Re: [RFC PATCH 2/4] rseq: Fix: Unregister rseq for CLONE_TLS

2019-09-10 Thread Mathieu Desnoyers
Of course, this patch title should read: rseq: Fix: Unregister rseq for CLONE_SETTLS - On Sep 11, 2019, at 1:27 AM, Mathieu Desnoyers mathieu.desnoy...@efficios.com wrote: > /* > * If parent process has a registered restartable sequences area, the > - * child inherits. Only applies

[RFC PATCH 2/4] rseq: Fix: Unregister rseq for CLONE_TLS

2019-09-10 Thread Mathieu Desnoyers
It has been reported by Google that rseq is not behaving properly with respect to clone when CLONE_VM is used without CLONE_THREAD. It keeps the prior thread's rseq TLS registered when the TLS of the thread has moved, so the kernel deals with the wrong TLS. The approach of clearing the per