CVE-2025-23135

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> RISC-V: KVM: Teardown riscv specific bits after kvm_exit<br /> <br /> During a module removal, kvm_exit invokes arch specific disable<br /> call which disables AIA. However, we invoke aia_exit before kvm_exit<br /> resulting in the following warning. KVM kernel module can&amp;#39;t be inserted<br /> afterwards due to inconsistent state of IRQ.<br /> <br /> [25469.031389] percpu IRQ 31 still enabled on CPU0!<br /> [25469.031732] WARNING: CPU: 3 PID: 943 at kernel/irq/manage.c:2476 __free_percpu_irq+0xa2/0x150<br /> [25469.031804] Modules linked in: kvm(-)<br /> [25469.031848] CPU: 3 UID: 0 PID: 943 Comm: rmmod Not tainted 6.14.0-rc5-06947-g91c763118f47-dirty #2<br /> [25469.031905] Hardware name: riscv-virtio,qemu (DT)<br /> [25469.031928] epc : __free_percpu_irq+0xa2/0x150<br /> [25469.031976] ra : __free_percpu_irq+0xa2/0x150<br /> [25469.032197] epc : ffffffff8007db1e ra : ffffffff8007db1e sp : ff2000000088bd50<br /> [25469.032241] gp : ffffffff8131cef8 tp : ff60000080b96400 t0 : ff2000000088baf8<br /> [25469.032285] t1 : fffffffffffffffc t2 : 5249207570637265 s0 : ff2000000088bd90<br /> [25469.032329] s1 : ff60000098b21080 a0 : 037d527a15eb4f00 a1 : 037d527a15eb4f00<br /> [25469.032372] a2 : 0000000000000023 a3 : 0000000000000001 a4 : ffffffff8122dbf8<br /> [25469.032410] a5 : 0000000000000fff a6 : 0000000000000000 a7 : ffffffff8122dc10<br /> [25469.032448] s2 : ff60000080c22eb0 s3 : 0000000200000022 s4 : 000000000000001f<br /> [25469.032488] s5 : ff60000080c22e00 s6 : ffffffff80c351c0 s7 : 0000000000000000<br /> [25469.032582] s8 : 0000000000000003 s9 : 000055556b7fb490 s10: 00007ffff0e12fa0<br /> [25469.032621] s11: 00007ffff0e13e9a t3 : ffffffff81354ac7 t4 : ffffffff81354ac7<br /> [25469.032664] t5 : ffffffff81354ac8 t6 : ffffffff81354ac7<br /> [25469.032698] status: 0000000200000100 badaddr: ffffffff8007db1e cause: 0000000000000003<br /> [25469.032738] [] __free_percpu_irq+0xa2/0x150<br /> [25469.032797] [] free_percpu_irq+0x30/0x5e<br /> [25469.032856] [] kvm_riscv_aia_exit+0x40/0x42 [kvm]<br /> [25469.033947] [] cleanup_module+0x10/0x32 [kvm]<br /> [25469.035300] [] __riscv_sys_delete_module+0x18e/0x1fc<br /> [25469.035374] [] syscall_handler+0x3a/0x46<br /> [25469.035456] [] do_trap_ecall_u+0x72/0x134<br /> [25469.035536] [] handle_exception+0x148/0x156<br /> <br /> Invoke aia_exit and other arch specific cleanup functions after kvm_exit<br /> so that disable gets a chance to be called first before exit.