Messages in this thread | | | Date | Mon, 18 Jul 2022 11:31:33 +0200 | From | Peter Zijlstra <> | Subject | Re: [patch 02/38] x86/cpu: Use native_wrmsrl() in load_percpu_segment() |
| |
On Mon, Jul 18, 2022 at 10:55:29AM +0200, Thomas Gleixner wrote: > On Mon, Jul 18 2022 at 08:54, Thomas Gleixner wrote: > > On Mon, Jul 18 2022 at 07:11, Juergen Gross wrote: > >>> - switch_to_new_gdt(cpu); > >>> + switch_to_real_gdt(cpu); > >> > >> ... can't you use the paravirt variant of load_gdt in switch_to_real_gdt() ? > > > > That does not solve the problem of having a disagreement between GDT and > > GS_BASE. Let me dig into this some more. > > Bah. The real problem is __loadsegment_simple(gs, 0). After that GS_BASE > is 0. So any per CPU access before setting MSR_GS_BASE back to working > state is going into lala land. > > So it's not the GDT. It's the mov 0, %gs which makes stuff go south, but > as %gs is already 0, we can keep the paravirt load_gdt() and use > native_write_msr() and everything should be happy.
How is the ret from xen_load_gdt() not going to explode?
| |