4.17-stable review patch. If anyone has any objections, please let me know.
------------------
From: Vitaly Kuznetsov vkuznets@redhat.com
commit b062b794c7831a70bda4dfac202c1a9418e06ac0 upstream.
When we switched from doing rdmsr() to reading FS/GS base values from current->thread we completely forgot about legacy 32-bit userspaces which we still support in KVM (why?). task->thread.{fsbase,gsbase} are only synced for 64-bit processes, calling save_fsgs_for_kvm() and using its result from current is illegal for legacy processes.
There's no ARCH_SET_FS/GS prctls for legacy applications. Base MSRs are, however, not always equal to zero. Intel's manual says (3.4.4 Segment Loading Instructions in IA-32e Mode):
"In order to set up compatibility mode for an application, segment-load instructions (MOV to Sreg, POP Sreg) work normally in 64-bit mode. An entry is read from the system descriptor table (GDT or LDT) and is loaded in the hidden portion of the segment register. ... The hidden descriptor register fields for FS.base and GS.base are physically mapped to MSRs in order to load all address bits supported by a 64-bit implementation. "
The issue was found by strace test suite where 32-bit ioctl_kvm_run test started segfaulting.
Reported-by: Dmitry V. Levin ldv@altlinux.org Bisected-by: Masatake YAMATO yamato@redhat.com Fixes: 42b933b59721 ("x86/kvm/vmx: read MSR_{FS,KERNEL_GS}_BASE from current->thread") Cc: stable@vger.kernel.org Signed-off-by: Vitaly Kuznetsov vkuznets@redhat.com Signed-off-by: Paolo Bonzini pbonzini@redhat.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org
--- arch/x86/kvm/vmx.c | 25 +++++++++++++++++-------- 1 file changed, 17 insertions(+), 8 deletions(-)
--- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -2376,6 +2376,7 @@ static void vmx_save_host_state(struct k struct vcpu_vmx *vmx = to_vmx(vcpu); #ifdef CONFIG_X86_64 int cpu = raw_smp_processor_id(); + unsigned long fs_base, kernel_gs_base; #endif int i;
@@ -2391,12 +2392,20 @@ static void vmx_save_host_state(struct k vmx->host_state.gs_ldt_reload_needed = vmx->host_state.ldt_sel;
#ifdef CONFIG_X86_64 - save_fsgs_for_kvm(); - vmx->host_state.fs_sel = current->thread.fsindex; - vmx->host_state.gs_sel = current->thread.gsindex; -#else - savesegment(fs, vmx->host_state.fs_sel); - savesegment(gs, vmx->host_state.gs_sel); + if (likely(is_64bit_mm(current->mm))) { + save_fsgs_for_kvm(); + vmx->host_state.fs_sel = current->thread.fsindex; + vmx->host_state.gs_sel = current->thread.gsindex; + fs_base = current->thread.fsbase; + kernel_gs_base = current->thread.gsbase; + } else { +#endif + savesegment(fs, vmx->host_state.fs_sel); + savesegment(gs, vmx->host_state.gs_sel); +#ifdef CONFIG_X86_64 + fs_base = read_msr(MSR_FS_BASE); + kernel_gs_base = read_msr(MSR_KERNEL_GS_BASE); + } #endif if (!(vmx->host_state.fs_sel & 7)) { vmcs_write16(HOST_FS_SELECTOR, vmx->host_state.fs_sel); @@ -2416,10 +2425,10 @@ static void vmx_save_host_state(struct k savesegment(ds, vmx->host_state.ds_sel); savesegment(es, vmx->host_state.es_sel);
- vmcs_writel(HOST_FS_BASE, current->thread.fsbase); + vmcs_writel(HOST_FS_BASE, fs_base); vmcs_writel(HOST_GS_BASE, cpu_kernelmode_gs_base(cpu));
- vmx->msr_host_kernel_gs_base = current->thread.gsbase; + vmx->msr_host_kernel_gs_base = kernel_gs_base; if (is_long_mode(&vmx->vcpu)) wrmsrl(MSR_KERNEL_GS_BASE, vmx->msr_guest_kernel_gs_base); #else