ghsa-v2hx-rxf2-p233
Vulnerability from github
In the Linux kernel, the following vulnerability has been resolved:
uprobes: Reject the shared zeropage in uprobe_write_opcode()
We triggered the following crash in syzkaller tests:
BUG: Bad page state in process syz.7.38 pfn:1eff3 page: refcount:0 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x1eff3 flags: 0x3fffff00004004(referenced|reserved|node=0|zone=1|lastcpupid=0x1fffff) raw: 003fffff00004004 ffffe6c6c07bfcc8 ffffe6c6c07bfcc8 0000000000000000 raw: 0000000000000000 0000000000000000 00000000fffffffe 0000000000000000 page dumped because: PAGE_FLAGS_CHECK_AT_FREE flag(s) set Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-1ubuntu1.1 04/01/2014 Call Trace: dump_stack_lvl+0x32/0x50 bad_page+0x69/0xf0 free_unref_page_prepare+0x401/0x500 free_unref_page+0x6d/0x1b0 uprobe_write_opcode+0x460/0x8e0 install_breakpoint.part.0+0x51/0x80 register_for_each_vma+0x1d9/0x2b0 __uprobe_register+0x245/0x300 bpf_uprobe_multi_link_attach+0x29b/0x4f0 link_create+0x1e2/0x280 __sys_bpf+0x75f/0xac0 __x64_sys_bpf+0x1a/0x30 do_syscall_64+0x56/0x100 entry_SYSCALL_64_after_hwframe+0x78/0xe2
BUG: Bad rss-counter state mm:00000000452453e0 type:MM_FILEPAGES val:-1
The following syzkaller test case can be used to reproduce:
r2 = creat(&(0x7f0000000000)='./file0\x00', 0x8) write$nbd(r2, &(0x7f0000000580)=ANY=[], 0x10) r4 = openat(0xffffffffffffff9c, &(0x7f0000000040)='./file0\x00', 0x42, 0x0) mmap$IORING_OFF_SQ_RING(&(0x7f0000ffd000/0x3000)=nil, 0x3000, 0x0, 0x12, r4, 0x0) r5 = userfaultfd(0x80801) ioctl$UFFDIO_API(r5, 0xc018aa3f, &(0x7f0000000040)={0xaa, 0x20}) r6 = userfaultfd(0x80801) ioctl$UFFDIO_API(r6, 0xc018aa3f, &(0x7f0000000140)) ioctl$UFFDIO_REGISTER(r6, 0xc020aa00, &(0x7f0000000100)={{&(0x7f0000ffc000/0x4000)=nil, 0x4000}, 0x2}) ioctl$UFFDIO_ZEROPAGE(r5, 0xc020aa04, &(0x7f0000000000)={{&(0x7f0000ffd000/0x1000)=nil, 0x1000}}) r7 = bpf$PROG_LOAD(0x5, &(0x7f0000000140)={0x2, 0x3, &(0x7f0000000200)=ANY=[@ANYBLOB="1800000000120000000000000000000095"], &(0x7f0000000000)='GPL\x00', 0x7, 0x0, 0x0, 0x0, 0x0, '\x00', 0x0, @fallback=0x30, 0xffffffffffffffff, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x10, 0x0, @void, @value}, 0x94) bpf$BPF_LINK_CREATE_XDP(0x1c, &(0x7f0000000040)={r7, 0x0, 0x30, 0x1e, @val=@uprobe_multi={&(0x7f0000000080)='./file0\x00', &(0x7f0000000100)=[0x2], 0x0, 0x0, 0x1}}, 0x40)
The cause is that zero pfn is set to the PTE without increasing the RSS count in mfill_atomic_pte_zeropage() and the refcount of zero folio does not increase accordingly. Then, the operation on the same pfn is performed in uprobe_write_opcode()->__replace_page() to unconditional decrease the RSS count and old_folio's refcount.
Therefore, two bugs are introduced:
-
The RSS count is incorrect, when process exit, the check_mm() report error "Bad rss-count".
-
The reserved folio (zero folio) is freed when folio->refcount is zero, then free_pages_prepare->free_page_is_bad() report error "Bad page state".
There is more, the following warning could also theoretically be triggered:
__replace_page() -> ... -> folio_remove_rmap_pte() -> VM_WARN_ON_FOLIO(is_zero_folio(folio), folio)
Considering that uprobe hit on the zero folio is a very rare case, just reject zero old folio immediately after get_user_page_vma_remote().
[ mingo: Cleaned up the changelog ]
{
"affected": [],
"aliases": [
"CVE-2025-21881"
],
"database_specific": {
"cwe_ids": [],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2025-03-27T15:15:56Z",
"severity": null
},
"details": "In the Linux kernel, the following vulnerability has been resolved:\n\nuprobes: Reject the shared zeropage in uprobe_write_opcode()\n\nWe triggered the following crash in syzkaller tests:\n\n BUG: Bad page state in process syz.7.38 pfn:1eff3\n page: refcount:0 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x1eff3\n flags: 0x3fffff00004004(referenced|reserved|node=0|zone=1|lastcpupid=0x1fffff)\n raw: 003fffff00004004 ffffe6c6c07bfcc8 ffffe6c6c07bfcc8 0000000000000000\n raw: 0000000000000000 0000000000000000 00000000fffffffe 0000000000000000\n page dumped because: PAGE_FLAGS_CHECK_AT_FREE flag(s) set\n Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-1ubuntu1.1 04/01/2014\n Call Trace:\n \u003cTASK\u003e\n dump_stack_lvl+0x32/0x50\n bad_page+0x69/0xf0\n free_unref_page_prepare+0x401/0x500\n free_unref_page+0x6d/0x1b0\n uprobe_write_opcode+0x460/0x8e0\n install_breakpoint.part.0+0x51/0x80\n register_for_each_vma+0x1d9/0x2b0\n __uprobe_register+0x245/0x300\n bpf_uprobe_multi_link_attach+0x29b/0x4f0\n link_create+0x1e2/0x280\n __sys_bpf+0x75f/0xac0\n __x64_sys_bpf+0x1a/0x30\n do_syscall_64+0x56/0x100\n entry_SYSCALL_64_after_hwframe+0x78/0xe2\n\n BUG: Bad rss-counter state mm:00000000452453e0 type:MM_FILEPAGES val:-1\n\nThe following syzkaller test case can be used to reproduce:\n\n r2 = creat(\u0026(0x7f0000000000)=\u0027./file0\\x00\u0027, 0x8)\n write$nbd(r2, \u0026(0x7f0000000580)=ANY=[], 0x10)\n r4 = openat(0xffffffffffffff9c, \u0026(0x7f0000000040)=\u0027./file0\\x00\u0027, 0x42, 0x0)\n mmap$IORING_OFF_SQ_RING(\u0026(0x7f0000ffd000/0x3000)=nil, 0x3000, 0x0, 0x12, r4, 0x0)\n r5 = userfaultfd(0x80801)\n ioctl$UFFDIO_API(r5, 0xc018aa3f, \u0026(0x7f0000000040)={0xaa, 0x20})\n r6 = userfaultfd(0x80801)\n ioctl$UFFDIO_API(r6, 0xc018aa3f, \u0026(0x7f0000000140))\n ioctl$UFFDIO_REGISTER(r6, 0xc020aa00, \u0026(0x7f0000000100)={{\u0026(0x7f0000ffc000/0x4000)=nil, 0x4000}, 0x2})\n ioctl$UFFDIO_ZEROPAGE(r5, 0xc020aa04, \u0026(0x7f0000000000)={{\u0026(0x7f0000ffd000/0x1000)=nil, 0x1000}})\n r7 = bpf$PROG_LOAD(0x5, \u0026(0x7f0000000140)={0x2, 0x3, \u0026(0x7f0000000200)=ANY=[@ANYBLOB=\"1800000000120000000000000000000095\"], \u0026(0x7f0000000000)=\u0027GPL\\x00\u0027, 0x7, 0x0, 0x0, 0x0, 0x0, \u0027\\x00\u0027, 0x0, @fallback=0x30, 0xffffffffffffffff, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x10, 0x0, @void, @value}, 0x94)\n bpf$BPF_LINK_CREATE_XDP(0x1c, \u0026(0x7f0000000040)={r7, 0x0, 0x30, 0x1e, @val=@uprobe_multi={\u0026(0x7f0000000080)=\u0027./file0\\x00\u0027, \u0026(0x7f0000000100)=[0x2], 0x0, 0x0, 0x1}}, 0x40)\n\nThe cause is that zero pfn is set to the PTE without increasing the RSS\ncount in mfill_atomic_pte_zeropage() and the refcount of zero folio does\nnot increase accordingly. Then, the operation on the same pfn is performed\nin uprobe_write_opcode()-\u003e__replace_page() to unconditional decrease the\nRSS count and old_folio\u0027s refcount.\n\nTherefore, two bugs are introduced:\n\n 1. The RSS count is incorrect, when process exit, the check_mm() report\n error \"Bad rss-count\".\n\n 2. The reserved folio (zero folio) is freed when folio-\u003erefcount is zero,\n then free_pages_prepare-\u003efree_page_is_bad() report error\n \"Bad page state\".\n\nThere is more, the following warning could also theoretically be triggered:\n\n __replace_page()\n -\u003e ...\n -\u003e folio_remove_rmap_pte()\n -\u003e VM_WARN_ON_FOLIO(is_zero_folio(folio), folio)\n\nConsidering that uprobe hit on the zero folio is a very rare case, just\nreject zero old folio immediately after get_user_page_vma_remote().\n\n[ mingo: Cleaned up the changelog ]",
"id": "GHSA-v2hx-rxf2-p233",
"modified": "2025-03-27T15:31:11Z",
"published": "2025-03-27T15:31:11Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2025-21881"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/0b6f19714588cf2366b0364234f97ba963688f63"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/13cca2b73e2b0ec3ea6d6615d615395621d22752"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/54011fc94422f094eaf47555284de70a4bc32bb9"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/bddf10d26e6e5114e7415a0e442ec6f51a559468"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/c4cb2bfa99513311886c1eb5c1c2ac26f3338a6e"
}
],
"schema_version": "1.4.0",
"severity": []
}
Sightings
| Author | Source | Type | Date |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or seen somewhere by the user.
- Confirmed: The vulnerability is confirmed from an analyst perspective.
- Exploited: This vulnerability was exploited and seen by the user reporting the sighting.
- Patched: This vulnerability was successfully patched by the user reporting the sighting.
- Not exploited: This vulnerability was not exploited or seen by the user reporting the sighting.
- Not confirmed: The user expresses doubt about the veracity of the vulnerability.
- Not patched: This vulnerability was not successfully patched by the user reporting the sighting.