kernel/linux.git/net/xdp, branch v6.1.168

xsk: Fix race condition in AF_XDP generic RX path

2026-02-19T15:25:19+00:00

[ Upstream commit a1356ac7749cafc4e27aa62c0c4604b5dca4983e ] Move rx_lock from xsk_socket to xsk_buff_pool. Fix synchronization for shared umem mode in generic RX path where multiple sockets share single xsk_buff_pool. RX queue is exclusive to xsk_socket, while FILL queue can be shared between multiple sockets. This could result in race condition where two CPU cores access RX path of two different sockets sharing the same umem. Protect both queues by acquiring spinlock in shared xsk_buff_pool. Lock contention may be minimized in the future by some per-thread FQ buffering. It's safe and necessary to move spin_lock_bh(rx_lock) after xsk_rcv_check(): * xs->pool and spinlock_init is synchronized by xsk_bind() -> xsk_is_bound() memory barriers. * xsk_rcv_check() may return true at the moment of xsk_release() or xsk_unbind_dev(), however this will not cause any data races or race conditions. xsk_unbind_dev() removes xdp socket from all maps and waits for completion of all outstanding rx operations. Packets in RX path will either complete safely or drop. Signed-off-by: Eryk Kubanski Fixes: bf0bdd1343efb ("xdp: fix race on generic receive path") Acked-by: Magnus Karlsson Link: https://patch.msgid.link/20250416101908.10919-1-e.kubanski@partner.samsung.com Signed-off-by: Jakub Kicinski [ Conflict is resolved when backporting this fix. ] Signed-off-by: Jianqiang kang Signed-off-by: Greg Kroah-Hartman

xsk: fix an integer overflow in xp_create_and_assign_umem()

2025-03-28T20:59:01+00:00

commit 559847f56769037e5b2e0474d3dbff985b98083d upstream. Since the i and pool->chunk_size variables are of type 'u32', their product can wrap around and then be cast to 'u64'. This can lead to two different XDP buffers pointing to the same memory area. Found by InfoTeCS on behalf of Linux Verification Center (linuxtesting.org) with SVACE. Fixes: 94033cd8e73b ("xsk: Optimize for aligned case") Cc: stable@vger.kernel.org Signed-off-by: Ilia Gavrilov Link: https://patch.msgid.link/20250313085007.3116044-1-Ilia.Gavrilov@infotecs.ru Signed-off-by: Paolo Abeni Signed-off-by: Greg Kroah-Hartman

xsk: fix OOB map writes when deleting elements

2024-12-14T18:54:36+00:00

commit 32cd3db7de97c0c7a018756ce66244342fd583f0 upstream. Jordy says: " In the xsk_map_delete_elem function an unsigned integer (map->max_entries) is compared with a user-controlled signed integer (k). Due to implicit type conversion, a large unsigned value for map->max_entries can bypass the intended bounds check: if (k >= map->max_entries) return -EINVAL; This allows k to hold a negative value (between -2147483648 and -2), which is then used as an array index in m->xsk_map[k], which results in an out-of-bounds access. spin_lock_bh(&m->lock); map_entry = &m->xsk_map[k]; // Out-of-bounds map_entry old_xs = unrcu_pointer(xchg(map_entry, NULL)); // Oob write if (old_xs) xsk_map_sock_delete(old_xs, map_entry); spin_unlock_bh(&m->lock); The xchg operation can then be used to cause an out-of-bounds write. Moreover, the invalid map_entry passed to xsk_map_sock_delete can lead to further memory corruption. " It indeed results in following splat: [76612.897343] BUG: unable to handle page fault for address: ffffc8fc2e461108 [76612.904330] #PF: supervisor write access in kernel mode [76612.909639] #PF: error_code(0x0002) - not-present page [76612.914855] PGD 0 P4D 0 [76612.917431] Oops: Oops: 0002 [#1] PREEMPT SMP [76612.921859] CPU: 11 UID: 0 PID: 10318 Comm: a.out Not tainted 6.12.0-rc1+ #470 [76612.929189] Hardware name: Intel Corporation S2600WFT/S2600WFT, BIOS SE5C620.86B.02.01.0008.031920191559 03/19/2019 [76612.939781] RIP: 0010:xsk_map_delete_elem+0x2d/0x60 [76612.944738] Code: 00 00 41 54 55 53 48 63 2e 3b 6f 24 73 38 4c 8d a7 f8 00 00 00 48 89 fb 4c 89 e7 e8 2d bf 05 00 48 8d b4 eb 00 01 00 00 31 ff <48> 87 3e 48 85 ff 74 05 e8 16 ff ff ff 4c 89 e7 e8 3e bc 05 00 31 [76612.963774] RSP: 0018:ffffc9002e407df8 EFLAGS: 00010246 [76612.969079] RAX: 0000000000000000 RBX: ffffc9002e461000 RCX: 0000000000000000 [76612.976323] RDX: 0000000000000001 RSI: ffffc8fc2e461108 RDI: 0000000000000000 [76612.983569] RBP: ffffffff80000001 R08: 0000000000000000 R09: 0000000000000007 [76612.990812] R10: ffffc9002e407e18 R11: ffff888108a38858 R12: ffffc9002e4610f8 [76612.998060] R13: ffff888108a38858 R14: 00007ffd1ae0ac78 R15: ffffc9002e4610c0 [76613.005303] FS: 00007f80b6f59740(0000) GS:ffff8897e0ec0000(0000) knlGS:0000000000000000 [76613.013517] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [76613.019349] CR2: ffffc8fc2e461108 CR3: 000000011e3ef001 CR4: 00000000007726f0 [76613.026595] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [76613.033841] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [76613.041086] PKRU: 55555554 [76613.043842] Call Trace: [76613.046331] [76613.048468] ? __die+0x20/0x60 [76613.051581] ? page_fault_oops+0x15a/0x450 [76613.055747] ? search_extable+0x22/0x30 [76613.059649] ? search_bpf_extables+0x5f/0x80 [76613.063988] ? exc_page_fault+0xa9/0x140 [76613.067975] ? asm_exc_page_fault+0x22/0x30 [76613.072229] ? xsk_map_delete_elem+0x2d/0x60 [76613.076573] ? xsk_map_delete_elem+0x23/0x60 [76613.080914] __sys_bpf+0x19b7/0x23c0 [76613.084555] __x64_sys_bpf+0x1a/0x20 [76613.088194] do_syscall_64+0x37/0xb0 [76613.091832] entry_SYSCALL_64_after_hwframe+0x4b/0x53 [76613.096962] RIP: 0033:0x7f80b6d1e88d [76613.100592] Code: 5b 41 5c c3 66 0f 1f 84 00 00 00 00 00 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 73 b5 0f 00 f7 d8 64 89 01 48 [76613.119631] RSP: 002b:00007ffd1ae0ac68 EFLAGS: 00000206 ORIG_RAX: 0000000000000141 [76613.131330] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f80b6d1e88d [76613.142632] RDX: 0000000000000098 RSI: 00007ffd1ae0ad20 RDI: 0000000000000003 [76613.153967] RBP: 00007ffd1ae0adc0 R08: 0000000000000000 R09: 0000000000000000 [76613.166030] R10: 00007f80b6f77040 R11: 0000000000000206 R12: 00007ffd1ae0aed8 [76613.177130] R13: 000055ddf42ce1e9 R14: 000055ddf42d0d98 R15: 00007f80b6fab040 [76613.188129] Fix this by simply changing key type from int to u32. Fixes: fbfc504a24f5 ("bpf: introduce new bpf AF_XDP map type BPF_MAP_TYPE_XSKMAP") CC: stable@vger.kernel.org Reported-by: Jordy Zomer Suggested-by: Jordy Zomer Reviewed-by: Toke Høiland-Jørgensen Acked-by: John Fastabend Signed-off-by: Maciej Fijalkowski Link: https://lore.kernel.org/r/20241122121030.716788-2-maciej.fijalkowski@intel.com Signed-off-by: Alexei Starovoitov Signed-off-by: Greg Kroah-Hartman

xsk: always clear DMA mapping information when unmapping the pool

2024-12-14T18:54:29+00:00

[ Upstream commit ac9a48a6f1610b094072b815e884e1668aea4401 ] When the umem is shared, the DMA mapping is also shared between the xsk pools, therefore it should stay valid as long as at least 1 user remains. However, the pool also keeps the copies of DMA-related information that are initialized in the same way in xp_init_dma_info(), but cleared by xp_dma_unmap() only for the last remaining pool, this causes the problems below. The first one is that the commit adbf5a42341f ("ice: remove af_xdp_zc_qps bitmap") relies on pool->dev to determine the presence of a ZC pool on a given queue, avoiding internal bookkeeping. This works perfectly fine if the UMEM is not shared, but reliably fails otherwise as stated in the linked report. The second one is pool->dma_pages which is dynamically allocated and only freed in xp_dma_unmap(), this leads to a small memory leak. kmemleak does not catch it, but by printing the allocation results after terminating the userspace program it is possible to see that all addresses except the one belonging to the last detached pool are still accessible through the kmemleak dump functionality. Always clear the DMA mapping information from the pool and free pool->dma_pages when unmapping the pool, so that the only difference between results of the last remaining user's call and the ones before would be the destruction of the DMA mapping. Fixes: adbf5a42341f ("ice: remove af_xdp_zc_qps bitmap") Fixes: 921b68692abb ("xsk: Enable sharing of dma mappings") Reported-by: Alasdair McWilliam Closes: https://lore.kernel.org/PA4P194MB10056F208AF221D043F57A3D86512@PA4P194MB1005.EURP194.PROD.OUTLOOK.COM Acked-by: Maciej Fijalkowski Signed-off-by: Larysa Zaremba Link: https://lore.kernel.org/r/20241122112912.89881-1-larysa.zaremba@intel.com Signed-off-by: Alexei Starovoitov Signed-off-by: Sasha Levin

sock_diag: add module pointer to "struct sock_diag_handler"

2024-12-14T18:53:32+00:00

[ Upstream commit 114b4bb1cc19239b272d52ebbe156053483fe2f8 ] Following patch is going to use RCU instead of sock_diag_table_mutex acquisition. This patch is a preparation, no change of behavior yet. Signed-off-by: Eric Dumazet Reviewed-by: Guillaume Nault Reviewed-by: Kuniyuki Iwashima Reviewed-by: Willem de Bruijn Signed-off-by: Paolo Abeni Stable-dep-of: eb02688c5c45 ("ipv6: release nexthop on device removal") Signed-off-by: Sasha Levin

xsk: validate user input for XDP_{UMEM|COMPLETION}_FILL_RING

2024-04-17T09:18:23+00:00

[ Upstream commit 237f3cf13b20db183d3706d997eedc3c49eacd44 ] syzbot reported an illegal copy in xsk_setsockopt() [1] Make sure to validate setsockopt() @optlen parameter. [1] BUG: KASAN: slab-out-of-bounds in copy_from_sockptr_offset include/linux/sockptr.h:49 [inline] BUG: KASAN: slab-out-of-bounds in copy_from_sockptr include/linux/sockptr.h:55 [inline] BUG: KASAN: slab-out-of-bounds in xsk_setsockopt+0x909/0xa40 net/xdp/xsk.c:1420 Read of size 4 at addr ffff888028c6cde3 by task syz-executor.0/7549 CPU: 0 PID: 7549 Comm: syz-executor.0 Not tainted 6.8.0-syzkaller-08951-gfe46a7dd189e #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/27/2024 Call Trace: __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0x241/0x360 lib/dump_stack.c:114 print_address_description mm/kasan/report.c:377 [inline] print_report+0x169/0x550 mm/kasan/report.c:488 kasan_report+0x143/0x180 mm/kasan/report.c:601 copy_from_sockptr_offset include/linux/sockptr.h:49 [inline] copy_from_sockptr include/linux/sockptr.h:55 [inline] xsk_setsockopt+0x909/0xa40 net/xdp/xsk.c:1420 do_sock_setsockopt+0x3af/0x720 net/socket.c:2311 __sys_setsockopt+0x1ae/0x250 net/socket.c:2334 __do_sys_setsockopt net/socket.c:2343 [inline] __se_sys_setsockopt net/socket.c:2340 [inline] __x64_sys_setsockopt+0xb5/0xd0 net/socket.c:2340 do_syscall_64+0xfb/0x240 entry_SYSCALL_64_after_hwframe+0x6d/0x75 RIP: 0033:0x7fb40587de69 Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 e1 20 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b0 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007fb40665a0c8 EFLAGS: 00000246 ORIG_RAX: 0000000000000036 RAX: ffffffffffffffda RBX: 00007fb4059abf80 RCX: 00007fb40587de69 RDX: 0000000000000005 RSI: 000000000000011b RDI: 0000000000000006 RBP: 00007fb4058ca47a R08: 0000000000000002 R09: 0000000000000000 R10: 0000000020001980 R11: 0000000000000246 R12: 0000000000000000 R13: 000000000000000b R14: 00007fb4059abf80 R15: 00007fff57ee4d08 Allocated by task 7549: kasan_save_stack mm/kasan/common.c:47 [inline] kasan_save_track+0x3f/0x80 mm/kasan/common.c:68 poison_kmalloc_redzone mm/kasan/common.c:370 [inline] __kasan_kmalloc+0x98/0xb0 mm/kasan/common.c:387 kasan_kmalloc include/linux/kasan.h:211 [inline] __do_kmalloc_node mm/slub.c:3966 [inline] __kmalloc+0x233/0x4a0 mm/slub.c:3979 kmalloc include/linux/slab.h:632 [inline] __cgroup_bpf_run_filter_setsockopt+0xd2f/0x1040 kernel/bpf/cgroup.c:1869 do_sock_setsockopt+0x6b4/0x720 net/socket.c:2293 __sys_setsockopt+0x1ae/0x250 net/socket.c:2334 __do_sys_setsockopt net/socket.c:2343 [inline] __se_sys_setsockopt net/socket.c:2340 [inline] __x64_sys_setsockopt+0xb5/0xd0 net/socket.c:2340 do_syscall_64+0xfb/0x240 entry_SYSCALL_64_after_hwframe+0x6d/0x75 The buggy address belongs to the object at ffff888028c6cde0 which belongs to the cache kmalloc-8 of size 8 The buggy address is located 1 bytes to the right of allocated 2-byte region [ffff888028c6cde0, ffff888028c6cde2) The buggy address belongs to the physical page: page:ffffea0000a31b00 refcount:1 mapcount:0 mapping:0000000000000000 index:0xffff888028c6c9c0 pfn:0x28c6c anon flags: 0xfff00000000800(slab|node=0|zone=1|lastcpupid=0x7ff) page_type: 0xffffffff() raw: 00fff00000000800 ffff888014c41280 0000000000000000 dead000000000001 raw: ffff888028c6c9c0 0000000080800057 00000001ffffffff 0000000000000000 page dumped because: kasan: bad access detected page_owner tracks the page as allocated page last allocated via order 0, migratetype Unmovable, gfp_mask 0x112cc0(GFP_USER|__GFP_NOWARN|__GFP_NORETRY), pid 6648, tgid 6644 (syz-executor.0), ts 133906047828, free_ts 133859922223 set_page_owner include/linux/page_owner.h:31 [inline] post_alloc_hook+0x1ea/0x210 mm/page_alloc.c:1533 prep_new_page mm/page_alloc.c:1540 [inline] get_page_from_freelist+0x33ea/0x3580 mm/page_alloc.c:3311 __alloc_pages+0x256/0x680 mm/page_alloc.c:4569 __alloc_pages_node include/linux/gfp.h:238 [inline] alloc_pages_node include/linux/gfp.h:261 [inline] alloc_slab_page+0x5f/0x160 mm/slub.c:2175 allocate_slab mm/slub.c:2338 [inline] new_slab+0x84/0x2f0 mm/slub.c:2391 ___slab_alloc+0xc73/0x1260 mm/slub.c:3525 __slab_alloc mm/slub.c:3610 [inline] __slab_alloc_node mm/slub.c:3663 [inline] slab_alloc_node mm/slub.c:3835 [inline] __do_kmalloc_node mm/slub.c:3965 [inline] __kmalloc_node+0x2db/0x4e0 mm/slub.c:3973 kmalloc_node include/linux/slab.h:648 [inline] __vmalloc_area_node mm/vmalloc.c:3197 [inline] __vmalloc_node_range+0x5f9/0x14a0 mm/vmalloc.c:3392 __vmalloc_node mm/vmalloc.c:3457 [inline] vzalloc+0x79/0x90 mm/vmalloc.c:3530 bpf_check+0x260/0x19010 kernel/bpf/verifier.c:21162 bpf_prog_load+0x1667/0x20f0 kernel/bpf/syscall.c:2895 __sys_bpf+0x4ee/0x810 kernel/bpf/syscall.c:5631 __do_sys_bpf kernel/bpf/syscall.c:5738 [inline] __se_sys_bpf kernel/bpf/syscall.c:5736 [inline] __x64_sys_bpf+0x7c/0x90 kernel/bpf/syscall.c:5736 do_syscall_64+0xfb/0x240 entry_SYSCALL_64_after_hwframe+0x6d/0x75 page last free pid 6650 tgid 6647 stack trace: reset_page_owner include/linux/page_owner.h:24 [inline] free_pages_prepare mm/page_alloc.c:1140 [inline] free_unref_page_prepare+0x95d/0xa80 mm/page_alloc.c:2346 free_unref_page_list+0x5a3/0x850 mm/page_alloc.c:2532 release_pages+0x2117/0x2400 mm/swap.c:1042 tlb_batch_pages_flush mm/mmu_gather.c:98 [inline] tlb_flush_mmu_free mm/mmu_gather.c:293 [inline] tlb_flush_mmu+0x34d/0x4e0 mm/mmu_gather.c:300 tlb_finish_mmu+0xd4/0x200 mm/mmu_gather.c:392 exit_mmap+0x4b6/0xd40 mm/mmap.c:3300 __mmput+0x115/0x3c0 kernel/fork.c:1345 exit_mm+0x220/0x310 kernel/exit.c:569 do_exit+0x99e/0x27e0 kernel/exit.c:865 do_group_exit+0x207/0x2c0 kernel/exit.c:1027 get_signal+0x176e/0x1850 kernel/signal.c:2907 arch_do_signal_or_restart+0x96/0x860 arch/x86/kernel/signal.c:310 exit_to_user_mode_loop kernel/entry/common.c:105 [inline] exit_to_user_mode_prepare include/linux/entry-common.h:328 [inline] __syscall_exit_to_user_mode_work kernel/entry/common.c:201 [inline] syscall_exit_to_user_mode+0xc9/0x360 kernel/entry/common.c:212 do_syscall_64+0x10a/0x240 arch/x86/entry/common.c:89 entry_SYSCALL_64_after_hwframe+0x6d/0x75 Memory state around the buggy address: ffff888028c6cc80: fa fc fc fc fa fc fc fc fa fc fc fc fa fc fc fc ffff888028c6cd00: fa fc fc fc fa fc fc fc 00 fc fc fc 06 fc fc fc >ffff888028c6cd80: fa fc fc fc fa fc fc fc fa fc fc fc 02 fc fc fc ^ ffff888028c6ce00: fa fc fc fc fa fc fc fc fa fc fc fc fa fc fc fc ffff888028c6ce80: fa fc fc fc fa fc fc fc fa fc fc fc fa fc fc fc Fixes: 423f38329d26 ("xsk: add umem fill queue support and mmap") Reported-by: syzbot Signed-off-by: Eric Dumazet Cc: "Björn Töpel" Cc: Magnus Karlsson Cc: Maciej Fijalkowski Cc: Jonathan Lemon Acked-by: Daniel Borkmann Link: https://lore.kernel.org/r/20240404202738.3634547-1-edumazet@google.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin

xsk: Skip polling event check for unbound socket

2023-12-13T17:39:08+00:00

[ Upstream commit e4d008d49a7135214e0ee70537405b6a069e3a3f ] In xsk_poll(), checking available events and setting mask bits should be executed only when a socket has been bound. Setting mask bits for unbound socket is meaningless. Currently, it checks events even when xsk_check_common() failed. To prevent this, we move goto location (skip_tx) after that checking. Fixes: 1596dae2f17e ("xsk: check IFF_UP earlier in Tx path") Signed-off-by: Yewon Choi Signed-off-by: Daniel Borkmann Acked-by: Magnus Karlsson Link: https://lore.kernel.org/bpf/20231201061048.GA1510@libra05 Signed-off-by: Sasha Levin

xsk: Fix xsk_diag use-after-free error during socket cleanup

2023-09-19T10:28:01+00:00

[ Upstream commit 3e019d8a05a38abb5c85d4f1e85fda964610aa14 ] Fix a use-after-free error that is possible if the xsk_diag interface is used after the socket has been unbound from the device. This can happen either due to the socket being closed or the device disappearing. In the early days of AF_XDP, the way we tested that a socket was not bound to a device was to simply check if the netdevice pointer in the xsk socket structure was NULL. Later, a better system was introduced by having an explicit state variable in the xsk socket struct. For example, the state of a socket that is on the way to being closed and has been unbound from the device is XSK_UNBOUND. The commit in the Fixes tag below deleted the old way of signalling that a socket is unbound, setting dev to NULL. This in the belief that all code using the old way had been exterminated. That was unfortunately not true as the xsk diagnostics code was still using the old way and thus does not work as intended when a socket is going down. Fix this by introducing a test against the state variable. If the socket is in the state XSK_UNBOUND, simply abort the diagnostic's netlink operation. Fixes: 18b1ab7aa76b ("xsk: Fix race at socket teardown") Reported-by: syzbot+822d1359297e2694f873@syzkaller.appspotmail.com Signed-off-by: Magnus Karlsson Signed-off-by: Daniel Borkmann Tested-by: syzbot+822d1359297e2694f873@syzkaller.appspotmail.com Tested-by: Maciej Fijalkowski Reviewed-by: Maciej Fijalkowski Link: https://lore.kernel.org/bpf/20230831100119.17408-1-magnus.karlsson@gmail.com Signed-off-by: Sasha Levin

xsk: fix refcount underflow in error path

2023-08-16T16:27:27+00:00

commit 85c2c79a07302fe68a1ad5cc449458cc559e314d upstream. Fix a refcount underflow problem reported by syzbot that can happen when a system is running out of memory. If xp_alloc_tx_descs() fails, and it can only fail due to not having enough memory, then the error path is triggered. In this error path, the refcount of the pool is decremented as it has incremented before. However, the reference to the pool in the socket was not nulled. This means that when the socket is closed later, the socket teardown logic will think that there is a pool attached to the socket and try to decrease the refcount again, leading to a refcount underflow. I chose this fix as it involved adding just a single line. Another option would have been to move xp_get_pool() and the assignment of xs->pool to after the if-statement and using xs_umem->pool instead of xs->pool in the whole if-statement resulting in somewhat simpler code, but this would have led to much more churn in the code base perhaps making it harder to backport. Fixes: ba3beec2ec1d ("xsk: Fix possible crash when multiple sockets are created") Reported-by: syzbot+8ada0057e69293a05fd4@syzkaller.appspotmail.com Signed-off-by: Magnus Karlsson Link: https://lore.kernel.org/r/20230809142843.13944-1-magnus.karlsson@gmail.com Signed-off-by: Martin KaFai Lau Signed-off-by: Greg Kroah-Hartman

net: annotate data-races around sk->sk_mark

2023-08-11T10:08:14+00:00

[ Upstream commit 3c5b4d69c358a9275a8de98f87caf6eda644b086 ] sk->sk_mark is often read while another thread could change the value. Fixes: 4a19ec5800fc ("[NET]: Introducing socket mark socket option.") Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller Signed-off-by: Sasha Levin