PT-2026-98608 · Linux · Linux
CVE-2026-97944
·
Published
2026-09-25
·
Updated
2026-09-25
None
No severity ratings or metrics are available. When they are, we'll update the corresponding info on the page.
In the Linux kernel, the following vulnerability has been resolved:
x86/cfi: Fix FineIBT hash offset in cfi get func hash()
The switch of the FineIBT preamble from "subl $hash, %r10d" to the
shorter "subl $hash, %eax" moved the hash immediate from offset 7 to
offset 5 of the preamble. fineibt preamble hash was updated to match,
but the open-coded offset in cfi get func hash() was missed and it
still reads the hash at offset 7.
cfi get func hash() is used by the BPF JIT to give a struct ops
trampoline the CFI hash of the stub function it stands in for. With
FineIBT the trampoline now gets the upper half of the real hash
followed by the first two bytes of the next instruction, so the first
indirect call from the kernel into a struct ops program,
tcp init congestion control() calling ->init() of a BPF congestion
control for example, fails the FineIBT check and the kernel dies with
a CFI failure.
Move the FineIBT preamble template and its offset defines above
cfi get func hash() and use fineibt preamble hash there, so every
reader of the preamble shares one definition of its layout. The
CFI FINEIBT arm is only built with CONFIG FINEIBT, the only
configuration in which cfi mode can take that value.
cfi get func arity() does not need the same treatment: the bhi args
call whose displacement it reads still ends at the function address.
Found an issue in the description? Have something to add? Feel free to write us 👾
Related Identifiers
Affected Products
Linux