Oracle Linux: CVE-2024-26603: ELSA-2024-3618: kernel update (MODERATE) (Multiple Advisories)

Description

In the Linux kernel, the following vulnerability has been resolved: x86/fpu: Stop relying on userspace for info to fault in xsave buffer Before this change, the expected size of the user space buffer was taken from fx_sw->xstate_size. fx_sw->xstate_size can be changed from user-space, so it is possible construct a sigreturn frame where: * fx_sw->xstate_size is smaller than the size required by valid bits in fx_sw->xfeatures. * user-space unmaps parts of the sigrame fpu buffer so that not all of the buffer required by xrstor is accessible. In this case, xrstor tries to restore and accesses the unmapped area which results in a fault. But fault_in_readable succeeds because buf + fx_sw->xstate_size is within the still mapped area, so it goes back and tries xrstor again. It will spin in this loop forever. Instead, fault in the maximum size which can be touched by XRSTOR (taken from fpstate->user_size). [ dhansen: tweak subject / changelog ] A flaw was found in the Linux kernel's x86/fpu module, which revolves around an issue with relying on user space for critical information regarding the xsave buffer. In the affected scenario, the expected size of the user space buffer is derived from user-controlled data, specifically, fx_sw->xstate_size. By manipulating this value, an attacker could construct a malicious sigreturn frame where the indicated size is smaller than required by valid bits in fx_sw->xfeatures. Furthermore, the attacker could unmap portions of the fpu buffer in the user space, rendering them inaccessible to xrstor. This manipulation leads to a situation where xrstor repeatedly attempts to restore and access an unmapped area, causing a fault. However, the fault_in_readable function erroneously succeeds because the accessed region, buf + fx_sw->xstate_size, remains within the mapped area. Consequently, the system enters a perpetual loop as xrstor continually retries the operation.