similar to: [PATCH][VT]Make 32-bit VMX guest work on 64-bit host

The patch enables x86_32 PAE unmodified guests on 64-bit Xen when the hvm feature is present. We tested only Linux at this point, and we''ll improve the functionality as we test other guests. The SVM needs the equivalent changes to the vmc.c to get this functionality working, but this patch does not break the build. Signed-off-by: Jun Nakajima <jun.nakajima@intel.com>

To avoid MSR save/restore at every VM exit/entry time, we restore the x86_64 specific MSRs at domain switch time if modified. In VMX domains, we modify those upon requests from the guests to that end. Note that IA32_EFER.LME and IA32_EFER.LMA are saved/restored by H/W on every VM exit. For the usual domains (i.e. dom0 and domU), those MSRs are not modified once set at initialization time, so we

Hi, This patch fixed the size problem in the MMIO decoder. Signed-off-by: Chengyuan Li <chengyuan.li@intel.com> Thanks, Chengyuan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel

The patch attached is to set effective memory type for EPT according to the VT-d snoop control capability, and also includes some cleansup for EPT & VT-d both enabled. Signed-off-by: Zhai, Edwin Edwin.Zhai@intel.com<mailto:Edwin.Zhai@intel.com> Signed-off-by: Xin, Xiaohui xiaohui.xin@intel.com<mailto:xiaohui.xin@intel.com> _______________________________________________

The patch removes the duplicate spin_lock_irqsave to spin_lock in the Vt-d code path. The duplicate spin_lock_irqsave() flushes the original EFLAGS saved, and thus disable the local irq. Signed-off-by: Xin, Xiaohui Xiaohui.xin@intel.com Signed-off-by: Tian, Kevin <Kevin.Tian@intel.com> _______________________________________________ Xen-devel mailing list

Keir, It''s a resending patch for the timeout mechanism to deal with the shared interrupt issue for vt-d enabled hvm guest. We modify the patch following your comments last time and make some other small fix: 1) We don''t touch the locking around the hvm_dpci_eoi(). 2) Remove the HZ from the TIME_OUT_PERIOD macro which may confuse others. 3) Add some

Set guest selectors to be null if it''s not valid when switch to protected-mode in the guest firmware. Signed-off-by: Chengyuan Li <chengyuan.li@intel.com> Signed-off-by: Asit Mallick <asit.k.mallick@intel.com> Thanks, Chengyuan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel

Intercepting this MSR is pointless - The swapgs instruction does not cause a vmexit, so the cached result of this is potentially stale after the next guest instruction. It is correctly saved and restored on vcpu context switch. Furthermore, 64bit Windows writes to this MSR on every thread context switch, so interception causes a substantial performance hit. From: Paul Durrant

Currently the mmio_operands assumes writing to memory when operand 0 is register or immediate, this is false for cmp/test opcode. This patch resolve this problem, please review. Signed-off-by: Yunhong Jiang <yunhong.jiang@intel.com> Signed-off-by: Jun Nakajima <jun.nakajima@intel.com> Thanks Yunhong Jiang _______________________________________________ Xen-devel mailing list

This patch adds CR8 virtualization. It''s the initial patch for booting HVM x64 Windows guest, and just let every CR8 access issues vmexit. And later we will do acceleration to it. Signed-off-by: Xiaohui Xin <xiaohui.xin@intel.com> _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel

Hi Kashyap, Here I want to test and see how much I/O performance could be improved by virtio-blk data plane feature. Yes, this is a dirty way to enable it :-), You mentioned using newer libvirt, is there parameter similar "x-data-plane=on" in new version libvirt? Please let know me you if you have better solution. Regards, CY. On Mon, Apr 20, 2015 at 10:51 PM, Kashyap Chamarthy

Hi Keir, this patch changes common code of xen hypervisor, but seems we have to, because current call to do_boot_vcpu is always from domains, however, for SMP VMX guest AP bring up, we need call it from xen hypervisor, so we''d like to split it. What''s your opinion? _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com

This patch removes the ignored "Unconditional I/O exiting" setting in VMCS for the vmx guest since "Use I/O bitmaps" setting is being used. Signed-off by Himanshu Raj (rhim@cc.gatech.edu) diff -r 19c55935580f xen/include/asm-x86/hvm/vmx/vmx.h --- a/xen/include/asm-x86/hvm/vmx/vmx.h Wed Apr 12 17:53:38 2006 +++ b/xen/include/asm-x86/hvm/vmx/vmx.h Thu Apr 13 02:31:25 2006 @@

Hello, i have 2 problems with the virtualization of windows on a xen- domain. The first one is the vnc connection. I''ve set everything up, but if i try to do a xm console XX, it says something about xenconsole tty file not found, and if i look up the qemu-dm-XXXX.log there is just one line that says : "/path/to/xen/bin/qemu-dm invalid option -- ''-

This patch clean up VMX builder. Since now we have vmxloader, old code loading linux kernel directly is not needed. 1. remove linux parameters. 2. remove init ramdisk. 3. remove e820 memmap config file parse. Also e820 map handling code is changed in vmx_setup_platform. 4. remove provisional GDT, and construct_init_guest_vmcs is updated greatly. 5. remove some ugly tailing space. Thanks Keir for

VT is enabled in BIOS and ''xm info'' shows that hvm can be created like below: xen_caps : xen-3.0-x86_64 xen-3.0-x86_32p hvm-3.0-x86_32 hvm-3.0-x86_32p hvm -3.0-x86_64 Several Windows VMs have been created on this server and running perfectly. However, I wonder that it does not show anything when commands show VT-x availability such as ''xm dmesg | grep VMX(or

Jun suggested to use another elegant way, pls ignore this one. -Xin >SMP VMX guest AP bring up need call do_boot_vcpu in hypervisor. >Current call to do_boot_vcpu always from domain, however, for SMP VMX >guest AP bring up, it is called from xen hypervisor. >Signed-off-by: Xin Li <xin.b.li@intel.com> > >diff -r f069a06e650f xen/common/domain.c >--- a/xen/common/domain.c

Fix config file parsing for VMX domains. If we define "vif" in the config file, image.py will raise error "vmx: missing vbd configuration". The reason is "vif" is dealt with as a "vbd" device. This patch fixes this issue by dealing with "vbd" and "vif " separately, removing "macaddr" arg and parsing mac address from

Hi, I was running libvirt 1.2.2 and qemu 2.0 which are from icehouse cloud archive repo on ubuntu 12.04, "virsh list --all" showed correct instances after launching several VMs, Then I just downgraded libvirt to 0.9.13 and qemu to 1.0, but kept the VMs running. However, when I run "virsh list --all" after downgrade, I can't see any VMs. Seems that the old version libvirt

> -----Original Message----- > From: xen-devel-bounces@lists.xensource.com > [mailto:xen-devel-bounces@lists.xensource.com] On Behalf Of > Thorolf Godawa > Sent: 02 February 2006 16:33 > To: Charles Coffing > Cc: xen-devel@lists.xensource.com > Subject: Re: [Xen-devel] No VMX-guests on Intel Xeon/VT > > Hi Chuck and all others, > > > yeah, its going