ghsa-x3cp-9cgr-m6gc
Vulnerability from github
In the Linux kernel, the following vulnerability has been resolved:
KVM: x86: Mark target gfn of emulated atomic instruction as dirty
When emulating an atomic access on behalf of the guest, mark the target gfn dirty if the CMPXCHG by KVM is attempted and doesn't fault. This fixes a bug where KVM effectively corrupts guest memory during live migration by writing to guest memory without informing userspace that the page is dirty.
Marking the page dirty got unintentionally dropped when KVM's emulated CMPXCHG was converted to do a user access. Before that, KVM explicitly mapped the guest page into kernel memory, and marked the page dirty during the unmap phase.
Mark the page dirty even if the CMPXCHG fails, as the old data is written back on failure, i.e. the page is still written. The value written is guaranteed to be the same because the operation is atomic, but KVM's ABI is that all writes are dirty logged regardless of the value written. And more importantly, that's what KVM did before the buggy commit.
Huge kudos to the folks on the Cc list (and many others), who did all the actual work of triaging and debugging.
base-commit: 6769ea8da8a93ed4630f1ce64df6aafcaabfce64
{
"affected": [],
"aliases": [
"CVE-2024-35804"
],
"database_specific": {
"cwe_ids": [],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2024-05-17T14:15:13Z",
"severity": null
},
"details": "In the Linux kernel, the following vulnerability has been resolved:\n\nKVM: x86: Mark target gfn of emulated atomic instruction as dirty\n\nWhen emulating an atomic access on behalf of the guest, mark the target\ngfn dirty if the CMPXCHG by KVM is attempted and doesn\u0027t fault. This\nfixes a bug where KVM effectively corrupts guest memory during live\nmigration by writing to guest memory without informing userspace that the\npage is dirty.\n\nMarking the page dirty got unintentionally dropped when KVM\u0027s emulated\nCMPXCHG was converted to do a user access. Before that, KVM explicitly\nmapped the guest page into kernel memory, and marked the page dirty during\nthe unmap phase.\n\nMark the page dirty even if the CMPXCHG fails, as the old data is written\nback on failure, i.e. the page is still written. The value written is\nguaranteed to be the same because the operation is atomic, but KVM\u0027s ABI\nis that all writes are dirty logged regardless of the value written. And\nmore importantly, that\u0027s what KVM did before the buggy commit.\n\nHuge kudos to the folks on the Cc list (and many others), who did all the\nactual work of triaging and debugging.\n\nbase-commit: 6769ea8da8a93ed4630f1ce64df6aafcaabfce64",
"id": "GHSA-x3cp-9cgr-m6gc",
"modified": "2024-05-17T15:31:09Z",
"published": "2024-05-17T15:31:09Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2024-35804"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/225d587a073584946c05c9b7651d637bd45c0c71"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/726374dde5d608b15b9756bd52b6fc283fda7a06"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/910c57dfa4d113aae6571c2a8b9ae8c430975902"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/9d1b22e573a3789ed1f32033ee709106993ba551"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/a9bd6bb6f02bf7132c1ab192ba62bbfa52df7d66"
}
],
"schema_version": "1.4.0",
"severity": []
}