[PATCH] KVM: x86: always set accessed bit in shadow PTEs

Commit 7a1638ce4220 ("nEPT: Redefine EPT-specific link_shadow_page()", 2013-08-05) says: Since nEPT doesn't support A/D bit, we should not set those bit when building the shadow page table. but this is not necessary. Even though nEPT doesn't support A/D bits, and hence the vmcs12 EPT

Re: [PATCH] KVM: PPC: Exit guest upon fatal machine check exception

On Friday 13 November 2015 01:08 PM, Thomas Huth wrote: > On 13/11/15 07:26, Aravinda Prasad wrote: >> >> On Friday 13 November 2015 07:20 AM, David Gibson wrote: >>> On Thu, Nov 12, 2015 at 11:22:29PM +0530, Aravinda Prasad wrote: > [...] So thinking whether qemu should explicitly enable

Re: [PATCH] KVM: PPC: Exit guest upon fatal machine check exception

On Friday 13 November 2015 01:08 PM, Thomas Huth wrote: > On 13/11/15 07:26, Aravinda Prasad wrote: >> >> On Friday 13 November 2015 07:20 AM, David Gibson wrote: >>> On Thu, Nov 12, 2015 at 11:22:29PM +0530, Aravinda Prasad wrote: > [...] So thinking whether qemu should explicitly enable

Re: [PATCH 2/3] KVM: x86: request interrupt window when IRQ chip is split

On 12/11/2015 20:08, Matt Gingell wrote: > + bool req_int_win = !pic_in_kernel(vcpu->kvm) && > vcpu->run->request_interrupt_window; Hey, this is the new form of dm_request_for_irq_injection I have just written about, in the review to patch 3. So it's probably best to swap

Re: [PATCH 1/3] KVM: x86: set KVM_REQ_EVENT on local interupt request from user space

On 12/11/2015 20:08, Matt Gingell wrote: > Set KVM_REQ_EVENT when a PIC in user space injects a local interrupt. > > Currently a request is only made when neither the PIC nor the APIC is in > the kernel, which is not sufficient in the split IRQ chip case. > > This addresses a problem in QEMU

Re: [PATCH 3/3] KVM: x86: fix ready_for_interrupt reporting in split IRQ chip case

On 12/11/2015 20:07, Matt Gingell wrote: > This patch adds a call to kvm_arch_interrupt_allowed to ensure ready for > interrupt is reported to user space correctly. This addresses a problem > observed in QEMU when kvm->ready_for_interrupt is set but the x86 > interrupt flag is clear. > >

Re: [PATCH kvm-unit-tests] x86: tsc: add test for RDTSCP

On 13/11/2015 02:48, Wanpeng Li wrote: > > diff --git a/x86/unittests.cfg b/x86/unittests.cfg > > index ac652fc..fd7e9b2 100644 > > --- a/x86/unittests.cfg > > +++ b/x86/unittests.cfg > > @@ -114,6 +114,7 @@ file = sieve.flat > > > > [tsc] > > file = tsc.flat > > +extra_params = -cpu

Re: [PATCH] KVM: x86: expose MSR_TSC_AUX to userspace

> Paolo, under what circumstances (which versions of Windows? Anything > special running in the guest?) has this failure happened? I'd like to repro > this, I'm not sure if we've observed it before. We saw it with migration under Windows 10, nothing special running in the guest. It's very hard

Re: [PATCH net-next RFC V3 0/3] basic busy polling support for vhost_net

On 11/12/2015 08:02 PM, Felipe Franciosi wrote: > Hi Jason, > > I understand your busy loop timeout is quite conservative at 50us. Did you > try any other values? I've also tried 20us. And results shows 50us was better in: - very small packet tx (e.g 64bytes at most 46% improvement) - TCP_RR

Re: [Qemu-devel] [PATCH v3 1/3] target-i386: add a subsection for migrating vcpu's TSC rate

On Fri, Nov 13, 2015 at 10:23:54AM +0800, Haozhong Zhang wrote: > On 11/11/15 22:27, Haozhong Zhang wrote: > > On 11/11/15 12:16, Eduardo Habkost wrote: > [...] > > > > diff --git a/hw/i386/pc_q35.c b/hw/i386/pc_q35.c > > > > index 2f8f396..858ed69 100644 > > > > --- a/hw/i386/pc_q35.c > > > > +++

[GIT PULL] VFIO updates for v4.4-rc1

Hi Linus, The following changes since commit 32b88194f71d6ae7768a29f87fbba454728273ee: Linux 4.3-rc7 (2015-10-25 10:39:47 +0900) are available in the git repository at: git://github.com/awilliam/linux-vfio.git tags/vfio-v4.4-rc1 for you to fetch changes up to

Re: [PATCH v7 23/35] nvdimm: implement NVDIMM device abstract

On 02.11.2015 12:13, Xiao Guangrong wrote: Introduce "nvdimm" device which is based on dimm device type 128K memory region which is the minimum namespace label size required by NVDIMM Namespace Spec locates at the end of backend memory device is reserved for label data We can use "-m

Macvtap Loses Connectivity

Hi, I'm not subscribed to the list so please forgive this intrusion… I am a new user of KVM. I'm having an issue where connectivity is lost if the network a macvtap interface is connected to goes down. I've tried to describe the issue in detail in the following blog post. I was hoping to get

Re: [PATCH] KVM: x86: expose MSR_TSC_AUX to userspace

On Fri, Nov 13, 2015 at 1:42 AM, Paolo Bonzini wrote: > >> Paolo, under what circumstances (which versions of Windows? Anything >> special running in the guest?) has this failure happened? I'd like to repro >> this, I'm not sure if we've observed it before. > > We saw it with

Re: [PATCH 05/10] KVM: x86: MMU: Use for_each_rmap_spte macro instead of pte_list_walk()

On Thu, Nov 12, 2015 at 08:52:45PM +0900, Takuya Yoshikawa wrote: > kvm_mmu_mark_parents_unsync() alone uses pte_list_walk(), witch does > nearly the same as the for_each_rmap_spte macro. The only difference > is that is_shadow_present_pte() checks cannot be placed there because >

Re: [PATCH 3/3] KVM: x86: fix ready_for_interrupt reporting in split IRQ chip case

On Fri, Nov 13, 2015 at 12:52 AM, Paolo Bonzini wrote: > > > On 12/11/2015 20:07, Matt Gingell wrote: >> This patch adds a call to kvm_arch_interrupt_allowed to ensure ready for >> interrupt is reported to user space correctly. This addresses a problem >> observed in QEMU

Re: [PATCH 06/10] KVM: x86: MMU: Consolidate WARN_ON/BUG_ON checks for reverse-mapped sptes

On Thu, Nov 12, 2015 at 08:53:43PM +0900, Takuya Yoshikawa wrote: > At some call sites of rmap_get_first() and rmap_get_next(), BUG_ON is > placed right after the call to detect unrelated sptes which must not be > found in the reverse-mapping list. > > Move this check in rmap_get_first/next() so

Re: [PATCH] virtio_ring: Shadow available ring flags & index

On Wed, Nov 11, 2015 at 02:34:33PM +0200, Michael S. Tsirkin wrote: > On Tue, Nov 10, 2015 at 04:21:07PM -0800, Venkatesh Srinivas wrote: > > Improves cacheline transfer flow of available ring header. > > > > Virtqueues are implemented as a pair of rings, one producer->consumer > > avail ring and

[PATCH 1/2] KVM: kvm_is_visible_gfn can be boolean

This patch makes kvm_is_visible_gfn return bool due to this particular function only using either one or zero as its return value. No functional change. Signed-off-by: Yaowei Bai --- include/linux/kvm_host.h | 2 +- virt/kvm/kvm_main.c | 6 +++--- 2 files

[PATCH 2/2] KVM: kvm_para_has_feature can be boolean

This patch makes kvm_para_has_feature return bool due to this particular function only using either one or zero as its return value. No functional change. Signed-off-by: Yaowei Bai --- include/linux/kvm_para.h | 6 ++ 1 file changed, 2 insertions(+), 4

Re: [PATCH 2/2] arm: kvm: Fix STRICT_MM_TYPECHECK errors

On 11 November 2015 at 03:03, Laura Abbott wrote: > > PAGE_S2_DEVICE is a pgprot val and needs to be accessed using the proper > accessors. Switch to these accessors to avoid errors with > STRICT_MM_TYPECHECK. > > Signed-off-by: Laura Abbott

Re: [PATCH 09/10 RFC] KVM: x86: MMU: Move parent_pte handling from kvm_mmu_get_page() to link_shadow_page()

On 13/11/2015 03:15, Takuya Yoshikawa wrote: > Actually, I don't understand why this is named kvm_mmu_put_page() for > just removing parent_pte pointer from the sp->parent_ptes pointer chain. Because it undoes kvm_mmu_get_page, I guess. :) > >> On to kvm_mmu_get_page... >> >> if