Re: Windows fails to boot after rebase to QEMU master

[Claudio Fontana <cfontana@suse.de>]

On 5/27/21 12:53 PM, Claudio Fontana wrote:
> On 5/27/21 11:48 AM, Claudio Fontana wrote:
>> On 5/27/21 11:15 AM, Dr. David Alan Gilbert wrote:
>>> * Philippe Mathieu-Daudé (philmd@redhat.com) wrote:
>>>> On 5/27/21 10:31 AM, Dr. David Alan Gilbert wrote:
>>>>> * Claudio Fontana (cfontana@suse.de) wrote:
>>>>>> On 5/26/21 9:30 PM, Dr. David Alan Gilbert wrote:
>>>>>>> * Michael S. Tsirkin (mst@redhat.com) wrote:
>>>>>>>> On Fri, May 21, 2021 at 11:17:19AM +0200, Siddharth Chandrasekaran wrote:
>>>>>>>>> After a rebase to QEMU master, I am having trouble booting windows VMs.
>>>>>>>>> Git bisect indicates commit f5cc5a5c1686 ("i386: split cpu accelerators
>>>>>>>>> from cpu.c, using AccelCPUClass") to have introduced the issue. I spent
>>>>>>>>> some time looking at into it yesterday without much luck.
>>>>>>>>>
>>>>>>>>> Steps to reproduce:
>>>>>>>>>
>>>>>>>>>     $ ./configure --enable-kvm --disable-xen --target-list=x86_64-softmmu --enable-debug
>>>>>>>>>     $ make -j `nproc`
>>>>>>>>>     $ ./build/x86_64-softmmu/qemu-system-x86_64 \
>>>>>>>>>         -cpu host,hv_synic,hv_vpindex,hv_time,hv_runtime,hv_stimer,hv_crash \
>>>>>>>>>         -enable-kvm \
>>>>>>>>>         -name test,debug-threads=on \
>>>>>>>>>         -smp 1,threads=1,cores=1,sockets=1 \
>>>>>>>>>         -m 4G \
>>>>>>>>>         -net nic -net user \
>>>>>>>>>         -boot d,menu=on \
>>>>>>>>>         -usbdevice tablet \
>>>>>>>>>         -vnc :3 \
>>>>>>>>>         -machine q35,smm=on \
>>>>>>>>>         -drive if=pflash,format=raw,readonly=on,unit=0,file="../OVMF_CODE.secboot.fd" \
>>>>>>>>>         -drive if=pflash,format=raw,unit=1,file="../OVMF_VARS.secboot.fd" \
>>>>>>>>>         -global ICH9-LPC.disable_s3=1 \
>>>>>>>>>         -global driver=cfi.pflash01,property=secure,value=on \
>>>>>>>>>         -cdrom "../Windows_Server_2016_14393.ISO" \
>>>>>>>>>         -drive file="../win_server_2016.qcow2",format=qcow2,if=none,id=rootfs_drive \
>>>>>>>>>         -device ahci,id=ahci \
>>>>>>>>>         -device ide-hd,drive=rootfs_drive,bus=ahci.0
>>>>>>>>>
>>>>>>>>> If the issue is not obvious, I'd like some pointers on how to go about
>>>>>>>>> fixing this issue.
>>>>>>>>>
>>>>>>>>> ~ Sid.
>>>>>>>>>
>>>>>>>>
>>>>>>>> At a guess this commit inadvertently changed something in the CPU ID.
>>>>>>>> I'd start by using a linux guest to dump cpuid before and after the
>>>>>>>> change.
>>>>>>>
>>>>>>> I've not had a chance to do that yet, however I did just end up with a
>>>>>>> bisect of a linux guest failure bisecting to the same patch:
>>>>>>>
>>>>>>> [dgilbert@dgilbert-t580 qemu]$ git bisect bad
>>>>>>> f5cc5a5c168674f84bf061cdb307c2d25fba5448 is the first bad commit
>>>>>>> commit f5cc5a5c168674f84bf061cdb307c2d25fba5448
>>>>>>> Author: Claudio Fontana <cfontana@suse.de>
>>>>>>> Date:   Mon Mar 22 14:27:40 2021 +0100
>>>>>>>
>>>>>>>     i386: split cpu accelerators from cpu.c, using AccelCPUClass
>>>>>>>     
>>>>>>>     i386 is the first user of AccelCPUClass, allowing to split
>>>>>>>     cpu.c into:
>>>>>>>     
>>>>>>>     cpu.c            cpuid and common x86 cpu functionality
>>>>>>>     host-cpu.c       host x86 cpu functions and "host" cpu type
>>>>>>>     kvm/kvm-cpu.c    KVM x86 AccelCPUClass
>>>>>>>     hvf/hvf-cpu.c    HVF x86 AccelCPUClass
>>>>>>>     tcg/tcg-cpu.c    TCG x86 AccelCPUClass
>>>>
>>>> Well this is a big commit... I'm not custom to x86 target, and am
>>>> having hard time following the cpu host/max change.
>>>>
>>>> Is it working when you use '-cpu max,...' instead of '-cpu host,'?
>>>
>>> No; and in fact the cpuid's are almost entirely different with and
>>> without this patch! (both with -cpu host).  It looks like with this
>>> patch we're getting the cpuid for the TCG cpuid rather than the host:
>>>
>>> Prior to this patch:
>>> :/# cat /proc/cpuinfo
>>> processor       : 0
>>> vendor_id       : GenuineIntel
>>> cpu family      : 6
>>> model           : 142
>>> model name      : Intel(R) Core(TM) i7-8650U CPU @ 1.90GHz
>>> stepping        : 10
>>> microcode       : 0xe0
>>> cpu MHz         : 2111.998
>>> cache size      : 16384 KB
>>> physical id     : 0
>>> siblings        : 1
>>> core id         : 0
>>> cpu cores       : 1
>>> apicid          : 0
>>> initial apicid  : 0
>>> fpu             : yes
>>> fpu_exception   : yes
>>> cpuid level     : 22
>>> wp              : yes
>>> flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss syscall nx pdpe1gb rdtscp lm constant
>>> _tsc arch_perfmon rep_good nopl xtopology cpuid tsc_known_freq pni pclmulqdq vmx ssse3 fma cx16 pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_tim
>>> er aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch cpuid_fault invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid
>>> ept_ad fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm mpx rdseed adx smap clflushopt xsaveopt xsavec xgetbv1 xsaves arat umip md_clear arch_ca
>>> pabilities
>>> vmx flags       : vnmi preemption_timer invvpid ept_x_only ept_ad ept_1gb flexpriority tsc_offset vtpr mtf vapic ept vpid unrestricted_guest shadow_vmcs pml
>>> bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs taa srbds
>>> bogomips        : 4223.99
>>> clflush size    : 64
>>> cache_alignment : 64
>>> address sizes   : 39 bits physical, 48 bits virtual
>>> power management:
>>>
>>> With this patch:
>>> processor       : 0
>>> vendor_id       : AuthenticAMD
>>> cpu family      : 6
>>> model           : 6
>>> model name      : QEMU TCG CPU version 2.5+
>>> stepping        : 3
>>> cpu MHz         : 2111.998
>>> cache size      : 512 KB
>>> physical id     : 0
>>> siblings        : 1
>>> core id         : 0
>>> cpu cores       : 1
>>> apicid          : 0
>>> initial apicid  : 0
>>> fpu             : yes
>>> fpu_exception   : yes
>>> cpuid level     : 13
>>> wp              : yes
>>> flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss syscall nx pdpe1gb rdtscp lm nopl cpu
>>> id tsc_known_freq pni pclmulqdq vmx ssse3 fma cx16 pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_
>>> lm abm 3dnowprefetch invpcid_single ssbd ibrs ibpb stibp vmmcall fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm mpx rdseed adx smap clflushopt
>>>  xsaveopt xsavec xgetbv1 xsaves arat umip md_clear arch_capabilities
>>> bugs            : fxsave_leak sysret_ss_attrs spectre_v1 spectre_v2 spec_store_bypass taa
>>> bogomips        : 4223.99
>>> TLB size        : 1024 4K pages
>>> clflush size    : 64
>>> cache_alignment : 64
>>> address sizes   : 40 bits physical, 48 bits virtual
>>> power management:
>>>
>>> cpuid.f5cc5a5c16
>>>
>>> CPU 0:
>>>    0x00000000 0x00: eax=0x0000000d ebx=0x68747541 ecx=0x444d4163 edx=0x69746e65
>>>    0x00000001 0x00: eax=0x00000663 ebx=0x00000800 ecx=0xfffab223 edx=0x0f8bfbff
>>>    0x00000002 0x00: eax=0x00000001 ebx=0x00000000 ecx=0x0000004d edx=0x002c307d
>>>    0x00000003 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x00000004 0x00: eax=0x00000121 ebx=0x01c0003f ecx=0x0000003f edx=0x00000001
>>>    0x00000004 0x01: eax=0x00000122 ebx=0x01c0003f ecx=0x0000003f edx=0x00000001
>>>    0x00000004 0x02: eax=0x00000143 ebx=0x03c0003f ecx=0x00000fff edx=0x00000001
>>>    0x00000004 0x03: eax=0x00000163 ebx=0x03c0003f ecx=0x00003fff edx=0x00000006
>>>    0x00000005 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000003 edx=0x00000000
>>>    0x00000006 0x00: eax=0x00000004 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x00000007 0x00: eax=0x00000000 ebx=0x009c4fbb ecx=0x00000004 edx=0xac000400
>>>    0x00000008 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x00000009 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x0000000a 0x00: eax=0x07300402 ebx=0x00000000 ecx=0x00000000 edx=0x00008603
>>>    0x0000000b 0x00: eax=0x00000000 ebx=0x00000001 ecx=0x00000100 edx=0x00000000
>>>    0x0000000b 0x01: eax=0x00000000 ebx=0x00000001 ecx=0x00000201 edx=0x00000000
>>>    0x0000000c 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x0000000d 0x00: eax=0x0000001f ebx=0x00000440 ecx=0x00000440 edx=0x00000000
>>>    0x0000000d 0x01: eax=0x0000000f ebx=0x000003c0 ecx=0x00000000 edx=0x00000000
>>>    0x0000000d 0x02: eax=0x00000100 ebx=0x00000240 ecx=0x00000000 edx=0x00000000
>>>    0x0000000d 0x03: eax=0x00000040 ebx=0x000003c0 ecx=0x00000000 edx=0x00000000
>>>    0x0000000d 0x04: eax=0x00000040 ebx=0x00000400 ecx=0x00000000 edx=0x00000000
>>>    0x40000000 0x00: eax=0x40000001 ebx=0x4b4d564b ecx=0x564b4d56 edx=0x0000004d
>>>    0x40000001 0x00: eax=0x01007afb ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x80000000 0x00: eax=0x80000008 ebx=0x68747541 ecx=0x444d4163 edx=0x69746e65
>>>    0x80000001 0x00: eax=0x00000663 ebx=0x00000000 ecx=0x00000121 edx=0x2d93fbff
>>>    0x80000002 0x00: eax=0x554d4551 ebx=0x47435420 ecx=0x55504320 edx=0x72657620
>>>    0x80000003 0x00: eax=0x6e6f6973 ebx=0x352e3220 ecx=0x0000002b edx=0x00000000
>>>    0x80000004 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x80000005 0x00: eax=0x01ff01ff ebx=0x01ff01ff ecx=0x40020140 edx=0x40020140
>>>    0x80000006 0x00: eax=0x00000000 ebx=0x42004200 ecx=0x02008140 edx=0x00808140
>>>    0x80000007 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x80000008 0x00: eax=0x00003028 ebx=0x0100d000 ecx=0x00000000 edx=0x00000000
>>>    0x80860000 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0xc0000000 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>
>>>
>>> cpuid.0ac2b19743
>>>
>>> CPU 0:
>>>    0x00000000 0x00: eax=0x00000016 ebx=0x756e6547 ecx=0x6c65746e edx=0x49656e69
>>>    0x00000001 0x00: eax=0x000806ea ebx=0x00000800 ecx=0xfffab223 edx=0x0f8bfbff
>>>    0x00000002 0x00: eax=0x00000001 ebx=0x00000000 ecx=0x0000004d edx=0x002c307d
>>>    0x00000003 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x00000004 0x00: eax=0x00000121 ebx=0x01c0003f ecx=0x0000003f edx=0x00000001
>>>    0x00000004 0x01: eax=0x00000122 ebx=0x01c0003f ecx=0x0000003f edx=0x00000001
>>>    0x00000004 0x02: eax=0x00000143 ebx=0x03c0003f ecx=0x00000fff edx=0x00000001
>>>    0x00000004 0x03: eax=0x00000163 ebx=0x03c0003f ecx=0x00003fff edx=0x00000006
>>>    0x00000005 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000003 edx=0x00000000
>>>    0x00000006 0x00: eax=0x00000004 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x00000007 0x00: eax=0x00000000 ebx=0x009c4fbb ecx=0x00000004 edx=0xac000400
>>>    0x00000008 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x00000009 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x0000000a 0x00: eax=0x07300402 ebx=0x00000000 ecx=0x00000000 edx=0x00008603
>>>    0x0000000b 0x00: eax=0x00000000 ebx=0x00000001 ecx=0x00000100 edx=0x00000000
>>>    0x0000000b 0x01: eax=0x00000000 ebx=0x00000001 ecx=0x00000201 edx=0x00000000
>>>    0x0000000c 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x0000000d 0x00: eax=0x0000001f ebx=0x00000440 ecx=0x00000440 edx=0x00000000
>>>    0x0000000d 0x01: eax=0x0000000f ebx=0x000003c0 ecx=0x00000000 edx=0x00000000
>>>    0x0000000d 0x02: eax=0x00000100 ebx=0x00000240 ecx=0x00000000 edx=0x00000000
>>>    0x0000000d 0x03: eax=0x00000040 ebx=0x000003c0 ecx=0x00000000 edx=0x00000000
>>>    0x0000000d 0x04: eax=0x00000040 ebx=0x00000400 ecx=0x00000000 edx=0x00000000
>>>    0x0000000e 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x0000000f 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x00000010 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x00000011 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x00000012 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x00000013 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x00000014 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x00000015 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x00000016 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x40000000 0x00: eax=0x40000001 ebx=0x4b4d564b ecx=0x564b4d56 edx=0x0000004d
>>>    0x40000001 0x00: eax=0x01007afb ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x80000000 0x00: eax=0x80000008 ebx=0x756e6547 ecx=0x6c65746e edx=0x49656e69
>>>    0x80000001 0x00: eax=0x000806ea ebx=0x00000000 ecx=0x00000121 edx=0x2c100800
>>>    0x80000002 0x00: eax=0x65746e49 ebx=0x2952286c ecx=0x726f4320 edx=0x4d542865
>>>    0x80000003 0x00: eax=0x37692029 ebx=0x3536382d ecx=0x43205530 edx=0x40205550
>>>    0x80000004 0x00: eax=0x392e3120 ebx=0x7a484730 ecx=0x00000000 edx=0x00000000
>>>    0x80000005 0x00: eax=0x01ff01ff ebx=0x01ff01ff ecx=0x40020140 edx=0x40020140
>>>    0x80000006 0x00: eax=0x00000000 ebx=0x42004200 ecx=0x02008140 edx=0x00808140
>>>    0x80000007 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0x80000008 0x00: eax=0x00003027 ebx=0x0100d000 ecx=0x00000000 edx=0x00000000
>>>    0x80860000 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>    0xc0000000 0x00: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
>>>
>>
>> I started looking at it.
>>
>> Claudio
>>
> 
> I wonder how I missed this, the initialization functions for max_x86_cpu_initfn and kvm_cpu_max_instance_init end up being called in the wrong order.
> 


Just to check whether this is actually the issue we are talking about, Sid et al, could you try this?


diff --git a/target/i386/cpu.c b/target/i386/cpu.c
index c496bfa1c2..810c46281b 100644
--- a/target/i386/cpu.c
+++ b/target/i386/cpu.c
@@ -4252,6 +4252,7 @@ static void max_x86_cpu_initfn(Object *obj)
     object_property_set_str(OBJECT(cpu), "model-id",
                             "QEMU TCG CPU version " QEMU_HW_VERSION,
                             &error_abort);
+    accel_cpu_instance_init(CPU(cpu));
 }

 static const TypeInfo max_x86_cpu_type_info = {
------------------------------------------------------------------------------------------

Does this band-aid happen to cover-up the issue?


I need to think about the proper fix though, any suggestions Paolo, Eduardo?

The pickle here is that we'd need to call the accelerator specific initialization of the x86 accel-cpu only after the x86 cpu subclass initfn,
otherwise the accel-specific cpu initialization code has no chance to see the subclass (max) trying to set ->max_features.

C.