Re: [PATCH v4] kvm: better MWAIT emulation for guests

2017-03-16 Thread Wanpeng Li
2017-03-16 4:13 GMT+08:00 Radim Krčmář : > 2017-03-15 21:28+0200, Michael S. Tsirkin: >> Guests running Mac OS 5, 6, and 7 (Leopard through Lion) have a problem: >> unless explicitly provided with kernel command line argument >> "idlehalt=0" they'd implicitly assume MONITOR and MWAIT availability,

Re: [PATCH v4] kvm: better MWAIT emulation for guests

2017-03-16 Thread kbuild test robot
Hi Michael, [auto build test WARNING on kvm/linux-next] [also build test WARNING on v4.11-rc2 next-20170310] [if your patch is applied to the wrong git tree, please drop us a note to help improve the system] url: https://github.com/0day-ci/linux/commits/Michael-S-Tsirkin/kvm-better-MWAIT-emu

Re: [PATCH v4] kvm: better MWAIT emulation for guests

2017-03-16 Thread kbuild test robot
Hi Michael, [auto build test ERROR on kvm/linux-next] [also build test ERROR on v4.11-rc2 next-20170310] [if your patch is applied to the wrong git tree, please drop us a note to help improve the system] url: https://github.com/0day-ci/linux/commits/Michael-S-Tsirkin/kvm-better-MWAIT-emulati

Re: [PATCH v4] kvm: better MWAIT emulation for guests

2017-03-15 Thread Gabriel L. Somlo
On Wed, Mar 15, 2017 at 09:46:18PM +0100, Radim Krčmář wrote: > 2017-03-15 16:21-0400, Gabriel L. Somlo: > > On Wed, Mar 15, 2017 at 09:13:49PM +0100, Radim Krčmář wrote: > >> 2017-03-15 21:28+0200, Michael S. Tsirkin: > >> > Guests running Mac OS 5, 6, and 7 (Leopard through Lion) have a problem:

Re: [PATCH v4] kvm: better MWAIT emulation for guests

2017-03-15 Thread Radim Krčmář
2017-03-15 16:21-0400, Gabriel L. Somlo: > On Wed, Mar 15, 2017 at 09:13:49PM +0100, Radim Krčmář wrote: >> 2017-03-15 21:28+0200, Michael S. Tsirkin: >> > Guests running Mac OS 5, 6, and 7 (Leopard through Lion) have a problem: >> > unless explicitly provided with kernel command line argument >> >

Re: [PATCH v4] kvm: better MWAIT emulation for guests

2017-03-15 Thread Gabriel L. Somlo
On Wed, Mar 15, 2017 at 09:13:49PM +0100, Radim Krčmář wrote: > 2017-03-15 21:28+0200, Michael S. Tsirkin: > > Guests running Mac OS 5, 6, and 7 (Leopard through Lion) have a problem: > > unless explicitly provided with kernel command line argument > > "idlehalt=0" they'd implicitly assume MONITOR

Re: [PATCH v4] kvm: better MWAIT emulation for guests

2017-03-15 Thread Gabriel L. Somlo
On Wed, Mar 15, 2017 at 04:21:41PM -0400, Gabriel L. Somlo wrote: > > > > - do you see VM exits on the "hung" VCPU? > > how would I go about looking ? > > > - what is your CPU model? > > $ cat /proc/cpuinfo > ... > processor : 3 > vendor_id : GenuineIntel > cpu family : 6 > m

Re: [PATCH v4] kvm: better MWAIT emulation for guests

2017-03-15 Thread Radim Krčmář
2017-03-15 21:28+0200, Michael S. Tsirkin: > Guests running Mac OS 5, 6, and 7 (Leopard through Lion) have a problem: > unless explicitly provided with kernel command line argument > "idlehalt=0" they'd implicitly assume MONITOR and MWAIT availability, > without checking CPUID. > > We currently em

[PATCH v4] kvm: better MWAIT emulation for guests

2017-03-15 Thread Michael S. Tsirkin
Guests running Mac OS 5, 6, and 7 (Leopard through Lion) have a problem: unless explicitly provided with kernel command line argument "idlehalt=0" they'd implicitly assume MONITOR and MWAIT availability, without checking CPUID. We currently emulate that as a NOP but on VMX we can do better: let gu