Re: [PATCH] KVM: x86: fix bogus warning about reserved bits

[Borislav Petkov <bp@alien8.de>]

On Wed, Sep 23, 2015 at 11:36:47AM +0200, Paolo Bonzini wrote:
> And another patch, which both cranks up the debugging a bit and
> tries another fix:
> 
> diff --git a/arch/x86/kvm/cpuid.h b/arch/x86/kvm/cpuid.h
> index dd05b9cef6ae..b2f49bb15ba1 100644
> --- a/arch/x86/kvm/cpuid.h
> +++ b/arch/x86/kvm/cpuid.h
> @@ -105,8 +105,15 @@ static inline bool guest_cpuid_has_x2apic(struct kvm_vcpu *vcpu)
>  static inline bool guest_cpuid_is_amd(struct kvm_vcpu *vcpu)
>  {
>  	struct kvm_cpuid_entry2 *best;
> +	static bool first;
>  
>  	best = kvm_find_cpuid_entry(vcpu, 0, 0);
> +	if (first && best) {
> +		printk("cpuid(0).ebx = %x\n", best->ebx);
> +		first = false;
> +	} else if (first)
> +		printk_ratelimited("cpuid(0) not initialized yet\n");
> +
>  	return best && best->ebx == X86EMUL_CPUID_VENDOR_AuthenticAMD_ebx;

Do I see it correctly that that "first" thing is never true?

In any case, I changed it to initialize to true but still no output from
that function.

[  102.448438] walk_shadow_page_get_mmio_spte: detect reserved bits on spte, addr 0xb8000 (level 4, 0xf0000000000f8)
[  102.458706] walk_shadow_page_get_mmio_spte: detect reserved bits on spte, addr 0xb8000 (level 3, 0xf000000000078)
[  102.468955] walk_shadow_page_get_mmio_spte: detect reserved bits on spte, addr 0xb8000 (level 2, 0xf000000000078)
[  102.479337] dump hierarchy:
[  102.482152] ------ spte 0x416edb027 level 4.
[  102.482154] ------ spte 0x416eda027 level 3.
[  102.482155] ------ spte 0x416ed5027 level 2.
[  102.482157] ------ spte 0xffff0000000b8f67 level 1.
[  102.482158] ------------[ cut here ]------------
[  102.482196] WARNING: CPU: 6 PID: 3550 at arch/x86/kvm/mmu.c:3396 handle_mmio_page_fault.part.57+0x1a/0x20 [kvm]()
[  102.482236] Modules linked in: tun sha256_ssse3 sha256_generic drbg binfmt_misc ipv6 vfat fat fuse dm_crypt dm_mod kvm_amd kvm crc32_pclmul aesni_intel aes_x86_64 lrw gf128mul glue_helper ablk_helper cryptd amd64_edac_mod fam15h_power k10temp edac_core amdkfd amd_iommu_v2 radeon acpi_cpufreq
[  102.482240] CPU: 6 PID: 3550 Comm: qemu-system-x86 Not tainted 4.3.0-rc2+ #1
[  102.482242] Hardware name: To be filled by O.E.M. To be filled by O.E.M./M5A97 EVO R2.0, BIOS 1503 01/16/2013
[  102.482249]  ffffffffa030c992 ffff880424b8fb78 ffffffff812c758a 0000000000000000
[  102.482253]  ffff880424b8fbb0 ffffffff810534c1 ffff8804160e0000 000000000000000f
[  102.482257]  00000000000b8000 0000000000000000 00000000ffffffff ffff880424b8fbc0
[  102.482259] Call Trace:
[  102.482268]  [<ffffffff812c758a>] dump_stack+0x4e/0x84
[  102.482273]  [<ffffffff810534c1>] warn_slowpath_common+0x91/0xd0
[  102.482276]  [<ffffffff810535ba>] warn_slowpath_null+0x1a/0x20
[  102.482306]  [<ffffffffa02deb0a>] handle_mmio_page_fault.part.57+0x1a/0x20 [kvm]
[  102.482334]  [<ffffffffa02e6420>] tdp_page_fault+0x2a0/0x2b0 [kvm]
[  102.482340]  [<ffffffff810a282d>] ? __lock_acquire+0x57d/0x17a0
[  102.482369]  [<ffffffffa02e0665>] kvm_mmu_page_fault+0x35/0x240 [kvm]
[  102.482376]  [<ffffffffa03656b8>] pf_interception+0x108/0x1d0 [kvm_amd]
[  102.482381]  [<ffffffffa0367d10>] handle_exit+0x150/0xa40 [kvm_amd]
[  102.482408]  [<ffffffffa02d7398>] ? kvm_arch_vcpu_ioctl_run+0x4c8/0x16f0 [kvm]
[  102.482435]  [<ffffffffa02d7403>] kvm_arch_vcpu_ioctl_run+0x533/0x16f0 [kvm]
[  102.482461]  [<ffffffffa02d7398>] ? kvm_arch_vcpu_ioctl_run+0x4c8/0x16f0 [kvm]
[  102.482466]  [<ffffffff816bde42>] ? mutex_lock_killable_nested+0x312/0x480
[  102.482485]  [<ffffffffa02be979>] ? kvm_vcpu_ioctl+0x79/0x6f0 [kvm]
[  102.482490]  [<ffffffff8107e133>] ? preempt_count_sub+0xb3/0x110
[  102.482509]  [<ffffffffa02bec3f>] kvm_vcpu_ioctl+0x33f/0x6f0 [kvm]
[  102.482515]  [<ffffffff811939d7>] do_vfs_ioctl+0x2d7/0x530
[  102.482519]  [<ffffffff8119f889>] ? __fget_light+0x29/0x90
[  102.482523]  [<ffffffff81193c7c>] SyS_ioctl+0x4c/0x90
[  102.482527]  [<ffffffff816c209b>] entry_SYSCALL_64_fastpath+0x16/0x73
[  102.482531] ---[ end trace b8899512fc52cf2e ]---

Thanks.

-- 
Regards/Gruss,
    Boris.

ECO tip #101: Trim your mails when you reply.