message doesn't bring much value, since the code path
is clear.
(This was discussed in-depth in [1])

Transform the netdev_WARN_ONCE() into a netdev_warn_once(), and print a
more well-behaved message, instead of a full WARN().

bnxt_en 0000:67:00.0 eth0: TS query for TX timer failed rc = fffffff5

Link: https://lore.kernel.org/all/ZbDj%2FFI4EJezcfd1@gmail.com/
Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Reviewed-by: Michael Chan <michael.chan@broadcom.com>
Fixes: 056bce63c469 ("bnxt_en: Make PTP TX timestamp HWRM query silent")
Link: https://lore.kernel.org/r/20240125134104.2045573-1-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agonet/smc: fix incorrect SMC-D link group matching logic
Wen Gu [Thu, 25 Jan 2024 12:39:16 +0000 (20:39 +0800)]
net/smc: fix incorrect SMC-D link group matching logic

The logic to determine if SMC-D link group matches is incorrect. The
correct logic should be that it only returns true when the GID is the
same, and the SMC-D device is the same and the extended GID is the same
(in the case of virtual ISM).

It can be fixed by adding brackets around the conditional (or ternary)
operator expression. But for better readability and maintainability, it
has been changed to an if-else statement.

Reported-by: Matthew Rosato <mjrosato@linux.ibm.com>
Closes: https://lore.kernel.org/r/13579588-eb9d-4626-a063-c0b77ed80f11@linux.ibm.com
Fixes: b40584d14570 ("net/smc: compatible with 128-bits extended GID of virtual ISM device")
Link: https://lore.kernel.org/r/13579588-eb9d-4626-a063-c0b77ed80f11@linux.ibm.com
Signed-off-by: Wen Gu <guwen@linux.alibaba.com>
Reviewed-by: Alexandra Winter <wintera@linux.ibm.com>
Link: https://lore.kernel.org/r/20240125123916.77928-1-guwen@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agoselftests: net: add missing config for big tcp tests
Paolo Abeni [Fri, 26 Jan 2024 15:32:36 +0000 (16:32 +0100)]
selftests: net: add missing config for big tcp tests

The big_tcp test-case requires a few kernel knobs currently
not specified in the net selftests config, causing the
following failure:

  # selftests: net: big_tcp.sh
  # Error: Failed to load TC action module.
  # We have an error talking to the kernel
...
  # Testing for BIG TCP:
  # CLI GSO | GW GRO | GW GSO | SER GRO
  # ./big_tcp.sh: line 107: test: !=: unary operator expected
...
  # on        on       on       on      : [FAIL_on_link1]

Add the missing configs

Fixes: 6bb382bcf742 ("selftests: add a selftest for big tcp")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Aaron Conole <aconole@redhat.com>
Acked-by: Xin Long <lucien.xin@gmail.com>
Link: https://lore.kernel.org/all/21630ecea872fea13f071342ac64ef52a991a9b5.1706282943.git.pabeni@redhat.com/
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agonet: phy: mediatek-ge-soc: sync driver with MediaTek SDK
Daniel Golle [Wed, 24 Jan 2024 05:18:23 +0000 (05:18 +0000)]
net: phy: mediatek-ge-soc: sync driver with MediaTek SDK

Sync initialization and calibration routines with MediaTek's reference
driver. Improves compliance and resolves link stability issues with
CH340 IoT devices connected to MT798x built-in PHYs.

Fixes: 98c485eaf509 ("net: phy: add driver for MediaTek SoC built-in GE PHYs")
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Link: https://lore.kernel.org/r/f2195279c234c0f618946424b8236026126bc595.1706071311.git.daniel@makrotopia.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agonet: ethernet: mtk_eth_soc: set DMA coherent mask to get PPE working
Daniel Golle [Wed, 24 Jan 2024 16:22:09 +0000 (16:22 +0000)]
net: ethernet: mtk_eth_soc: set DMA coherent mask to get PPE working

Set DMA coherent mask to 32-bit which makes PPE offloading engine start
working on BPi-R4 which got 4 GiB of RAM.

Fixes: 2d75891ebc09 ("net: ethernet: mtk_eth_soc: support 36-bit DMA addressing on MT7988")
Suggested-by: Elad Yifee <eladwf@users.github.com>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Link: https://lore.kernel.org/r/97e90925368b405f0974b9b15f1b7377c4a329ad.1706113251.git.daniel@makrotopia.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agoMerge branch 'nfp-flower-a-few-small-conntrack-offload-fixes'
Jakub Kicinski [Fri, 26 Jan 2024 01:17:30 +0000 (17:17 -0800)]
Merge branch 'nfp-flower-a-few-small-conntrack-offload-fixes'

Louis Peens says:

====================
nfp: flower: a few small conntrack offload fixes

This small series addresses two bugs in the nfp conntrack offloading
code.

The first patch is a check to prevent offloading for a case which is
currently not supported by the nfp.

The second patch fixes up parsing of layer4 mangling code so it can be
correctly offloaded. Since the masks are an inverse mask and we are
shifting it so it can be packed together with the destination we
effectively need to 'clear' the lower bits of the mask by setting it to
0xFFFF.
====================

Link: https://lore.kernel.org/r/20240124151909.31603-1-louis.peens@corigine.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agonfp: flower: fix hardware offload for the transfer layer port
Hui Zhou [Wed, 24 Jan 2024 15:19:09 +0000 (17:19 +0200)]
nfp: flower: fix hardware offload for the transfer layer port

The nfp driver will merge the tp source port and tp destination port
into one dword which the offset must be zero to do hardware offload.
However, the mangle action for the tp source port and tp destination
port is separated for tc ct action. Modify the mangle action for the
FLOW_ACT_MANGLE_HDR_TYPE_TCP and FLOW_ACT_MANGLE_HDR_TYPE_UDP to
satisfy the nfp driver offload check for the tp port.

The mangle action provides a 4B value for source, and a 4B value for
the destination, but only 2B of each contains the useful information.
For offload the 2B of each is combined into a single 4B word. Since the
incoming mask for the source is '0xFFFF<mask>' the shift-left will
throw away the 0xFFFF part. When this gets combined together in the
offload it will clear the destination field. Fix this by setting the
lower bits back to 0xFFFF, effectively doing a rotate-left operation on
the mask.

Fixes: 5cee92c6f57a ("nfp: flower: support hw offload for ct nat action")
CC: stable@vger.kernel.org # 6.1+
Signed-off-by: Hui Zhou <hui.zhou@corigine.com>
Signed-off-by: Louis Peens <louis.peens@corigine.com>
Link: https://lore.kernel.org/r/20240124151909.31603-3-louis.peens@corigine.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agonfp: flower: add hardware offload check for post ct entry
Hui Zhou [Wed, 24 Jan 2024 15:19:08 +0000 (17:19 +0200)]
nfp: flower: add hardware offload check for post ct entry

The nfp offload flow pay will not allocate a mask id when the out port
is openvswitch internal port. This is because these flows are used to
configure the pre_tun table and are never actually send to the firmware
as an add-flow message. When a tc rule which action contains ct and
the post ct entry's out port is openvswitch internal port, the merge
offload flow pay with the wrong mask id of 0 will be send to the
firmware. Actually, the nfp can not support hardware offload for this
situation, so return EOPNOTSUPP.

Fixes: bd0fe7f96a3c ("nfp: flower-ct: add zone table entry when handling pre/post_ct flows")
CC: stable@vger.kernel.org # 5.14+
Signed-off-by: Hui Zhou <hui.zhou@corigine.com>
Signed-off-by: Louis Peens <louis.peens@corigine.com>
Link: https://lore.kernel.org/r/20240124151909.31603-2-louis.peens@corigine.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agogve: Fix skb truesize underestimation
Praveen Kaligineedi [Wed, 24 Jan 2024 16:10:25 +0000 (08:10 -0800)]
gve: Fix skb truesize underestimation

For a skb frag with a newly allocated copy page, the true size is
incorrectly set to packet buffer size. It should be set to PAGE_SIZE
instead.

Fixes: 82fd151d38d9 ("gve: Reduce alloc and copy costs in the GQ rx path")
Signed-off-by: Praveen Kaligineedi <pkaligineedi@google.com>
Link: https://lore.kernel.org/r/20240124161025.1819836-1-pkaligineedi@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agoselftests/net/lib: update busywait timeout value
Hangbin Liu [Wed, 24 Jan 2024 06:13:44 +0000 (14:13 +0800)]
selftests/net/lib: update busywait timeout value

The busywait timeout value is a millisecond, not a second. So the
current setting 2 is too small. On slow/busy host (or VMs) the
current timeout can expire even on "correct" execution, causing random
failures. Let's copy the WAIT_TIMEOUT from forwarding/lib.sh and set
BUSYWAIT_TIMEOUT here.

Fixes: 25ae948b4478 ("selftests/net: add lib.sh")
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240124061344.1864484-1-liuhangbin@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agoselftests: tcp_ao: set the timeout to 2 minutes
Jakub Kicinski [Wed, 24 Jan 2024 23:36:30 +0000 (15:36 -0800)]
selftests: tcp_ao: set the timeout to 2 minutes

The default timeout for tests is 45sec, bench-lookups_ipv6
seems to take around 50sec when running in a VM without
HW acceleration. Give it a 2x margin and set the timeout
to 120sec.

Fixes: d1066c9c58d4 ("selftests/net: Add test/benchmark for removing MKTs")
Reviewed-by: Dmitry Safonov <0x7f454c46@gmail.com>
Link: https://lore.kernel.org/r/20240124233630.1977708-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agoMerge branch 'selftests-net-a-few-fixes'
Jakub Kicinski [Thu, 25 Jan 2024 23:59:24 +0000 (15:59 -0800)]
Merge branch 'selftests-net-a-few-fixes'

Paolo Abeni says:

====================
selftests: net: a few fixes

This series address self-tests failures for udp gro-related tests.

The first patch addresses the main problem I observe locally - the XDP
program required by such tests, xdp_dummy, is currently build in the
ebpf self-tests directory, not available if/when the user targets net
only. Arguably is more a refactor than a fix, but still targeting net
to hopefully

The second patch fixes the integration of such tests with the build
system.

Patch 3/3 fixes sporadic failures due to races.

Tested with:

make -C tools/testing/selftests/ TARGETS=net install
./tools/testing/selftests/kselftest_install/run_kselftest.sh \
-t "net:udpgro_bench.sh net:udpgro.sh net:udpgro_fwd.sh \
    net:udpgro_frglist.sh net:veth.sh"

no failures.
====================

Link: https://lore.kernel.org/r/cover.1706131762.git.pabeni@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agoselftests: net: explicitly wait for listener ready
Paolo Abeni [Wed, 24 Jan 2024 21:33:22 +0000 (22:33 +0100)]
selftests: net: explicitly wait for listener ready

The UDP GRO forwarding test still hard-code an arbitrary pause
to wait for the UDP listener becoming ready in background.

That causes sporadic failures depending on the host load.

Replace the sleep with the existing helper waiting for the desired
port being exposed.

Fixes: a062260a9d5f ("selftests: net: add UDP GRO forwarding self-tests")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/4d58900fb09cef42749cfcf2ad7f4b91a97d225c.1706131762.git.pabeni@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agoselftests: net: included needed helper in the install targets
Paolo Abeni [Wed, 24 Jan 2024 21:33:21 +0000 (22:33 +0100)]
selftests: net: included needed helper in the install targets

The blamed commit below introduce a dependency in some net self-tests
towards a newly introduce helper script.

Such script is currently not included into the TEST_PROGS_EXTENDED list
and thus is not installed, causing failure for the relevant tests when
executed from the install dir.

Fix the issue updating the install targets.

Fixes: 3bdd9fd29cb0 ("selftests/net: synchronize udpgro tests' tx and rx connection")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/076e8758e21ff2061cc9f81640e7858df775f0a9.1706131762.git.pabeni@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agoselftests: net: remove dependency on ebpf tests
Paolo Abeni [Wed, 24 Jan 2024 21:33:20 +0000 (22:33 +0100)]
selftests: net: remove dependency on ebpf tests

Several net tests requires an XDP program build under the ebpf
directory, and error out if such program is not available.

That makes running successful net test hard, let's duplicate into the
net dir the [very small] program, re-using the existing rules to build
it, and finally dropping the bogus dependency.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/28e7af7c031557f691dc8045ee41dd549dd5e74c.1706131762.git.pabeni@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agoselftests: tcp_ao: add a config file
Jakub Kicinski [Wed, 24 Jan 2024 19:25:50 +0000 (11:25 -0800)]
selftests: tcp_ao: add a config file

Still a bit unclear whether each directory should have its own
config file, but assuming they should lets add one for tcp_ao.

The following tests still fail with this config in place:
 - rst_ipv4,
 - rst_ipv6,
 - bench-lookups_ipv6.
other 21 pass.

Fixes: d11301f65977 ("selftests/net: Add TCP-AO ICMPs accept test")
Reviewed-by: Dmitry Safonov <0x7f454c46@gmail.com>
Link: https://lore.kernel.org/r/20240124192550.1865743-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agoMerge tag 'net-6.8-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Linus Torvalds [Thu, 25 Jan 2024 18:58:35 +0000 (10:58 -0800)]
Merge tag 'net-6.8-rc2' of git://git./linux/kernel/git/netdev/net

Pull networking fixes from Paolo Abeni:
 "Including fixes from bpf, netfilter and WiFi.

  Jakub is doing a lot of work to include the self-tests in our CI, as a
  result a significant amount of self-tests related fixes is flowing in
  (and will likely continue in the next few weeks).

  Current release - regressions:

   - bpf: fix a kernel crash for the riscv 64 JIT

   - bnxt_en: fix memory leak in bnxt_hwrm_get_rings()

   - revert "net: macsec: use skb_ensure_writable_head_tail to expand
     the skb"

  Previous releases - regressions:

   - core: fix removing a namespace with conflicting altnames

   - tc/flower: fix chain template offload memory leak

   - tcp:
      - make sure init the accept_queue's spinlocks once
      - fix autocork on CPUs with weak memory model

   - udp: fix busy polling

   - mlx5e:
      - fix out-of-bound read in port timestamping
      - fix peer flow lists corruption

   - iwlwifi: fix a memory corruption

  Previous releases - always broken:

   - netfilter:
      - nft_chain_filter: handle NETDEV_UNREGISTER for inet/ingress
        basechain
      - nft_limit: reject configurations that cause integer overflow

   - bpf: fix bpf_xdp_adjust_tail() with XSK zero-copy mbuf, avoiding a
     NULL pointer dereference upon shrinking

   - llc: make llc_ui_sendmsg() more robust against bonding changes

   - smc: fix illegal rmb_desc access in SMC-D connection dump

   - dpll: fix pin dump crash for rebound module

   - bnxt_en: fix possible crash after creating sw mqprio TCs

   - hv_netvsc: calculate correct ring size when PAGE_SIZE is not 4kB

  Misc:

   - several self-tests fixes for better integration with the netdev CI

   - added several missing modules descriptions"

* tag 'net-6.8-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (88 commits)
  tsnep: Fix XDP_RING_NEED_WAKEUP for empty fill ring
  tsnep: Remove FCS for XDP data path
  net: fec: fix the unhandled context fault from smmu
  selftests: bonding: do not test arp/ns target with mode balance-alb/tlb
  fjes: fix memleaks in fjes_hw_setup
  i40e: update xdp_rxq_info::frag_size for ZC enabled Rx queue
  i40e: set xdp_rxq_info::frag_size
  xdp: reflect tail increase for MEM_TYPE_XSK_BUFF_POOL
  ice: update xdp_rxq_info::frag_size for ZC enabled Rx queue
  intel: xsk: initialize skb_frag_t::bv_offset in ZC drivers
  ice: remove redundant xdp_rxq_info registration
  i40e: handle multi-buffer packets that are shrunk by xdp prog
  ice: work on pre-XDP prog frag count
  xsk: fix usage of multi-buffer BPF helpers for ZC XDP
  xsk: make xsk_buff_pool responsible for clearing xdp_buff::flags
  xsk: recycle buffer in case Rx queue was full
  net: fill in MODULE_DESCRIPTION()s for rvu_mbox
  net: fill in MODULE_DESCRIPTION()s for litex
  net: fill in MODULE_DESCRIPTION()s for fsl_pq_mdio
  net: fill in MODULE_DESCRIPTION()s for fec
  ...

7 months agoMerge tag 'ovl-fixes-6.8-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/overla...
Linus Torvalds [Thu, 25 Jan 2024 18:52:30 +0000 (10:52 -0800)]
Merge tag 'ovl-fixes-6.8-rc2' of git://git./linux/kernel/git/overlayfs/vfs

Pull overlayfs fix from Amir Goldstein:
 "Change the on-disk format for the new "xwhiteouts" feature introduced
  in v6.7

  The change reduces unneeded overhead of an extra getxattr per readdir.
  The only user of the "xwhiteout" feature is the external composefs
  tool, which has been updated to support the new on-disk format.

  This change is also designated for 6.7.y"

* tag 'ovl-fixes-6.8-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/overlayfs/vfs:
  ovl: mark xwhiteouts directory with overlay.opaque='x'

7 months agoMerge tag 'vfs-6.8-rc2.netfs' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs
Linus Torvalds [Thu, 25 Jan 2024 18:41:29 +0000 (10:41 -0800)]
Merge tag 'vfs-6.8-rc2.netfs' of git://git./linux/kernel/git/vfs/vfs

Pull netfs fixes from Christian Brauner:
 "This contains various fixes for the netfs work merged earlier this
  cycle:

  afs:
   - Fix locking imbalance in afs_proc_addr_prefs_show()
   - Remove afs_dynroot_d_revalidate() which is redundant
   - Fix error handling during lookup
   - Hide sillyrenames from userspace. This fixes a race between
     silly-rename files being created/removed and userspace iterating
     over directory entries
   - Don't use unnecessary folio_*() functions

  cifs:
   - Don't use unnecessary folio_*() functions

  cachefiles:
   - erofs: Fix Null dereference when cachefiles are not doing
     ondemand-mode
   - Update mailing list

  netfs library:
   - Add Jeff Layton as reviewer
   - Update mailing list
   - Fix a error checking in netfs_perform_write()
   - fscache: Check error before dereferencing
   - Don't use unnecessary folio_*() functions"

* tag 'vfs-6.8-rc2.netfs' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
  afs: Fix missing/incorrect unlocking of RCU read lock
  afs: Remove afs_dynroot_d_revalidate() as it is redundant
  afs: Fix error handling with lookup via FS.InlineBulkStatus
  afs: Hide silly-rename files from userspace
  cachefiles, erofs: Fix NULL deref in when cachefiles is not doing ondemand-mode
  netfs: Fix a NULL vs IS_ERR() check in netfs_perform_write()
  netfs, fscache: Prevent Oops in fscache_put_cache()
  cifs: Don't use certain unnecessary folio_*() functions
  afs: Don't use certain unnecessary folio_*() functions
  netfs: Don't use certain unnecessary folio_*() functions
  netfs: Add Jeff Layton as reviewer
  netfs, cachefiles: Change mailing list

7 months agoMerge tag 'nfsd-6.8-1' of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux
Linus Torvalds [Thu, 25 Jan 2024 18:26:52 +0000 (10:26 -0800)]
Merge tag 'nfsd-6.8-1' of git://git./linux/kernel/git/cel/linux

Pull nfsd fixes from Chuck Lever:

 - Fix in-kernel RPC UDP transport

 - Fix NFSv4.0 RELEASE_LOCKOWNER

* tag 'nfsd-6.8-1' of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux:
  nfsd: fix RELEASE_LOCKOWNER
  SUNRPC: use request size to initialize bio_vec in svc_udp_sendto()

7 months agoMerge tag 'urgent-rcu.2024.01.24a' of https://github.com/neeraju/linux
Linus Torvalds [Thu, 25 Jan 2024 18:21:21 +0000 (10:21 -0800)]
Merge tag 'urgent-rcu.2024.01.24a' of https://github.com/neeraju/linux

Pull RCU fix from Neeraj Upadhyay:
 "This fixes RCU grace period stalls, which are observed when an
  outgoing CPU's quiescent state reporting results in wakeup of one of
  the grace period kthreads, to complete the grace period.

  If those kthreads have SCHED_FIFO policy, the wake up can indirectly
  arm the RT bandwith timer to the local offline CPU.

  Earlier migration of the hrtimers from the CPU introduced in commit
  5c0930ccaad5 ("hrtimers: Push pending hrtimers away from outgoing CPU
  earlier") results in this timer getting ignored.

  If the RCU grace period kthreads are waiting for RT bandwidth to be
  available, they may never be actually scheduled, resulting in RCU
  stall warnings"

* tag 'urgent-rcu.2024.01.24a' of https://github.com/neeraju/linux:
  rcu: Defer RCU kthreads wakeup when CPU is dying

7 months agoMerge branch 'tsnep-xdp-fixes'
Paolo Abeni [Thu, 25 Jan 2024 10:59:44 +0000 (11:59 +0100)]
Merge branch 'tsnep-xdp-fixes'

Gerhard Engleder says:

====================
tsnep: XDP fixes

Found two driver specific problems during XDP and XSK testing.
====================

Link: https://lore.kernel.org/r/20240123200918.61219-1-gerhard@engleder-embedded.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
7 months agotsnep: Fix XDP_RING_NEED_WAKEUP for empty fill ring
Gerhard Engleder [Tue, 23 Jan 2024 20:09:18 +0000 (21:09 +0100)]
tsnep: Fix XDP_RING_NEED_WAKEUP for empty fill ring

The fill ring of the XDP socket may contain not enough buffers to
completey fill the RX queue during socket creation. In this case the
flag XDP_RING_NEED_WAKEUP is not set as this flag is only set if the RX
queue is not completely filled during polling.

Set XDP_RING_NEED_WAKEUP flag also if RX queue is not completely filled
during XDP socket creation.

Fixes: 3fc2333933fd ("tsnep: Add XDP socket zero-copy RX support")
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
7 months agotsnep: Remove FCS for XDP data path
Gerhard Engleder [Tue, 23 Jan 2024 20:09:17 +0000 (21:09 +0100)]
tsnep: Remove FCS for XDP data path

The RX data buffer includes the FCS. The FCS is already stripped for the
normal data path. But for the XDP data path the FCS is included and
acts like additional/useless data.

Remove the FCS from the RX data buffer also for XDP.

Fixes: 65b28c810035 ("tsnep: Add XDP RX support")
Fixes: 3fc2333933fd ("tsnep: Add XDP socket zero-copy RX support")
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
7 months agoMerge tag 'mlx5-fixes-2024-01-24' of git://git.kernel.org/pub/scm/linux/kernel/git...
Paolo Abeni [Thu, 25 Jan 2024 10:42:27 +0000 (11:42 +0100)]
Merge tag 'mlx5-fixes-2024-01-24' of git://git./linux/kernel/git/saeed/linux

Saeed Mahameed says:

====================
mlx5 fixes 2024-01-24

This series provides bug fixes to mlx5 driver.
Please pull and let me know if there is any problem.

* tag 'mlx5-fixes-2024-01-24' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux:
  net/mlx5e: fix a potential double-free in fs_any_create_groups
  net/mlx5e: fix a double-free in arfs_create_groups
  net/mlx5e: Ignore IPsec replay window values on sender side
  net/mlx5e: Allow software parsing when IPsec crypto is enabled
  net/mlx5: Use mlx5 device constant for selecting CQ period mode for ASO
  net/mlx5: DR, Can't go to uplink vport on RX rule
  net/mlx5: DR, Use the right GVMI number for drop action
  net/mlx5: Bridge, fix multicast packets sent to uplink
  net/mlx5: Fix a WARN upon a callback command failure
  net/mlx5e: Fix peer flow lists handling
  net/mlx5e: Fix inconsistent hairpin RQT sizes
  net/mlx5e: Fix operation precedence bug in port timestamping napi_poll context
  net/mlx5: Fix query of sd_group field
  net/mlx5e: Use the correct lag ports number when creating TISes
====================

Link: https://lore.kernel.org/r/20240124081855.115410-1-saeed@kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
7 months agoMerge tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf
Paolo Abeni [Thu, 25 Jan 2024 10:30:31 +0000 (11:30 +0100)]
Merge tag 'for-netdev' of https://git./linux/kernel/git/bpf/bpf

Daniel Borkmann says:

====================
pull-request: bpf 2024-01-25

The following pull-request contains BPF updates for your *net* tree.

We've added 12 non-merge commits during the last 2 day(s) which contain
a total of 13 files changed, 190 insertions(+), 91 deletions(-).

The main changes are:

1) Fix bpf_xdp_adjust_tail() in context of XSK zero-copy drivers which
   support XDP multi-buffer. The former triggered a NULL pointer
   dereference upon shrinking, from Maciej Fijalkowski & Tirthendu Sarkar.

2) Fix a bug in riscv64 BPF JIT which emitted a wrong prologue and
   epilogue for struct_ops programs, from Pu Lehui.

* tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf:
  i40e: update xdp_rxq_info::frag_size for ZC enabled Rx queue
  i40e: set xdp_rxq_info::frag_size
  xdp: reflect tail increase for MEM_TYPE_XSK_BUFF_POOL
  ice: update xdp_rxq_info::frag_size for ZC enabled Rx queue
  intel: xsk: initialize skb_frag_t::bv_offset in ZC drivers
  ice: remove redundant xdp_rxq_info registration
  i40e: handle multi-buffer packets that are shrunk by xdp prog
  ice: work on pre-XDP prog frag count
  xsk: fix usage of multi-buffer BPF helpers for ZC XDP
  xsk: make xsk_buff_pool responsible for clearing xdp_buff::flags
  xsk: recycle buffer in case Rx queue was full
  riscv, bpf: Fix unpredictable kernel crash about RV64 struct_ops
====================

Link: https://lore.kernel.org/r/20240125084416.10876-1-daniel@iogearbox.net
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
7 months agonet: fec: fix the unhandled context fault from smmu
Shenwei Wang [Tue, 23 Jan 2024 16:51:41 +0000 (10:51 -0600)]
net: fec: fix the unhandled context fault from smmu

When repeatedly changing the interface link speed using the command below:

ethtool -s eth0 speed 100 duplex full
ethtool -s eth0 speed 1000 duplex full

The following errors may sometimes be reported by the ARM SMMU driver:

[ 5395.035364] fec 5b040000.ethernet eth0: Link is Down
[ 5395.039255] arm-smmu 51400000.iommu: Unhandled context fault:
fsr=0x402, iova=0x00000000, fsynr=0x100001, cbfrsynra=0x852, cb=2
[ 5398.108460] fec 5b040000.ethernet eth0: Link is Up - 100Mbps/Full -
flow control off

It is identified that the FEC driver does not properly stop the TX queue
during the link speed transitions, and this results in the invalid virtual
I/O address translations from the SMMU and causes the context faults.

Fixes: dbc64a8ea231 ("net: fec: move calls to quiesce/resume packet processing out of fec_restart()")
Signed-off-by: Shenwei Wang <shenwei.wang@nxp.com>
Link: https://lore.kernel.org/r/20240123165141.2008104-1-shenwei.wang@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
7 months agoselftests: bonding: do not test arp/ns target with mode balance-alb/tlb
Hangbin Liu [Tue, 23 Jan 2024 07:59:17 +0000 (15:59 +0800)]
selftests: bonding: do not test arp/ns target with mode balance-alb/tlb

The prio_arp/ns tests hard code the mode to active-backup. At the same
time, The balance-alb/tlb modes do not support arp/ns target. So remove
the prio_arp/ns tests from the loop and only test active-backup mode.

Fixes: 481b56e0391e ("selftests: bonding: re-format bond option tests")
Reported-by: Jay Vosburgh <jay.vosburgh@canonical.com>
Closes: https://lore.kernel.org/netdev/17415.1705965957@famine/
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Acked-by: Jay Vosburgh <jay.vosburgh@canonical.com>
Link: https://lore.kernel.org/r/20240123075917.1576360-1-liuhangbin@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
7 months agoMerge tag 'nf-24-01-24' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf
Jakub Kicinski [Thu, 25 Jan 2024 05:03:16 +0000 (21:03 -0800)]
Merge tag 'nf-24-01-24' of git://git./linux/kernel/git/netfilter/nf

Pablo Neira Ayuso says:

====================
Netfilter fixes for net

The following patchset contains Netfilter fixes for net:

1) Update nf_tables kdoc to keep it in sync with the code, from George Guo.

2) Handle NETDEV_UNREGISTER event for inet/ingress basechain.

3) Reject configuration that cause nft_limit to overflow,
   from Florian Westphal.

4) Restrict anonymous set/map names to 16 bytes, from Florian Westphal.

5) Disallow to encode queue number and error in verdicts. This reverts
   a patch which seems to have introduced an early attempt to support for
   nfqueue maps, which is these days supported via nft_queue expression.

6) Sanitize family via .validate for expressions that explicitly refer
   to NF_INET_* hooks.

* tag 'nf-24-01-24' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf:
  netfilter: nf_tables: validate NFPROTO_* family
  netfilter: nf_tables: reject QUEUE/DROP verdict parameters
  netfilter: nf_tables: restrict anonymous set and map names to 16 bytes
  netfilter: nft_limit: reject configurations that cause integer overflow
  netfilter: nft_chain_filter: handle NETDEV_UNREGISTER for inet/ingress basechain
  netfilter: nf_tables: cleanup documentation
====================

Link: https://lore.kernel.org/r/20240124191248.75463-1-pablo@netfilter.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agofjes: fix memleaks in fjes_hw_setup
Zhipeng Lu [Mon, 22 Jan 2024 17:24:42 +0000 (01:24 +0800)]
fjes: fix memleaks in fjes_hw_setup

In fjes_hw_setup, it allocates several memory and delay the deallocation
to the fjes_hw_exit in fjes_probe through the following call chain:

fjes_probe
  |-> fjes_hw_init
        |-> fjes_hw_setup
  |-> fjes_hw_exit

However, when fjes_hw_setup fails, fjes_hw_exit won't be called and thus
all the resources allocated in fjes_hw_setup will be leaked. In this
patch, we free those resources in fjes_hw_setup and prevents such leaks.

Fixes: 2fcbca687702 ("fjes: platform_driver's .probe and .remove routine")
Signed-off-by: Zhipeng Lu <alexious@zju.edu.cn>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240122172445.3841883-1-alexious@zju.edu.cn
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agoMerge tag 'ceph-for-6.8-rc2' of https://github.com/ceph/ceph-client
Linus Torvalds [Thu, 25 Jan 2024 00:59:52 +0000 (16:59 -0800)]
Merge tag 'ceph-for-6.8-rc2' of https://github.com/ceph/ceph-client

Pull ceph fixes from Ilya Dryomov:
 "A fix to avoid triggering an assert in some cases where RBD exclusive
  mappings are involved and a deprecated API cleanup"

* tag 'ceph-for-6.8-rc2' of https://github.com/ceph/ceph-client:
  rbd: don't move requests to the running list on errors
  rbd: remove usage of the deprecated ida_simple_*() API

7 months agoMerge tag 'integrity-v6.8-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/zohar...
Linus Torvalds [Thu, 25 Jan 2024 00:51:59 +0000 (16:51 -0800)]
Merge tag 'integrity-v6.8-rc1' of git://git./linux/kernel/git/zohar/linux-integrity

Pull integrity fix from Mimi Zohar:
 "Revert patch that required user-provided key data, since keys can be
  created from kernel-generated random numbers"

* tag 'integrity-v6.8-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/zohar/linux-integrity:
  Revert "KEYS: encrypted: Add check for strsep"

7 months agoMerge branch 'net-bpf_xdp_adjust_tail-and-intel-mbuf-fixes'
Alexei Starovoitov [Thu, 25 Jan 2024 00:24:07 +0000 (16:24 -0800)]
Merge branch 'net-bpf_xdp_adjust_tail-and-intel-mbuf-fixes'

Maciej Fijalkowski says:

====================
net: bpf_xdp_adjust_tail() and Intel mbuf fixes

Hey,

after a break followed by dealing with sickness, here is a v6 that makes
bpf_xdp_adjust_tail() actually usable for ZC drivers that support XDP
multi-buffer. Since v4 I tried also using bpf_xdp_adjust_tail() with
positive offset which exposed yet another issues, which can be observed
by increased commit count when compared to v3.

John, in the end I think we should remove handling
MEM_TYPE_XSK_BUFF_POOL from __xdp_return(), but it is out of the scope
for fixes set, IMHO.

Thanks,
Maciej

v6:
- add acks [Magnus]
- fix spelling mistakes [Magnus]
- avoid touching xdp_buff in xp_alloc_{reused,new_from_fq}() [Magnus]
- s/shrink_data/bpf_xdp_shrink_data [Jakub]
- remove __shrink_data() [Jakub]
- check retvals from __xdp_rxq_info_reg() [Magnus]

v5:
- pick correct version of patch 5 [Simon]
- elaborate a bit more on what patch 2 fixes

v4:
- do not clear frags flag when deleting tail; xsk_buff_pool now does
  that
- skip some NULL tests for xsk_buff_get_tail [Martin, John]
- address problems around registering xdp_rxq_info
- fix bpf_xdp_frags_increase_tail() for ZC mbuf

v3:
- add acks
- s/xsk_buff_tail_del/xsk_buff_del_tail
- address i40e as well (thanks Tirthendu)

v2:
- fix !CONFIG_XDP_SOCKETS builds
- add reviewed-by tag to patch 3
====================

Link: https://lore.kernel.org/r/20240124191602.566724-1-maciej.fijalkowski@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
7 months agoi40e: update xdp_rxq_info::frag_size for ZC enabled Rx queue
Maciej Fijalkowski [Wed, 24 Jan 2024 19:16:02 +0000 (20:16 +0100)]
i40e: update xdp_rxq_info::frag_size for ZC enabled Rx queue

Now that i40e driver correctly sets up frag_size in xdp_rxq_info, let us
make it work for ZC multi-buffer as well. i40e_ring::rx_buf_len for ZC
is being set via xsk_pool_get_rx_frame_size() and this needs to be
propagated up to xdp_rxq_info.

Fixes: 1c9ba9c14658 ("i40e: xsk: add RX multi-buffer support")
Acked-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://lore.kernel.org/r/20240124191602.566724-12-maciej.fijalkowski@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
7 months agoi40e: set xdp_rxq_info::frag_size
Maciej Fijalkowski [Wed, 24 Jan 2024 19:16:01 +0000 (20:16 +0100)]
i40e: set xdp_rxq_info::frag_size

i40e support XDP multi-buffer so it is supposed to use
__xdp_rxq_info_reg() instead of xdp_rxq_info_reg() and set the
frag_size. It can not be simply converted at existing callsite because
rx_buf_len could be un-initialized, so let us register xdp_rxq_info
within i40e_configure_rx_ring(), which happen to be called with already
initialized rx_buf_len value.

Commit 5180ff1364bc ("i40e: use int for i40e_status") converted 'err' to
int, so two variables to deal with return codes are not needed within
i40e_configure_rx_ring(). Remove 'ret' and use 'err' to handle status
from xdp_rxq_info registration.

Fixes: e213ced19bef ("i40e: add support for XDP multi-buffer Rx")
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://lore.kernel.org/r/20240124191602.566724-11-maciej.fijalkowski@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
7 months agoxdp: reflect tail increase for MEM_TYPE_XSK_BUFF_POOL
Maciej Fijalkowski [Wed, 24 Jan 2024 19:16:00 +0000 (20:16 +0100)]
xdp: reflect tail increase for MEM_TYPE_XSK_BUFF_POOL

XSK ZC Rx path calculates the size of data that will be posted to XSK Rx
queue via subtracting xdp_buff::data_end from xdp_buff::data.

In bpf_xdp_frags_increase_tail(), when underlying memory type of
xdp_rxq_info is MEM_TYPE_XSK_BUFF_POOL, add offset to data_end in tail
fragment, so that later on user space will be able to take into account
the amount of bytes added by XDP program.

Fixes: 24ea50127ecf ("xsk: support mbuf on ZC RX")
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://lore.kernel.org/r/20240124191602.566724-10-maciej.fijalkowski@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
7 months agoice: update xdp_rxq_info::frag_size for ZC enabled Rx queue
Maciej Fijalkowski [Wed, 24 Jan 2024 19:15:59 +0000 (20:15 +0100)]
ice: update xdp_rxq_info::frag_size for ZC enabled Rx queue

Now that ice driver correctly sets up frag_size in xdp_rxq_info, let us
make it work for ZC multi-buffer as well. ice_rx_ring::rx_buf_len for ZC
is being set via xsk_pool_get_rx_frame_size() and this needs to be
propagated up to xdp_rxq_info.

Use a bigger hammer and instead of unregistering only xdp_rxq_info's
memory model, unregister it altogether and register it again and have
xdp_rxq_info with correct frag_size value.

Fixes: 1bbc04de607b ("ice: xsk: add RX multi-buffer support")
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://lore.kernel.org/r/20240124191602.566724-9-maciej.fijalkowski@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
7 months agointel: xsk: initialize skb_frag_t::bv_offset in ZC drivers
Maciej Fijalkowski [Wed, 24 Jan 2024 19:15:58 +0000 (20:15 +0100)]
intel: xsk: initialize skb_frag_t::bv_offset in ZC drivers

Ice and i40e ZC drivers currently set offset of a frag within
skb_shared_info to 0, which is incorrect. xdp_buffs that come from
xsk_buff_pool always have 256 bytes of a headroom, so they need to be
taken into account to retrieve xdp_buff::data via skb_frag_address().
Otherwise, bpf_xdp_frags_increase_tail() would be starting its job from
xdp_buff::data_hard_start which would result in overwriting existing
payload.

Fixes: 1c9ba9c14658 ("i40e: xsk: add RX multi-buffer support")
Fixes: 1bbc04de607b ("ice: xsk: add RX multi-buffer support")
Acked-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://lore.kernel.org/r/20240124191602.566724-8-maciej.fijalkowski@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
7 months agoice: remove redundant xdp_rxq_info registration
Maciej Fijalkowski [Wed, 24 Jan 2024 19:15:57 +0000 (20:15 +0100)]
ice: remove redundant xdp_rxq_info registration

xdp_rxq_info struct can be registered by drivers via two functions -
xdp_rxq_info_reg() and __xdp_rxq_info_reg(). The latter one allows
drivers that support XDP multi-buffer to set up xdp_rxq_info::frag_size
which in turn will make it possible to grow the packet via
bpf_xdp_adjust_tail() BPF helper.

Currently, ice registers xdp_rxq_info in two spots:
1) ice_setup_rx_ring() // via xdp_rxq_info_reg(), BUG
2) ice_vsi_cfg_rxq()   // via __xdp_rxq_info_reg(), OK

Cited commit under fixes tag took care of setting up frag_size and
updated registration scheme in 2) but it did not help as
1) is called before 2) and as shown above it uses old registration
function. This means that 2) sees that xdp_rxq_info is already
registered and never calls __xdp_rxq_info_reg() which leaves us with
xdp_rxq_info::frag_size being set to 0.

To fix this misbehavior, simply remove xdp_rxq_info_reg() call from
ice_setup_rx_ring().

Fixes: 2fba7dc5157b ("ice: Add support for XDP multi-buffer on Rx side")
Acked-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://lore.kernel.org/r/20240124191602.566724-7-maciej.fijalkowski@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
7 months agoi40e: handle multi-buffer packets that are shrunk by xdp prog
Tirthendu Sarkar [Wed, 24 Jan 2024 19:15:56 +0000 (20:15 +0100)]
i40e: handle multi-buffer packets that are shrunk by xdp prog

XDP programs can shrink packets by calling the bpf_xdp_adjust_tail()
helper function. For multi-buffer packets this may lead to reduction of
frag count stored in skb_shared_info area of the xdp_buff struct. This
results in issues with the current handling of XDP_PASS and XDP_DROP
cases.

For XDP_PASS, currently skb is being built using frag count of
xdp_buffer before it was processed by XDP prog and thus will result in
an inconsistent skb when frag count gets reduced by XDP prog. To fix
this, get correct frag count while building the skb instead of using
pre-obtained frag count.

For XDP_DROP, current page recycling logic will not reuse the page but
instead will adjust the pagecnt_bias so that the page can be freed. This
again results in inconsistent behavior as the page refcnt has already
been changed by the helper while freeing the frag(s) as part of
shrinking the packet. To fix this, only adjust pagecnt_bias for buffers
that are stillpart of the packet post-xdp prog run.

Fixes: e213ced19bef ("i40e: add support for XDP multi-buffer Rx")
Reported-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Tirthendu Sarkar <tirthendu.sarkar@intel.com>
Link: https://lore.kernel.org/r/20240124191602.566724-6-maciej.fijalkowski@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
7 months agoice: work on pre-XDP prog frag count
Maciej Fijalkowski [Wed, 24 Jan 2024 19:15:55 +0000 (20:15 +0100)]
ice: work on pre-XDP prog frag count

Fix an OOM panic in XDP_DRV mode when a XDP program shrinks a
multi-buffer packet by 4k bytes and then redirects it to an AF_XDP
socket.

Since support for handling multi-buffer frames was added to XDP, usage
of bpf_xdp_adjust_tail() helper within XDP program can free the page
that given fragment occupies and in turn decrease the fragment count
within skb_shared_info that is embedded in xdp_buff struct. In current
ice driver codebase, it can become problematic when page recycling logic
decides not to reuse the page. In such case, __page_frag_cache_drain()
is used with ice_rx_buf::pagecnt_bias that was not adjusted after
refcount of page was changed by XDP prog which in turn does not drain
the refcount to 0 and page is never freed.

To address this, let us store the count of frags before the XDP program
was executed on Rx ring struct. This will be used to compare with
current frag count from skb_shared_info embedded in xdp_buff. A smaller
value in the latter indicates that XDP prog freed frag(s). Then, for
given delta decrement pagecnt_bias for XDP_DROP verdict.

While at it, let us also handle the EOP frag within
ice_set_rx_bufs_act() to make our life easier, so all of the adjustments
needed to be applied against freed frags are performed in the single
place.

Fixes: 2fba7dc5157b ("ice: Add support for XDP multi-buffer on Rx side")
Acked-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://lore.kernel.org/r/20240124191602.566724-5-maciej.fijalkowski@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
7 months agoxsk: fix usage of multi-buffer BPF helpers for ZC XDP
Maciej Fijalkowski [Wed, 24 Jan 2024 19:15:54 +0000 (20:15 +0100)]
xsk: fix usage of multi-buffer BPF helpers for ZC XDP

Currently when packet is shrunk via bpf_xdp_adjust_tail() and memory
type is set to MEM_TYPE_XSK_BUFF_POOL, null ptr dereference happens:

[1136314.192256] BUG: kernel NULL pointer dereference, address:
0000000000000034
[1136314.203943] #PF: supervisor read access in kernel mode
[1136314.213768] #PF: error_code(0x0000) - not-present page
[1136314.223550] PGD 0 P4D 0
[1136314.230684] Oops: 0000 [#1] PREEMPT SMP NOPTI
[1136314.239621] CPU: 8 PID: 54203 Comm: xdpsock Not tainted 6.6.0+ #257
[1136314.250469] Hardware name: Intel Corporation S2600WFT/S2600WFT,
BIOS SE5C620.86B.02.01.0008.031920191559 03/19/2019
[1136314.265615] RIP: 0010:__xdp_return+0x6c/0x210
[1136314.274653] Code: ad 00 48 8b 47 08 49 89 f8 a8 01 0f 85 9b 01 00 00 0f 1f 44 00 00 f0 41 ff 48 34 75 32 4c 89 c7 e9 79 cd 80 ff 83 fe 03 75 17 <f6> 41 34 01 0f 85 02 01 00 00 48 89 cf e9 22 cc 1e 00 e9 3d d2 86
[1136314.302907] RSP: 0018:ffffc900089f8db0 EFLAGS: 00010246
[1136314.312967] RAX: ffffc9003168aed0 RBX: ffff8881c3300000 RCX:
0000000000000000
[1136314.324953] RDX: 0000000000000000 RSI: 0000000000000003 RDI:
ffffc9003168c000
[1136314.336929] RBP: 0000000000000ae0 R08: 0000000000000002 R09:
0000000000010000
[1136314.348844] R10: ffffc9000e495000 R11: 0000000000000040 R12:
0000000000000001
[1136314.360706] R13: 0000000000000524 R14: ffffc9003168aec0 R15:
0000000000000001
[1136314.373298] FS:  00007f8df8bbcb80(0000) GS:ffff8897e0e00000(0000)
knlGS:0000000000000000
[1136314.386105] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[1136314.396532] CR2: 0000000000000034 CR3: 00000001aa912002 CR4:
00000000007706f0
[1136314.408377] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
0000000000000000
[1136314.420173] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7:
0000000000000400
[1136314.431890] PKRU: 55555554
[1136314.439143] Call Trace:
[1136314.446058]  <IRQ>
[1136314.452465]  ? __die+0x20/0x70
[1136314.459881]  ? page_fault_oops+0x15b/0x440
[1136314.468305]  ? exc_page_fault+0x6a/0x150
[1136314.476491]  ? asm_exc_page_fault+0x22/0x30
[1136314.484927]  ? __xdp_return+0x6c/0x210
[1136314.492863]  bpf_xdp_adjust_tail+0x155/0x1d0
[1136314.501269]  bpf_prog_ccc47ae29d3b6570_xdp_sock_prog+0x15/0x60
[1136314.511263]  ice_clean_rx_irq_zc+0x206/0xc60 [ice]
[1136314.520222]  ? ice_xmit_zc+0x6e/0x150 [ice]
[1136314.528506]  ice_napi_poll+0x467/0x670 [ice]
[1136314.536858]  ? ttwu_do_activate.constprop.0+0x8f/0x1a0
[1136314.546010]  __napi_poll+0x29/0x1b0
[1136314.553462]  net_rx_action+0x133/0x270
[1136314.561619]  __do_softirq+0xbe/0x28e
[1136314.569303]  do_softirq+0x3f/0x60

This comes from __xdp_return() call with xdp_buff argument passed as
NULL which is supposed to be consumed by xsk_buff_free() call.

To address this properly, in ZC case, a node that represents the frag
being removed has to be pulled out of xskb_list. Introduce
appropriate xsk helpers to do such node operation and use them
accordingly within bpf_xdp_adjust_tail().

Fixes: 24ea50127ecf ("xsk: support mbuf on ZC RX")
Acked-by: Magnus Karlsson <magnus.karlsson@intel.com> # For the xsk header part
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://lore.kernel.org/r/20240124191602.566724-4-maciej.fijalkowski@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
7 months agoxsk: make xsk_buff_pool responsible for clearing xdp_buff::flags
Maciej Fijalkowski [Wed, 24 Jan 2024 19:15:53 +0000 (20:15 +0100)]
xsk: make xsk_buff_pool responsible for clearing xdp_buff::flags

XDP multi-buffer support introduced XDP_FLAGS_HAS_FRAGS flag that is
used by drivers to notify data path whether xdp_buff contains fragments
or not. Data path looks up mentioned flag on first buffer that occupies
the linear part of xdp_buff, so drivers only modify it there. This is
sufficient for SKB and XDP_DRV modes as usually xdp_buff is allocated on
stack or it resides within struct representing driver's queue and
fragments are carried via skb_frag_t structs. IOW, we are dealing with
only one xdp_buff.

ZC mode though relies on list of xdp_buff structs that is carried via
xsk_buff_pool::xskb_list, so ZC data path has to make sure that
fragments do *not* have XDP_FLAGS_HAS_FRAGS set. Otherwise,
xsk_buff_free() could misbehave if it would be executed against xdp_buff
that carries a frag with XDP_FLAGS_HAS_FRAGS flag set. Such scenario can
take place when within supplied XDP program bpf_xdp_adjust_tail() is
used with negative offset that would in turn release the tail fragment
from multi-buffer frame.

Calling xsk_buff_free() on tail fragment with XDP_FLAGS_HAS_FRAGS would
result in releasing all the nodes from xskb_list that were produced by
driver before XDP program execution, which is not what is intended -
only tail fragment should be deleted from xskb_list and then it should
be put onto xsk_buff_pool::free_list. Such multi-buffer frame will never
make it up to user space, so from AF_XDP application POV there would be
no traffic running, however due to free_list getting constantly new
nodes, driver will be able to feed HW Rx queue with recycled buffers.
Bottom line is that instead of traffic being redirected to user space,
it would be continuously dropped.

To fix this, let us clear the mentioned flag on xsk_buff_pool side
during xdp_buff initialization, which is what should have been done
right from the start of XSK multi-buffer support.

Fixes: 1bbc04de607b ("ice: xsk: add RX multi-buffer support")
Fixes: 1c9ba9c14658 ("i40e: xsk: add RX multi-buffer support")
Fixes: 24ea50127ecf ("xsk: support mbuf on ZC RX")
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://lore.kernel.org/r/20240124191602.566724-3-maciej.fijalkowski@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
7 months agoxsk: recycle buffer in case Rx queue was full
Maciej Fijalkowski [Wed, 24 Jan 2024 19:15:52 +0000 (20:15 +0100)]
xsk: recycle buffer in case Rx queue was full

Add missing xsk_buff_free() call when __xsk_rcv_zc() failed to produce
descriptor to XSK Rx queue.

Fixes: 24ea50127ecf ("xsk: support mbuf on ZC RX")
Acked-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://lore.kernel.org/r/20240124191602.566724-2-maciej.fijalkowski@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
7 months agoMerge branch 'fix-module_description-for-net-p2'
Jakub Kicinski [Wed, 24 Jan 2024 23:12:55 +0000 (15:12 -0800)]
Merge branch 'fix-module_description-for-net-p2'

Breno Leitao says:

====================
Fix MODULE_DESCRIPTION() for net (p2)

There are hundreds of network modules that misses MODULE_DESCRIPTION(),
causing a warnning when compiling with W=1. Example:

        WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/net/arcnet/com90io.o
        WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/net/arcnet/arc-rimi.o
        WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/net/arcnet/com20020.o

This part2 of the patchset focus on the drivers/net/ethernet drivers.
There are still some missing warnings in drivers/net/ethernet that will
be fixed in an upcoming patchset.

v1: https://lore.kernel.org/all/20240122184543.2501493-2-leitao@debian.org/
====================

Link: https://lore.kernel.org/r/20240123190332.677489-1-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agonet: fill in MODULE_DESCRIPTION()s for rvu_mbox
Breno Leitao [Tue, 23 Jan 2024 19:03:31 +0000 (11:03 -0800)]
net: fill in MODULE_DESCRIPTION()s for rvu_mbox

W=1 builds now warn if module is built without a MODULE_DESCRIPTION().
Add descriptions to the Marvel RVU mbox driver.

Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://lore.kernel.org/r/20240123190332.677489-11-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agonet: fill in MODULE_DESCRIPTION()s for litex
Breno Leitao [Tue, 23 Jan 2024 19:03:30 +0000 (11:03 -0800)]
net: fill in MODULE_DESCRIPTION()s for litex

W=1 builds now warn if module is built without a MODULE_DESCRIPTION().
Add descriptions to the LiteX Liteeth Ethernet device.

Signed-off-by: Breno Leitao <leitao@debian.org>
Acked-by: Gabriel Somlo <gsomlo@gmail.com>
Link: https://lore.kernel.org/r/20240123190332.677489-10-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agonet: fill in MODULE_DESCRIPTION()s for fsl_pq_mdio
Breno Leitao [Tue, 23 Jan 2024 19:03:29 +0000 (11:03 -0800)]
net: fill in MODULE_DESCRIPTION()s for fsl_pq_mdio

W=1 builds now warn if module is built without a MODULE_DESCRIPTION().
Add descriptions to the Freescale PQ MDIO driver.

Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://lore.kernel.org/r/20240123190332.677489-9-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agonet: fill in MODULE_DESCRIPTION()s for fec
Breno Leitao [Tue, 23 Jan 2024 19:03:28 +0000 (11:03 -0800)]
net: fill in MODULE_DESCRIPTION()s for fec

W=1 builds now warn if module is built without a MODULE_DESCRIPTION().
Add descriptions to the FEC (MPC8xx) Ethernet controller.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Wei Fang <wei.fang@nxp.com>
Link: https://lore.kernel.org/r/20240123190332.677489-8-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agonet: fill in MODULE_DESCRIPTION()s for enetc
Breno Leitao [Tue, 23 Jan 2024 19:03:27 +0000 (11:03 -0800)]
net: fill in MODULE_DESCRIPTION()s for enetc

W=1 builds now warn if module is built without a MODULE_DESCRIPTION().
Add descriptions to the NXP ENETC Ethernet driver.

Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://lore.kernel.org/r/20240123190332.677489-7-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agonet: fill in MODULE_DESCRIPTION()s for nps_enet
Breno Leitao [Tue, 23 Jan 2024 19:03:26 +0000 (11:03 -0800)]
net: fill in MODULE_DESCRIPTION()s for nps_enet

W=1 builds now warn if module is built without a MODULE_DESCRIPTION().
Add descriptions to the EZchip NPS ethernet driver.

Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://lore.kernel.org/r/20240123190332.677489-6-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agonet: fill in MODULE_DESCRIPTION()s for ep93xxx_eth
Breno Leitao [Tue, 23 Jan 2024 19:03:25 +0000 (11:03 -0800)]
net: fill in MODULE_DESCRIPTION()s for ep93xxx_eth

W=1 builds now warn if module is built without a MODULE_DESCRIPTION().
Add descriptions to the Cirrus EP93xx ethernet driver.

Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://lore.kernel.org/r/20240123190332.677489-5-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agonet: fill in MODULE_DESCRIPTION()s for liquidio
Breno Leitao [Tue, 23 Jan 2024 19:03:24 +0000 (11:03 -0800)]
net: fill in MODULE_DESCRIPTION()s for liquidio

W=1 builds now warn if module is built without a MODULE_DESCRIPTION().
Add descriptions to the Cavium Liquidio.

Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://lore.kernel.org/r/20240123190332.677489-4-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agonet: fill in MODULE_DESCRIPTION()s for Broadcom bgmac
Breno Leitao [Tue, 23 Jan 2024 19:03:23 +0000 (11:03 -0800)]
net: fill in MODULE_DESCRIPTION()s for Broadcom bgmac

W=1 builds now warn if module is built without a MODULE_DESCRIPTION().
Add descriptions to the Broadcom iProc GBit driver.

Signed-off-by: Breno Leitao <leitao@debian.org>
Acked-by: Florian Fainelli <florian.fainelli@broadcom.com>
Link: https://lore.kernel.org/r/20240123190332.677489-3-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agonet: fill in MODULE_DESCRIPTION()s for 8390
Breno Leitao [Tue, 23 Jan 2024 19:03:22 +0000 (11:03 -0800)]
net: fill in MODULE_DESCRIPTION()s for 8390

W=1 builds now warn if module is built without a MODULE_DESCRIPTION().
Add descriptions to all the good old 8390 modules and drivers.

Signed-off-by: Breno Leitao <leitao@debian.org>
CC: geert@linux-m68k.org
Link: https://lore.kernel.org/r/20240123190332.677489-2-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agoselftests: netdevsim: fix the udp_tunnel_nic test
Jakub Kicinski [Tue, 23 Jan 2024 06:05:29 +0000 (22:05 -0800)]
selftests: netdevsim: fix the udp_tunnel_nic test

This test is missing a whole bunch of checks for interface
renaming and one ifup. Presumably it was only used on a system
with renaming disabled and NetworkManager running.

Fixes: 91f430b2c49d ("selftests: net: add a test for UDP tunnel info infra")
Acked-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240123060529.1033912-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agoselftests: net: fix rps_default_mask with >32 CPUs
Jakub Kicinski [Mon, 22 Jan 2024 19:58:15 +0000 (11:58 -0800)]
selftests: net: fix rps_default_mask with >32 CPUs

If there is more than 32 cpus the bitmask will start to contain
commas, leading to:

./rps_default_mask.sh: line 36: [: 00000000,00000000: integer expression expected

Remove the commas, bash doesn't interpret leading zeroes as oct
so that should be good enough. Switch to bash, Simon reports that
not all shells support this type of substitution.

Fixes: c12e0d5f267d ("self-tests: introduce self-tests for RPS default mask")
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240122195815.638997-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agoMerge tag 'execve-v6.8-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/kees...
Linus Torvalds [Wed, 24 Jan 2024 21:32:29 +0000 (13:32 -0800)]
Merge tag 'execve-v6.8-rc2' of git://git./linux/kernel/git/kees/linux

Pull execve fixes from Kees Cook:

 - Fix error handling in begin_new_exec() (Bernd Edlinger)

 - MAINTAINERS: specifically mention ELF (Alexey Dobriyan)

 - Various cleanups related to earlier open() (Askar Safin, Kees Cook)

* tag 'execve-v6.8-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
  exec: Distinguish in_execve from in_exec
  exec: Fix error handling in begin_new_exec()
  exec: Add do_close_execat() helper
  exec: remove useless comment
  ELF, MAINTAINERS: specifically mention ELF

7 months agouselib: remove use of __FMODE_EXEC
Linus Torvalds [Wed, 24 Jan 2024 21:12:20 +0000 (13:12 -0800)]
uselib: remove use of __FMODE_EXEC

Jann Horn points out that uselib() really shouldn't trigger the new
FMODE_EXEC logic introduced by commit 4759ff71f23e ("exec: __FMODE_EXEC
instead of in_execve for LSMs").

In fact, it shouldn't even have ever triggered the old pre-existing
logic for __FMODE_EXEC (like the NFS code that makes executables not
need read permissions).  Unlike a real execve(), that can work even with
files that are purely executable by the user (not readable), uselib()
has that MAY_READ requirement becasue it's really just a convenience
wrapper around mmap() for legacy shared libraries.

The whole FMODE_EXEC bit was originally introduced by commit
b500531e6f5f ("[PATCH] Introduce FMODE_EXEC file flag"), primarily to
give ETXTBUSY error returns for distributed filesystems.

It has since grown a few other warts (like that NFS thing), but there
really isn't any reason to use it for uselib(), and now that we are
trying to use it to replace the horrid 'tsk->in_execve' flag, it's
actively wrong.

Of course, as Jann Horn also points out, nobody should be enabling
CONFIG_USELIB in the first place in this day and age, but that's a
different discussion entirely.

Reported-by: Jann Horn <jannh@google.com>
Fixes: 4759ff71f23e ("exec: __FMODE_EXEC instead of in_execve for LSMs")
Cc: Kees Cook <keescook@chromium.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 months agoRevert "KEYS: encrypted: Add check for strsep"
Mimi Zohar [Wed, 24 Jan 2024 19:21:44 +0000 (14:21 -0500)]
Revert "KEYS: encrypted: Add check for strsep"

This reverts commit b4af096b5df5dd131ab796c79cedc7069d8f4882.

New encrypted keys are created either from kernel-generated random
numbers or user-provided decrypted data.  Revert the change requiring
user-provided decrypted data.

Reported-by: Vishal Verma <vishal.l.verma@intel.com>
Signed-off-by: Mimi Zohar <zohar@linux.ibm.com>
7 months agonet: mvpp2: clear BM pool before initialization
Jenishkumar Maheshbhai Patel [Fri, 19 Jan 2024 03:59:14 +0000 (19:59 -0800)]
net: mvpp2: clear BM pool before initialization

Register value persist after booting the kernel using
kexec which results in kernel panic. Thus clear the
BM pool registers before initialisation to fix the issue.

Fixes: 3f518509dedc ("ethernet: Add new driver for Marvell Armada 375 network unit")
Signed-off-by: Jenishkumar Maheshbhai Patel <jpatel2@marvell.com>
Reviewed-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Link: https://lore.kernel.org/r/20240119035914.2595665-1-jpatel2@marvell.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agonet: stmmac: Wait a bit for the reset to take effect
Bernd Edlinger [Mon, 22 Jan 2024 18:19:09 +0000 (19:19 +0100)]
net: stmmac: Wait a bit for the reset to take effect

otherwise the synopsys_id value may be read out wrong,
because the GMAC_VERSION register might still be in reset
state, for at least 1 us after the reset is de-asserted.

Add a wait for 10 us before continuing to be on the safe side.

> From what have you got that delay value?

Just try and error, with very old linux versions and old gcc versions
the synopsys_id was read out correctly most of the time (but not always),
with recent linux versions and recnet gcc versions it was read out
wrongly most of the time, but again not always.
I don't have access to the VHDL code in question, so I cannot
tell why it takes so long to get the correct values, I also do not
have more than a few hardware samples, so I cannot tell how long
this timeout must be in worst case.
Experimentally I can tell that the register is read several times
as zero immediately after the reset is de-asserted, also adding several
no-ops is not enough, adding a printk is enough, also udelay(1) seems to
be enough but I tried that not very often, and I have not access to many
hardware samples to be 100% sure about the necessary delay.
And since the udelay here is only executed once per device instance,
it seems acceptable to delay the boot for 10 us.

BTW: my hardware's synopsys id is 0x37.

Fixes: c5e4ddbdfa11 ("net: stmmac: Add support for optional reset control")
Signed-off-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Serge Semin <fancer.lancer@gmail.com>
Link: https://lore.kernel.org/r/AS8P193MB1285A810BD78C111E7F6AA34E4752@AS8P193MB1285.EURP193.PROD.OUTLOOK.COM
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
7 months agosamples/cgroup: add .gitignore file for generated samples
Linus Torvalds [Wed, 24 Jan 2024 19:52:40 +0000 (11:52 -0800)]
samples/cgroup: add .gitignore file for generated samples

Make 'git status' quietly happy again after a full allmodconfig build.

Fixes: 60433a9d038d ("samples: introduce new samples subdir for cgroup")
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 months agoexec: Distinguish in_execve from in_exec
Kees Cook [Wed, 24 Jan 2024 19:15:33 +0000 (11:15 -0800)]
exec: Distinguish in_execve from in_exec

Just to help distinguish the fs->in_exec flag from the current->in_execve
flag, add comments in check_unsafe_exec() and copy_fs() for more
context. Also note that in_execve is only used by TOMOYO now.

Cc: Kentaro Takeda <takedakn@nttdata.co.jp>
Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Jan Kara <jack@suse.cz>
Cc: Eric Biederman <ebiederm@xmission.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: linux-fsdevel@vger.kernel.org
Cc: linux-mm@kvack.org
Signed-off-by: Kees Cook <keescook@chromium.org>
7 months agoexec: Check __FMODE_EXEC instead of in_execve for LSMs
Kees Cook [Wed, 24 Jan 2024 19:22:32 +0000 (11:22 -0800)]
exec: Check __FMODE_EXEC instead of in_execve for LSMs

After commit 978ffcbf00d8 ("execve: open the executable file before
doing anything else"), current->in_execve was no longer in sync with the
open(). This broke AppArmor and TOMOYO which depend on this flag to
distinguish "open" operations from being "exec" operations.

Instead of moving around in_execve, switch to using __FMODE_EXEC, which
is where the "is this an exec?" intent is stored. Note that TOMOYO still
uses in_execve around cred handling.

Reported-by: Kevin Locke <kevin@kevinlocke.name>
Closes: https://lore.kernel.org/all/ZbE4qn9_h14OqADK@kevinlocke.name
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Fixes: 978ffcbf00d8 ("execve: open the executable file before doing anything else")
Cc: Josh Triplett <josh@joshtriplett.org>
Cc: John Johansen <john.johansen@canonical.com>
Cc: Paul Moore <paul@paul-moore.com>
Cc: James Morris <jmorris@namei.org>
Cc: Serge E. Hallyn <serge@hallyn.com>
Cc: Kentaro Takeda <takedakn@nttdata.co.jp>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Jan Kara <jack@suse.cz>
Cc: Eric Biederman <ebiederm@xmission.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: <linux-fsdevel@vger.kernel.org>
Cc: <linux-mm@kvack.org>
Cc: <apparmor@lists.ubuntu.com>
Cc: <linux-security-module@vger.kernel.org>
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 months agonetfilter: nf_tables: validate NFPROTO_* family
Pablo Neira Ayuso [Tue, 23 Jan 2024 15:38:25 +0000 (16:38 +0100)]
netfilter: nf_tables: validate NFPROTO_* family

Several expressions explicitly refer to NF_INET_* hook definitions
from expr->ops->validate, however, family is not validated.

Bail out with EOPNOTSUPP in case they are used from unsupported
families.

Fixes: 0ca743a55991 ("netfilter: nf_tables: add compatibility layer for x_tables")
Fixes: a3c90f7a2323 ("netfilter: nf_tables: flow offload expression")
Fixes: 2fa841938c64 ("netfilter: nf_tables: introduce routing expression")
Fixes: 554ced0a6e29 ("netfilter: nf_tables: add support for native socket matching")
Fixes: ad49d86e07a4 ("netfilter: nf_tables: Add synproxy support")
Fixes: 4ed8eb6570a4 ("netfilter: nf_tables: Add native tproxy support")
Fixes: 6c47260250fc ("netfilter: nf_tables: add xfrm expression")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
7 months agonetfilter: nf_tables: reject QUEUE/DROP verdict parameters
Florian Westphal [Sat, 20 Jan 2024 21:50:04 +0000 (22:50 +0100)]
netfilter: nf_tables: reject QUEUE/DROP verdict parameters

This reverts commit e0abdadcc6e1.

core.c:nf_hook_slow assumes that the upper 16 bits of NF_DROP
verdicts contain a valid errno, i.e. -EPERM, -EHOSTUNREACH or similar,
or 0.

Due to the reverted commit, its possible to provide a positive
value, e.g. NF_ACCEPT (1), which results in use-after-free.

Its not clear to me why this commit was made.

NF_QUEUE is not used by nftables; "queue" rules in nftables
will result in use of "nft_queue" expression.

If we later need to allow specifiying errno values from userspace
(do not know why), this has to call NF_DROP_GETERR and check that
"err <= 0" holds true.

Fixes: e0abdadcc6e1 ("netfilter: nf_tables: accept QUEUE/DROP verdict parameters")
Cc: stable@vger.kernel.org
Reported-by: Notselwyn <notselwyn@pwning.tech>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
7 months agonetfilter: nf_tables: restrict anonymous set and map names to 16 bytes
Florian Westphal [Fri, 19 Jan 2024 12:34:32 +0000 (13:34 +0100)]
netfilter: nf_tables: restrict anonymous set and map names to 16 bytes

nftables has two types of sets/maps, one where userspace defines the
name, and anonymous sets/maps, where userspace defines a template name.

For the latter, kernel requires presence of exactly one "%d".
nftables uses "__set%d" and "__map%d" for this.  The kernel will
expand the format specifier and replaces it with the smallest unused
number.

As-is, userspace could define a template name that allows to move
the set name past the 256 bytes upperlimit (post-expansion).

I don't see how this could be a problem, but I would prefer if userspace
cannot do this, so add a limit of 16 bytes for the '%d' template name.

16 bytes is the old total upper limit for set names that existed when
nf_tables was merged initially.

Fixes: 387454901bd6 ("netfilter: nf_tables: Allow set names of up to 255 chars")
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
7 months agonetfilter: nft_limit: reject configurations that cause integer overflow
Florian Westphal [Fri, 19 Jan 2024 12:11:32 +0000 (13:11 +0100)]
netfilter: nft_limit: reject configurations that cause integer overflow

Reject bogus configs where internal token counter wraps around.
This only occurs with very very large requests, such as 17gbyte/s.

Its better to reject this rather than having incorrect ratelimit.

Fixes: d2168e849ebf ("netfilter: nft_limit: add per-byte limiting")
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
7 months agonetfilter: nft_chain_filter: handle NETDEV_UNREGISTER for inet/ingress basechain
Pablo Neira Ayuso [Thu, 18 Jan 2024 09:56:26 +0000 (10:56 +0100)]
netfilter: nft_chain_filter: handle NETDEV_UNREGISTER for inet/ingress basechain

Remove netdevice from inet/ingress basechain in case NETDEV_UNREGISTER
event is reported, otherwise a stale reference to netdevice remains in
the hook list.

Fixes: 60a3815da702 ("netfilter: add inet ingress support")
Cc: stable@vger.kernel.org
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
7 months agonetfilter: nf_tables: cleanup documentation
George Guo [Tue, 26 Dec 2023 09:42:42 +0000 (17:42 +0800)]
netfilter: nf_tables: cleanup documentation

- Correct comments for nlpid, family, udlen and udata in struct nft_table,
  and afinfo is no longer a member of enum nft_set_class.

- Add comment for data in struct nft_set_elem.

- Add comment for flags in struct nft_ctx.

- Add comments for timeout in struct nft_set_iter, and flags is not a
  member of struct nft_set_iter, remove the comment for it.

- Add comments for commit, abort, estimate and gc_init in struct
  nft_set_ops.

- Add comments for pending_update, num_exprs, exprs and catchall_list
  in struct nft_set.

- Add comment for ext_len in struct nft_set_ext_tmpl.

- Add comment for inner_ops in struct nft_expr_type.

- Add comments for clone, destroy_clone, reduce, gc, offload,
  offload_action, offload_stats in struct nft_expr_ops.

- Add comments for blob_gen_0, blob_gen_1, bound, genmask, udlen, udata,
  blob_next in struct nft_chain.

- Add comment for flags in struct nft_base_chain.

- Add comments for udlen, udata in struct nft_object.

- Add comment for type in struct nft_object_ops.

- Add comment for hook_list in struct nft_flowtable, and remove comments
  for dev_name and ops which are not members of struct nft_flowtable.

Signed-off-by: George Guo <guodongtai@kylinos.cn>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
7 months agorcu: Defer RCU kthreads wakeup when CPU is dying
Frederic Weisbecker [Mon, 18 Dec 2023 23:19:15 +0000 (00:19 +0100)]
rcu: Defer RCU kthreads wakeup when CPU is dying

When the CPU goes idle for the last time during the CPU down hotplug
process, RCU reports a final quiescent state for the current CPU. If
this quiescent state propagates up to the top, some tasks may then be
woken up to complete the grace period: the main grace period kthread
and/or the expedited main workqueue (or kworker).

If those kthreads have a SCHED_FIFO policy, the wake up can indirectly
arm the RT bandwith timer to the local offline CPU. Since this happens
after hrtimers have been migrated at CPUHP_AP_HRTIMERS_DYING stage, the
timer gets ignored. Therefore if the RCU kthreads are waiting for RT
bandwidth to be available, they may never be actually scheduled.

This triggers TREE03 rcutorture hangs:

 rcu: INFO: rcu_preempt self-detected stall on CPU
 rcu:     4-...!: (1 GPs behind) idle=9874/1/0x4000000000000000 softirq=0/0 fqs=20 rcuc=21071 jiffies(starved)
 rcu:     (t=21035 jiffies g=938281 q=40787 ncpus=6)
 rcu: rcu_preempt kthread starved for 20964 jiffies! g938281 f0x0 RCU_GP_WAIT_FQS(5) ->state=0x0 ->cpu=0
 rcu:     Unless rcu_preempt kthread gets sufficient CPU time, OOM is now expected behavior.
 rcu: RCU grace-period kthread stack dump:
 task:rcu_preempt     state:R  running task     stack:14896 pid:14    tgid:14    ppid:2      flags:0x00004000
 Call Trace:
  <TASK>
  __schedule+0x2eb/0xa80
  schedule+0x1f/0x90
  schedule_timeout+0x163/0x270
  ? __pfx_process_timeout+0x10/0x10
  rcu_gp_fqs_loop+0x37c/0x5b0
  ? __pfx_rcu_gp_kthread+0x10/0x10
  rcu_gp_kthread+0x17c/0x200
  kthread+0xde/0x110
  ? __pfx_kthread+0x10/0x10
  ret_from_fork+0x2b/0x40
  ? __pfx_kthread+0x10/0x10
  ret_from_fork_asm+0x1b/0x30
  </TASK>

The situation can't be solved with just unpinning the timer. The hrtimer
infrastructure and the nohz heuristics involved in finding the best
remote target for an unpinned timer would then also need to handle
enqueues from an offline CPU in the most horrendous way.

So fix this on the RCU side instead and defer the wake up to an online
CPU if it's too late for the local one.

Reported-by: Paul E. McKenney <paulmck@kernel.org>
Fixes: 5c0930ccaad5 ("hrtimers: Push pending hrtimers away from outgoing CPU earlier")
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Neeraj Upadhyay (AMD) <neeraj.iitr10@gmail.com>
7 months agoMerge tag 'fbdev-for-6.8-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller...
Linus Torvalds [Wed, 24 Jan 2024 16:55:51 +0000 (08:55 -0800)]
Merge tag 'fbdev-for-6.8-rc2' of git://git./linux/kernel/git/deller/linux-fbdev

Pull fbdev fixes and cleanups from Helge Deller:
 "A crash fix in stifb which was missed to be included in the drm-misc
  tree, two checks to prevent wrong userspace input in sisfb and
  savagefb and two trivial printk cleanups:

   - stifb: Fix crash in stifb_blank()

   - savage/sis: Error out if pixclock equals zero

   - minor trivial cleanups"

* tag 'fbdev-for-6.8-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/linux-fbdev:
  fbdev: stifb: Fix crash in stifb_blank()
  fbcon: Fix incorrect printed function name in fbcon_prepare_logo()
  fbdev: sis: Error out if pixclock equals zero
  fbdev: savage: Error out if pixclock equals zero
  fbdev: vt8500lcdfb: Remove unnecessary print function dev_err()

7 months agonfsd: fix RELEASE_LOCKOWNER
NeilBrown [Mon, 22 Jan 2024 03:58:16 +0000 (14:58 +1100)]
nfsd: fix RELEASE_LOCKOWNER

The test on so_count in nfsd4_release_lockowner() is nonsense and
harmful.  Revert to using check_for_locks(), changing that to not sleep.

First: harmful.
As is documented in the kdoc comment for nfsd4_release_lockowner(), the
test on so_count can transiently return a false positive resulting in a
return of NFS4ERR_LOCKS_HELD when in fact no locks are held.  This is
clearly a protocol violation and with the Linux NFS client it can cause
incorrect behaviour.

If RELEASE_LOCKOWNER is sent while some other thread is still
processing a LOCK request which failed because, at the time that request
was received, the given owner held a conflicting lock, then the nfsd
thread processing that LOCK request can hold a reference (conflock) to
the lock owner that causes nfsd4_release_lockowner() to return an
incorrect error.

The Linux NFS client ignores that NFS4ERR_LOCKS_HELD error because it
never sends NFS4_RELEASE_LOCKOWNER without first releasing any locks, so
it knows that the error is impossible.  It assumes the lock owner was in
fact released so it feels free to use the same lock owner identifier in
some later locking request.

When it does reuse a lock owner identifier for which a previous RELEASE
failed, it will naturally use a lock_seqid of zero.  However the server,
which didn't release the lock owner, will expect a larger lock_seqid and
so will respond with NFS4ERR_BAD_SEQID.

So clearly it is harmful to allow a false positive, which testing
so_count allows.

The test is nonsense because ... well... it doesn't mean anything.

so_count is the sum of three different counts.
1/ the set of states listed on so_stateids
2/ the set of active vfs locks owned by any of those states
3/ various transient counts such as for conflicting locks.

When it is tested against '2' it is clear that one of these is the
transient reference obtained by find_lockowner_str_locked().  It is not
clear what the other one is expected to be.

In practice, the count is often 2 because there is precisely one state
on so_stateids.  If there were more, this would fail.

In my testing I see two circumstances when RELEASE_LOCKOWNER is called.
In one case, CLOSE is called before RELEASE_LOCKOWNER.  That results in
all the lock states being removed, and so the lockowner being discarded
(it is removed when there are no more references which usually happens
when the lock state is discarded).  When nfsd4_release_lockowner() finds
that the lock owner doesn't exist, it returns success.

The other case shows an so_count of '2' and precisely one state listed
in so_stateid.  It appears that the Linux client uses a separate lock
owner for each file resulting in one lock state per lock owner, so this
test on '2' is safe.  For another client it might not be safe.

So this patch changes check_for_locks() to use the (newish)
find_any_file_locked() so that it doesn't take a reference on the
nfs4_file and so never calls nfsd_file_put(), and so never sleeps.  With
this check is it safe to restore the use of check_for_locks() rather
than testing so_count against the mysterious '2'.

Fixes: ce3c4ad7f4ce ("NFSD: Fix possible sleep during nfsd4_release_lockowner()")
Signed-off-by: NeilBrown <neilb@suse.de>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Cc: stable@vger.kernel.org # v6.2+
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
7 months agonet/mlx5e: fix a potential double-free in fs_any_create_groups
Dinghao Liu [Tue, 28 Nov 2023 09:29:01 +0000 (17:29 +0800)]
net/mlx5e: fix a potential double-free in fs_any_create_groups

When kcalloc() for ft->g succeeds but kvzalloc() for in fails,
fs_any_create_groups() will free ft->g. However, its caller
fs_any_create_table() will free ft->g again through calling
mlx5e_destroy_flow_table(), which will lead to a double-free.
Fix this by setting ft->g to NULL in fs_any_create_groups().

Fixes: 0f575c20bf06 ("net/mlx5e: Introduce Flow Steering ANY API")
Signed-off-by: Dinghao Liu <dinghao.liu@zju.edu.cn>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
7 months agonet/mlx5e: fix a double-free in arfs_create_groups
Zhipeng Lu [Wed, 17 Jan 2024 07:17:36 +0000 (15:17 +0800)]
net/mlx5e: fix a double-free in arfs_create_groups

When `in` allocated by kvzalloc fails, arfs_create_groups will free
ft->g and return an error. However, arfs_create_table, the only caller of
arfs_create_groups, will hold this error and call to
mlx5e_destroy_flow_table, in which the ft->g will be freed again.

Fixes: 1cabe6b0965e ("net/mlx5e: Create aRFS flow tables")
Signed-off-by: Zhipeng Lu <alexious@zju.edu.cn>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
7 months agonet/mlx5e: Ignore IPsec replay window values on sender side
Leon Romanovsky [Sun, 26 Nov 2023 09:08:10 +0000 (11:08 +0200)]
net/mlx5e: Ignore IPsec replay window values on sender side

XFRM stack doesn't prevent from users to configure replay window
in TX side and strongswan sets replay_window to be 1. It causes
to failures in validation logic when trying to offload the SA.

Replay window is not relevant in TX side and should be ignored.

Fixes: cded6d80129b ("net/mlx5e: Store replay window in XFRM attributes")
Signed-off-by: Aya Levin <ayal@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
7 months agonet/mlx5e: Allow software parsing when IPsec crypto is enabled
Leon Romanovsky [Tue, 12 Dec 2023 11:52:55 +0000 (13:52 +0200)]
net/mlx5e: Allow software parsing when IPsec crypto is enabled

All ConnectX devices have software parsing capability enabled, but it is
more correct to set allow_swp only if capability exists, which for IPsec
means that crypto offload is supported.

Fixes: 2451da081a34 ("net/mlx5: Unify device IPsec capabilities check")
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
7 months agonet/mlx5: Use mlx5 device constant for selecting CQ period mode for ASO
Rahul Rameshbabu [Tue, 28 Nov 2023 22:01:54 +0000 (14:01 -0800)]
net/mlx5: Use mlx5 device constant for selecting CQ period mode for ASO

mlx5 devices have specific constants for choosing the CQ period mode. These
constants do not have to match the constants used by the kernel software
API for DIM period mode selection.

Fixes: cdd04f4d4d71 ("net/mlx5: Add support to create SQ and CQ for ASO")
Signed-off-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Reviewed-by: Jianbo Liu <jianbol@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
7 months agonet/mlx5: DR, Can't go to uplink vport on RX rule
Yevgeny Kliteynik [Sun, 17 Dec 2023 11:20:36 +0000 (13:20 +0200)]
net/mlx5: DR, Can't go to uplink vport on RX rule

Go-To-Vport action on RX is not allowed when the vport is uplink.
In such case, the packet should be dropped.

Fixes: 9db810ed2d37 ("net/mlx5: DR, Expose steering action functionality")
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Erez Shitrit <erezsh@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
7 months agonet/mlx5: DR, Use the right GVMI number for drop action
Yevgeny Kliteynik [Sun, 17 Dec 2023 09:24:08 +0000 (11:24 +0200)]
net/mlx5: DR, Use the right GVMI number for drop action

When FW provides ICM addresses for drop RX/TX, the provided capability
is 64 bits that contain its GVMI as well as the ICM address itself.
In case of TX DROP this GVMI is different from the GVMI that the
domain is operating on.

This patch fixes the action to use these GVMI IDs, as provided by FW.

Fixes: 9db810ed2d37 ("net/mlx5: DR, Expose steering action functionality")
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
7 months agonet/mlx5: Bridge, fix multicast packets sent to uplink
Moshe Shemesh [Sat, 30 Dec 2023 20:40:37 +0000 (22:40 +0200)]
net/mlx5: Bridge, fix multicast packets sent to uplink

To enable multicast packets which are offloaded in bridge multicast
offload mode to be sent also to uplink, FTE bit uplink_hairpin_en should
be set. Add this bit to FTE for the bridge multicast offload rules.

Fixes: 18c2916cee12 ("net/mlx5: Bridge, snoop igmp/mld packets")
Signed-off-by: Moshe Shemesh <moshe@nvidia.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
7 months agonet/mlx5: Fix a WARN upon a callback command failure
Yishai Hadas [Sun, 31 Dec 2023 13:19:50 +0000 (15:19 +0200)]
net/mlx5: Fix a WARN upon a callback command failure

The below WARN [1] is reported once a callback command failed.

As a callback runs under an interrupt context, needs to use the IRQ
save/restore variant.

[1]
DEBUG_LOCKS_WARN_ON(lockdep_hardirq_context())
WARNING: CPU: 15 PID: 0 at kernel/locking/lockdep.c:4353
              lockdep_hardirqs_on_prepare+0x11b/0x180
Modules linked in: vhost_net vhost tap mlx5_vfio_pci
vfio_pci vfio_pci_core vfio_iommu_type1 vfio mlx5_vdpa vringh
vhost_iotlb vdpa nfnetlink_cttimeout openvswitch nsh ip6table_mangle
ip6table_nat ip6table_filter ip6_tables iptable_mangle
xt_conntrackxt_MASQUERADE nf_conntrack_netlink nfnetlink
xt_addrtype iptable_nat nf_nat br_netfilter rpcsec_gss_krb5
auth_rpcgss oid_registry overlay rpcrdma rdma_ucm ib_iser libiscsi
scsi_transport_iscsi rdma_cm iw_cm ib_umad ib_ipoib ib_cm
mlx5_ib ib_uverbs ib_core fuse mlx5_core
CPU: 15 PID: 0 Comm: swapper/15 Tainted: G        W 6.7.0-rc4+ #1587
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS
rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014
RIP: 0010:lockdep_hardirqs_on_prepare+0x11b/0x180
Code: 00 5b c3 c3 e8 e6 0d 58 00 85 c0 74 d6 8b 15 f0 c3
      76 01 85 d2 75 cc 48 c7 c6 04 a5 3b 82 48 c7 c7 f1
      e9 39 82 e8 95 12 f9 ff <0f> 0b 5b c3 e8 bc 0d 58 00
      85 c0 74 ac 8b 3d c6 c3 76 01 85 ff 75
RSP: 0018:ffffc900003ecd18 EFLAGS: 00010086
RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000027
RDX: 0000000000000000 RSI: ffff88885fbdb880 RDI: ffff88885fbdb888
RBP: 00000000ffffff87 R08: 0000000000000000 R09: 0000000000000001
R10: 0000000000000000 R11: 284e4f5f4e524157 R12: 00000000002c9aa1
R13: ffff88810aace980 R14: ffff88810aace9b8 R15: 0000000000000003
FS:  0000000000000000(0000) GS:ffff88885fbc0000(0000)
knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f731436f4c8 CR3: 000000010aae6001 CR4: 0000000000372eb0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <IRQ>
? __warn+0x81/0x170
? lockdep_hardirqs_on_prepare+0x11b/0x180
? report_bug+0xf8/0x1c0
? handle_bug+0x3f/0x70
? exc_invalid_op+0x13/0x60
? asm_exc_invalid_op+0x16/0x20
? lockdep_hardirqs_on_prepare+0x11b/0x180
? lockdep_hardirqs_on_prepare+0x11b/0x180
trace_hardirqs_on+0x4a/0xa0
raw_spin_unlock_irq+0x24/0x30
cmd_status_err+0xc0/0x1a0 [mlx5_core]
cmd_status_err+0x1a0/0x1a0 [mlx5_core]
mlx5_cmd_exec_cb_handler+0x24/0x40 [mlx5_core]
mlx5_cmd_comp_handler+0x129/0x4b0 [mlx5_core]
cmd_comp_notifier+0x1a/0x20 [mlx5_core]
notifier_call_chain+0x3e/0xe0
atomic_notifier_call_chain+0x5f/0x130
mlx5_eq_async_int+0xe7/0x200 [mlx5_core]
notifier_call_chain+0x3e/0xe0
atomic_notifier_call_chain+0x5f/0x130
irq_int_handler+0x11/0x20 [mlx5_core]
__handle_irq_event_percpu+0x99/0x220
? tick_irq_enter+0x5d/0x80
handle_irq_event_percpu+0xf/0x40
handle_irq_event+0x3a/0x60
handle_edge_irq+0xa2/0x1c0
__common_interrupt+0x55/0x140
common_interrupt+0x7d/0xa0
</IRQ>
<TASK>
asm_common_interrupt+0x22/0x40
RIP: 0010:default_idle+0x13/0x20
Code: c0 08 00 00 00 4d 29 c8 4c 01 c7 4c 29 c2 e9 72 ff
ff ff cc cc cc cc 8b 05 ea 08 25 01 85 c0 7e 07 0f 00 2d 7f b0 26 00 fb
f4 <fa> c3 90 66 2e 0f 1f 84 00 00 00 00 00 65 48 8b 04 25 80 d0 02 00
RSP: 0018:ffffc9000010fec8 EFLAGS: 00000242
RAX: 0000000000000001 RBX: 000000000000000f RCX: 4000000000000000
RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffffffff811c410c
RBP: ffffffff829478c0 R08: 0000000000000001 R09: 0000000000000001
R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000
R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
? do_idle+0x1ec/0x210
default_idle_call+0x6c/0x90
do_idle+0x1ec/0x210
cpu_startup_entry+0x26/0x30
start_secondary+0x11b/0x150
secondary_startup_64_no_verify+0x165/0x16b
</TASK>
irq event stamp: 833284
hardirqs last  enabled at (833283): [<ffffffff811c410c>]
do_idle+0x1ec/0x210
hardirqs last disabled at (833284): [<ffffffff81daf9ef>]
common_interrupt+0xf/0xa0
softirqs last  enabled at (833224): [<ffffffff81dc199f>]
__do_softirq+0x2bf/0x40e
softirqs last disabled at (833177): [<ffffffff81178ddf>]
irq_exit_rcu+0x7f/0xa0

Fixes: 34f46ae0d4b3 ("net/mlx5: Add command failures data to debugfs")
Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>