CVE-2025-38033

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> x86/Kconfig: make CFI_AUTO_DEFAULT depend on !RUST or Rust &gt;= 1.88<br /> <br /> Calling core::fmt::write() from rust code while FineIBT is enabled<br /> results in a kernel panic:<br /> <br /> [ 4614.199779] kernel BUG at arch/x86/kernel/cet.c:132!<br /> [ 4614.205343] Oops: invalid opcode: 0000 [#1] PREEMPT SMP NOPTI<br /> [ 4614.211781] CPU: 2 UID: 0 PID: 6057 Comm: dmabuf_dump Tainted: G U O 6.12.17-android16-0-g6ab38c534a43 #1 9da040f27673ec3945e23b998a0f8bd64c846599<br /> [ 4614.227832] Tainted: [U]=USER, [O]=OOT_MODULE<br /> [ 4614.241247] RIP: 0010:do_kernel_cp_fault+0xea/0xf0<br /> ...<br /> [ 4614.398144] RIP: 0010:_RNvXs5_NtNtNtCs3o2tGsuHyou_4core3fmt3num3impyNtB9_7Display3fmt+0x0/0x20<br /> [ 4614.407792] Code: 48 f7 df 48 0f 48 f9 48 89 f2 89 c6 5d e9 18 fd ff ff 0f 1f 84 00 00 00 00 00 f3 0f 1e fa 41 81 ea 14 61 af 2c 74 03 0f 0b 90 0f 1f 00 55 48 89 e5 48 89 f2 48 8b 3f be 01 00 00 00 5d e9 e7<br /> [ 4614.428775] RSP: 0018:ffffb95acfa4ba68 EFLAGS: 00010246<br /> [ 4614.434609] RAX: 0000000000000000 RBX: 0000000000000010 RCX: 0000000000000000<br /> [ 4614.442587] RDX: 0000000000000007 RSI: ffffb95acfa4ba70 RDI: ffffb95acfa4bc88<br /> [ 4614.450557] RBP: ffffb95acfa4bae0 R08: ffff0a00ffffff05 R09: 0000000000000070<br /> [ 4614.458527] R10: 0000000000000000 R11: ffffffffab67eaf0 R12: ffffb95acfa4bcc8<br /> [ 4614.466493] R13: ffffffffac5d50f0 R14: 0000000000000000 R15: 0000000000000000<br /> [ 4614.474473] ? __cfi__RNvXs5_NtNtNtCs3o2tGsuHyou_4core3fmt3num3impyNtB9_7Display3fmt+0x10/0x10<br /> [ 4614.484118] ? _RNvNtCs3o2tGsuHyou_4core3fmt5write+0x1d2/0x250<br /> <br /> This happens because core::fmt::write() calls<br /> core::fmt::rt::Argument::fmt(), which currently has CFI disabled:<br /> <br /> library/core/src/fmt/rt.rs:<br /> 171 // FIXME: Transmuting formatter in new and indirectly branching to/calling<br /> 172 // it here is an explicit CFI violation.<br /> 173 #[allow(inline_no_sanitize)]<br /> 174 #[no_sanitize(cfi, kcfi)]<br /> 175 #[inline]<br /> 176 pub(super) unsafe fn fmt(&amp;self, f: &amp;mut Formatter) -&gt; Result {<br /> <br /> This causes a Control Protection exception, because FineIBT has sealed<br /> off the original function&amp;#39;s endbr64.<br /> <br /> This makes rust currently incompatible with FineIBT. Add a Kconfig<br /> dependency that prevents FineIBT from getting turned on by default<br /> if rust is enabled.<br /> <br /> [ Rust 1.88.0 (scheduled for 2025-06-26) should have this fixed [1],<br /> and thus we relaxed the condition with Rust &gt;= 1.88.<br /> <br /> When `objtool` lands checking for this with e.g. [2], the plan is<br /> to ideally run that in upstream Rust&amp;#39;s CI to prevent regressions<br /> early [3], since we do not control `core`&amp;#39;s source code.<br /> <br /> Alice tested the Rust PR backported to an older compiler.<br /> <br /> Peter would like that Rust provides a stable `core` which can be<br /> pulled into the kernel: "Relying on that much out of tree code is<br /> &amp;#39;unfortunate&amp;#39;".<br /> <br /> - Miguel ]<br /> <br /> [ Reduced splat. - Miguel ]