Vulnerabilities

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> drm/amdkfd: Fix double release compute pasid<br /> <br /> If kfd_process_device_init_vm returns failure after vm is converted to<br /> compute vm and vm-&gt;pasid set to compute pasid, KFD will not take<br /> pdd-&gt;drm_file reference. As a result, drm close file handler maybe<br /> called to release the compute pasid before KFD process destroy worker to<br /> release the same pasid and set vm-&gt;pasid to zero, this generates below<br /> WARNING backtrace and NULL pointer access.<br /> <br /> Add helper amdgpu_amdkfd_gpuvm_set_vm_pasid and call it at the last step<br /> of kfd_process_device_init_vm, to ensure vm pasid is the original pasid<br /> if acquiring vm failed or is the compute pasid with pdd-&gt;drm_file<br /> reference taken to avoid double release same pasid.<br /> <br /> amdgpu: Failed to create process VM object<br /> ida_free called for id=32770 which is not allocated.<br /> WARNING: CPU: 57 PID: 72542 at ../lib/idr.c:522 ida_free+0x96/0x140<br /> RIP: 0010:ida_free+0x96/0x140<br /> Call Trace:<br /> amdgpu_pasid_free_delayed+0xe1/0x2a0 [amdgpu]<br /> amdgpu_driver_postclose_kms+0x2d8/0x340 [amdgpu]<br /> drm_file_free.part.13+0x216/0x270 [drm]<br /> drm_close_helper.isra.14+0x60/0x70 [drm]<br /> drm_release+0x6e/0xf0 [drm]<br /> __fput+0xcc/0x280<br /> ____fput+0xe/0x20<br /> task_work_run+0x96/0xc0<br /> do_exit+0x3d0/0xc10<br /> <br /> BUG: kernel NULL pointer dereference, address: 0000000000000000<br /> RIP: 0010:ida_free+0x76/0x140<br /> Call Trace:<br /> amdgpu_pasid_free_delayed+0xe1/0x2a0 [amdgpu]<br /> amdgpu_driver_postclose_kms+0x2d8/0x340 [amdgpu]<br /> drm_file_free.part.13+0x216/0x270 [drm]<br /> drm_close_helper.isra.14+0x60/0x70 [drm]<br /> drm_release+0x6e/0xf0 [drm]<br /> __fput+0xcc/0x280<br /> ____fput+0xe/0x20<br /> task_work_run+0x96/0xc0<br /> do_exit+0x3d0/0xc10

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> mtd: core: fix possible resource leak in init_mtd()<br /> <br /> I got the error report while inject fault in init_mtd():<br /> <br /> sysfs: cannot create duplicate filename &amp;#39;/devices/virtual/bdi/mtd-0&amp;#39;<br /> Call Trace:<br /> <br /> dump_stack_lvl+0x67/0x83<br /> sysfs_warn_dup+0x60/0x70<br /> sysfs_create_dir_ns+0x109/0x120<br /> kobject_add_internal+0xce/0x2f0<br /> kobject_add+0x98/0x110<br /> device_add+0x179/0xc00<br /> device_create_groups_vargs+0xf4/0x100<br /> device_create+0x7b/0xb0<br /> bdi_register_va.part.13+0x58/0x2d0<br /> bdi_register+0x9b/0xb0<br /> init_mtd+0x62/0x171 [mtd]<br /> do_one_initcall+0x6c/0x3c0<br /> do_init_module+0x58/0x222<br /> load_module+0x268e/0x27d0<br /> __do_sys_finit_module+0xd5/0x140<br /> do_syscall_64+0x37/0x90<br /> entry_SYSCALL_64_after_hwframe+0x63/0xcd<br /> <br /> kobject_add_internal failed for mtd-0 with -EEXIST, don&amp;#39;t try to register<br /> things with the same name in the same directory.<br /> Error registering mtd class or bdi: -17<br /> <br /> If init_mtdchar() fails in init_mtd(), mtd_bdi will not be unregistered,<br /> as a result, we can&amp;#39;t load the mtd module again, to fix this by calling<br /> bdi_unregister(mtd_bdi) after out_procfs label.

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> qlcnic: prevent -&gt;dcb use-after-free on qlcnic_dcb_enable() failure<br /> <br /> adapter-&gt;dcb would get silently freed inside qlcnic_dcb_enable() in<br /> case qlcnic_dcb_attach() would return an error, which always happens<br /> under OOM conditions. This would lead to use-after-free because both<br /> of the existing callers invoke qlcnic_dcb_get_info() on the obtained<br /> pointer, which is potentially freed at that point.<br /> <br /> Propagate errors from qlcnic_dcb_enable(), and instead free the dcb<br /> pointer at callsite using qlcnic_dcb_free(). This also removes the now<br /> unused qlcnic_clear_dcb_ops() helper, which was a simple wrapper around<br /> kfree() also causing memory leaks for partially initialized dcb.<br /> <br /> Found by Linux Verification Center (linuxtesting.org) with the SVACE<br /> static analysis tool.

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> ocfs2: fix memory leak in ocfs2_stack_glue_init()<br /> <br /> ocfs2_table_header should be free in ocfs2_stack_glue_init() if<br /> ocfs2_sysfs_init() failed, otherwise kmemleak will report memleak.<br /> <br /> BUG: memory leak<br /> unreferenced object 0xffff88810eeb5800 (size 128):<br /> comm "modprobe", pid 4507, jiffies 4296182506 (age 55.888s)<br /> hex dump (first 32 bytes):<br /> c0 40 14 a0 ff ff ff ff 00 00 00 00 01 00 00 00 .@..............<br /> 01 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................<br /> backtrace:<br /> [] __register_sysctl_table+0xca/0xef0<br /> [] 0xffffffffa0050037<br /> [] do_one_initcall+0xdb/0x480<br /> [] do_init_module+0x1cf/0x680<br /> [] load_module+0x6441/0x6f20<br /> [] __do_sys_finit_module+0x12f/0x1c0<br /> [] do_syscall_64+0x3f/0x90<br /> [] entry_SYSCALL_64_after_hwframe+0x63/0xcd

Rejected reason: This CVE ID has been rejected or withdrawn by its CVE Numbering Authority.

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> kcm: annotate data-races around kcm-&gt;rx_psock<br /> <br /> kcm-&gt;rx_psock can be read locklessly in kcm_rfree().<br /> Annotate the read and writes accordingly.<br /> <br /> We do the same for kcm-&gt;rx_wait in the following patch.<br /> <br /> syzbot reported:<br /> BUG: KCSAN: data-race in kcm_rfree / unreserve_rx_kcm<br /> <br /> write to 0xffff888123d827b8 of 8 bytes by task 2758 on cpu 1:<br /> unreserve_rx_kcm+0x72/0x1f0 net/kcm/kcmsock.c:313<br /> kcm_rcv_strparser+0x2b5/0x3a0 net/kcm/kcmsock.c:373<br /> __strp_recv+0x64c/0xd20 net/strparser/strparser.c:301<br /> strp_recv+0x6d/0x80 net/strparser/strparser.c:335<br /> tcp_read_sock+0x13e/0x5a0 net/ipv4/tcp.c:1703<br /> strp_read_sock net/strparser/strparser.c:358 [inline]<br /> do_strp_work net/strparser/strparser.c:406 [inline]<br /> strp_work+0xe8/0x180 net/strparser/strparser.c:415<br /> process_one_work+0x3d3/0x720 kernel/workqueue.c:2289<br /> worker_thread+0x618/0xa70 kernel/workqueue.c:2436<br /> kthread+0x1a9/0x1e0 kernel/kthread.c:376<br /> ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:306<br /> <br /> read to 0xffff888123d827b8 of 8 bytes by task 5859 on cpu 0:<br /> kcm_rfree+0x14c/0x220 net/kcm/kcmsock.c:181<br /> skb_release_head_state+0x8e/0x160 net/core/skbuff.c:841<br /> skb_release_all net/core/skbuff.c:852 [inline]<br /> __kfree_skb net/core/skbuff.c:868 [inline]<br /> kfree_skb_reason+0x5c/0x260 net/core/skbuff.c:891<br /> kfree_skb include/linux/skbuff.h:1216 [inline]<br /> kcm_recvmsg+0x226/0x2b0 net/kcm/kcmsock.c:1161<br /> ____sys_recvmsg+0x16c/0x2e0<br /> ___sys_recvmsg net/socket.c:2743 [inline]<br /> do_recvmmsg+0x2f1/0x710 net/socket.c:2837<br /> __sys_recvmmsg net/socket.c:2916 [inline]<br /> __do_sys_recvmmsg net/socket.c:2939 [inline]<br /> __se_sys_recvmmsg net/socket.c:2932 [inline]<br /> __x64_sys_recvmmsg+0xde/0x160 net/socket.c:2932<br /> do_syscall_x64 arch/x86/entry/common.c:50 [inline]<br /> do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80<br /> entry_SYSCALL_64_after_hwframe+0x63/0xcd<br /> <br /> value changed: 0xffff88812971ce00 -&gt; 0x0000000000000000<br /> <br /> Reported by Kernel Concurrency Sanitizer on:<br /> CPU: 0 PID: 5859 Comm: syz-executor.3 Not tainted 6.0.0-syzkaller-12189-g19d17ab7c68b-dirty #0<br /> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 09/22/2022

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> drm/msm/dp: fix bridge lifetime<br /> <br /> Device-managed resources allocated post component bind must be tied to<br /> the lifetime of the aggregate DRM device or they will not necessarily be<br /> released when binding of the aggregate device is deferred.<br /> <br /> This can lead resource leaks or failure to bind the aggregate device<br /> when binding is later retried and a second attempt to allocate the<br /> resources is made.<br /> <br /> For the DP bridges, previously allocated bridges will leak on probe<br /> deferral.<br /> <br /> Fix this by amending the DP parser interface and tying the lifetime of<br /> the bridge device to the DRM device rather than DP platform device.<br /> <br /> Patchwork: https://patchwork.freedesktop.org/patch/502667/

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> btrfs: do not BUG_ON() on ENOMEM when dropping extent items for a range<br /> <br /> If we get -ENOMEM while dropping file extent items in a given range, at<br /> btrfs_drop_extents(), due to failure to allocate memory when attempting to<br /> increment the reference count for an extent or drop the reference count,<br /> we handle it with a BUG_ON(). This is excessive, instead we can simply<br /> abort the transaction and return the error to the caller. In fact most<br /> callers of btrfs_drop_extents(), directly or indirectly, already abort<br /> the transaction if btrfs_drop_extents() returns any error.<br /> <br /> Also, we already have error paths at btrfs_drop_extents() that may return<br /> -ENOMEM and in those cases we abort the transaction, like for example<br /> anything that changes the b+tree may return -ENOMEM due to a failure to<br /> allocate a new extent buffer when COWing an existing extent buffer, such<br /> as a call to btrfs_duplicate_item() for example.<br /> <br /> So replace the BUG_ON() calls with proper logic to abort the transaction<br /> and return the error.

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> wifi: libertas: fix memory leak in lbs_init_adapter()<br /> <br /> When kfifo_alloc() failed in lbs_init_adapter(), cmd buffer is not<br /> released. Add free memory to processing error path.

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> io_uring/msg_ring: Fix NULL pointer dereference in io_msg_send_fd()<br /> <br /> Syzkaller produced the below call trace:<br /> <br /> BUG: KASAN: null-ptr-deref in io_msg_ring+0x3cb/0x9f0<br /> Write of size 8 at addr 0000000000000070 by task repro/16399<br /> <br /> CPU: 0 PID: 16399 Comm: repro Not tainted 6.1.0-rc1 #28<br /> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.0-2.el7<br /> Call Trace:<br /> <br /> dump_stack_lvl+0xcd/0x134<br /> ? io_msg_ring+0x3cb/0x9f0<br /> kasan_report+0xbc/0xf0<br /> ? io_msg_ring+0x3cb/0x9f0<br /> kasan_check_range+0x140/0x190<br /> io_msg_ring+0x3cb/0x9f0<br /> ? io_msg_ring_prep+0x300/0x300<br /> io_issue_sqe+0x698/0xca0<br /> io_submit_sqes+0x92f/0x1c30<br /> __do_sys_io_uring_enter+0xae4/0x24b0<br /> ....<br /> RIP: 0033:0x7f2eaf8f8289<br /> RSP: 002b:00007fff40939718 EFLAGS: 00000246 ORIG_RAX: 00000000000001aa<br /> RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f2eaf8f8289<br /> RDX: 0000000000000000 RSI: 0000000000006f71 RDI: 0000000000000004<br /> RBP: 00007fff409397a0 R08: 0000000000000000 R09: 0000000000000039<br /> R10: 0000000000000000 R11: 0000000000000246 R12: 00000000004006d0<br /> R13: 00007fff40939880 R14: 0000000000000000 R15: 0000000000000000<br /> <br /> Kernel panic - not syncing: panic_on_warn set ...<br /> <br /> We don&amp;#39;t have a NULL check on file_ptr in io_msg_send_fd() function,<br /> so when file_ptr is NUL src_file is also NULL and get_file()<br /> dereferences a NULL pointer and leads to above crash.<br /> <br /> Add a NULL check to fix this issue.

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> UM: cpuinfo: Fix a warning for CONFIG_CPUMASK_OFFSTACK<br /> <br /> When CONFIG_CPUMASK_OFFSTACK and CONFIG_DEBUG_PER_CPU_MAPS is selected,<br /> cpu_max_bits_warn() generates a runtime warning similar as below while<br /> we show /proc/cpuinfo. Fix this by using nr_cpu_ids (the runtime limit)<br /> instead of NR_CPUS to iterate CPUs.<br /> <br /> [ 3.052463] ------------[ cut here ]------------<br /> [ 3.059679] WARNING: CPU: 3 PID: 1 at include/linux/cpumask.h:108 show_cpuinfo+0x5e8/0x5f0<br /> [ 3.070072] Modules linked in: efivarfs autofs4<br /> [ 3.076257] CPU: 0 PID: 1 Comm: systemd Not tainted 5.19-rc5+ #1052<br /> [ 3.099465] Stack : 9000000100157b08 9000000000f18530 9000000000cf846c 9000000100154000<br /> [ 3.109127] 9000000100157a50 0000000000000000 9000000100157a58 9000000000ef7430<br /> [ 3.118774] 90000001001578e8 0000000000000040 0000000000000020 ffffffffffffffff<br /> [ 3.128412] 0000000000aaaaaa 1ab25f00eec96a37 900000010021de80 900000000101c890<br /> [ 3.138056] 0000000000000000 0000000000000000 0000000000000000 0000000000aaaaaa<br /> [ 3.147711] ffff8000339dc220 0000000000000001 0000000006ab4000 0000000000000000<br /> [ 3.157364] 900000000101c998 0000000000000004 9000000000ef7430 0000000000000000<br /> [ 3.167012] 0000000000000009 000000000000006c 0000000000000000 0000000000000000<br /> [ 3.176641] 9000000000d3de08 9000000001639390 90000000002086d8 00007ffff0080286<br /> [ 3.186260] 00000000000000b0 0000000000000004 0000000000000000 0000000000071c1c<br /> [ 3.195868] ...<br /> [ 3.199917] Call Trace:<br /> [ 3.203941] [] show_stack+0x38/0x14c<br /> [ 3.210666] [] dump_stack_lvl+0x60/0x88<br /> [ 3.217625] [] __warn+0xd0/0x100<br /> [ 3.223958] [] warn_slowpath_fmt+0x7c/0xcc<br /> [ 3.231150] [] show_cpuinfo+0x5e8/0x5f0<br /> [ 3.238080] [] seq_read_iter+0x354/0x4b4<br /> [ 3.245098] [] new_sync_read+0x17c/0x1c4<br /> [ 3.252114] [] vfs_read+0x138/0x1d0<br /> [ 3.258694] [] ksys_read+0x70/0x100<br /> [ 3.265265] [] do_syscall+0x7c/0x94<br /> [ 3.271820] [] handle_syscall+0xc4/0x160<br /> [ 3.281824] ---[ end trace 8b484262b4b8c24c ]---

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> pnode: terminate at peers of source<br /> <br /> The propagate_mnt() function handles mount propagation when creating<br /> mounts and propagates the source mount tree @source_mnt to all<br /> applicable nodes of the destination propagation mount tree headed by<br /> @dest_mnt.<br /> <br /> Unfortunately it contains a bug where it fails to terminate at peers of<br /> @source_mnt when looking up copies of the source mount that become<br /> masters for copies of the source mount tree mounted on top of slaves in<br /> the destination propagation tree causing a NULL dereference.<br /> <br /> Once the mechanics of the bug are understood it&amp;#39;s easy to trigger.<br /> Because of unprivileged user namespaces it is available to unprivileged<br /> users.<br /> <br /> While fixing this bug we&amp;#39;ve gotten confused multiple times due to<br /> unclear terminology or missing concepts. So let&amp;#39;s start this with some<br /> clarifications:<br /> <br /> * The terms "master" or "peer" denote a shared mount. A shared mount<br /> belongs to a peer group.<br /> <br /> * A peer group is a set of shared mounts that propagate to each other.<br /> They are identified by a peer group id. The peer group id is available<br /> in @shared_mnt-&gt;mnt_group_id.<br /> Shared mounts within the same peer group have the same peer group id.<br /> The peers in a peer group can be reached via @shared_mnt-&gt;mnt_share.<br /> <br /> * The terms "slave mount" or "dependent mount" denote a mount that<br /> receives propagation from a peer in a peer group. IOW, shared mounts<br /> may have slave mounts and slave mounts have shared mounts as their<br /> master. Slave mounts of a given peer in a peer group are listed on<br /> that peers slave list available at @shared_mnt-&gt;mnt_slave_list.<br /> <br /> * The term "master mount" denotes a mount in a peer group. IOW, it<br /> denotes a shared mount or a peer mount in a peer group. The term<br /> "master mount" - or "master" for short - is mostly used when talking<br /> in the context of slave mounts that receive propagation from a master<br /> mount. A master mount of a slave identifies the closest peer group a<br /> slave mount receives propagation from. The master mount of a slave can<br /> be identified via @slave_mount-&gt;mnt_master. Different slaves may point<br /> to different masters in the same peer group.<br /> <br /> * Multiple peers in a peer group can have non-empty -&gt;mnt_slave_lists.<br /> Non-empty -&gt;mnt_slave_lists of peers don&amp;#39;t intersect. Consequently, to<br /> ensure all slave mounts of a peer group are visited the<br /> -&gt;mnt_slave_lists of all peers in a peer group have to be walked.<br /> <br /> * Slave mounts point to a peer in the closest peer group they receive<br /> propagation from via @slave_mnt-&gt;mnt_master (see above). Together with<br /> these peers they form a propagation group (see below). The closest<br /> peer group can thus be identified through the peer group id<br /> @slave_mnt-&gt;mnt_master-&gt;mnt_group_id of the peer/master that a slave<br /> mount receives propagation from.<br /> <br /> * A shared-slave mount is a slave mount to a peer group pg1 while also<br /> a peer in another peer group pg2. IOW, a peer group may receive<br /> propagation from another peer group.<br /> <br /> If a peer group pg1 is a slave to another peer group pg2 then all<br /> peers in peer group pg1 point to the same peer in peer group pg2 via<br /> -&gt;mnt_master. IOW, all peers in peer group pg1 appear on the same<br /> -&gt;mnt_slave_list. IOW, they cannot be slaves to different peer groups.<br /> <br /> * A pure slave mount is a slave mount that is a slave to a peer group<br /> but is not a peer in another peer group.<br /> <br /> * A propagation group denotes the set of mounts consisting of a single<br /> peer group pg1 and all slave mounts and shared-slave mounts that point<br /> to a peer in that peer group via -&gt;mnt_master. IOW, all slave mounts<br /> such that @slave_mnt-&gt;mnt_master-&gt;mnt_group_id is equal to<br /> @shared_mnt-&gt;mnt_group_id.<br /> <br /> The concept of a propagation group makes it easier to talk about a<br /> single propagation level in a propagation tree.<br /> <br /> For example, in propagate_mnt() the immediate peers of @dest_mnt and<br /> all slaves of @dest_mnt&amp;#39;s peer group form a propagation group pr<br /> ---truncated---