Wind River Support Network

Description

In the Linux kernel, the following vulnerability has been resolved:EOL][EOL]fs/eventpoll: fix endless busy loop after timeout has expired[EOL][EOL]After commit 0a65bc27bd64 ("eventpoll: Set epoll timeout if it's in[EOL]the future"), the following program would immediately enter a busy[EOL]loop in the kernel:[EOL][EOL]```[EOL]int main() {[EOL]  int e = epoll_create1(0);[EOL]  struct epoll_event event = {.events = EPOLLIN};[EOL]  epoll_ctl(e, EPOLL_CTL_ADD, 0, &event);[EOL]  const struct timespec timeout = {.tv_nsec = 1};[EOL]  epoll_pwait2(e, &event, 1, &timeout, 0);[EOL]}[EOL]```[EOL][EOL]This happens because the given (non-zero) timeout of 1 nanosecond[EOL]usually expires before ep_poll() is entered and then[EOL]ep_schedule_timeout() returns false, but `timed_out` is never set[EOL]because the code line that sets it is skipped.  This quickly turns[EOL]into a soft lockup, RCU stalls and deadlocks, inflicting severe[EOL]headaches to the whole system.[EOL][EOL]When the timeout has expired, we don't need to schedule a hrtimer, but[EOL]we should set the `timed_out` variable.  Therefore, I suggest moving[EOL]the ep_schedule_timeout() check into the `timed_out` expression[EOL]instead of skipping it.[EOL][EOL]brauner: Note that there was an earlier fix by Joe Damato in response to[EOL]my bug report in [1].

CREATE(Triage):(User=lchen-cn) [CVE-2025-38017 (https://nvd.nist.gov/vuln/detail/CVE-2025-38017)