| Message ID | 20190115072549.10198-1-juergh@canonical.com |
|---|---|
| State | New |
| Headers | show |
| Series | [SRU,Xenial,PULL] x86/kvm: Backport fixup and missing commits (LP: #1811646) | expand |
On 15.01.19 08:25, Juerg Haefliger wrote: > == SRU Justification == > > The update to stable 4.4.168 contains a couple of x86/kvm related commits that have later fixup commits in upstream. Also, there are a couple of commits in stable 4.4. that we're missing. > > == Fix == > > Cherry-pick or backport the relevant commits. > > == Regression Potential == > > Low. The relevant commits have been in upstream for quite a while. > > == Test Case == > > Not available. It's unclear what the impact is of not having those fixups but it's clear from upstream's commit messages that they fix real issues. > > Compile-tested all architectres. Ran release regression tests (locally). > > Signed-off-by: Juerg Haefliger <juergh@canonical.com> > --- > > The following changes since commit 6aea40b78942cf00d2c8a9e2e204e7950f988a10: > > Linux 4.4.170 (2019-01-14 09:43:46 +0100) > > are available in the Git repository at: > > git://git.launchpad.net/~juergh/+git/xenial-linux lp1811646 > > for you to fetch changes up to 04474027f01c02570ab512b0a127ad1e377e220d: > > UBUNTU: SAUCE: KVM: Move code fragments, cleanup and re-indent (2019-01-14 13:07:48 +0100) > > ---------------------------------------------------------------- > Jim Mattson (4): > kvm: nVMX: VMCLEAR an active shadow VMCS after last use > kvm: x86: IA32_ARCH_CAPABILITIES is always supported > kvm: svm: Ensure an IBPB on all affected CPUs when freeing a vmcb > kvm: vmx: Scrub hardware GPRs at VM-exit > > Juerg Haefliger (2): > UBUNTU: SAUCE: [Fix] x86/KVM/VMX: Add L1D flush logic > UBUNTU: SAUCE: KVM: Move code fragments, cleanup and re-indent > > KarimAllah Ahmed (1): > X86/nVMX: Properly set spec_ctrl and pred_cmd before merging MSRs > > Paolo Bonzini (3): > KVM: x86: avoid vmalloc(0) in the KVM_SET_CPUID > KVM/VMX: Optimize vmx_vcpu_run() and svm_vcpu_run() by marking the RDMSR path as unlikely() > KVM: VMX: fixes for vmentry_l1d_flush module parameter > > Thomas Gleixner (1): > KVM: x86: SVM: Call x86_spec_ctrl_set_guest/host() with interrupts disabled > > Tom Lendacky (1): > KVM: SVM: Add MSR-based feature support for serializing LFENCE > > Wanpeng Li (1): > KVM: X86: Allow userspace to define the microcode version > > arch/x86/include/asm/kvm_host.h | 1 + > arch/x86/kvm/cpuid.c | 27 +++++++----- > arch/x86/kvm/cpuid.h | 44 ++++++++++---------- > arch/x86/kvm/svm.c | 91 ++++++++++++++++++++++++++++++++++------- > arch/x86/kvm/vmx.c | 78 +++++++++++++++++++++++------------ > arch/x86/kvm/x86.c | 12 +++++- > 6 files changed, 179 insertions(+), 74 deletions(-) > Mainly based on testing and it being limited to kvm. Speaking of, your testing, was that done inside of VMs running on a modified host? And Intel CPU only or also AMD`(for the svm changes)? Acked-by: Stefan Bader <stefan.bader@canonical.com>
On Fri, 1 Feb 2019 09:56:06 +0100 Stefan Bader <stefan.bader@canonical.com> wrote: > On 15.01.19 08:25, Juerg Haefliger wrote: > > == SRU Justification == > > > > The update to stable 4.4.168 contains a couple of x86/kvm related commits that have later fixup commits in upstream. Also, there are a couple of commits in stable 4.4. that we're missing. > > > > == Fix == > > > > Cherry-pick or backport the relevant commits. > > > > == Regression Potential == > > > > Low. The relevant commits have been in upstream for quite a while. > > > > == Test Case == > > > > Not available. It's unclear what the impact is of not having those fixups but it's clear from upstream's commit messages that they fix real issues. > > > > Compile-tested all architectres. Ran release regression tests (locally). > > > > Signed-off-by: Juerg Haefliger <juergh@canonical.com> > > --- > > > > The following changes since commit 6aea40b78942cf00d2c8a9e2e204e7950f988a10: > > > > Linux 4.4.170 (2019-01-14 09:43:46 +0100) > > > > are available in the Git repository at: > > > > git://git.launchpad.net/~juergh/+git/xenial-linux lp1811646 > > > > for you to fetch changes up to 04474027f01c02570ab512b0a127ad1e377e220d: > > > > UBUNTU: SAUCE: KVM: Move code fragments, cleanup and re-indent (2019-01-14 13:07:48 +0100) > > > > ---------------------------------------------------------------- > > Jim Mattson (4): > > kvm: nVMX: VMCLEAR an active shadow VMCS after last use > > kvm: x86: IA32_ARCH_CAPABILITIES is always supported > > kvm: svm: Ensure an IBPB on all affected CPUs when freeing a vmcb > > kvm: vmx: Scrub hardware GPRs at VM-exit > > > > Juerg Haefliger (2): > > UBUNTU: SAUCE: [Fix] x86/KVM/VMX: Add L1D flush logic > > UBUNTU: SAUCE: KVM: Move code fragments, cleanup and re-indent > > > > KarimAllah Ahmed (1): > > X86/nVMX: Properly set spec_ctrl and pred_cmd before merging MSRs > > > > Paolo Bonzini (3): > > KVM: x86: avoid vmalloc(0) in the KVM_SET_CPUID > > KVM/VMX: Optimize vmx_vcpu_run() and svm_vcpu_run() by marking the RDMSR path as unlikely() > > KVM: VMX: fixes for vmentry_l1d_flush module parameter > > > > Thomas Gleixner (1): > > KVM: x86: SVM: Call x86_spec_ctrl_set_guest/host() with interrupts disabled > > > > Tom Lendacky (1): > > KVM: SVM: Add MSR-based feature support for serializing LFENCE > > > > Wanpeng Li (1): > > KVM: X86: Allow userspace to define the microcode version > > > > arch/x86/include/asm/kvm_host.h | 1 + > > arch/x86/kvm/cpuid.c | 27 +++++++----- > > arch/x86/kvm/cpuid.h | 44 ++++++++++---------- > > arch/x86/kvm/svm.c | 91 ++++++++++++++++++++++++++++++++++------- > > arch/x86/kvm/vmx.c | 78 +++++++++++++++++++++++------------ > > arch/x86/kvm/x86.c | 12 +++++- > > 6 files changed, 179 insertions(+), 74 deletions(-) > > > Mainly based on testing and it being limited to kvm. Speaking of, your testing, > was that done inside of VMs running on a modified host? Just running the release regression tests on a local baremetal Xeon system. Quite frankly I'm not sure how much that exercises the relevant KVM portion of the kernel. > And Intel CPU only or > also AMD`(for the svm changes)? Intel only. ...Juerg > Acked-by: Stefan Bader <stefan.bader@canonical.com> >
On 1/15/19 8:25 AM, Juerg Haefliger wrote: > == SRU Justification == > > The update to stable 4.4.168 contains a couple of x86/kvm related commits that have later fixup commits in upstream. Also, there are a couple of commits in stable 4.4. that we're missing. > > == Fix == > > Cherry-pick or backport the relevant commits. > > == Regression Potential == > > Low. The relevant commits have been in upstream for quite a while. > > == Test Case == > > Not available. It's unclear what the impact is of not having those fixups but it's clear from upstream's commit messages that they fix real issues. > > Compile-tested all architectres. Ran release regression tests (locally). > > Signed-off-by: Juerg Haefliger <juergh@canonical.com> > --- > > The following changes since commit 6aea40b78942cf00d2c8a9e2e204e7950f988a10: > > Linux 4.4.170 (2019-01-14 09:43:46 +0100) > > are available in the Git repository at: > > git://git.launchpad.net/~juergh/+git/xenial-linux lp1811646 > > for you to fetch changes up to 04474027f01c02570ab512b0a127ad1e377e220d: > > UBUNTU: SAUCE: KVM: Move code fragments, cleanup and re-indent (2019-01-14 13:07:48 +0100) > > ---------------------------------------------------------------- > Jim Mattson (4): > kvm: nVMX: VMCLEAR an active shadow VMCS after last use > kvm: x86: IA32_ARCH_CAPABILITIES is always supported > kvm: svm: Ensure an IBPB on all affected CPUs when freeing a vmcb > kvm: vmx: Scrub hardware GPRs at VM-exit > > Juerg Haefliger (2): > UBUNTU: SAUCE: [Fix] x86/KVM/VMX: Add L1D flush logic > UBUNTU: SAUCE: KVM: Move code fragments, cleanup and re-indent > > KarimAllah Ahmed (1): > X86/nVMX: Properly set spec_ctrl and pred_cmd before merging MSRs > > Paolo Bonzini (3): > KVM: x86: avoid vmalloc(0) in the KVM_SET_CPUID > KVM/VMX: Optimize vmx_vcpu_run() and svm_vcpu_run() by marking the RDMSR path as unlikely() > KVM: VMX: fixes for vmentry_l1d_flush module parameter > > Thomas Gleixner (1): > KVM: x86: SVM: Call x86_spec_ctrl_set_guest/host() with interrupts disabled > > Tom Lendacky (1): > KVM: SVM: Add MSR-based feature support for serializing LFENCE > > Wanpeng Li (1): > KVM: X86: Allow userspace to define the microcode version > > arch/x86/include/asm/kvm_host.h | 1 + > arch/x86/kvm/cpuid.c | 27 +++++++----- > arch/x86/kvm/cpuid.h | 44 ++++++++++---------- > arch/x86/kvm/svm.c | 91 ++++++++++++++++++++++++++++++++++------- > arch/x86/kvm/vmx.c | 78 +++++++++++++++++++++++------------ > arch/x86/kvm/x86.c | 12 +++++- > 6 files changed, 179 insertions(+), 74 deletions(-) > Acked-by: Kleber Sacilotto de Souza <kleber.souza@canonical.com>
On 1/15/19 8:25 AM, Juerg Haefliger wrote: > == SRU Justification == > > The update to stable 4.4.168 contains a couple of x86/kvm related commits that have later fixup commits in upstream. Also, there are a couple of commits in stable 4.4. that we're missing. > > == Fix == > > Cherry-pick or backport the relevant commits. > > == Regression Potential == > > Low. The relevant commits have been in upstream for quite a while. > > == Test Case == > > Not available. It's unclear what the impact is of not having those fixups but it's clear from upstream's commit messages that they fix real issues. > > Compile-tested all architectres. Ran release regression tests (locally). > > Signed-off-by: Juerg Haefliger <juergh@canonical.com> > --- > > The following changes since commit 6aea40b78942cf00d2c8a9e2e204e7950f988a10: > > Linux 4.4.170 (2019-01-14 09:43:46 +0100) > > are available in the Git repository at: > > git://git.launchpad.net/~juergh/+git/xenial-linux lp1811646 > > for you to fetch changes up to 04474027f01c02570ab512b0a127ad1e377e220d: > > UBUNTU: SAUCE: KVM: Move code fragments, cleanup and re-indent (2019-01-14 13:07:48 +0100) > > ---------------------------------------------------------------- > Jim Mattson (4): > kvm: nVMX: VMCLEAR an active shadow VMCS after last use > kvm: x86: IA32_ARCH_CAPABILITIES is always supported > kvm: svm: Ensure an IBPB on all affected CPUs when freeing a vmcb > kvm: vmx: Scrub hardware GPRs at VM-exit > > Juerg Haefliger (2): > UBUNTU: SAUCE: [Fix] x86/KVM/VMX: Add L1D flush logic > UBUNTU: SAUCE: KVM: Move code fragments, cleanup and re-indent > > KarimAllah Ahmed (1): > X86/nVMX: Properly set spec_ctrl and pred_cmd before merging MSRs > > Paolo Bonzini (3): > KVM: x86: avoid vmalloc(0) in the KVM_SET_CPUID > KVM/VMX: Optimize vmx_vcpu_run() and svm_vcpu_run() by marking the RDMSR path as unlikely() > KVM: VMX: fixes for vmentry_l1d_flush module parameter > > Thomas Gleixner (1): > KVM: x86: SVM: Call x86_spec_ctrl_set_guest/host() with interrupts disabled > > Tom Lendacky (1): > KVM: SVM: Add MSR-based feature support for serializing LFENCE > > Wanpeng Li (1): > KVM: X86: Allow userspace to define the microcode version > > arch/x86/include/asm/kvm_host.h | 1 + > arch/x86/kvm/cpuid.c | 27 +++++++----- > arch/x86/kvm/cpuid.h | 44 ++++++++++---------- > arch/x86/kvm/svm.c | 91 ++++++++++++++++++++++++++++++++++------- > arch/x86/kvm/vmx.c | 78 +++++++++++++++++++++++------------ > arch/x86/kvm/x86.c | 12 +++++- > 6 files changed, 179 insertions(+), 74 deletions(-) > Applied to xenial/master-next branch. Thanks, Kleber
== SRU Justification == The update to stable 4.4.168 contains a couple of x86/kvm related commits that have later fixup commits in upstream. Also, there are a couple of commits in stable 4.4. that we're missing. == Fix == Cherry-pick or backport the relevant commits. == Regression Potential == Low. The relevant commits have been in upstream for quite a while. == Test Case == Not available. It's unclear what the impact is of not having those fixups but it's clear from upstream's commit messages that they fix real issues. Compile-tested all architectres. Ran release regression tests (locally). Signed-off-by: Juerg Haefliger <juergh@canonical.com> --- The following changes since commit 6aea40b78942cf00d2c8a9e2e204e7950f988a10: Linux 4.4.170 (2019-01-14 09:43:46 +0100) are available in the Git repository at: git://git.launchpad.net/~juergh/+git/xenial-linux lp1811646 for you to fetch changes up to 04474027f01c02570ab512b0a127ad1e377e220d: UBUNTU: SAUCE: KVM: Move code fragments, cleanup and re-indent (2019-01-14 13:07:48 +0100) ---------------------------------------------------------------- Jim Mattson (4): kvm: nVMX: VMCLEAR an active shadow VMCS after last use kvm: x86: IA32_ARCH_CAPABILITIES is always supported kvm: svm: Ensure an IBPB on all affected CPUs when freeing a vmcb kvm: vmx: Scrub hardware GPRs at VM-exit Juerg Haefliger (2): UBUNTU: SAUCE: [Fix] x86/KVM/VMX: Add L1D flush logic UBUNTU: SAUCE: KVM: Move code fragments, cleanup and re-indent KarimAllah Ahmed (1): X86/nVMX: Properly set spec_ctrl and pred_cmd before merging MSRs Paolo Bonzini (3): KVM: x86: avoid vmalloc(0) in the KVM_SET_CPUID KVM/VMX: Optimize vmx_vcpu_run() and svm_vcpu_run() by marking the RDMSR path as unlikely() KVM: VMX: fixes for vmentry_l1d_flush module parameter Thomas Gleixner (1): KVM: x86: SVM: Call x86_spec_ctrl_set_guest/host() with interrupts disabled Tom Lendacky (1): KVM: SVM: Add MSR-based feature support for serializing LFENCE Wanpeng Li (1): KVM: X86: Allow userspace to define the microcode version arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/cpuid.c | 27 +++++++----- arch/x86/kvm/cpuid.h | 44 ++++++++++---------- arch/x86/kvm/svm.c | 91 ++++++++++++++++++++++++++++++++++------- arch/x86/kvm/vmx.c | 78 +++++++++++++++++++++++------------ arch/x86/kvm/x86.c | 12 +++++- 6 files changed, 179 insertions(+), 74 deletions(-)