Re: [PATCH 09/12] docs: ioctl-number: update virtualization entries

From: Randy Dunlap

Date: Mon Nov 16 2026 - 17:21:33 EST




On 11/16/26 6:26 AM, Sean Christopherson wrote:
> On Sat, Nov 14, 2026, Randy Dunlap wrote:
>> Update the list of files where virtualization ioctls are defined.
>>
>> Signed-off-by: Randy Dunlap <rdunlap@xxxxxxxxxxxxx>
>> ---
>> [Since having multiple maintainers trying to merge these would
>> cause some conflicts, it would be more manageable for Jon to
>> take all of these through the docs-next tree.]
>>
>> Cc: Jonathan Corbet <corbet@xxxxxxx>
>> Cc: Shuah Khan <skhan@xxxxxxxxxxxxxxxxxxx>
>> Cc: linux-doc@xxxxxxxxxxxxxxx
>> Cc: Juergen Gross <jgross@xxxxxxxx>
>> Cc: Stefano Stabellini <sstabellini@xxxxxxxxxx>
>> Cc: xen-devel@xxxxxxxxxxxxxxxxxxxx
>> Cc: Hans de Goede <hansg@xxxxxxxxxx>
>> Cc: Arnd Bergmann <arnd@xxxxxxxx>
>> Cc: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
>> Cc: "Michael S. Tsirkin" <mst@xxxxxxxxxx>
>> Cc: Jason Wang <jasowangio@xxxxxxxxx>
>> Cc: Eugenio Pérez <eperezma@xxxxxxxxxx>
>> Cc: Sean Christopherson <seanjc@xxxxxxxxxx>
>> Cc: Paolo Bonzini <pbonzini@xxxxxxxxxx>
>> Cc: kvm@xxxxxxxxxxxxxxx
>> Cc: virtualization@xxxxxxxxxxxxxxx
>>
>> Documentation/userspace-api/ioctl/ioctl-number.rst | 6 ++++++
>> 1 file changed, 6 insertions(+)
>>
>> --- linux-next-20261113.orig/Documentation/userspace-api/ioctl/ioctl-number.rst
>> +++ linux-next-20261113/Documentation/userspace-api/ioctl/ioctl-number.rst
>> @@ -200,6 +200,7 @@ Code Seq# Include File
>> 'M' 01-03 drivers/scsi/megaraid/megaraid_sas.h
>> 'M' 00-03 drivers/video/fbdev/mb862xx/mb862xxfb.h
>> 'M' 00-0F drivers/video/fsl-diu-fb.h conflict!
>> +'M' 01-03 xen/interface/xen-mca.h conflict!
>> 'N' 00-1F drivers/usb/scanner.h
>> 'N' 40-7F uapi/linux/nvme_ioctl.h
>> 'N' 80-83 uapi/linux/ntsync.h and uapi/linux/nvme_ioctl.h conflict!
>> @@ -239,6 +240,7 @@ Code Seq# Include File
>> 'U' 64 drivers/usb/misc/usbtest.c conflict!
>> 'V' all linux/vt.h conflict!
>> 'V' all linux/videodev2.h conflict!
>> +'V' 00-1F linux/vboxguest.h conflict!
>> 'V' C0 linux/ivtvfb.h conflict!
>> 'V' C0 linux/ivtv.h conflict!
>> 'V' C0 media/si4713.h conflict!
>> @@ -357,6 +359,7 @@ Code Seq# Include File
>> '|' 80-9F samples/ Any sample and example drivers
>> 0x80 00-1F linux/fb.h
>> 0x81 00-1F linux/vduse.h
>> + drivers/vdpa/vdpa_user/vduse_dev.c
>> 0x89 00-07 asm-generic/sockios.h
>> 0x89 06-07 linux/sockios.h
>> 0x89 0B-DF linux/sockios.h
>> @@ -398,8 +401,11 @@ Code Seq# Include File
>> <mailto:kvm@xxxxxxxxxxxxxxx>
>> 0xAE 40-FF linux/kvm.h Kernel-based Virtual Machine
>> <mailto:kvm@xxxxxxxxxxxxxxx>
>> +0xAE C6 arch/x86/kvm/x86.c X86 KVM
>
> I'd rather omit this entry. 0xAE C6 is covered by the entry above, and the horror
> in arch/x86/kvm/x86.c is purely a 32-bit compatibility alias for an ioctl that is
> defined in kvm.h (the structure payload defines an pointer instead of a __u64).

OK.

> If it's a sticking point, e.g. this will trip automated tooling over and over, then
> I vote to move the relevant definitions to uapi headers, e.g.

I don't know of anything like that.

--
~Randy