Vulnerabilities

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> fbdev: fbcon: fix out-of-bounds read in err_out of fbcon_do_set_font()<br /> <br /> When fbcon_do_set_font() fails (e.g., due to a memory allocation failure<br /> inside vc_resize() under heavy memory pressure), it jumps to the `err_out`<br /> label to roll back the console state. However, the current rollback logic<br /> forgets to restore the `hi_font` state, leading to a severe state machine<br /> corruption.<br /> <br /> Earlier in the function, `set_vc_hi_font()` might be called to change<br /> `vc-&gt;vc_hi_font_mask` and mutate the screen buffer. If `vc_resize()`<br /> subsequently fails, the `err_out` path restores `vc_font.charcount`<br /> but entirely skips rolling back the `vc_hi_font_mask` and the screen<br /> buffer.<br /> <br /> This mismatch leaves the terminal in a desynchronized state. Because<br /> `vc_hi_font_mask` remains set, the VT subsystem will still accept<br /> character indices greater than 255 from userspace and write them to the<br /> screen buffer. Subsequent rendering calls (e.g., `fbcon_putcs()`) will<br /> then use these inflated indices to access the reverted, 256-character<br /> font array, leading to a deterministic out-of-bounds read and potential<br /> kernel memory disclosure.<br /> <br /> Fix this by adding the missing rollback logic for the `hi_font` mask<br /> and screen buffer in the error path.

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> fbdev: Fix fb_new_modelist to prevent null-ptr-deref in fb_videomode_to_var<br /> <br /> info-&gt;var, a framebuffer&amp;#39;s current mode, is expected to have a matching<br /> entry in info-&gt;modelist. var_to_display() relies on this and treats a<br /> failed fb_match_mode() as "This should not happen". fb_set_var() keeps it<br /> true by adding the mode to the list on every change, and<br /> do_register_framebuffer() does the same at registration.<br /> <br /> store_modes() replaces the modelist from userspace. fb_new_modelist()<br /> validates the new modes but does not check that info-&gt;var still has a<br /> match. It relies on fbcon_new_modelist() to re-point consoles, but that<br /> only handles consoles mapped to the framebuffer. With fbcon unbound there<br /> are none, so info-&gt;var is left describing a mode that is no longer in the<br /> list.<br /> <br /> A later console takeover runs var_to_display(), where fb_match_mode()<br /> returns NULL and leaves fb_display[i].mode NULL. fbcon_switch() passes it<br /> to display_to_var(), and fb_videomode_to_var() dereferences the NULL mode.<br /> <br /> Keep the current mode in the list in fb_new_modelist(), the same way<br /> fb_set_var() does.

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> ntfs: serialize volume label accesses<br /> <br /> Protect vol-&gt;volume_label with a mutex and snaphost the label before<br /> copy_to_user. This prevent a use-after-free when FS_IOC_SETFSLABEL<br /> replaces the vol-&gt;volume_label and FS_IOC_GETTSLABEL reads it<br /> concurrently.

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> KVM: SVM: Fix page overflow in sev_dbg_crypt() for ENCRYPT path<br /> <br /> In sev_dbg_crypt(), the per-iteration transfer length is bounded by<br /> the source page offset (PAGE_SIZE - s_off) but not by the destination<br /> page offset (PAGE_SIZE - d_off). When d_off &gt; s_off, the encrypt<br /> path (__sev_dbg_encrypt_user) performs a read-modify-write using a<br /> single-page intermediate buffer (dst_tpage):<br /> <br /> 1. __sev_dbg_decrypt() expands the size to round_up(len + (d_off &amp; 15), 16)<br /> before issuing the PSP command. If len + (d_off &amp; 15) &gt; PAGE_SIZE,<br /> the PSP writes beyond the end of the 4096-byte dst_tpage allocation.<br /> <br /> 2. The subsequent memcpy()/copy_from_user() into<br /> page_address(dst_tpage) + (d_off &amp; 15) of &amp;#39;len&amp;#39; bytes overflows<br /> by up to 15 bytes under the same condition.<br /> <br /> Trigger example: s_off = 0, d_off = 1, debug.len = PAGE_SIZE -<br /> the PSP is instructed to write round_up(4097, 16) = 4112 bytes to<br /> a 4096-byte buffer.<br /> <br /> Fix by also bounding len by (PAGE_SIZE - d_off), the same check that<br /> sev_send_update_data() already performs for its single-page guest<br /> region.<br /> <br /> ==================================================================<br /> BUG: KASAN: slab-use-after-free in sev_dbg_crypt+0x993/0xd10 [kvm_amd]<br /> Write of size 4095 at addr ff110062293bb009 by task sev_dbg_test/228214<br /> <br /> CPU: 96 UID: 0 PID: 228214 Comm: sev_dbg_test Tainted: G U W 7.0.0-smp--5ce9b0c48211-dbg #156 PREEMPTLAZY<br /> Tainted: [U]=USER, [W]=WARN<br /> Hardware name: Google Astoria/astoria, BIOS 0.20250817.1-0 08/25/2025<br /> Call Trace:<br /> <br /> dump_stack_lvl+0x54/0x70<br /> print_report+0xbc/0x260<br /> kasan_report+0xa2/0xd0<br /> kasan_check_range+0x25f/0x2c0<br /> __asan_memcpy+0x40/0x70<br /> sev_dbg_crypt+0x993/0xd10 [kvm_amd]<br /> sev_mem_enc_ioctl+0x33c/0x450 [kvm_amd]<br /> kvm_vm_ioctl+0x65d/0x6d0 [kvm]<br /> __se_sys_ioctl+0xb2/0x100<br /> do_syscall_64+0xe8/0x870<br /> entry_SYSCALL_64_after_hwframe+0x4b/0x53<br /> <br /> <br /> The buggy address belongs to the physical page:<br /> page: refcount:1 mapcount:0 mapping:0000000000000000 index:0x7fe72b6a0 pfn:0x62293bb<br /> memcg:ff11000112827d82<br /> flags: 0x1400000000000000(node=1|zone=1)<br /> raw: 1400000000000000 0000000000000000 dead000000000122 0000000000000000<br /> raw: 00000007fe72b6a0 0000000000000000 00000001ffffffff ff11000112827d82<br /> page dumped because: kasan: bad access detected<br /> <br /> Memory state around the buggy address:<br /> ff110062293bbf00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00<br /> ff110062293bbf80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00<br /> &gt;ff110062293bc000: fa fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc<br /> ^<br /> ff110062293bc080: fa fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc<br /> ff110062293bc100: fa fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc<br /> ==================================================================<br /> Disabling lock debugging due to kernel taint<br /> <br /> [sean: add sample KASAN splat, Fixes, and stable@]

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> 9p: avoid putting oldfid in p9_client_walk() error path<br /> <br /> When p9_client_walk() is called with clone set to false, fid aliases<br /> oldfid. If the walk subsequently fails after the request has been sent,<br /> the error path jumps to clunk_fid, which currently calls p9_fid_put(fid)<br /> unconditionally.<br /> <br /> This drops a reference to oldfid even though ownership of oldfid remains<br /> with the caller. If this is the last reference, oldfid can be clunked and<br /> destroyed while the caller still expects it to be valid. A later use or<br /> put of oldfid can then trigger a use-after-free or refcount underflow.<br /> <br /> Fix this by only putting fid in the clunk_fid error path when it does not<br /> alias oldfid, matching the existing guard in the error path below.<br /> <br /> This can be triggered when a multi-component walk is split into multiple<br /> p9_client_walk() calls and a later non-cloning walk fails. A reproducer<br /> and refcount warning logs are available on request.

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> ksmbd: fix out-of-bounds read in smb_check_perm_dacl()<br /> <br /> The permission-check ACE walk in smb_check_perm_dacl() validates the ACE<br /> header size and caps sid.num_subauth at SID_MAX_SUB_AUTHORITIES, but it<br /> never checks that ace-&gt;size is actually large enough to contain<br /> num_subauth sub-authorities before compare_sids() dereferences them.<br /> <br /> CIFS_SID_BASE_SIZE covers the SID header up to but excluding the<br /> sub_auth[] array, and offsetof(struct smb_ace, sid) is the ACE header,<br /> so the existing guards only guarantee the 8-byte SID base, i.e. zero<br /> sub-authorities. compare_sids() then reads ace-&gt;sid.sub_auth[i] for<br /> i num_subauth, ace-&gt;sid.num_subauth). The local<br /> comparison SIDs (sid_everyone, sid_unix_NFS_mode, and the id_to_sid()<br /> result) always have at least one sub-authority, and an attacker controls<br /> the ACE revision and authority bytes (which lie within the in-bounds SID<br /> base), so they can match one of those SIDs and force the sub_auth read.<br /> <br /> A crafted ACE with size == 16 and num_subauth &gt;= 1 placed at the tail of<br /> the security descriptor therefore causes a heap out-of-bounds read of up<br /> to SID_MAX_SUB_AUTHORITIES * sizeof(__le32) bytes past the pntsd<br /> allocation. The security descriptor is loaded by ksmbd_vfs_get_sd_xattr()<br /> into a buffer sized exactly to the on-disk data (kzalloc(sd_size) in<br /> ndr_decode_v4_ntacl()), so the read lands past the allocation. The<br /> malformed descriptor can be stored verbatim via SMB2_SET_INFO (the DACL<br /> is not normalised before being written to the security.NTACL xattr) and<br /> the read fires on a subsequent SMB2_CREATE access check, making this<br /> reachable by an authenticated client on a share that uses ACL xattrs.<br /> <br /> Add the missing num_subauth-versus-ace_size check, mirroring the<br /> identical guards already present in the sibling parsers parse_dacl() and<br /> smb_inherit_dacl().

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> NFSv4/pNFS: reject zero-length r_addr in nfs4_decode_mp_ds_addr<br /> <br /> nfs4_decode_mp_ds_addr() decodes the r_netid and r_addr opaques of a<br /> netaddr4 from a GETDEVICEINFO multipath-DS body, then immediately<br /> calls strrchr(buf, &amp;#39;.&amp;#39;) to locate the port separator. Both decodes<br /> use xdr_stream_decode_string_dup(), and the current code checks only<br /> "nlen

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> NFSv4/flexfiles: reject zero filehandle version count<br /> <br /> ff_layout_alloc_lseg() decodes the filehandle-version array count<br /> from the flexfiles layout body. The value is used as the count for<br /> kzalloc_objs(), and the current code only rejects NULL.<br /> <br /> A zero count yields ZERO_SIZE_PTR, which can be stored in<br /> dss_info-&gt;fh_versions even though later flexfiles paths assume that at<br /> least one filehandle version exists.<br /> <br /> Reject fh_count == 0 before the allocation, matching the existing zero<br /> version_count validation in the flexfiles GETDEVICEINFO parser.<br /> <br /> A QEMU/KASAN run with a malformed flexfiles layout hit:<br /> <br /> KASAN: null-ptr-deref in range [0x0000000000000010-0x0000000000000017]<br /> RIP: 0010:ff_layout_encode_ff_layoutupdate.isra.0+0x15f/0x750<br /> ff_layout_encode_layoutreturn+0x683/0x970<br /> nfs4_xdr_enc_layoutreturn+0x278/0x3a0<br /> Kernel panic - not syncing: Fatal exception<br /> <br /> The patched kernel rejects the malformed layout without KASAN/oops/panic,<br /> and a valid fh_count=1 regression still opens, reads, and unmounts cleanly.

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> nfsd: reset write verifier on deferred writeback errors<br /> <br /> nfsd_vfs_write() and nfsd_commit() both call filemap_check_wb_err() to<br /> detect deferred writeback errors, but neither rotates the server&amp;#39;s write<br /> verifier (nn-&gt;writeverf) when this check fails. Every other<br /> durable-storage-failure path in these functions calls<br /> commit_reset_write_verifier() before returning an error.<br /> <br /> The missing rotation means clients holding UNSTABLE write data under the<br /> current verifier will COMMIT, receive the unchanged verifier back, and<br /> conclude their data is durable — silently dropping data that failed<br /> writeback. This violates the UNSTABLE+COMMIT durability contract<br /> (RFC 1813 §3.3.7, RFC 8881 §18.32).<br /> <br /> Add commit_reset_write_verifier() calls at both filemap_check_wb_err()<br /> error sites, matching the pattern used by adjacent error paths in the<br /> same functions. The helper already filters -EAGAIN and -ESTALE<br /> internally, so the calls are unconditionally safe.

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> nfsd: avoid leaking pre-allocated openowner on unconfirmed retry race<br /> <br /> When find_or_alloc_open_stateowner() encounters an unconfirmed owner, it<br /> calls release_openowner() and sets oo = NULL. Control then falls through<br /> past the `if (oo)` guard -- which would have freed any pre-allocated<br /> `new` -- and unconditionally executes `new = alloc_stateowner(...)`. If<br /> `new` was already allocated on a prior iteration, the pointer is<br /> silently overwritten and the previous allocation (slab object + owner<br /> name buffer) is leaked.<br /> <br /> This requires a race: two NFSv4.0 OPEN threads with the same owner<br /> string, where a concurrent thread inserts a new unconfirmed owner into<br /> the hash between retry iterations. The window is narrow but repeatable<br /> under adversarial conditions.<br /> <br /> Fix by adding `goto retry` after `oo = NULL` so the already-allocated<br /> `new` is reused on the next iteration rather than overwritten.

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> nfsd: fix dead ACL conflict guard in nfsd4_create<br /> <br /> nfsd4_create() steals create-&gt;cr_dpacl/cr_pacl into the local<br /> nfsd_attrs via the designated initializer, then immediately sets the<br /> source pointers to NULL. The subsequent conflict guard tests the<br /> already-nilled source fields, making it permanently dead code:<br /> <br /> if (create-&gt;cr_acl) {<br /> if (create-&gt;cr_dpacl || create-&gt;cr_pacl) /* always false */<br /> <br /> When a client encodes both FATTR4_WORD0_ACL and<br /> FATTR4_WORD2_POSIX_{DEFAULT,ACCESS}_ACL in the same CREATE fattr<br /> bitmap, nfsd4_acl_to_attr() overwrites attrs.na_pacl/na_dpacl without<br /> releasing the originals, leaking two posix_acl slab objects per<br /> request. Repeated requests cause unbounded slab exhaustion.<br /> <br /> Fix by checking attrs.na_dpacl/na_pacl (the stolen values) instead of<br /> the nilled create-&gt;cr_dpacl/cr_pacl, matching the correct pattern<br /> already used in nfsd4_setattr().

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> nfsd: fix posix_acl leak and ignored error in nfsd4_create_file<br /> <br /> nfsd4_create_file() has two bugs in its ACL handling:<br /> <br /> The return value of nfsd4_acl_to_attr() is silently discarded. When<br /> the NFSv4-to-POSIX ACL conversion fails (e.g., -EINVAL for<br /> unsupported ACE types), the file is created without any ACL and the<br /> client receives NFS4_OK. This violates RFC 7530/8881 which require<br /> the server to reject unsupported attributes on CREATE.<br /> <br /> When start_creating() fails after ACL attributes have been populated<br /> in attrs (either via nfsd4_acl_to_attr or via ownership transfer from<br /> open-&gt;op_dpacl/op_pacl), the function jumps to out_write which skips<br /> nfsd_attrs_free(). The posix_acl allocations are leaked. A client<br /> can trigger this repeatedly with OPEN(CREATE), ACL attributes, and an<br /> invalid filename (e.g., longer than NAME_MAX).<br /> <br /> Fix both by capturing the nfsd4_acl_to_attr() return value and by<br /> changing the early error paths to jump to out instead of out_write.<br /> Initialize child to ERR_PTR(-EINVAL) so that end_creating() is safe<br /> to call even if start_creating() was never reached.