Messages in this thread | | | From | Kees Cook <> | Subject | Re: [PATCH] cfi: Fix __cfi_slowpath_diag RCU usage with cpuidle | Date | Mon, 13 Jun 2022 09:26:10 -0700 |
| |
On Tue, 31 May 2022 10:59:10 -0700, Sami Tolvanen wrote: > RCU_NONIDLE usage during __cfi_slowpath_diag can result in an invalid > RCU state in the cpuidle code path: > > WARNING: CPU: 1 PID: 0 at kernel/rcu/tree.c:613 rcu_eqs_enter+0xe4/0x138 > ... > Call trace: > rcu_eqs_enter+0xe4/0x138 > rcu_idle_enter+0xa8/0x100 > cpuidle_enter_state+0x154/0x3a8 > cpuidle_enter+0x3c/0x58 > do_idle.llvm.6590768638138871020+0x1f4/0x2ec > cpu_startup_entry+0x28/0x2c > secondary_start_kernel+0x1b8/0x220 > __secondary_switched+0x94/0x98 > > [...]
Applied to for-next/hardening, thanks!
[1/1] cfi: Fix __cfi_slowpath_diag RCU usage with cpuidle https://git.kernel.org/kees/c/57cd6d157eb4
-- Kees Cook
| |