CVE-2022-49049 in Linux
Summary
by MITRE • 02/26/2025
In the Linux kernel, the following vulnerability has been resolved:
mm/secretmem: fix panic when growing a memfd_secret
When one tries to grow an existing memfd_secret with ftruncate, one gets a panic [1]. For example, doing the following reliably induces the
panic:
fd = memfd_secret();
ftruncate(fd, 10); ptr = mmap(NULL, 10, PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0); strcpy(ptr, "123456789");
munmap(ptr, 10); ftruncate(fd, 20);
The basic reason for this is, when we grow with ftruncate, we call down into simple_setattr, and then truncate_inode_pages_range, and eventually we try to zero part of the memory. The normal truncation code does this via the direct map (i.e., it calls page_address() and hands that to memset()).
For memfd_secret though, we specifically don't map our pages via the direct map (i.e. we call set_direct_map_invalid_noflush() on every fault). So the address returned by page_address() isn't useful, and when we try to memset() with it we panic.
This patch avoids the panic by implementing a custom setattr for memfd_secret, which detects resizes specifically (setting the size for the first time works just fine, since there are no existing pages to try to zero), and rejects them with EINVAL.
One could argue growing should be supported, but I think that will require a significantly more lengthy change. So, I propose a minimal fix for the benefit of stable kernels, and then perhaps to extend memfd_secret to support growing in a separate patch.
[1]:
BUG: unable to handle page fault for address: ffffa0a889277028 #PF: supervisor write access in kernel mode #PF: error_code(0x0002) - not-present page PGD afa01067 P4D afa01067 PUD 83f909067 PMD 83f8bf067 PTE 800ffffef6d88060 Oops: 0002 [#1] PREEMPT SMP DEBUG_PAGEALLOC PTI
CPU: 0 PID: 281 Comm: repro Not tainted 5.17.0-dbg-DEV #1 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014 RIP: 0010:memset_erms+0x9/0x10 Code: c1 e9 03 40 0f b6 f6 48 b8 01 01 01 01 01 01 01 01 48 0f af c6 f3 48 ab 89 d1 f3 aa 4c 89 c8 c3 90 49 89 f9 40 88 f0 48 89 d1 aa 4c 89 c8 c3 90 49 89 fa 40 0f b6 ce 48 b8 01 01 01 01 01 01 RSP: 0018:ffffb932c09afbf0 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffffda63c4249dc0 RCX: 0000000000000fd8 RDX: 0000000000000fd8 RSI: 0000000000000000 RDI: ffffa0a889277028 RBP: ffffb932c09afc00 R08: 0000000000001000 R09: ffffa0a889277028 R10: 0000000000020023 R11: 0000000000000000 R12: ffffda63c4249dc0 R13: ffffa0a890d70d98 R14: 0000000000000028 R15: 0000000000000fd8 FS: 00007f7294899580(0000) GS:ffffa0af9bc00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: ffffa0a889277028 CR3: 0000000107ef6006 CR4: 0000000000370ef0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: ? zero_user_segments+0x82/0x190 truncate_inode_partial_folio+0xd4/0x2a0 truncate_inode_pages_range+0x380/0x830 truncate_setsize+0x63/0x80 simple_setattr+0x37/0x60 notify_change+0x3d8/0x4d0 do_sys_ftruncate+0x162/0x1d0 __x64_sys_ftruncate+0x1c/0x20 do_syscall_64+0x44/0xa0 entry_SYSCALL_64_after_hwframe+0x44/0xae Modules linked in: xhci_pci xhci_hcd virtio_net net_failover failover virtio_blk virtio_balloon uhci_hcd ohci_pci ohci_hcd evdev ehci_pci ehci_hcd 9pnet_virtio 9p netfs 9pnet CR2: ffffa0a889277028
[[email protected]: secretmem_iops can be static]
Signed-off-by: kernel test robot [[email protected]: return EINVAL]
VulDB is the best source for vulnerability data and more expert information about this specific topic.
Analysis
by VulDB Data Team • 10/15/2025
The vulnerability described in CVE-2022-49049 affects the Linux kernel's implementation of memfd_secret, a specialized memory file descriptor type designed for secure memory sharing. This flaw manifests as a kernel panic when attempting to increase the size of an existing memfd_secret file using the ftruncate system call. The issue stems from the fundamental architectural differences between regular memory mappings and memfd_secret objects, which are intentionally isolated from the standard direct map mechanism for security purposes.
The technical root cause lies in how the kernel handles file truncation operations for memfd_secret objects. When ftruncate is called on a memfd_secret, the system invokes the standard truncation path through simple_setattr and truncate_inode_pages_range functions. These functions attempt to zero out the newly allocated memory regions by using page_address() to obtain direct virtual addresses for memory operations. However, memfd_secret objects deliberately avoid using the direct map by calling set_direct_map_invalid_noflush() on every memory fault, making the addresses returned by page_address() invalid for use with memset operations. This mismatch leads to a page fault when the kernel attempts to write to an unmapped memory region, resulting in an immediate system crash.
This vulnerability directly relates to CWE-125: Out-of-bounds Read and CWE-787: Out-of-bounds Write, as the system attempts to access memory outside of its allocated bounds due to improper handling of memory mapping structures. The issue also aligns with ATT&CK technique T1059.007: Command and Scripting Interpreter: Python, where malicious actors could potentially leverage such a kernel panic to cause system instability or denial of service, though the direct exploitation path is limited to kernel-level operations. The panic occurs in the kernel's memory management subsystem, specifically in the memset_erms function which is part of the kernel's memory zeroing mechanism.
The operational impact of this vulnerability is significant for systems utilizing memfd_secret for secure memory operations, as it renders the ftruncate functionality unreliable and potentially destabilizing. Systems that depend on growing memfd_secret objects for dynamic memory allocation could experience unexpected kernel panics, leading to service interruptions and potential data loss. The fix implemented in this patch addresses the immediate issue by rejecting resize operations with an EINVAL error code, preventing the kernel from entering the problematic code path while maintaining the integrity of existing memfd_secret objects. This approach follows the principle of least privilege and minimal change, ensuring that existing functionality remains intact while preventing the specific crash condition. The solution aligns with kernel development best practices for maintaining stability in long-term support releases, as suggested by the patch author's recommendation to implement a more comprehensive solution in a separate update rather than introducing a complex change that could affect other kernel subsystems. The fix specifically targets the resize operation without affecting the initial creation or other valid memfd_secret operations, preserving the security properties of the mechanism while eliminating the crash scenario.