x86info: dump kvm cpuid's
Subsystems (2)
Michael S. Tsirkin
maintainerThe following makes 'x86info -r' dump kvm cpu ids
(signature+features) when running in a vm.
On the guest we see the signature and the features:
eax in: 0x40000000, eax = 00000000 ebx = 4b4d564b ecx = 564b4d56 edx = 0000004d eax in: 0x40000001, eax = 0100007b ebx = 00000000 ecx = 00000000 edx = 00000000
On the host it just adds a couple of zero lines:
eax in: 0x40000000, eax = 00000000 ebx = 00000000 ecx = 00000000 edx = 00000000 eax in: 0x40000001, eax = 00000000 ebx = 00000000 ecx = 00000000 edx = 00000000
---
Dave - not sure whether there's a mailing list for x86info. The patch is on top of the master branch ingit://git.codemonkey.org.uk/x86info.git
Thanks!
x86info.c +1 -0
@@ -44,6 +44,7 @@ static void display_detailed_info(struct cpudata *cpu)if (cpu->maxei2 >=0xC0000000 )dump_raw_cpuid (cpu->number,0xC0000000 , cpu->maxei2); +dump_raw_cpuid (cpu->number,0x40000000 ,0x40000001 ); }if (show_cacheinfo) {
Quoted 11 lines · Michael S. Tsirkin
The following makes 'x86info -r' dump kvm cpu ids
(signature+features) when running in a vm.On the guest we see the signature and the features:
eax in:0x40000000, eax = 00000000 ebx =4b4d564becx =564b4d56edx =0000004deax in:0x40000001, eax =0100007bebx = 00000000 ecx = 00000000 edx = 00000000On the host it just adds a couple of zero lines:
eax in:0x40000000, eax = 00000000 ebx = 00000000 ecx = 00000000 edx = 00000000 eax in:0x40000001, eax = 00000000 ebx = 00000000 ecx = 00000000 edx = 00000000
This is too KVM specific. Other hypervisors may use more cpuid leafs. As far as I see Hyper-V uses 5 and use cpuid.0x40000000.eax as max cpuid leaf available. Haven't checked Xen or VMWare.
Quoted 15 lines
Signed-off-by Michael S. Tsirkin <[email protected]>---
Dave - not sure whether there's a mailing list for x86info. The patch is on top of the master branch in
git://git.codemonkey.org.uk/x86info.gitThanks!
x86info.c +0
-0
x86info.c static void display_detailed_info(struct cpudata *cpu) · line 50
Quoted 7 lines
x86info.c +1
-0@@ -44,6 +44,7 @@ static void display_detailed_info(struct cpudata *cpu)if (cpu->maxei2 >=0xC0000000 )dump_raw_cpuid (cpu->number,0xC0000000 , cpu->maxei2); +dump_raw_cpuid (cpu->number,0x40000000 ,0x40000001 ); }if (show_cacheinfo) {
--
Gleb.
Michael S. Tsirkin
maintainerQuoted 8 lines · Gleb Natapov
Quoted 7 lines · Michael S. Tsirkin
The following makes 'x86info -r' dump kvm cpu ids
(signature+features) when running in a vm.On the guest we see the signature and the features:
eax in:0x40000000, eax = 00000000 ebx =4b4d564becx =564b4d56edx =0000004deax in:0x40000001, eax =0100007bebx = 00000000 ecx = 00000000 edx = 00000000
Quoted 5 lines
Quoted 4 lines
On the host it just adds a couple of zero lines:
eax in:0x40000000, eax = 00000000 ebx = 00000000 ecx = 00000000 edx = 00000000 eax in:0x40000001, eax = 00000000 ebx = 00000000 ecx = 00000000 edx = 00000000This is too KVM specific.
That's what I have. I scratch my own itch.
Quoted 1 line
Other hypervisors may use more cpuid leafs.
But not less so no harm's done.
Quoted 2 lines
As far as I see Hyper-V uses 5 and use cpuid.0x40000000.eax as max cpuid leaf available. Haven't checked Xen or VMWare.
I don't think guessing at the CPU behaviour from linux source is the right thing to do.
I Cc'd some addresses found in MAINTAINERS in the linux kernel. This will give more people the opportunity to ask for their stuff to be added, if they care.
Quoted 25 lines
Quoted 15 lines
Signed-off-by Michael S. Tsirkin <[email protected]>---
Dave - not sure whether there's a mailing list for x86info. The patch is on top of the master branch in
git://git.codemonkey.org.uk/x86info.gitThanks!
x86info.c +0
-0x86info.c static void display_detailed_info(struct cpudata *cpu) · line 50
Quoted 7 lines
x86info.c +1
-0@@ -44,6 +44,7 @@ static void display_detailed_info(struct cpudata *cpu)if (cpu->maxei2 >=0xC0000000 )dump_raw_cpuid (cpu->number,0xC0000000 , cpu->maxei2); +dump_raw_cpuid (cpu->number,0x40000000 ,0x40000001 ); }if (show_cacheinfo) {--
Gleb.
Quoted 19 lines · Michael S. Tsirkin
Quoted 8 lines · Gleb Natapov
Quoted 7 lines · Michael S. Tsirkin
The following makes 'x86info -r' dump kvm cpu ids
(signature+features) when running in a vm.On the guest we see the signature and the features:
eax in:0x40000000, eax = 00000000 ebx =4b4d564becx =564b4d56edx =0000004deax in:0x40000001, eax =0100007bebx = 00000000 ecx = 00000000 edx = 00000000Quoted 5 lines
Quoted 4 lines
On the host it just adds a couple of zero lines:
eax in:0x40000000, eax = 00000000 ebx = 00000000 ecx = 00000000 edx = 00000000 eax in:0x40000001, eax = 00000000 ebx = 00000000 ecx = 00000000 edx = 00000000This is too KVM specific.
That's what I have. I scratch my own itch.
Quoted 1 line
Other hypervisors may use more cpuid leafs.
Quoted 8 lines
But not less so no harm's done.
Quoted 2 lines
As far as I see Hyper-V uses 5 and use cpuid.0x40000000.eax as max cpuid leaf available. Haven't checked Xen or VMWare.
I don't think guessing at the CPU behaviour from linux source is the right thing to do.
That is guessing from Hyper-V specification. The best kind of guess.
http://msdn.microsoft.com/en-us/library/windows/hardware/ff542700%28v=vs.85%29.aspx
Quoted 29 lines
I Cc'd some addresses found in
MAINTAINERSin the linux kernel. This will give more people the opportunity to ask for their stuff to be added, if they care.Quoted 25 lines
Quoted 15 lines
Signed-off-by Michael S. Tsirkin <[email protected]>---
Dave - not sure whether there's a mailing list for x86info. The patch is on top of the master branch in
git://git.codemonkey.org.uk/x86info.gitThanks!
x86info.c +0
-0x86info.c static void display_detailed_info(struct cpudata *cpu) · line 50
Quoted 7 lines
x86info.c +1
-0@@ -44,6 +44,7 @@ static void display_detailed_info(struct cpudata *cpu)if (cpu->maxei2 >=0xC0000000 )dump_raw_cpuid (cpu->number,0xC0000000 , cpu->maxei2); +dump_raw_cpuid (cpu->number,0x40000000 ,0x40000001 ); }if (show_cacheinfo) {--
Gleb.
--
Gleb.
Quoted 13 lines · Michael S. Tsirkin
Quoted 12 lines · Gleb Natapov
Quoted 11 lines · Michael S. Tsirkin
The following makes 'x86info -r' dump kvm cpu ids
(signature+features) when running in a vm.On the guest we see the signature and the features:
eax in:0x40000000, eax = 00000000 ebx =4b4d564becx =564b4d56edx =0000004deax in:0x40000001, eax =0100007bebx = 00000000 ecx = 00000000 edx = 00000000On the host it just adds a couple of zero lines:
eax in:0x40000000, eax = 00000000 ebx = 00000000 ecx = 00000000 edx = 00000000 eax in:0x40000001, eax = 00000000 ebx = 00000000 ecx = 00000000 edx = 00000000
Quoted 10 lines
Quoted 1 line
This is too KVM specific.
That's what I have. I scratch my own itch.
Quoted 1 line
Other hypervisors may use more cpuid leafs.
But not less so no harm's done.
Quoted 2 lines
As far as I see Hyper-V uses 5 and use cpuid.0x40000000.eax as max cpuid leaf available. Haven't checked Xen or VMWare.
Xen does the same, documentation in the Xen public interfaces header:
http://xenbits.xen.org/docs/unstable/hypercall/include,public,arch-x86,cpuid.h.html.
If compat mode for another h/v is enabled then those leaves will appear at 0x40000000 and Xen's will be bumped up, so a fully Xen aware set of drivers (or detection routine, etc) should check at 0x100 intervals until 0x40010000 for the appropriate signatures (I realise that the docs are somewhat lacking in this regard, I should cook up a patch).
Ian.
Michael S. Tsirkin
maintainerQuoted 20 lines · Ian Campbell
Quoted 19 lines · Michael S. Tsirkin
Quoted 8 lines · Gleb Natapov
Quoted 7 lines · Michael S. Tsirkin
The following makes 'x86info -r' dump kvm cpu ids
(signature+features) when running in a vm.On the guest we see the signature and the features:
eax in:0x40000000, eax = 00000000 ebx =4b4d564becx =564b4d56edx =0000004deax in:0x40000001, eax =0100007bebx = 00000000 ecx = 00000000 edx = 00000000Quoted 5 lines
Quoted 4 lines
On the host it just adds a couple of zero lines:
eax in:0x40000000, eax = 00000000 ebx = 00000000 ecx = 00000000 edx = 00000000 eax in:0x40000001, eax = 00000000 ebx = 00000000 ecx = 00000000 edx = 00000000This is too KVM specific.
That's what I have. I scratch my own itch.
Quoted 1 line
Other hypervisors may use more cpuid leafs.
Quoted 7 lines
Quoted 4 lines
But not less so no harm's done.
Quoted 2 lines
As far as I see Hyper-V uses 5 and use cpuid.0x40000000.eax as max cpuid leaf available. Haven't checked Xen or VMWare.
Xen does the same, documentation in the Xen public interfaces header:
http://xenbits.xen.org/docs/unstable/hypercall/include,public,arch-x86,cpuid.h.html.
So ack to my patch?
Quoted 7 lines
If compat mode for another h/v is enabled then those leaves will appear at
0x40000000and Xen's will be bumped up, so a fully Xen aware set of drivers (or detection routine, etc) should check at0x100intervals until0x40010000for the appropriate signatures (I realise that the docs are somewhat lacking in this regard, I should cook up a patch).Ian.
How does guest know that the data at 0x40000100 makes sense?
Signature · 1 line
MST
Quoted 23 lines · Michael S. Tsirkin
Quoted 22 lines · Ian Campbell
Quoted 21 lines · Michael S. Tsirkin
Quoted 8 lines · Gleb Natapov
Quoted 7 lines · Michael S. Tsirkin
The following makes 'x86info -r' dump kvm cpu ids
(signature+features) when running in a vm.On the guest we see the signature and the features:
eax in:0x40000000, eax = 00000000 ebx =4b4d564becx =564b4d56edx =0000004deax in:0x40000001, eax =0100007bebx = 00000000 ecx = 00000000 edx = 00000000Quoted 5 lines
Quoted 4 lines
On the host it just adds a couple of zero lines:
eax in:0x40000000, eax = 00000000 ebx = 00000000 ecx = 00000000 edx = 00000000 eax in:0x40000001, eax = 00000000 ebx = 00000000 ecx = 00000000 edx = 00000000This is too KVM specific.
That's what I have. I scratch my own itch.
Quoted 1 line
Other hypervisors may use more cpuid leafs.
But not less so no harm's done.
Quoted 7 lines
Quoted 5 lines
Quoted 2 lines
Quoted 2 lines
As far as I see Hyper-V uses 5 and use cpuid.0x40000000.eax as max cpuid leaf available. Haven't checked Xen or VMWare.
Xen does the same, documentation in the Xen public interfaces header:
http://xenbits.xen.org/docs/unstable/hypercall/include,public,arch-x86,cpuid.h.html.So ack to my patch?
I didn't see the patch, where should I be looking?
Quoted 9 lines
Quoted 7 lines
If compat mode for another h/v is enabled then those leaves will appear at
0x40000000and Xen's will be bumped up, so a fully Xen aware set of drivers (or detection routine, etc) should check at0x100intervals until0x40010000for the appropriate signatures (I realise that the docs are somewhat lacking in this regard, I should cook up a patch).Ian.
How does guest know that the data at
0x40000100makes sense?
http://xenbits.xen.org/docs/unstable/hypercall/include,public,arch-x86,cpuid.h.html
EBX, ECX and EDX contain a signature "XenVMMXenVMM". I'm fairly certain that hyperv has it's own magic number here.
Ian.
Quoted 23 lines · Michael S. Tsirkin
Quoted 22 lines · Ian Campbell
Quoted 21 lines · Michael S. Tsirkin
Quoted 8 lines · Gleb Natapov
Quoted 7 lines · Michael S. Tsirkin
The following makes 'x86info -r' dump kvm cpu ids
(signature+features) when running in a vm.On the guest we see the signature and the features:
eax in:0x40000000, eax = 00000000 ebx =4b4d564becx =564b4d56edx =0000004deax in:0x40000001, eax =0100007bebx = 00000000 ecx = 00000000 edx = 00000000Quoted 5 lines
Quoted 4 lines
On the host it just adds a couple of zero lines:
eax in:0x40000000, eax = 00000000 ebx = 00000000 ecx = 00000000 edx = 00000000 eax in:0x40000001, eax = 00000000 ebx = 00000000 ecx = 00000000 edx = 00000000This is too KVM specific.
That's what I have. I scratch my own itch.
Quoted 1 line
Other hypervisors may use more cpuid leafs.
But not less so no harm's done.
Quoted 7 lines
Quoted 5 lines
Quoted 2 lines
Quoted 2 lines
As far as I see Hyper-V uses 5 and use cpuid.0x40000000.eax as max cpuid leaf available. Haven't checked Xen or VMWare.
Xen does the same, documentation in the Xen public interfaces header:
http://xenbits.xen.org/docs/unstable/hypercall/include,public,arch-x86,cpuid.h.html.So ack to my patch?
I didn't see the patch, where should I be looking?
Quoted 9 lines
Quoted 7 lines
If compat mode for another h/v is enabled then those leaves will appear at
0x40000000and Xen's will be bumped up, so a fully Xen aware set of drivers (or detection routine, etc) should check at0x100intervals until0x40010000for the appropriate signatures (I realise that the docs are somewhat lacking in this regard, I should cook up a patch).Ian.
How does guest know that the data at
0x40000100makes sense?
http://xenbits.xen.org/docs/unstable/hypercall/include,public,arch-x86,cpuid.h.html
EBX, ECX and EDX contain a signature "XenVMMXenVMM". I'm fairly certain that hyperv has it's own magic number here.
Ian.
Michael S. Tsirkin
maintainerQuoted 20 lines · Ian Campbell
Quoted 19 lines · Michael S. Tsirkin
Quoted 8 lines · Gleb Natapov
Quoted 7 lines · Michael S. Tsirkin
The following makes 'x86info -r' dump kvm cpu ids
(signature+features) when running in a vm.On the guest we see the signature and the features:
eax in:0x40000000, eax = 00000000 ebx =4b4d564becx =564b4d56edx =0000004deax in:0x40000001, eax =0100007bebx = 00000000 ecx = 00000000 edx = 00000000Quoted 5 lines
Quoted 4 lines
On the host it just adds a couple of zero lines:
eax in:0x40000000, eax = 00000000 ebx = 00000000 ecx = 00000000 edx = 00000000 eax in:0x40000001, eax = 00000000 ebx = 00000000 ecx = 00000000 edx = 00000000This is too KVM specific.
That's what I have. I scratch my own itch.
Quoted 1 line
Other hypervisors may use more cpuid leafs.
Quoted 7 lines
Quoted 4 lines
But not less so no harm's done.
Quoted 2 lines
As far as I see Hyper-V uses 5 and use cpuid.0x40000000.eax as max cpuid leaf available. Haven't checked Xen or VMWare.
Xen does the same, documentation in the Xen public interfaces header:
http://xenbits.xen.org/docs/unstable/hypercall/include,public,arch-x86,cpuid.h.html.
So ack to my patch?
Quoted 7 lines
If compat mode for another h/v is enabled then those leaves will appear at
0x40000000and Xen's will be bumped up, so a fully Xen aware set of drivers (or detection routine, etc) should check at0x100intervals until0x40010000for the appropriate signatures (I realise that the docs are somewhat lacking in this regard, I should cook up a patch).Ian.
How does guest know that the data at 0x40000100 makes sense?
Signature · 1 line
MST