Re: regression bisected; KVM: entry failed, hardware error 0x80000021
On 25/12/2014 08:46, Chen, Tiejun wrote: diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index f528343..910bc48 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -672,6 +672,7 @@ static void update_memslots(struct kvm_memslots *slots, WARN_ON(mslots[i].id != id); if (!new-npages) { new-base_gfn = 0; + new-flags = 0; if (mslots[i].npages) slots-used_slots--; } else { Why is this assignment needed? Paolo -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: regression bisected; KVM: entry failed, hardware error 0x80000021
Chen, Tiejun wrote: On 2014/12/24 19:02, Jamie Heilman wrote: Chen, Tiejun wrote: On 2014/12/23 15:26, Jamie Heilman wrote: Chen, Tiejun wrote: On 2014/12/23 9:50, Chen, Tiejun wrote: On 2014/12/22 17:23, Jamie Heilman wrote: KVM internal error. Suberror: 1 emulation failure EAX=000de494 EBX= ECX= EDX=0cfd ESI=0059 EDI= EBP= ESP=6fb4 EIP=000f15c1 EFL=00010016 [AP-] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES =0010 00c09300 DPL=0 DS [-WA] CS =0008 00c09b00 DPL=0 CS32 [-RA] SS =0010 00c09300 DPL=0 DS [-WA] DS =0010 00c09300 DPL=0 DS [-WA] FS =0010 00c09300 DPL=0 DS [-WA] GS =0010 00c09300 DPL=0 DS [-WA] LDT= 8200 DPL=0 LDT TR = 8b00 DPL=0 TSS32-busy GDT= 000f6be8 0037 IDT= 000f6c26 CR0=6011 CR2= CR3= CR4= DR0= DR1= DR2= DR3= DR6=0ff0 DR7=0400 EFER= Code=e8 ae fc ff ff 89 f2 a8 10 89 d8 75 0a b9 41 15 ff ff ff d1 5b 5e c3 5b 5e e9 76 ff ff ff b0 11 e6 20 e6 a0 b0 08 e6 21 b0 70 e6 a1 b0 04 e6 21 b0 02 FWIW, I get the same thing with 34a1cd60d17 reverted. Maybe there are two bugs, maybe there's more to this first one. I can repro this So if my understanding is correct, this is probably another bug. And especially, I already saw the same log in another thread, Cleaning up the KVM clock. Maybe you can continue to `git bisect` to locate that bad commit. Looks just now Andy found that commit, 0e60b0799fedc495a5c57dbd669de3c10d72edd2 kvm: change memslot sorting rule from size to GFN, maybe you can try to revert this to try yours again. That doesn't revert cleanly for me, and I don't have much time to fiddle with it until the 24th---so checked out the commit before it (d4ae84a0), applied your patch, built, and yes, everything works fine at that point. I'll probably have time for another full bisection later, assuming things aren't ironed out already by then. 3.18.0-rc3-00120-gd4ae84a0 + vmx reorder msr writes patch = OK 3.18.0-rc3-00121-g0e60b07 + vmx reorder msr writes patch = emulation failure So that certainly points to 0e60b0799fedc495a5c57dbd669de3c10d72edd2 as well. Could you try this to fix your last error? Running qemu-system-x86_64 -machine pc,accel=kvm -nodefaults works, my real (headless) kvm guests work, but this new patch makes running qemu-system-x86_64 -machine pc,accel=kvm fail again, this time with Are you sure? From my test based on 3.19-rc1 that it owns top commit, aa39477b5692611b91ac9455ae588738852b3f60 just plus my previous patch, kvm: x86: vmx: reorder some msr writing I already can execute such a command successfully, qemu-system-x86_64 -machine pc,accel=kvm -m 2048 -smp 2 -hda ubuntu.img And your log below seems not to relate mem_slot issue we're discussing, I guess you need to update qemu as well. Yes, I'm sure. But I also found my new patch just work out Andy's next case, its really bringing a new issue in !next case. So I tried to refine that patch again as follows, This latest patch (again, after fixing all the whitespace so it actually applies), does the trick. Both qemu-system-x86_64 -machine pc,accel=kvm and qemu-system-x86_64 -machine pc,accel=kvm -nodefaults work for me now without any of the aforementioned warnings from the host. Signed-off-by: Tiejun Chen tiejun.c...@intel.com --- virt/kvm/kvm_main.c | 5 - 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index f528343..910bc48 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -672,6 +672,7 @@ static void update_memslots(struct kvm_memslots *slots, WARN_ON(mslots[i].id != id); if (!new-npages) { new-base_gfn = 0; + new-flags = 0; if (mslots[i].npages) slots-used_slots--; } else { @@ -688,7 +689,9 @@ static void update_memslots(struct kvm_memslots *slots, i++; } while (i 0 - new-base_gfn mslots[i - 1].base_gfn) { + ((new-base_gfn mslots[i - 1].base_gfn) || +(!new-base_gfn + !mslots[i - 1].base_gfn !mslots[i - 1].npages))) { mslots[i] = mslots[i - 1]; slots-id_to_index[mslots[i].id] = i; i--; Tiejun errors in the host to the tune of: [ cut here ] WARNING: CPU: 1 PID: 3901 at arch/x86/kvm/x86.c:6575 kvm_arch_vcpu_ioctl_run+0xd63/0xe5b [kvm]() Modules linked in: nfsv4 cpufreq_userspace cpufreq_stats cpufreq_powersave cpufreq_ondemand cpufreq_conservative autofs4 fan nfsd auth_rpcgss nfs lockd grace fscache sunrpc
Re: regression bisected; KVM: entry failed, hardware error 0x80000021
On 2014/12/25 18:52, Jamie Heilman wrote: Chen, Tiejun wrote: On 2014/12/24 19:02, Jamie Heilman wrote: Chen, Tiejun wrote: On 2014/12/23 15:26, Jamie Heilman wrote: Chen, Tiejun wrote: On 2014/12/23 9:50, Chen, Tiejun wrote: On 2014/12/22 17:23, Jamie Heilman wrote: KVM internal error. Suberror: 1 emulation failure EAX=000de494 EBX= ECX= EDX=0cfd ESI=0059 EDI= EBP= ESP=6fb4 EIP=000f15c1 EFL=00010016 [AP-] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES =0010 00c09300 DPL=0 DS [-WA] CS =0008 00c09b00 DPL=0 CS32 [-RA] SS =0010 00c09300 DPL=0 DS [-WA] DS =0010 00c09300 DPL=0 DS [-WA] FS =0010 00c09300 DPL=0 DS [-WA] GS =0010 00c09300 DPL=0 DS [-WA] LDT= 8200 DPL=0 LDT TR = 8b00 DPL=0 TSS32-busy GDT= 000f6be8 0037 IDT= 000f6c26 CR0=6011 CR2= CR3= CR4= DR0= DR1= DR2= DR3= DR6=0ff0 DR7=0400 EFER= Code=e8 ae fc ff ff 89 f2 a8 10 89 d8 75 0a b9 41 15 ff ff ff d1 5b 5e c3 5b 5e e9 76 ff ff ff b0 11 e6 20 e6 a0 b0 08 e6 21 b0 70 e6 a1 b0 04 e6 21 b0 02 FWIW, I get the same thing with 34a1cd60d17 reverted. Maybe there are two bugs, maybe there's more to this first one. I can repro this So if my understanding is correct, this is probably another bug. And especially, I already saw the same log in another thread, Cleaning up the KVM clock. Maybe you can continue to `git bisect` to locate that bad commit. Looks just now Andy found that commit, 0e60b0799fedc495a5c57dbd669de3c10d72edd2 kvm: change memslot sorting rule from size to GFN, maybe you can try to revert this to try yours again. That doesn't revert cleanly for me, and I don't have much time to fiddle with it until the 24th---so checked out the commit before it (d4ae84a0), applied your patch, built, and yes, everything works fine at that point. I'll probably have time for another full bisection later, assuming things aren't ironed out already by then. 3.18.0-rc3-00120-gd4ae84a0 + vmx reorder msr writes patch = OK 3.18.0-rc3-00121-g0e60b07 + vmx reorder msr writes patch = emulation failure So that certainly points to 0e60b0799fedc495a5c57dbd669de3c10d72edd2 as well. Could you try this to fix your last error? Running qemu-system-x86_64 -machine pc,accel=kvm -nodefaults works, my real (headless) kvm guests work, but this new patch makes running qemu-system-x86_64 -machine pc,accel=kvm fail again, this time with Are you sure? From my test based on 3.19-rc1 that it owns top commit, aa39477b5692611b91ac9455ae588738852b3f60 just plus my previous patch, kvm: x86: vmx: reorder some msr writing I already can execute such a command successfully, qemu-system-x86_64 -machine pc,accel=kvm -m 2048 -smp 2 -hda ubuntu.img And your log below seems not to relate mem_slot issue we're discussing, I guess you need to update qemu as well. Yes, I'm sure. But I also found my new patch just work out Andy's next case, its really bringing a new issue in !next case. So I tried to refine that patch again as follows, This latest patch (again, after fixing all the whitespace so it actually Next time I guess I need to post that as a attached file :) applies), does the trick. Both qemu-system-x86_64 -machine pc,accel=kvm and qemu-system-x86_64 -machine pc,accel=kvm -nodefaults work for me now without any of the aforementioned warnings from the host. Sounds great and thanks for your test again. Tiejun Signed-off-by: Tiejun Chen tiejun.c...@intel.com --- virt/kvm/kvm_main.c | 5 - 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index f528343..910bc48 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -672,6 +672,7 @@ static void update_memslots(struct kvm_memslots *slots, WARN_ON(mslots[i].id != id); if (!new-npages) { new-base_gfn = 0; + new-flags = 0; if (mslots[i].npages) slots-used_slots--; } else { @@ -688,7 +689,9 @@ static void update_memslots(struct kvm_memslots *slots, i++; } while (i 0 - new-base_gfn mslots[i - 1].base_gfn) { + ((new-base_gfn mslots[i - 1].base_gfn) || +(!new-base_gfn + !mslots[i - 1].base_gfn !mslots[i - 1].npages))) { mslots[i] = mslots[i - 1]; slots-id_to_index[mslots[i].id] = i; i--; Tiejun errors in the host to the tune of: [ cut here ] WARNING: CPU: 1 PID: 3901 at arch/x86/kvm/x86.c:6575 kvm_arch_vcpu_ioctl_run+0xd63/0xe5b [kvm]() Modules linked in: nfsv4 cpufreq_userspace cpufreq_stats cpufreq_powersave cpufreq_ondemand
Re: regression bisected; KVM: entry failed, hardware error 0x80000021
On 2014/12/23 15:26, Jamie Heilman wrote: Chen, Tiejun wrote: On 2014/12/23 9:50, Chen, Tiejun wrote: On 2014/12/22 17:23, Jamie Heilman wrote: Chen, Tiejun wrote: On 2014/12/21 20:46, Jamie Heilman wrote: With v3.19-rc1 when I run qemu-system-x86_64 -machine pc,accel=kvm I get: KVM: entry failed, hardware error 0x8021 Looks some MSR writing issues such a failed entry. If you're running a guest on an Intel machine without unrestricted mode support, the failure can be most likely due to the guest entering an invalid state for Intel VT. For example, the guest maybe running in big real mode which is not supported on less recent Intel processors. EAX= EBX= ECX= EDX=0663 ESI= EDI= EBP= ESP= EIP=e05b EFL=00010002 [---] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES = 9300 CS =f000 000f 9b00 SS = 9300 DS = 9300 FS = 9300 GS = 9300 LDT= 8200 TR = 8b00 GDT= IDT= CR0=6010 CR2= CR3= CR4= DR0= DR1= DR2= DR3= DR6=0ff0 DR7=0400 EFER= And I don't see any obvious wrong as well. Any valuable info from dmesg? With the simple qemu command above, on 3.18.1 I see: kern.info: kvm: zapping shadow pages for mmio generation wraparound when I fire up a full guest that's actually useful I get: kern.info: kvm: zapping shadow pages for mmio generation wraparound kern.err: kvm [4073]: vcpu0 disabled perfctr wrmsr: 0xc1 data 0x On 3.18.0-rc3-00042-g34a1cd6 nothing appears in the dmesg, just the message I mention above to stderr. Same thing with a stock 3.19.0-rc1. Once I apply your patch the simple test command produces the same zapping shadow pages messages as 3.18.1, and a test guest of a Debian Jessie image (w/stock distro kernel) produces the same thing with disabled perfctr wrmsr message. However, it doesn't look like Sorry I'm not sure if I understood current status. Looks 3.19-rc1 my patch just fix that error above, KVM: entry failed, hardware error 0x8021 ... Right? I'm entirely out of the woods, because one of my other guest VMs with a custom kernel that works great under 3.18.1 now fails to run. Nothing in dmesg, but here's the stderr: But even you revert 34a1cd60d17 or just apply my patch, something else introduced between 3.18.1 and 3.19-rc1 led this error below, right? KVM internal error. Suberror: 1 emulation failure EAX=000de494 EBX= ECX= EDX=0cfd ESI=0059 EDI= EBP= ESP=6fb4 EIP=000f15c1 EFL=00010016 [AP-] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES =0010 00c09300 DPL=0 DS [-WA] CS =0008 00c09b00 DPL=0 CS32 [-RA] SS =0010 00c09300 DPL=0 DS [-WA] DS =0010 00c09300 DPL=0 DS [-WA] FS =0010 00c09300 DPL=0 DS [-WA] GS =0010 00c09300 DPL=0 DS [-WA] LDT= 8200 DPL=0 LDT TR = 8b00 DPL=0 TSS32-busy GDT= 000f6be8 0037 IDT= 000f6c26 CR0=6011 CR2= CR3= CR4= DR0= DR1= DR2= DR3= DR6=0ff0 DR7=0400 EFER= Code=e8 ae fc ff ff 89 f2 a8 10 89 d8 75 0a b9 41 15 ff ff ff d1 5b 5e c3 5b 5e e9 76 ff ff ff b0 11 e6 20 e6 a0 b0 08 e6 21 b0 70 e6 a1 b0 04 e6 21 b0 02 FWIW, I get the same thing with 34a1cd60d17 reverted. Maybe there are two bugs, maybe there's more to this first one. I can repro this So if my understanding is correct, this is probably another bug. And especially, I already saw the same log in another thread, Cleaning up the KVM clock. Maybe you can continue to `git bisect` to locate that bad commit. Looks just now Andy found that commit, 0e60b0799fedc495a5c57dbd669de3c10d72edd2 kvm: change memslot sorting rule from size to GFN, maybe you can try to revert this to try yours again. That doesn't revert cleanly for me, and I don't have much time to fiddle with it until the 24th---so checked out the commit before it (d4ae84a0), applied your patch, built, and yes, everything works fine at that point. I'll probably have time for another full bisection later, assuming things aren't ironed out already by then. Could you try this to fix your last error? Signed-off-by: Tiejun Chen tiejun.c...@intel.com --- virt/kvm/kvm_main.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index f528343..a2d928c 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -672,6 +672,7 @@ static void update_memslots(struct kvm_memslots *slots,
Re: regression bisected; KVM: entry failed, hardware error 0x80000021
Chen, Tiejun wrote: On 2014/12/23 15:26, Jamie Heilman wrote: Chen, Tiejun wrote: On 2014/12/23 9:50, Chen, Tiejun wrote: On 2014/12/22 17:23, Jamie Heilman wrote: KVM internal error. Suberror: 1 emulation failure EAX=000de494 EBX= ECX= EDX=0cfd ESI=0059 EDI= EBP= ESP=6fb4 EIP=000f15c1 EFL=00010016 [AP-] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES =0010 00c09300 DPL=0 DS [-WA] CS =0008 00c09b00 DPL=0 CS32 [-RA] SS =0010 00c09300 DPL=0 DS [-WA] DS =0010 00c09300 DPL=0 DS [-WA] FS =0010 00c09300 DPL=0 DS [-WA] GS =0010 00c09300 DPL=0 DS [-WA] LDT= 8200 DPL=0 LDT TR = 8b00 DPL=0 TSS32-busy GDT= 000f6be8 0037 IDT= 000f6c26 CR0=6011 CR2= CR3= CR4= DR0= DR1= DR2= DR3= DR6=0ff0 DR7=0400 EFER= Code=e8 ae fc ff ff 89 f2 a8 10 89 d8 75 0a b9 41 15 ff ff ff d1 5b 5e c3 5b 5e e9 76 ff ff ff b0 11 e6 20 e6 a0 b0 08 e6 21 b0 70 e6 a1 b0 04 e6 21 b0 02 FWIW, I get the same thing with 34a1cd60d17 reverted. Maybe there are two bugs, maybe there's more to this first one. I can repro this So if my understanding is correct, this is probably another bug. And especially, I already saw the same log in another thread, Cleaning up the KVM clock. Maybe you can continue to `git bisect` to locate that bad commit. Looks just now Andy found that commit, 0e60b0799fedc495a5c57dbd669de3c10d72edd2 kvm: change memslot sorting rule from size to GFN, maybe you can try to revert this to try yours again. That doesn't revert cleanly for me, and I don't have much time to fiddle with it until the 24th---so checked out the commit before it (d4ae84a0), applied your patch, built, and yes, everything works fine at that point. I'll probably have time for another full bisection later, assuming things aren't ironed out already by then. 3.18.0-rc3-00120-gd4ae84a0 + vmx reorder msr writes patch = OK 3.18.0-rc3-00121-g0e60b07 + vmx reorder msr writes patch = emulation failure So that certainly points to 0e60b0799fedc495a5c57dbd669de3c10d72edd2 as well. Could you try this to fix your last error? Running qemu-system-x86_64 -machine pc,accel=kvm -nodefaults works, my real (headless) kvm guests work, but this new patch makes running qemu-system-x86_64 -machine pc,accel=kvm fail again, this time with errors in the host to the tune of: [ cut here ] WARNING: CPU: 1 PID: 3901 at arch/x86/kvm/x86.c:6575 kvm_arch_vcpu_ioctl_run+0xd63/0xe5b [kvm]() Modules linked in: nfsv4 cpufreq_userspace cpufreq_stats cpufreq_powersave cpufreq_ondemand cpufreq_conservative autofs4 fan nfsd auth_rpcgss nfs lockd grace fscache sunrpc bridge stp llc vhost_net tun vhost macvtap macvlan fuse cbc dm_crypt usb_storage snd_hda_codec_analog snd_hda_codec_generic kvm_intel kvm tg3 ptp pps_core sr_mod snd_hda_intel snd_hda_controller snd_hda_codec snd_hwdep snd_pcm snd_timer snd sg dcdbas cdrom psmouse soundcore floppy evdev xfs dm_mod raid1 md_mod CPU: 1 PID: 3901 Comm: qemu-system-x86 Not tainted 3.19.0-rc1-00011-g53262d1-dirty #1 Hardware name: Dell Inc. Precision WorkStation T3400 /0TP412, BIOS A14 04/30/2012 7e052328 8800c25ffcf8 813defbe 8800c25ffd38 8103b517 8800c25ffd28 a019bdec 8800caf1d000 8800c2774800 Call Trace: [813defbe] dump_stack+0x4c/0x6e [8103b517] warn_slowpath_common+0x97/0xb1 [a019bdec] ? kvm_arch_vcpu_ioctl_run+0xd63/0xe5b [kvm] [8103b60b] warn_slowpath_null+0x15/0x17 [a019bdec] kvm_arch_vcpu_ioctl_run+0xd63/0xe5b [kvm] [a02308b9] ? vmcs_load+0x20/0x62 [kvm_intel] [a0231e03] ? vmx_vcpu_load+0x140/0x16a [kvm_intel] [a0196ba3] ? kvm_arch_vcpu_load+0x15c/0x161 [kvm] [a018d8b1] kvm_vcpu_ioctl+0x189/0x4bd [kvm] [8104647a] ? do_sigtimedwait+0x12f/0x189 [810ea316] do_vfs_ioctl+0x370/0x436 [810f24f2] ? __fget+0x67/0x72 [810ea41b] SyS_ioctl+0x3f/0x5e [813e34d2] system_call_fastpath+0x12/0x17 ---[ end trace 46abac932fb3b4a1 ]--- [ cut here ] WARNING: CPU: 1 PID: 3901 at arch/x86/kvm/x86.c:6575 kvm_arch_vcpu_ioctl_run+0xd63/0xe5b [kvm]() Modules linked in: nfsv4 cpufreq_userspace cpufreq_stats cpufreq_powersave cpufreq_ondemand cpufreq_conservative autofs4 fan nfsd auth_rpcgss nfs lockd grace fscache sunrpc bridge stp llc vhost_net tun vhost macvtap macvlan fuse cbc dm_crypt usb_storage snd_hda_codec_analog snd_hda_codec_generic kvm_intel kvm tg3 ptp pps_core sr_mod snd_hda_intel snd_hda_controller snd_hda_codec snd_hwdep snd_pcm snd_timer snd sg dcdbas cdrom psmouse soundcore
Re: regression bisected; KVM: entry failed, hardware error 0x80000021
On 24/12/2014 12:02, Jamie Heilman wrote: Running qemu-system-x86_64 -machine pc,accel=kvm -nodefaults works, my real (headless) kvm guests work, but this new patch makes running qemu-system-x86_64 -machine pc,accel=kvm fail again, this time with errors in the host to the tune of: [ cut here ] WARNING: CPU: 1 PID: 3901 at arch/x86/kvm/x86.c:6575 kvm_arch_vcpu_ioctl_run+0xd63/0xe5b [kvm]() Modules linked in: nfsv4 cpufreq_userspace cpufreq_stats cpufreq_powersave cpufreq_ondemand cpufreq_conservative autofs4 fan nfsd auth_rpcgss nfs lockd grace fscache sunrpc bridge stp llc vhost_net tun vhost macvtap macvlan fuse cbc dm_crypt usb_storage snd_hda_codec_analog snd_hda_codec_generic kvm_intel kvm tg3 ptp pps_core sr_mod snd_hda_intel snd_hda_controller snd_hda_codec snd_hwdep snd_pcm snd_timer snd sg dcdbas cdrom psmouse soundcore floppy evdev xfs dm_mod raid1 md_mod CPU: 1 PID: 3901 Comm: qemu-system-x86 Not tainted 3.19.0-rc1-00011-g53262d1-dirty #1 Hardware name: Dell Inc. Precision WorkStation T3400 /0TP412, BIOS A14 04/30/2012 7e052328 8800c25ffcf8 813defbe 8800c25ffd38 8103b517 8800c25ffd28 a019bdec 8800caf1d000 8800c2774800 Call Trace: [813defbe] dump_stack+0x4c/0x6e [8103b517] warn_slowpath_common+0x97/0xb1 [a019bdec] ? kvm_arch_vcpu_ioctl_run+0xd63/0xe5b [kvm] [8103b60b] warn_slowpath_null+0x15/0x17 [a019bdec] kvm_arch_vcpu_ioctl_run+0xd63/0xe5b [kvm] [a02308b9] ? vmcs_load+0x20/0x62 [kvm_intel] [a0231e03] ? vmx_vcpu_load+0x140/0x16a [kvm_intel] [a0196ba3] ? kvm_arch_vcpu_load+0x15c/0x161 [kvm] [a018d8b1] kvm_vcpu_ioctl+0x189/0x4bd [kvm] [8104647a] ? do_sigtimedwait+0x12f/0x189 [810ea316] do_vfs_ioctl+0x370/0x436 [810f24f2] ? __fget+0x67/0x72 [810ea41b] SyS_ioctl+0x3f/0x5e [813e34d2] system_call_fastpath+0x12/0x17 ---[ end trace 46abac932fb3b4a1 ]--- [ cut here ] WARNING: CPU: 1 PID: 3901 at arch/x86/kvm/x86.c:6575 kvm_arch_vcpu_ioctl_run+0xd63/0xe5b [kvm]() Modules linked in: nfsv4 cpufreq_userspace cpufreq_stats cpufreq_powersave cpufreq_ondemand cpufreq_conservative autofs4 fan nfsd auth_rpcgss nfs lockd grace fscache sunrpc bridge stp llc vhost_net tun vhost macvtap macvlan fuse cbc dm_crypt usb_storage snd_hda_codec_analog snd_hda_codec_generic kvm_intel kvm tg3 ptp pps_core sr_mod snd_hda_intel snd_hda_controller snd_hda_codec snd_hwdep snd_pcm snd_timer snd sg dcdbas cdrom psmouse soundcore floppy evdev xfs dm_mod raid1 md_mod CPU: 1 PID: 3901 Comm: qemu-system-x86 Tainted: GW 3.19.0-rc1-00011-g53262d1-dirty #1 Hardware name: Dell Inc. Precision WorkStation T3400 /0TP412, BIOS A14 04/30/2012 7e052328 8800c25ffcf8 813defbe 8800c25ffd38 8103b517 8800c25ffd28 a019bdec 8800caf1d000 8800c2774800 Call Trace: [813defbe] dump_stack+0x4c/0x6e [8103b517] warn_slowpath_common+0x97/0xb1 [a019bdec] ? kvm_arch_vcpu_ioctl_run+0xd63/0xe5b [kvm] [8103b60b] warn_slowpath_null+0x15/0x17 [a019bdec] kvm_arch_vcpu_ioctl_run+0xd63/0xe5b [kvm] [a02308b9] ? vmcs_load+0x20/0x62 [kvm_intel] [a0231e03] ? vmx_vcpu_load+0x140/0x16a [kvm_intel] [a0196ba3] ? kvm_arch_vcpu_load+0x15c/0x161 [kvm] [a018d8b1] kvm_vcpu_ioctl+0x189/0x4bd [kvm] [8104647a] ? do_sigtimedwait+0x12f/0x189 [810ea316] do_vfs_ioctl+0x370/0x436 [810f24f2] ? __fget+0x67/0x72 [810ea41b] SyS_ioctl+0x3f/0x5e [813e34d2] system_call_fastpath+0x12/0x17 ---[ end trace 46abac932fb3b4a2 ]--- over and over and over ad nauseum, or until I kill the qemu command, it also eats a core's worth of cpu. Yeah, I'm fairly sure that the second hunk of Tiejun's patch is not correct, but he's on the right track. I hope to post a fix today, else on the 27th or 29th. Paolo -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: regression bisected; KVM: entry failed, hardware error 0x80000021
On 2014/12/24 19:02, Jamie Heilman wrote: Chen, Tiejun wrote: On 2014/12/23 15:26, Jamie Heilman wrote: Chen, Tiejun wrote: On 2014/12/23 9:50, Chen, Tiejun wrote: On 2014/12/22 17:23, Jamie Heilman wrote: KVM internal error. Suberror: 1 emulation failure EAX=000de494 EBX= ECX= EDX=0cfd ESI=0059 EDI= EBP= ESP=6fb4 EIP=000f15c1 EFL=00010016 [AP-] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES =0010 00c09300 DPL=0 DS [-WA] CS =0008 00c09b00 DPL=0 CS32 [-RA] SS =0010 00c09300 DPL=0 DS [-WA] DS =0010 00c09300 DPL=0 DS [-WA] FS =0010 00c09300 DPL=0 DS [-WA] GS =0010 00c09300 DPL=0 DS [-WA] LDT= 8200 DPL=0 LDT TR = 8b00 DPL=0 TSS32-busy GDT= 000f6be8 0037 IDT= 000f6c26 CR0=6011 CR2= CR3= CR4= DR0= DR1= DR2= DR3= DR6=0ff0 DR7=0400 EFER= Code=e8 ae fc ff ff 89 f2 a8 10 89 d8 75 0a b9 41 15 ff ff ff d1 5b 5e c3 5b 5e e9 76 ff ff ff b0 11 e6 20 e6 a0 b0 08 e6 21 b0 70 e6 a1 b0 04 e6 21 b0 02 FWIW, I get the same thing with 34a1cd60d17 reverted. Maybe there are two bugs, maybe there's more to this first one. I can repro this So if my understanding is correct, this is probably another bug. And especially, I already saw the same log in another thread, Cleaning up the KVM clock. Maybe you can continue to `git bisect` to locate that bad commit. Looks just now Andy found that commit, 0e60b0799fedc495a5c57dbd669de3c10d72edd2 kvm: change memslot sorting rule from size to GFN, maybe you can try to revert this to try yours again. That doesn't revert cleanly for me, and I don't have much time to fiddle with it until the 24th---so checked out the commit before it (d4ae84a0), applied your patch, built, and yes, everything works fine at that point. I'll probably have time for another full bisection later, assuming things aren't ironed out already by then. 3.18.0-rc3-00120-gd4ae84a0 + vmx reorder msr writes patch = OK 3.18.0-rc3-00121-g0e60b07 + vmx reorder msr writes patch = emulation failure So that certainly points to 0e60b0799fedc495a5c57dbd669de3c10d72edd2 as well. Could you try this to fix your last error? Running qemu-system-x86_64 -machine pc,accel=kvm -nodefaults works, my real (headless) kvm guests work, but this new patch makes running qemu-system-x86_64 -machine pc,accel=kvm fail again, this time with Are you sure? From my test based on 3.19-rc1 that it owns top commit, aa39477b5692611b91ac9455ae588738852b3f60 just plus my previous patch, kvm: x86: vmx: reorder some msr writing I already can execute such a command successfully, qemu-system-x86_64 -machine pc,accel=kvm -m 2048 -smp 2 -hda ubuntu.img And your log below seems not to relate mem_slot issue we're discussing, I guess you need to update qemu as well. But I also found my new patch just work out Andy's next case, its really bringing a new issue in !next case. So I tried to refine that patch again as follows, Signed-off-by: Tiejun Chen tiejun.c...@intel.com --- virt/kvm/kvm_main.c | 5 - 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index f528343..910bc48 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -672,6 +672,7 @@ static void update_memslots(struct kvm_memslots *slots, WARN_ON(mslots[i].id != id); if (!new-npages) { new-base_gfn = 0; + new-flags = 0; if (mslots[i].npages) slots-used_slots--; } else { @@ -688,7 +689,9 @@ static void update_memslots(struct kvm_memslots *slots, i++; } while (i 0 - new-base_gfn mslots[i - 1].base_gfn) { + ((new-base_gfn mslots[i - 1].base_gfn) || +(!new-base_gfn + !mslots[i - 1].base_gfn !mslots[i - 1].npages))) { mslots[i] = mslots[i - 1]; slots-id_to_index[mslots[i].id] = i; i--; Tiejun errors in the host to the tune of: [ cut here ] WARNING: CPU: 1 PID: 3901 at arch/x86/kvm/x86.c:6575 kvm_arch_vcpu_ioctl_run+0xd63/0xe5b [kvm]() Modules linked in: nfsv4 cpufreq_userspace cpufreq_stats cpufreq_powersave cpufreq_ondemand cpufreq_conservative autofs4 fan nfsd auth_rpcgss nfs lockd grace fscache sunrpc bridge stp llc vhost_net tun vhost macvtap macvlan fuse cbc dm_crypt usb_storage snd_hda_codec_analog snd_hda_codec_generic kvm_intel kvm tg3 ptp pps_core sr_mod snd_hda_intel snd_hda_controller snd_hda_codec snd_hwdep snd_pcm snd_timer snd sg dcdbas cdrom psmouse soundcore floppy evdev xfs dm_mod raid1 md_mod CPU: 1 PID: 3901 Comm: qemu-system-x86 Not tainted 3.19.0-rc1-00011-g53262d1-dirty #1 Hardware
Re: regression bisected; KVM: entry failed, hardware error 0x80000021
On 2014/12/24 19:11, Paolo Bonzini wrote: On 24/12/2014 12:02, Jamie Heilman wrote: Running qemu-system-x86_64 -machine pc,accel=kvm -nodefaults works, my real (headless) kvm guests work, but this new patch makes running qemu-system-x86_64 -machine pc,accel=kvm fail again, this time with errors in the host to the tune of: [ cut here ] WARNING: CPU: 1 PID: 3901 at arch/x86/kvm/x86.c:6575 kvm_arch_vcpu_ioctl_run+0xd63/0xe5b [kvm]() Modules linked in: nfsv4 cpufreq_userspace cpufreq_stats cpufreq_powersave cpufreq_ondemand cpufreq_conservative autofs4 fan nfsd auth_rpcgss nfs lockd grace fscache sunrpc bridge stp llc vhost_net tun vhost macvtap macvlan fuse cbc dm_crypt usb_storage snd_hda_codec_analog snd_hda_codec_generic kvm_intel kvm tg3 ptp pps_core sr_mod snd_hda_intel snd_hda_controller snd_hda_codec snd_hwdep snd_pcm snd_timer snd sg dcdbas cdrom psmouse soundcore floppy evdev xfs dm_mod raid1 md_mod CPU: 1 PID: 3901 Comm: qemu-system-x86 Not tainted 3.19.0-rc1-00011-g53262d1-dirty #1 Hardware name: Dell Inc. Precision WorkStation T3400 /0TP412, BIOS A14 04/30/2012 7e052328 8800c25ffcf8 813defbe 8800c25ffd38 8103b517 8800c25ffd28 a019bdec 8800caf1d000 8800c2774800 Call Trace: [813defbe] dump_stack+0x4c/0x6e [8103b517] warn_slowpath_common+0x97/0xb1 [a019bdec] ? kvm_arch_vcpu_ioctl_run+0xd63/0xe5b [kvm] [8103b60b] warn_slowpath_null+0x15/0x17 [a019bdec] kvm_arch_vcpu_ioctl_run+0xd63/0xe5b [kvm] [a02308b9] ? vmcs_load+0x20/0x62 [kvm_intel] [a0231e03] ? vmx_vcpu_load+0x140/0x16a [kvm_intel] [a0196ba3] ? kvm_arch_vcpu_load+0x15c/0x161 [kvm] [a018d8b1] kvm_vcpu_ioctl+0x189/0x4bd [kvm] [8104647a] ? do_sigtimedwait+0x12f/0x189 [810ea316] do_vfs_ioctl+0x370/0x436 [810f24f2] ? __fget+0x67/0x72 [810ea41b] SyS_ioctl+0x3f/0x5e [813e34d2] system_call_fastpath+0x12/0x17 ---[ end trace 46abac932fb3b4a1 ]--- [ cut here ] WARNING: CPU: 1 PID: 3901 at arch/x86/kvm/x86.c:6575 kvm_arch_vcpu_ioctl_run+0xd63/0xe5b [kvm]() Modules linked in: nfsv4 cpufreq_userspace cpufreq_stats cpufreq_powersave cpufreq_ondemand cpufreq_conservative autofs4 fan nfsd auth_rpcgss nfs lockd grace fscache sunrpc bridge stp llc vhost_net tun vhost macvtap macvlan fuse cbc dm_crypt usb_storage snd_hda_codec_analog snd_hda_codec_generic kvm_intel kvm tg3 ptp pps_core sr_mod snd_hda_intel snd_hda_controller snd_hda_codec snd_hwdep snd_pcm snd_timer snd sg dcdbas cdrom psmouse soundcore floppy evdev xfs dm_mod raid1 md_mod CPU: 1 PID: 3901 Comm: qemu-system-x86 Tainted: GW 3.19.0-rc1-00011-g53262d1-dirty #1 Hardware name: Dell Inc. Precision WorkStation T3400 /0TP412, BIOS A14 04/30/2012 7e052328 8800c25ffcf8 813defbe 8800c25ffd38 8103b517 8800c25ffd28 a019bdec 8800caf1d000 8800c2774800 Call Trace: [813defbe] dump_stack+0x4c/0x6e [8103b517] warn_slowpath_common+0x97/0xb1 [a019bdec] ? kvm_arch_vcpu_ioctl_run+0xd63/0xe5b [kvm] [8103b60b] warn_slowpath_null+0x15/0x17 [a019bdec] kvm_arch_vcpu_ioctl_run+0xd63/0xe5b [kvm] [a02308b9] ? vmcs_load+0x20/0x62 [kvm_intel] [a0231e03] ? vmx_vcpu_load+0x140/0x16a [kvm_intel] [a0196ba3] ? kvm_arch_vcpu_load+0x15c/0x161 [kvm] [a018d8b1] kvm_vcpu_ioctl+0x189/0x4bd [kvm] [8104647a] ? do_sigtimedwait+0x12f/0x189 [810ea316] do_vfs_ioctl+0x370/0x436 [810f24f2] ? __fget+0x67/0x72 [810ea41b] SyS_ioctl+0x3f/0x5e [813e34d2] system_call_fastpath+0x12/0x17 ---[ end trace 46abac932fb3b4a2 ]--- over and over and over ad nauseum, or until I kill the qemu command, it also eats a core's worth of cpu. Such a message above seems to be out of our mem_slot issue, I'm not 100% sure but actually I can run this case, qemu-system-x86_64 -machine pc,accel=kvm -m 2048 -smp 2 -hda ubuntu.img Just one patch, kvm: x86: vmx: reorder some msr writing, is needed here. So I guess you guy can try to pull your 3.19-rc1 + that patch, and also pull qemu. Yeah, I'm fairly sure that the second hunk of Tiejun's patch is not correct, but he's on the right track. I hope to post a fix today, else Yeah, looks that will broken !next case then I regenerate that again post into another email. Now at lease I myself can run Andy's next case and a normal case, qemu-system-x86_64 -machine pc,accel=kvm, at the same time. But if I'm missing something please correct me directly :) Tiejun -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: regression bisected; KVM: entry failed, hardware error 0x80000021
Chen, Tiejun wrote: On 2014/12/21 20:46, Jamie Heilman wrote: With v3.19-rc1 when I run qemu-system-x86_64 -machine pc,accel=kvm I get: KVM: entry failed, hardware error 0x8021 Looks some MSR writing issues such a failed entry. If you're running a guest on an Intel machine without unrestricted mode support, the failure can be most likely due to the guest entering an invalid state for Intel VT. For example, the guest maybe running in big real mode which is not supported on less recent Intel processors. EAX= EBX= ECX= EDX=0663 ESI= EDI= EBP= ESP= EIP=e05b EFL=00010002 [---] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES = 9300 CS =f000 000f 9b00 SS = 9300 DS = 9300 FS = 9300 GS = 9300 LDT= 8200 TR = 8b00 GDT= IDT= CR0=6010 CR2= CR3= CR4= DR0= DR1= DR2= DR3= DR6=0ff0 DR7=0400 EFER= And I don't see any obvious wrong as well. Any valuable info from dmesg? With the simple qemu command above, on 3.18.1 I see: kern.info: kvm: zapping shadow pages for mmio generation wraparound when I fire up a full guest that's actually useful I get: kern.info: kvm: zapping shadow pages for mmio generation wraparound kern.err: kvm [4073]: vcpu0 disabled perfctr wrmsr: 0xc1 data 0x On 3.18.0-rc3-00042-g34a1cd6 nothing appears in the dmesg, just the message I mention above to stderr. Same thing with a stock 3.19.0-rc1. Once I apply your patch the simple test command produces the same zapping shadow pages messages as 3.18.1, and a test guest of a Debian Jessie image (w/stock distro kernel) produces the same thing with disabled perfctr wrmsr message. However, it doesn't look like I'm entirely out of the woods, because one of my other guest VMs with a custom kernel that works great under 3.18.1 now fails to run. Nothing in dmesg, but here's the stderr: KVM internal error. Suberror: 1 emulation failure EAX=000de494 EBX= ECX= EDX=0cfd ESI=0059 EDI= EBP= ESP=6fb4 EIP=000f15c1 EFL=00010016 [AP-] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES =0010 00c09300 DPL=0 DS [-WA] CS =0008 00c09b00 DPL=0 CS32 [-RA] SS =0010 00c09300 DPL=0 DS [-WA] DS =0010 00c09300 DPL=0 DS [-WA] FS =0010 00c09300 DPL=0 DS [-WA] GS =0010 00c09300 DPL=0 DS [-WA] LDT= 8200 DPL=0 LDT TR = 8b00 DPL=0 TSS32-busy GDT= 000f6be8 0037 IDT= 000f6c26 CR0=6011 CR2= CR3= CR4= DR0= DR1= DR2= DR3= DR6=0ff0 DR7=0400 EFER= Code=e8 ae fc ff ff 89 f2 a8 10 89 d8 75 0a b9 41 15 ff ff ff d1 5b 5e c3 5b 5e e9 76 ff ff ff b0 11 e6 20 e6 a0 b0 08 e6 21 b0 70 e6 a1 b0 04 e6 21 b0 02 FWIW, I get the same thing with 34a1cd60d17 reverted. Maybe there are two bugs, maybe there's more to this first one. I can repro this error with the command: qemu-system-x86_64 -machine pc,accel=kvm -nodefaults This is with QEMU emulator version 2.1.2 (Debian 1:2.1+dfsg-11), Copyright (c) 2003-2008 Fabrice Bellard The host system is: cpu family : 6 model : 23 model name : Intel(R) Core(TM)2 Duo CPU E8400 @ 3.00GHz stepping: 10 microcode : 0xa0b ... flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx lm constant_tsc arch_perfmon pebs bts rep_good nopl aperfmperf pni dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm sse4_1 xsave lahf_lm dtherm tpr_shadow vnmi flexpriority I bisected this back to: commit 34a1cd60d17f62c1f077c1478a6c2ca8c3d17af4 Author: Tiejun Chen tiejun.c...@intel.com Date: Tue Oct 28 10:14:48 2014 +0800 kvm: x86: vmx: move some vmx setting from vmx_init() to hardware_setup() Instead of vmx_init(), actually it would make reasonable sense to do anything specific to vmx hardware setting in vmx_x86_ops-hardware_setup(). This commit just reorders something but some MSR writing depend on previous status. Could you try this? I unmangled the expanded tabs, and applied this: diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index feb852b..96c84a8 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -5840,49 +5840,6 @@ static __init int hardware_setup(void) memset(vmx_msr_bitmap_legacy, 0xff, PAGE_SIZE); memset(vmx_msr_bitmap_longmode,
Re: regression bisected; KVM: entry failed, hardware error 0x80000021
On 2014/12/22 17:23, Jamie Heilman wrote: Chen, Tiejun wrote: On 2014/12/21 20:46, Jamie Heilman wrote: With v3.19-rc1 when I run qemu-system-x86_64 -machine pc,accel=kvm I get: KVM: entry failed, hardware error 0x8021 Looks some MSR writing issues such a failed entry. If you're running a guest on an Intel machine without unrestricted mode support, the failure can be most likely due to the guest entering an invalid state for Intel VT. For example, the guest maybe running in big real mode which is not supported on less recent Intel processors. EAX= EBX= ECX= EDX=0663 ESI= EDI= EBP= ESP= EIP=e05b EFL=00010002 [---] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES = 9300 CS =f000 000f 9b00 SS = 9300 DS = 9300 FS = 9300 GS = 9300 LDT= 8200 TR = 8b00 GDT= IDT= CR0=6010 CR2= CR3= CR4= DR0= DR1= DR2= DR3= DR6=0ff0 DR7=0400 EFER= And I don't see any obvious wrong as well. Any valuable info from dmesg? With the simple qemu command above, on 3.18.1 I see: kern.info: kvm: zapping shadow pages for mmio generation wraparound when I fire up a full guest that's actually useful I get: kern.info: kvm: zapping shadow pages for mmio generation wraparound kern.err: kvm [4073]: vcpu0 disabled perfctr wrmsr: 0xc1 data 0x On 3.18.0-rc3-00042-g34a1cd6 nothing appears in the dmesg, just the message I mention above to stderr. Same thing with a stock 3.19.0-rc1. Once I apply your patch the simple test command produces the same zapping shadow pages messages as 3.18.1, and a test guest of a Debian Jessie image (w/stock distro kernel) produces the same thing with disabled perfctr wrmsr message. However, it doesn't look like Sorry I'm not sure if I understood current status. Looks 3.19-rc1 my patch just fix that error above, KVM: entry failed, hardware error 0x8021 ... Right? I'm entirely out of the woods, because one of my other guest VMs with a custom kernel that works great under 3.18.1 now fails to run. Nothing in dmesg, but here's the stderr: But even you revert 34a1cd60d17 or just apply my patch, something else introduced between 3.18.1 and 3.19-rc1 led this error below, right? KVM internal error. Suberror: 1 emulation failure EAX=000de494 EBX= ECX= EDX=0cfd ESI=0059 EDI= EBP= ESP=6fb4 EIP=000f15c1 EFL=00010016 [AP-] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES =0010 00c09300 DPL=0 DS [-WA] CS =0008 00c09b00 DPL=0 CS32 [-RA] SS =0010 00c09300 DPL=0 DS [-WA] DS =0010 00c09300 DPL=0 DS [-WA] FS =0010 00c09300 DPL=0 DS [-WA] GS =0010 00c09300 DPL=0 DS [-WA] LDT= 8200 DPL=0 LDT TR = 8b00 DPL=0 TSS32-busy GDT= 000f6be8 0037 IDT= 000f6c26 CR0=6011 CR2= CR3= CR4= DR0= DR1= DR2= DR3= DR6=0ff0 DR7=0400 EFER= Code=e8 ae fc ff ff 89 f2 a8 10 89 d8 75 0a b9 41 15 ff ff ff d1 5b 5e c3 5b 5e e9 76 ff ff ff b0 11 e6 20 e6 a0 b0 08 e6 21 b0 70 e6 a1 b0 04 e6 21 b0 02 FWIW, I get the same thing with 34a1cd60d17 reverted. Maybe there are two bugs, maybe there's more to this first one. I can repro this So if my understanding is correct, this is probably another bug. And especially, I already saw the same log in another thread, Cleaning up the KVM clock. Maybe you can continue to `git bisect` to locate that bad commit. Thanks Tiejun error with the command: qemu-system-x86_64 -machine pc,accel=kvm -nodefaults This is with QEMU emulator version 2.1.2 (Debian 1:2.1+dfsg-11), Copyright (c) 2003-2008 Fabrice Bellard The host system is: cpu family : 6 model : 23 model name : Intel(R) Core(TM)2 Duo CPU E8400 @ 3.00GHz stepping: 10 microcode : 0xa0b ... flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx lm constant_tsc arch_perfmon pebs bts rep_good nopl aperfmperf pni dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm sse4_1 xsave lahf_lm dtherm tpr_shadow vnmi flexpriority I bisected this back to: commit 34a1cd60d17f62c1f077c1478a6c2ca8c3d17af4 Author: Tiejun Chen tiejun.c...@intel.com Date: Tue Oct 28 10:14:48 2014 +0800 kvm: x86: vmx: move some vmx setting from vmx_init() to hardware_setup() Instead of vmx_init(), actually it would make reasonable sense
Re: regression bisected; KVM: entry failed, hardware error 0x80000021
On 2014/12/23 9:50, Chen, Tiejun wrote: On 2014/12/22 17:23, Jamie Heilman wrote: Chen, Tiejun wrote: On 2014/12/21 20:46, Jamie Heilman wrote: With v3.19-rc1 when I run qemu-system-x86_64 -machine pc,accel=kvm I get: KVM: entry failed, hardware error 0x8021 Looks some MSR writing issues such a failed entry. If you're running a guest on an Intel machine without unrestricted mode support, the failure can be most likely due to the guest entering an invalid state for Intel VT. For example, the guest maybe running in big real mode which is not supported on less recent Intel processors. EAX= EBX= ECX= EDX=0663 ESI= EDI= EBP= ESP= EIP=e05b EFL=00010002 [---] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES = 9300 CS =f000 000f 9b00 SS = 9300 DS = 9300 FS = 9300 GS = 9300 LDT= 8200 TR = 8b00 GDT= IDT= CR0=6010 CR2= CR3= CR4= DR0= DR1= DR2= DR3= DR6=0ff0 DR7=0400 EFER= And I don't see any obvious wrong as well. Any valuable info from dmesg? With the simple qemu command above, on 3.18.1 I see: kern.info: kvm: zapping shadow pages for mmio generation wraparound when I fire up a full guest that's actually useful I get: kern.info: kvm: zapping shadow pages for mmio generation wraparound kern.err: kvm [4073]: vcpu0 disabled perfctr wrmsr: 0xc1 data 0x On 3.18.0-rc3-00042-g34a1cd6 nothing appears in the dmesg, just the message I mention above to stderr. Same thing with a stock 3.19.0-rc1. Once I apply your patch the simple test command produces the same zapping shadow pages messages as 3.18.1, and a test guest of a Debian Jessie image (w/stock distro kernel) produces the same thing with disabled perfctr wrmsr message. However, it doesn't look like Sorry I'm not sure if I understood current status. Looks 3.19-rc1 my patch just fix that error above, KVM: entry failed, hardware error 0x8021 ... Right? I'm entirely out of the woods, because one of my other guest VMs with a custom kernel that works great under 3.18.1 now fails to run. Nothing in dmesg, but here's the stderr: But even you revert 34a1cd60d17 or just apply my patch, something else introduced between 3.18.1 and 3.19-rc1 led this error below, right? KVM internal error. Suberror: 1 emulation failure EAX=000de494 EBX= ECX= EDX=0cfd ESI=0059 EDI= EBP= ESP=6fb4 EIP=000f15c1 EFL=00010016 [AP-] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES =0010 00c09300 DPL=0 DS [-WA] CS =0008 00c09b00 DPL=0 CS32 [-RA] SS =0010 00c09300 DPL=0 DS [-WA] DS =0010 00c09300 DPL=0 DS [-WA] FS =0010 00c09300 DPL=0 DS [-WA] GS =0010 00c09300 DPL=0 DS [-WA] LDT= 8200 DPL=0 LDT TR = 8b00 DPL=0 TSS32-busy GDT= 000f6be8 0037 IDT= 000f6c26 CR0=6011 CR2= CR3= CR4= DR0= DR1= DR2= DR3= DR6=0ff0 DR7=0400 EFER= Code=e8 ae fc ff ff 89 f2 a8 10 89 d8 75 0a b9 41 15 ff ff ff d1 5b 5e c3 5b 5e e9 76 ff ff ff b0 11 e6 20 e6 a0 b0 08 e6 21 b0 70 e6 a1 b0 04 e6 21 b0 02 FWIW, I get the same thing with 34a1cd60d17 reverted. Maybe there are two bugs, maybe there's more to this first one. I can repro this So if my understanding is correct, this is probably another bug. And especially, I already saw the same log in another thread, Cleaning up the KVM clock. Maybe you can continue to `git bisect` to locate that bad commit. Looks just now Andy found that commit, 0e60b0799fedc495a5c57dbd669de3c10d72edd2 kvm: change memslot sorting rule from size to GFN, maybe you can try to revert this to try yours again. Thanks Tiejun -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: regression bisected; KVM: entry failed, hardware error 0x80000021
Chen, Tiejun wrote: On 2014/12/23 9:50, Chen, Tiejun wrote: On 2014/12/22 17:23, Jamie Heilman wrote: Chen, Tiejun wrote: On 2014/12/21 20:46, Jamie Heilman wrote: With v3.19-rc1 when I run qemu-system-x86_64 -machine pc,accel=kvm I get: KVM: entry failed, hardware error 0x8021 Looks some MSR writing issues such a failed entry. If you're running a guest on an Intel machine without unrestricted mode support, the failure can be most likely due to the guest entering an invalid state for Intel VT. For example, the guest maybe running in big real mode which is not supported on less recent Intel processors. EAX= EBX= ECX= EDX=0663 ESI= EDI= EBP= ESP= EIP=e05b EFL=00010002 [---] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES = 9300 CS =f000 000f 9b00 SS = 9300 DS = 9300 FS = 9300 GS = 9300 LDT= 8200 TR = 8b00 GDT= IDT= CR0=6010 CR2= CR3= CR4= DR0= DR1= DR2= DR3= DR6=0ff0 DR7=0400 EFER= And I don't see any obvious wrong as well. Any valuable info from dmesg? With the simple qemu command above, on 3.18.1 I see: kern.info: kvm: zapping shadow pages for mmio generation wraparound when I fire up a full guest that's actually useful I get: kern.info: kvm: zapping shadow pages for mmio generation wraparound kern.err: kvm [4073]: vcpu0 disabled perfctr wrmsr: 0xc1 data 0x On 3.18.0-rc3-00042-g34a1cd6 nothing appears in the dmesg, just the message I mention above to stderr. Same thing with a stock 3.19.0-rc1. Once I apply your patch the simple test command produces the same zapping shadow pages messages as 3.18.1, and a test guest of a Debian Jessie image (w/stock distro kernel) produces the same thing with disabled perfctr wrmsr message. However, it doesn't look like Sorry I'm not sure if I understood current status. Looks 3.19-rc1 my patch just fix that error above, KVM: entry failed, hardware error 0x8021 ... Right? I'm entirely out of the woods, because one of my other guest VMs with a custom kernel that works great under 3.18.1 now fails to run. Nothing in dmesg, but here's the stderr: But even you revert 34a1cd60d17 or just apply my patch, something else introduced between 3.18.1 and 3.19-rc1 led this error below, right? KVM internal error. Suberror: 1 emulation failure EAX=000de494 EBX= ECX= EDX=0cfd ESI=0059 EDI= EBP= ESP=6fb4 EIP=000f15c1 EFL=00010016 [AP-] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES =0010 00c09300 DPL=0 DS [-WA] CS =0008 00c09b00 DPL=0 CS32 [-RA] SS =0010 00c09300 DPL=0 DS [-WA] DS =0010 00c09300 DPL=0 DS [-WA] FS =0010 00c09300 DPL=0 DS [-WA] GS =0010 00c09300 DPL=0 DS [-WA] LDT= 8200 DPL=0 LDT TR = 8b00 DPL=0 TSS32-busy GDT= 000f6be8 0037 IDT= 000f6c26 CR0=6011 CR2= CR3= CR4= DR0= DR1= DR2= DR3= DR6=0ff0 DR7=0400 EFER= Code=e8 ae fc ff ff 89 f2 a8 10 89 d8 75 0a b9 41 15 ff ff ff d1 5b 5e c3 5b 5e e9 76 ff ff ff b0 11 e6 20 e6 a0 b0 08 e6 21 b0 70 e6 a1 b0 04 e6 21 b0 02 FWIW, I get the same thing with 34a1cd60d17 reverted. Maybe there are two bugs, maybe there's more to this first one. I can repro this So if my understanding is correct, this is probably another bug. And especially, I already saw the same log in another thread, Cleaning up the KVM clock. Maybe you can continue to `git bisect` to locate that bad commit. Looks just now Andy found that commit, 0e60b0799fedc495a5c57dbd669de3c10d72edd2 kvm: change memslot sorting rule from size to GFN, maybe you can try to revert this to try yours again. That doesn't revert cleanly for me, and I don't have much time to fiddle with it until the 24th---so checked out the commit before it (d4ae84a0), applied your patch, built, and yes, everything works fine at that point. I'll probably have time for another full bisection later, assuming things aren't ironed out already by then. -- Jamie Heilman http://audible.transient.net/~jamie/ -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: regression bisected; KVM: entry failed, hardware error 0x80000021
On 2014/12/23 15:26, Jamie Heilman wrote: Chen, Tiejun wrote: On 2014/12/23 9:50, Chen, Tiejun wrote: On 2014/12/22 17:23, Jamie Heilman wrote: Chen, Tiejun wrote: On 2014/12/21 20:46, Jamie Heilman wrote: With v3.19-rc1 when I run qemu-system-x86_64 -machine pc,accel=kvm I get: KVM: entry failed, hardware error 0x8021 Looks some MSR writing issues such a failed entry. If you're running a guest on an Intel machine without unrestricted mode support, the failure can be most likely due to the guest entering an invalid state for Intel VT. For example, the guest maybe running in big real mode which is not supported on less recent Intel processors. EAX= EBX= ECX= EDX=0663 ESI= EDI= EBP= ESP= EIP=e05b EFL=00010002 [---] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES = 9300 CS =f000 000f 9b00 SS = 9300 DS = 9300 FS = 9300 GS = 9300 LDT= 8200 TR = 8b00 GDT= IDT= CR0=6010 CR2= CR3= CR4= DR0= DR1= DR2= DR3= DR6=0ff0 DR7=0400 EFER= And I don't see any obvious wrong as well. Any valuable info from dmesg? With the simple qemu command above, on 3.18.1 I see: kern.info: kvm: zapping shadow pages for mmio generation wraparound when I fire up a full guest that's actually useful I get: kern.info: kvm: zapping shadow pages for mmio generation wraparound kern.err: kvm [4073]: vcpu0 disabled perfctr wrmsr: 0xc1 data 0x On 3.18.0-rc3-00042-g34a1cd6 nothing appears in the dmesg, just the message I mention above to stderr. Same thing with a stock 3.19.0-rc1. Once I apply your patch the simple test command produces the same zapping shadow pages messages as 3.18.1, and a test guest of a Debian Jessie image (w/stock distro kernel) produces the same thing with disabled perfctr wrmsr message. However, it doesn't look like Sorry I'm not sure if I understood current status. Looks 3.19-rc1 my patch just fix that error above, KVM: entry failed, hardware error 0x8021 ... Right? I'm entirely out of the woods, because one of my other guest VMs with a custom kernel that works great under 3.18.1 now fails to run. Nothing in dmesg, but here's the stderr: But even you revert 34a1cd60d17 or just apply my patch, something else introduced between 3.18.1 and 3.19-rc1 led this error below, right? KVM internal error. Suberror: 1 emulation failure EAX=000de494 EBX= ECX= EDX=0cfd ESI=0059 EDI= EBP= ESP=6fb4 EIP=000f15c1 EFL=00010016 [AP-] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES =0010 00c09300 DPL=0 DS [-WA] CS =0008 00c09b00 DPL=0 CS32 [-RA] SS =0010 00c09300 DPL=0 DS [-WA] DS =0010 00c09300 DPL=0 DS [-WA] FS =0010 00c09300 DPL=0 DS [-WA] GS =0010 00c09300 DPL=0 DS [-WA] LDT= 8200 DPL=0 LDT TR = 8b00 DPL=0 TSS32-busy GDT= 000f6be8 0037 IDT= 000f6c26 CR0=6011 CR2= CR3= CR4= DR0= DR1= DR2= DR3= DR6=0ff0 DR7=0400 EFER= Code=e8 ae fc ff ff 89 f2 a8 10 89 d8 75 0a b9 41 15 ff ff ff d1 5b 5e c3 5b 5e e9 76 ff ff ff b0 11 e6 20 e6 a0 b0 08 e6 21 b0 70 e6 a1 b0 04 e6 21 b0 02 FWIW, I get the same thing with 34a1cd60d17 reverted. Maybe there are two bugs, maybe there's more to this first one. I can repro this So if my understanding is correct, this is probably another bug. And especially, I already saw the same log in another thread, Cleaning up the KVM clock. Maybe you can continue to `git bisect` to locate that bad commit. Looks just now Andy found that commit, 0e60b0799fedc495a5c57dbd669de3c10d72edd2 kvm: change memslot sorting rule from size to GFN, maybe you can try to revert this to try yours again. That doesn't revert cleanly for me, and I don't have much time to Yeah, I guess all associated commits should be reverted gradually. fiddle with it until the 24th---so checked out the commit before it (d4ae84a0), applied your patch, built, and yes, everything works fine Thanks for your test. I think I can submit this patch to fix one of yours problems and I'd like to add you as Reported-by Tested-by. Then we can step into another issue. And I'm trying to fetch 3.19-rc1 (because I'm always working on kvm/next.) to take a look at that but maybe Paolo is already going on that. Tiejun at that point. I'll probably have time for another full bisection later, assuming things aren't ironed out already by then. -- To
regression bisected; KVM: entry failed, hardware error 0x80000021
With v3.19-rc1 when I run qemu-system-x86_64 -machine pc,accel=kvm I get: KVM: entry failed, hardware error 0x8021 If you're running a guest on an Intel machine without unrestricted mode support, the failure can be most likely due to the guest entering an invalid state for Intel VT. For example, the guest maybe running in big real mode which is not supported on less recent Intel processors. EAX= EBX= ECX= EDX=0663 ESI= EDI= EBP= ESP= EIP=e05b EFL=00010002 [---] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES = 9300 CS =f000 000f 9b00 SS = 9300 DS = 9300 FS = 9300 GS = 9300 LDT= 8200 TR = 8b00 GDT= IDT= CR0=6010 CR2= CR3= CR4= DR0= DR1= DR2= DR3= DR6=0ff0 DR7=0400 EFER= Code=85 00 87 00 89 00 8b 00 00 00 86 00 88 00 8a 00 8c 00 00 90 2e 66 83 3e 30 6c 00 0f 85 e7 f2 31 c0 8e d0 66 bc 00 70 00 00 66 ba 31 2e 0f 00 e9 45 f1 This is with QEMU emulator version 2.1.2 (Debian 1:2.1+dfsg-11), Copyright (c) 2003-2008 Fabrice Bellard The host system is: cpu family : 6 model : 23 model name : Intel(R) Core(TM)2 Duo CPU E8400 @ 3.00GHz stepping: 10 microcode : 0xa0b ... flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx lm constant_tsc arch_perfmon pebs bts rep_good nopl aperfmperf pni dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm sse4_1 xsave lahf_lm dtherm tpr_shadow vnmi flexpriority I bisected this back to: commit 34a1cd60d17f62c1f077c1478a6c2ca8c3d17af4 Author: Tiejun Chen tiejun.c...@intel.com Date: Tue Oct 28 10:14:48 2014 +0800 kvm: x86: vmx: move some vmx setting from vmx_init() to hardware_setup() Instead of vmx_init(), actually it would make reasonable sense to do anything specific to vmx hardware setting in vmx_x86_ops-hardware_setup(). Signed-off-by: Tiejun Chen tiejun.c...@intel.com Signed-off-by: Paolo Bonzini pbonz...@redhat.com reverting this commit appears to fix the issue, in so far that I can run kvm again without the aforementioned error, but I haven't tested it thoroughly beyond that. Let me know if you need more information or testing. -- Jamie Heilman http://audible.transient.net/~jamie/ -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: regression bisected; KVM: entry failed, hardware error 0x80000021
On 2014/12/21 20:46, Jamie Heilman wrote: With v3.19-rc1 when I run qemu-system-x86_64 -machine pc,accel=kvm I get: KVM: entry failed, hardware error 0x8021 Looks some MSR writing issues such a failed entry. If you're running a guest on an Intel machine without unrestricted mode support, the failure can be most likely due to the guest entering an invalid state for Intel VT. For example, the guest maybe running in big real mode which is not supported on less recent Intel processors. EAX= EBX= ECX= EDX=0663 ESI= EDI= EBP= ESP= EIP=e05b EFL=00010002 [---] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES = 9300 CS =f000 000f 9b00 SS = 9300 DS = 9300 FS = 9300 GS = 9300 LDT= 8200 TR = 8b00 GDT= IDT= CR0=6010 CR2= CR3= CR4= DR0= DR1= DR2= DR3= DR6=0ff0 DR7=0400 EFER= And I don't see any obvious wrong as well. Any valuable info from dmesg? Code=85 00 87 00 89 00 8b 00 00 00 86 00 88 00 8a 00 8c 00 00 90 2e 66 83 3e 30 6c 00 0f 85 e7 f2 31 c0 8e d0 66 bc 00 70 00 00 66 ba 31 2e 0f 00 e9 45 f1 This is with QEMU emulator version 2.1.2 (Debian 1:2.1+dfsg-11), Copyright (c) 2003-2008 Fabrice Bellard The host system is: cpu family : 6 model : 23 model name : Intel(R) Core(TM)2 Duo CPU E8400 @ 3.00GHz stepping: 10 microcode : 0xa0b ... flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx lm constant_tsc arch_perfmon pebs bts rep_good nopl aperfmperf pni dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm sse4_1 xsave lahf_lm dtherm tpr_shadow vnmi flexpriority I bisected this back to: commit 34a1cd60d17f62c1f077c1478a6c2ca8c3d17af4 Author: Tiejun Chen tiejun.c...@intel.com Date: Tue Oct 28 10:14:48 2014 +0800 kvm: x86: vmx: move some vmx setting from vmx_init() to hardware_setup() Instead of vmx_init(), actually it would make reasonable sense to do anything specific to vmx hardware setting in vmx_x86_ops-hardware_setup(). This commit just reorders something but some MSR writing depend on previous status. Could you try this? diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index feb852b..96c84a8 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -5840,49 +5840,6 @@ static __init int hardware_setup(void) memset(vmx_msr_bitmap_legacy, 0xff, PAGE_SIZE); memset(vmx_msr_bitmap_longmode, 0xff, PAGE_SIZE); - vmx_disable_intercept_for_msr(MSR_FS_BASE, false); - vmx_disable_intercept_for_msr(MSR_GS_BASE, false); - vmx_disable_intercept_for_msr(MSR_KERNEL_GS_BASE, true); - vmx_disable_intercept_for_msr(MSR_IA32_SYSENTER_CS, false); - vmx_disable_intercept_for_msr(MSR_IA32_SYSENTER_ESP, false); - vmx_disable_intercept_for_msr(MSR_IA32_SYSENTER_EIP, false); - vmx_disable_intercept_for_msr(MSR_IA32_BNDCFGS, true); - - memcpy(vmx_msr_bitmap_legacy_x2apic, - vmx_msr_bitmap_legacy, PAGE_SIZE); - memcpy(vmx_msr_bitmap_longmode_x2apic, - vmx_msr_bitmap_longmode, PAGE_SIZE); - - if (enable_apicv) { - for (msr = 0x800; msr = 0x8ff; msr++) - vmx_disable_intercept_msr_read_x2apic(msr); - - /* According SDM, in x2apic mode, the whole id reg is used. -* But in KVM, it only use the highest eight bits. Need to -* intercept it */ - vmx_enable_intercept_msr_read_x2apic(0x802); - /* TMCCT */ - vmx_enable_intercept_msr_read_x2apic(0x839); - /* TPR */ - vmx_disable_intercept_msr_write_x2apic(0x808); - /* EOI */ - vmx_disable_intercept_msr_write_x2apic(0x80b); - /* SELF-IPI */ - vmx_disable_intercept_msr_write_x2apic(0x83f); - } - - if (enable_ept) { - kvm_mmu_set_mask_ptes(0ull, - (enable_ept_ad_bits) ? VMX_EPT_ACCESS_BIT : 0ull, - (enable_ept_ad_bits) ? VMX_EPT_DIRTY_BIT : 0ull, - 0ull, VMX_EPT_EXECUTABLE_MASK); - ept_set_mmio_spte_mask(); - kvm_enable_tdp(); - } else - kvm_disable_tdp(); - - update_ple_window_actual_max(); - if (setup_vmcs_config(vmcs_config) 0) { r = -EIO; goto out7; @@ -5945,6 +5902,49 @@ static __init int hardware_setup(void) if (nested) nested_vmx_setup_ctls_msrs(); +