Re: [Xen-devel] [PATCH] X86/VPMU: mask off uncore overflow bit on xeon phi knights landing

2016-12-12 Thread Tian, Kevin
> From: Jan Beulich [mailto:jbeul...@suse.com] > Sent: Friday, December 09, 2016 5:43 PM ing > > >>> On 09.12.16 at 04:17, wrote: > > IA32_PERF_GLOBAL_STATUS.OvfUncore (MSR 38EH, bit[61]) is always 0 and > > writing 1 to IA32_PERF_GLOBAL_OVF_CTRL.ClrOvfUncore (MSR 390H, bit[61]) > > signals #GP.

Re: [Xen-devel] [PATCH] X86/VPMU: mask off uncore overflow bit on xeon phi knights landing

2016-12-12 Thread Jan Beulich
>>> On 12.12.16 at 09:06, wrote: >> >>> On 12.12.16 at 07:51, wrote: >> > By the way, I think another place may need to do some modify as well. >> > >> > @@ -868,7 +868,7 @@ static int core2_vpmu_do_interrupt(struct >> > cpu_user_regs *regs) >> > if ( is_pmc_quirk ) >> > ha

Re: [Xen-devel] [PATCH] X86/VPMU: mask off uncore overflow bit on xeon phi knights landing

2016-12-12 Thread Kang, Luwei
> >>> On 12.12.16 at 07:51, wrote: > > By the way, I think another place may need to do some modify as well. > > > > @@ -868,7 +868,7 @@ static int core2_vpmu_do_interrupt(struct cpu_user_regs > > *regs) > > if ( is_pmc_quirk ) > > handle_pmc_quirk(msr_content); > >

Re: [Xen-devel] [PATCH] X86/VPMU: mask off uncore overflow bit on xeon phi knights landing

2016-12-11 Thread Jan Beulich
>>> On 12.12.16 at 07:51, wrote: > By the way, I think another place may need to do some modify as well. > > @@ -868,7 +868,7 @@ static int core2_vpmu_do_interrupt(struct cpu_user_regs > *regs) > if ( is_pmc_quirk ) > handle_pmc_quirk(msr_content); > core2_vpmu_cxt

Re: [Xen-devel] [PATCH] X86/VPMU: mask off uncore overflow bit on xeon phi knights landing

2016-12-11 Thread Kang, Luwei
> On 12/08/2016 10:17 PM, Luwei Kang wrote: > > IA32_PERF_GLOBAL_STATUS.OvfUncore (MSR 38EH, bit[61]) is always 0 and > > writing 1 to IA32_PERF_GLOBAL_OVF_CTRL.ClrOvfUncore (MSR 390H, > > bit[61]) signals #GP. > > Reference "Intel Xeon Phi Procssor x200 Product Family", document > > number 334646-

Re: [Xen-devel] [PATCH] X86/VPMU: mask off uncore overflow bit on xeon phi knights landing

2016-12-09 Thread Boris Ostrovsky
On 12/08/2016 10:17 PM, Luwei Kang wrote: > IA32_PERF_GLOBAL_STATUS.OvfUncore (MSR 38EH, bit[61]) is always 0 and > writing 1 to IA32_PERF_GLOBAL_OVF_CTRL.ClrOvfUncore (MSR 390H, bit[61]) > signals #GP. > Reference "Intel Xeon Phi Procssor x200 Product Family", document > number 334646-008. > > Sig

Re: [Xen-devel] [PATCH] X86/VPMU: mask off uncore overflow bit on xeon phi knights landing

2016-12-09 Thread Jan Beulich
>>> On 09.12.16 at 04:17, wrote: > IA32_PERF_GLOBAL_STATUS.OvfUncore (MSR 38EH, bit[61]) is always 0 and > writing 1 to IA32_PERF_GLOBAL_OVF_CTRL.ClrOvfUncore (MSR 390H, bit[61]) > signals #GP. > Reference "Intel Xeon Phi Procssor x200 Product Family", document > number 334646-008. I can see this

[Xen-devel] [PATCH] X86/VPMU: mask off uncore overflow bit on xeon phi knights landing

2016-12-08 Thread Luwei Kang
IA32_PERF_GLOBAL_STATUS.OvfUncore (MSR 38EH, bit[61]) is always 0 and writing 1 to IA32_PERF_GLOBAL_OVF_CTRL.ClrOvfUncore (MSR 390H, bit[61]) signals #GP. Reference "Intel Xeon Phi Procssor x200 Product Family", document number 334646-008. Signed-off-by: Luwei Kang --- xen/arch/x86/cpu/vpmu_inte