Re: [Xen-devel] [PATCH v2] x86/vmx: Fix vmentry failure because of invalid LER on Broadwell

2017-05-31 Thread Tian, Kevin
> From: Ross Lagerwall [mailto:ross.lagerw...@citrix.com] > Sent: Tuesday, May 30, 2017 10:05 PM > > Occasionally, on certain Broadwell CPUs MSR_IA32_LASTINTTOIP has been > observed to have the top three bits corrupted as though the MSR is using > the LBR_FORMAT_EIP_FLAGS_TSX format. This is

Re: [Xen-devel] [PATCH v2] x86/vmx: Fix vmentry failure because of invalid LER on Broadwell

2017-05-30 Thread Jan Beulich
>>> On 30.05.17 at 16:05, wrote: > --- a/xen/include/asm-x86/x86_64/page.h > +++ b/xen/include/asm-x86/x86_64/page.h > @@ -28,6 +28,9 @@ > #define PADDR_MASK ((1UL << PADDR_BITS)-1) > #define VADDR_MASK ((1UL << VADDR_BITS)-1) > > +#define

Re: [Xen-devel] [PATCH v2] x86/vmx: Fix vmentry failure because of invalid LER on Broadwell

2017-05-30 Thread Andrew Cooper
On 30/05/17 15:05, Ross Lagerwall wrote: > Occasionally, on certain Broadwell CPUs MSR_IA32_LASTINTTOIP has been > observed to have the top three bits corrupted as though the MSR is using > the LBR_FORMAT_EIP_FLAGS_TSX format. This is incorrect and causes a > vmentry failure -- the MSR should

[Xen-devel] [PATCH v2] x86/vmx: Fix vmentry failure because of invalid LER on Broadwell

2017-05-30 Thread Ross Lagerwall
Occasionally, on certain Broadwell CPUs MSR_IA32_LASTINTTOIP has been observed to have the top three bits corrupted as though the MSR is using the LBR_FORMAT_EIP_FLAGS_TSX format. This is incorrect and causes a vmentry failure -- the MSR should contain an offset into the current code segment. This