Re: [PATCH v3 2/7] x86/arch_prctl: Fix ARCH_GET_FS and ARCH_GET_GS

2016-04-08 Thread Andy Lutomirski
On Fri, Apr 8, 2016 at 12:13 AM, Ingo Molnar wrote: > > * Andy Lutomirski wrote: > >> ARCH_GET_FS and ARCH_GET_GS attempted to figure out the fsbase and >> gsbase respectively from saved thread state. This was wrong: fsbase >> and gsbase live in registers

Re: [PATCH v3 2/7] x86/arch_prctl: Fix ARCH_GET_FS and ARCH_GET_GS

2016-04-08 Thread Andy Lutomirski
On Fri, Apr 8, 2016 at 12:13 AM, Ingo Molnar wrote: > > * Andy Lutomirski wrote: > >> ARCH_GET_FS and ARCH_GET_GS attempted to figure out the fsbase and >> gsbase respectively from saved thread state. This was wrong: fsbase >> and gsbase live in registers while a thread is running, not in >>

Re: [PATCH v3 2/7] x86/arch_prctl: Fix ARCH_GET_FS and ARCH_GET_GS

2016-04-08 Thread Dmitry Safonov
2016-04-08 10:13 GMT+03:00 Ingo Molnar : > > * Andy Lutomirski wrote: > >> ARCH_GET_FS and ARCH_GET_GS attempted to figure out the fsbase and >> gsbase respectively from saved thread state. This was wrong: fsbase >> and gsbase live in registers while a thread

Re: [PATCH v3 2/7] x86/arch_prctl: Fix ARCH_GET_FS and ARCH_GET_GS

2016-04-08 Thread Dmitry Safonov
2016-04-08 10:13 GMT+03:00 Ingo Molnar : > > * Andy Lutomirski wrote: > >> ARCH_GET_FS and ARCH_GET_GS attempted to figure out the fsbase and >> gsbase respectively from saved thread state. This was wrong: fsbase >> and gsbase live in registers while a thread is running, not in >> memory. > > So

Re: [PATCH v3 2/7] x86/arch_prctl: Fix ARCH_GET_FS and ARCH_GET_GS

2016-04-08 Thread Ingo Molnar
* Andy Lutomirski wrote: > ARCH_GET_FS and ARCH_GET_GS attempted to figure out the fsbase and > gsbase respectively from saved thread state. This was wrong: fsbase > and gsbase live in registers while a thread is running, not in > memory. So I'm wondering, the current code

Re: [PATCH v3 2/7] x86/arch_prctl: Fix ARCH_GET_FS and ARCH_GET_GS

2016-04-08 Thread Ingo Molnar
* Andy Lutomirski wrote: > ARCH_GET_FS and ARCH_GET_GS attempted to figure out the fsbase and > gsbase respectively from saved thread state. This was wrong: fsbase > and gsbase live in registers while a thread is running, not in > memory. So I'm wondering, the current code looks totally