ghsa-pj67-hxcx-g74v
Vulnerability from github
In the Linux kernel, the following vulnerability has been resolved:
arm64: probes: Remove broken LDR (literal) uprobe support
The simulate_ldr_literal() and simulate_ldrsw_literal() functions are unsafe to use for uprobes. Both functions were originally written for use with kprobes, and access memory with plain C accesses. When uprobes was added, these were reused unmodified even though they cannot safely access user memory.
There are three key problems:
1) The plain C accesses do not have corresponding extable entries, and thus if they encounter a fault the kernel will treat these as unintentional accesses to user memory, resulting in a BUG() which will kill the kernel thread, and likely lead to further issues (e.g. lockup or panic()).
2) The plain C accesses are subject to HW PAN and SW PAN, and so when either is in use, any attempt to simulate an access to user memory will fault. Thus neither simulate_ldr_literal() nor simulate_ldrsw_literal() can do anything useful when simulating a user instruction on any system with HW PAN or SW PAN.
3) The plain C accesses are privileged, as they run in kernel context, and in practice can access a small range of kernel virtual addresses. The instructions they simulate have a range of +/-1MiB, and since the simulated instructions must itself be a user instructions in the TTBR0 address range, these can address the final 1MiB of the TTBR1 acddress range by wrapping downwards from an address in the first 1MiB of the TTBR0 address range.
In contemporary kernels the last 8MiB of TTBR1 address range is reserved, and accesses to this will always fault, meaning this is no worse than (1).
Historically, it was theoretically possible for the linear map or vmemmap to spill into the final 8MiB of the TTBR1 address range, but in practice this is extremely unlikely to occur as this would require either:
-
Having enough physical memory to fill the entire linear map all the way to the final 1MiB of the TTBR1 address range.
-
Getting unlucky with KASLR randomization of the linear map such that the populated region happens to overlap with the last 1MiB of the TTBR address range.
... and in either case if we were to spill into the final page there would be larger problems as the final page would alias with error pointers.
Practically speaking, (1) and (2) are the big issues. Given there have been no reports of problems since the broken code was introduced, it appears that no-one is relying on probing these instructions with uprobes.
Avoid these issues by not allowing uprobes on LDR (literal) and LDRSW (literal), limiting the use of simulate_ldr_literal() and simulate_ldrsw_literal() to kprobes. Attempts to place uprobes on LDR (literal) and LDRSW (literal) will be rejected as arm_probe_decode_insn() will return INSN_REJECTED. In future we can consider introducing working uprobes support for these instructions, but this will require more significant work.
{
"affected": [],
"aliases": [
"CVE-2024-50099"
],
"database_specific": {
"cwe_ids": [],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2024-11-05T18:15:13Z",
"severity": "MODERATE"
},
"details": "In the Linux kernel, the following vulnerability has been resolved:\n\narm64: probes: Remove broken LDR (literal) uprobe support\n\nThe simulate_ldr_literal() and simulate_ldrsw_literal() functions are\nunsafe to use for uprobes. Both functions were originally written for\nuse with kprobes, and access memory with plain C accesses. When uprobes\nwas added, these were reused unmodified even though they cannot safely\naccess user memory.\n\nThere are three key problems:\n\n1) The plain C accesses do not have corresponding extable entries, and\n thus if they encounter a fault the kernel will treat these as\n unintentional accesses to user memory, resulting in a BUG() which\n will kill the kernel thread, and likely lead to further issues (e.g.\n lockup or panic()).\n\n2) The plain C accesses are subject to HW PAN and SW PAN, and so when\n either is in use, any attempt to simulate an access to user memory\n will fault. Thus neither simulate_ldr_literal() nor\n simulate_ldrsw_literal() can do anything useful when simulating a\n user instruction on any system with HW PAN or SW PAN.\n\n3) The plain C accesses are privileged, as they run in kernel context,\n and in practice can access a small range of kernel virtual addresses.\n The instructions they simulate have a range of +/-1MiB, and since the\n simulated instructions must itself be a user instructions in the\n TTBR0 address range, these can address the final 1MiB of the TTBR1\n acddress range by wrapping downwards from an address in the first\n 1MiB of the TTBR0 address range.\n\n In contemporary kernels the last 8MiB of TTBR1 address range is\n reserved, and accesses to this will always fault, meaning this is no\n worse than (1).\n\n Historically, it was theoretically possible for the linear map or\n vmemmap to spill into the final 8MiB of the TTBR1 address range, but\n in practice this is extremely unlikely to occur as this would\n require either:\n\n * Having enough physical memory to fill the entire linear map all the\n way to the final 1MiB of the TTBR1 address range.\n\n * Getting unlucky with KASLR randomization of the linear map such\n that the populated region happens to overlap with the last 1MiB of\n the TTBR address range.\n\n ... and in either case if we were to spill into the final page there\n would be larger problems as the final page would alias with error\n pointers.\n\nPractically speaking, (1) and (2) are the big issues. Given there have\nbeen no reports of problems since the broken code was introduced, it\nappears that no-one is relying on probing these instructions with\nuprobes.\n\nAvoid these issues by not allowing uprobes on LDR (literal) and LDRSW\n(literal), limiting the use of simulate_ldr_literal() and\nsimulate_ldrsw_literal() to kprobes. Attempts to place uprobes on LDR\n(literal) and LDRSW (literal) will be rejected as\narm_probe_decode_insn() will return INSN_REJECTED. In future we can\nconsider introducing working uprobes support for these instructions, but\nthis will require more significant work.",
"id": "GHSA-pj67-hxcx-g74v",
"modified": "2024-11-12T18:30:51Z",
"published": "2024-11-05T18:32:11Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2024-50099"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/20cde998315a3d2df08e26079a3ea7501abce6db"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/3728b4eb27910ffedd173018279a970705f2e03a"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/9f1e7735474e7457a4d919a517900e46868ae5f6"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/acc450aa07099d071b18174c22a1119c57da8227"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/ad4bc35a6d22e9ff9b67d0d0c38bce654232f195"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/ae743deca78d9e4b7f4f60ad2f95e20e8ea057f9"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/bae792617a7e911477f67a3aff850ad4ddf51572"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/cc86f2e9876c8b5300238cec6bf0bd8c842078ee"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H",
"type": "CVSS_V3"
}
]
}
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.