Changes in Kconfig

840b740a35bf969734e0f2e44c21289fdd03079e - mshv: Add conditional VMBus dependency

Tue, 26 May 2026 16:13:04 +0200

mshv: Add conditional VMBus dependencyWhen the VMBus driver is not part of the kernel (CONFIG_HYPERV_VMBUS=n),the MSHV root driver fails to link:ERROR: modpost: "hv_vmbus_exists" [drivers/hv/mshv_root.ko] undefined!Fix this while meeting these requirements:* It must be possible to include the MSHV root driver without the VMBus driver. In such case, the MSHV root driver can be built-in to the kernel image, or it can be built as a separate module.* If both the MSHV root driver and the VMBus driver are present, the MSHV root driver and VMBus driver can both be built-in, or they can both be separate modules. Or the MSHV root driver can be a module while the VMBus driver can be built-in, but the reverse is disallowed. Regardless of the build choices, the VMBus driver must be loaded before the MSHV driver in order for the SynIC to be managed properly (see comments in the MSHV SynIC code).The fix has two parts:* Add a Kconfig entry for MSHV_ROOT to depend on HYPERV_VMBUS if HYPERV_VMBUS is present. The entry disallows MSHV_ROOT being built-in when HYPERV_VMBUS is a module, but without requiring that HYPERV_VMBUS be built.* Add a stub implementation of hv_vmbus_exists() for when the VMBus driver is not present so that the MSHV root driver has no module dependency on VMBus. When the VMBus driver *is* present, the module dependency ensures that the VMBus driver loads first when both are built as modules.Existing code ensures that the VMBus driver loads first if it isbuilt-in. The VMBus driver uses subsys_initcall(), which isinitcall level 4. The MSHV root driver uses module_init(), whichbecomes device_init() when built-in, and device_init() isinitcall level 6.Reported-by: Arnd Bergmann Closes: https://lore.kernel.org/all/20260520074044.923728-1-arnd@kernel.org/Signed-off-by: Michael Kelley Acked-by: Arnd Bergmann Reviewed-by: Jork Loeser Reviewed-by: Hardik Garg Signed-off-by: Wei Liu List of files: /linux/drivers/hv/Kconfig

d33db956c9618e7cb08c2520ce708437914214ec - hv: Select CONFIG_SYSFB only for CONFIG_HYPERV_VMBUS

Thu, 02 Apr 2026 11:09:15 +0200

hv: Select CONFIG_SYSFB only for CONFIG_HYPERV_VMBUSHyperv's sysfb access only exists in the VMBUS support. Thereforeonly select CONFIG_SYSFB for CONFIG_HYPERV_VMBUS. Avoids sysfb codeon systems that don't need it.Signed-off-by: Thomas Zimmermann Fixes: 96959283a58d ("Drivers: hv: Always select CONFIG_SYSFB for Hyper-V guests")Cc: Michael Kelley Cc: Saurabh Sengar Cc: Wei Liu Cc: "K. Y. Srinivasan" Cc: Haiyang Zhang Cc: Dexuan Cui Cc: Long Li Cc: linux-hyperv@vger.kernel.orgCc: # v6.16+Reviewed-by: Saurabh Sengar Link: https://patch.msgid.link/20260402092305.208728-2-tzimmermann@suse.de List of files: /linux/drivers/hv/Kconfig

b9a66cd5ccbb9fade15d0e427e19470d8ad35b75 - mshv: Add support for movable memory regions

Wed, 03 Dec 2025 22:41:09 +0100

mshv: Add support for movable memory regionsIntroduce support for movable memory regions in the Hyper-V root partitiondriver to improve memory management flexibility and enable advanced usecases such as dynamic memory remapping.Mirror the address space between the Linux root partition and guest VMsusing HMM. The root partition owns the memory, while guest VMs act asdevices with page tables managed via hypercalls. MSHV handles VP interceptsby invoking hmm_range_fault() and updating SLAT entries. When memory isreclaimed, HMM invalidates the relevant regions, prompting MSHV to clearSLAT entries; guest VMs will fault again on access.Integrate mmu_interval_notifier for movable regions, implement handlers forHMM faults and memory invalidation, and update memory region mapping logicto support movable regions.While MMU notifiers are commonly used in virtualization drivers, thisimplementation leverages HMM (Heterogeneous Memory Management) for itsspecialized functionality. HMM provides a framework for mirroring,invalidation, and fault handling, reducing boilerplate and improvingmaintainability compared to generic MMU notifiers.Signed-off-by: Stanislav Kinsburskii Reviewed-by: Nuno Das Neves Signed-off-by: Wei Liu List of files: /linux/drivers/hv/Kconfig

7bfe3b8ea6e30437e01fcb8e4f56ef6e4d986d0f - Drivers: hv: Introduce mshv_vtl driver

Thu, 13 Nov 2025 05:41:49 +0100

Drivers: hv: Introduce mshv_vtl driverProvide an interface for Virtual Machine Monitor like OpenVMM and itsuse as OpenHCL paravisor to control VTL0 (Virtual trust Level).Expose devices and support IOCTLs for features like VTL creation,VTL0 memory management, context switch, making hypercalls,mapping VTL0 address space to VTL2 userspace, getting new VMBusmessages and channel events in VTL2 etc.Co-developed-by: Roman Kisel Signed-off-by: Roman Kisel Co-developed-by: Saurabh Sengar Signed-off-by: Saurabh Sengar Reviewed-by: Michael Kelley Signed-off-by: Naman Jain Signed-off-by: Wei Liu List of files: /linux/drivers/hv/Kconfig

e3ec97c3abaf2fb68cc755cae3229288696b9f3d - Drivers: hv: Make CONFIG_HYPERV bool

Tue, 16 Sep 2025 01:46:04 +0200

Drivers: hv: Make CONFIG_HYPERV boolWith CONFIG_HYPERV and CONFIG_HYPERV_VMBUS separated, change CONFIG_HYPERVto bool from tristate. CONFIG_HYPERV now becomes the core Hyper-Vhypervisor support, such as hypercalls, clocks/timers, ConfidentialComputing setup, PCI passthru, etc. that doesn't involve VMBus or VMBusdevices.Signed-off-by: Mukesh Rathor Signed-off-by: Wei Liu List of files: /linux/drivers/hv/Kconfig

94b04355e6397a0a70b69c2571fa5c7d9990b835 - Drivers: hv: Add CONFIG_HYPERV_VMBUS option

Tue, 16 Sep 2025 01:46:03 +0200

Drivers: hv: Add CONFIG_HYPERV_VMBUS optionAt present VMBus driver is hinged off of CONFIG_HYPERV which entailslot of builtin code and encompasses too much. It's not always clearwhat depends on builtin hv code and what depends on VMBus. SettingCONFIG_HYPERV as a module and fudging the Makefile to switch to builtinadds even more confusion. VMBus is an independent module and should haveits own config option. Also, there are scenarios like baremetal dom0/rootwhere support is built in with CONFIG_HYPERV but without VMBus. Lastly,there are more features coming down that use CONFIG_HYPERV and add moredependencies on it.So, create a fine grained HYPERV_VMBUS option and update Kconfigs fordependency on VMBus.Signed-off-by: Mukesh Rathor Acked-by: Bjorn Helgaas # drivers/pciSigned-off-by: Wei Liu List of files: /linux/drivers/hv/Kconfig

4691db0704ac1c266377c99f00288a014fdb7af1 - x86/hyperv: Switch to msi_create_parent_irq_domain()

Fri, 18 Jul 2025 21:57:50 +0200

x86/hyperv: Switch to msi_create_parent_irq_domain()Move away from the legacy MSI domain setup, switch to usemsi_create_parent_irq_domain().While doing the conversion, I noticed that hv_irq_compose_msi_msg() isdoing more than it is supposed to (composing message content). Theinterrupt allocation bits should be moved into hv_msi_domain_alloc().However, I have no hardware to test this change, therefore I leave a TODOnote.Signed-off-by: Nam Cao Acked-by: Thomas Gleixner Tested-by: Nuno Das Neves Reviewed-by: Nuno Das Neves Signed-off-by: Wei Liu List of files: /linux/drivers/hv/Kconfig

c5eebe075e1129748f74e567da7bd8fbe77e485b - mshv: Use common "entry virt" APIs to do work in root before running guest

Thu, 28 Aug 2025 02:01:55 +0200

mshv: Use common "entry virt" APIs to do work in root before running guestUse the kernel's common "entry virt" APIs to handle pending work prior to(re)entering guest mode, now that the virt APIs don't have a superfluousdependency on KVM.No functional change intended.Signed-off-by: Sean Christopherson Tested-by: Nuno Das Neves Reviewed-by: Nuno Das Neves Signed-off-by: Wei Liu List of files: /linux/drivers/hv/Kconfig

7b89a44b2e8c7ba548e3ad5d5155a17279625335 - Drivers: hv: Select CONFIG_SYSFB only if EFI is enabled

Sat, 14 Jun 2025 01:00:59 +0200

Drivers: hv: Select CONFIG_SYSFB only if EFI is enabledCommit 96959283a58d ("Drivers: hv: Always select CONFIG_SYSFBfor Hyper-V guests") selects CONFIG_SYSFB for Hyper-V guestsso that screen_info is available to the VMBus driver to getthe location of the framebuffer in Generation 2 VMs. However,if CONFIG_HYPERV is enabled but CONFIG_EFI is not, a kernellink error results in ARM64 builds because screen_info isprovided by the EFI firmware interface. While configuringan ARM64 Hyper-V guest without EFI isn't useful since EFI isrequired to boot, the configuration is still possible andthe link error should be prevented.Fix this by making the selection of CONFIG_SYSFB conditionalon CONFIG_EFI being defined. For Generation 1 VMs on x86/x64,which don't use EFI, the additional condition is OK becausesuch VMs get the framebuffer information via a mechanismthat doesn't use screen_info.Fixes: 96959283a58d ("Drivers: hv: Always select CONFIG_SYSFB for Hyper-V guests")Reported-by: Arnd Bergmann Closes: https://lore.kernel.org/linux-hyperv/20250610091810.2638058-1-arnd@kernel.org/Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202506080820.1wmkQufc-lkp@intel.com/Signed-off-by: Michael Kelley Link: https://lore.kernel.org/stable/20250613230059.380483-1-mhklinux%40outlook.comReviewed-by: Roman Kisel Link: https://lore.kernel.org/r/20250613230059.380483-1-mhklinux@outlook.comSigned-off-by: Wei Liu Message-ID: <20250613230059.380483-1-mhklinux@outlook.com> List of files: /linux/drivers/hv/Kconfig

96959283a58d91ae20d025546f00e16f0a555208 - Drivers: hv: Always select CONFIG_SYSFB for Hyper-V guests

Tue, 20 May 2025 06:01:43 +0200

Drivers: hv: Always select CONFIG_SYSFB for Hyper-V guestsThe Hyper-V host provides guest VMs with a range of MMIO addressesthat guest VMBus drivers can use. The VMBus driver in Linux managesthat MMIO space, and allocates portions to drivers upon request. Aspart of managing that MMIO space in a Generation 2 VM, the VMBusdriver must reserve the portion of the MMIO space that Hyper-V hasdesignated for the synthetic frame buffer, and not allocate thisspace to VMBus drivers other than graphics framebuffer drivers. Thesynthetic frame buffer MMIO area is described by the screen_info datastructure that is passed to the Linux kernel at boot time, so theVMBus driver must access screen_info for Generation 2 VMs. (InGeneration 1 VMs, the framebuffer MMIO space is communicated tothe guest via a PCI pseudo-device, and access to screen_info isnot needed.)In commit a07b50d80ab6 ("hyperv: avoid dependency on screen_info")the VMBus driver's access to screen_info is restricted to whenCONFIG_SYSFB is enabled. CONFIG_SYSFB is typically enabled in kernelsbuilt for Hyper-V by virtue of having at least one of CONFIG_FB_EFI,CONFIG_FB_VESA, or CONFIG_SYSFB_SIMPLEFB enabled, so the restrictiondoesn't usually affect anything. But it's valid to have none of theseenabled, in which case CONFIG_SYSFB is not enabled, and the VMBus driveris unable to properly reserve the framebuffer MMIO space for graphicsframebuffer drivers. The framebuffer MMIO space may be assigned tosome other VMBus driver, with undefined results. As an example, ifa VM is using a PCI pass-thru NVMe controller to host the OS disk,the PCI NVMe controller is probed before any graphics devices, and theNVMe controller is assigned a portion of the framebuffer MMIO space.Hyper-V reports an error to Linux during the probe, and the OS diskfails to get setup. Then Linux fails to boot in the VM.Fix this by having CONFIG_HYPERV always select SYSFB. Then theVMBus driver in a Gen 2 VM can always reserve the MMIO space for thegraphics framebuffer driver, and prevent the undefined behavior. Butdon't select SYSFB when building for HYPERV_VTL_MODE as VTLs otherthan VTL 0 don't have a framebuffer and aren't subject to the issue.Adding SYSFB in such cases is harmless, but would increase the imagesize for no purpose.Fixes: a07b50d80ab6 ("hyperv: avoid dependency on screen_info")Signed-off-by: Michael Kelley Reviewed-by: Saurabh Sengar Link: https://lore.kernel.org/stable/20250520040143.6964-1-mhklinux%40outlook.comLink: https://lore.kernel.org/r/20250520040143.6964-1-mhklinux@outlook.comSigned-off-by: Wei Liu Message-ID: <20250520040143.6964-1-mhklinux@outlook.com> List of files: /linux/drivers/hv/Kconfig

f41ceff1748612b9e1538becc13dc2f1617d9d14 - Drivers: hv: Enable VTL mode for arm64

Mon, 28 Apr 2025 23:07:34 +0200

Drivers: hv: Enable VTL mode for arm64Kconfig dependencies for arm64 guests on Hyper-V require that beACPI enabled, and limit VTL mode to x86/x64. To enable VTL modeon arm64 as well, update the dependencies. Since VTL mode requiresDeviceTree instead of ACPI, don’t require arm64 guests on Hyper-Vto have ACPI unconditionally.Signed-off-by: Roman Kisel Reviewed-by: Michael Kelley Link: https://lore.kernel.org/r/20250428210742.435282-4-romank@linux.microsoft.comSigned-off-by: Wei Liu Message-ID: <20250428210742.435282-4-romank@linux.microsoft.com> List of files: /linux/drivers/hv/Kconfig

621191d709b14882270dfd8ea5d7d6cdfebe2c35 - Drivers: hv: Introduce mshv_root module to expose /dev/mshv to VMMs

Fri, 14 Mar 2025 20:28:56 +0100

Drivers: hv: Introduce mshv_root module to expose /dev/mshv to VMMsProvide a set of IOCTLs for creating and managing child partitions whenrunning as root partition on Hyper-V. The new driver is enabled viaCONFIG_MSHV_ROOT.A brief overview of the interface:MSHV_CREATE_PARTITION is the entry point, returning a file descriptorrepresenting a child partition. IOCTLs on this fd can be used to mapmemory, create VPs, etc.Creating a VP returns another file descriptor representing that VP whichin turn has another set of corresponding IOCTLs for running the VP,getting/setting state, etc.MSHV_ROOT_HVCALL is a generic "passthrough" hypercall IOCTL which can beused for a number of partition or VP hypercalls. This is for hypercallsthat do not affect any state in the kernel driver, such as getting andsetting VP registers and partition properties, translating addresses,etc. It is "passthrough" because the binary input and output for thehypercall is only interpreted by the VMM - the kernel driver doesnothing but insert the VP and partition id where necessary (which arealways in the same place), and execute the hypercall.Co-developed-by: Anirudh Rayabharam Signed-off-by: Anirudh Rayabharam Co-developed-by: Jinank Jain Signed-off-by: Jinank Jain Co-developed-by: Mukesh Rathor Signed-off-by: Mukesh Rathor Co-developed-by: Muminul Islam Signed-off-by: Muminul Islam Co-developed-by: Praveen K Paladugu Signed-off-by: Praveen K Paladugu Co-developed-by: Stanislav Kinsburskii Signed-off-by: Stanislav Kinsburskii Co-developed-by: Wei Liu Signed-off-by: Nuno Das Neves Reviewed-by: Roman Kisel Link: https://lore.kernel.org/r/1741980536-3865-11-git-send-email-nunodasneves@linux.microsoft.comSigned-off-by: Wei Liu Message-ID: <1741980536-3865-11-git-send-email-nunodasneves@linux.microsoft.com> List of files: /linux/drivers/hv/Kconfig

461fbbd036b11d755b50e2ef7c165859a0f908d5 - hyperv: Add CONFIG_MSHV_ROOT to gate root partition support

Fri, 21 Feb 2025 20:56:35 +0100

hyperv: Add CONFIG_MSHV_ROOT to gate root partition supportCONFIG_MSHV_ROOT allows kernels built to run as a normal Hyper-V guestto exclude the root partition code, which is expected to growsignificantly over time.This option is a tristate so future driver code can be built as a(m)odule, allowing faster development iteration cycles.If CONFIG_MSHV_ROOT is disabled, don't compile hv_proc.c, and stubhv_root_partition() to return false unconditionally. This allows thecompiler to optimize away root partition code blocks since they willbe disabled at compile time.In the case of booting as root partition *without* CONFIG_MSHV_ROOTenabled, print a critical error (the kernel will likely crash).Signed-off-by: Nuno Das Neves Reviewed-by: Easwar Hariharan Reviewed-by: Michael Kelley Link: https://lore.kernel.org/r/1740167795-13296-4-git-send-email-nunodasneves@linux.microsoft.comSigned-off-by: Wei Liu Message-ID: <1740167795-13296-4-git-send-email-nunodasneves@linux.microsoft.com> List of files: /linux/drivers/hv/Kconfig

2b4b90e053a29057fb05ba81acce26bddce8d404 - x86/hyperv: Use per cpu initial stack for vtl context

Sun, 03 Mar 2024 09:01:36 +0100

x86/hyperv: Use per cpu initial stack for vtl contextCurrently, the secondary CPUs in Hyper-V VTL context lack support forparallel startup. Therefore, relying on the single initial_stack fetchedfrom the current task structure suffices for all vCPUs.However, common initial_stack risks stack corruption when parallel startupis enabled. In order to facilitate parallel startup, use the initial_stackfrom the per CPU idle thread instead of the current task.Fixes: 3be1bc2fe9d2 ("x86/hyperv: VTL support for Hyper-V")Signed-off-by: Saurabh Sengar Reviewed-by: Michael Kelley Link: https://lore.kernel.org/r/1709452896-13342-1-git-send-email-ssengar@linux.microsoft.comSigned-off-by: Wei Liu Message-ID: <1709452896-13342-1-git-send-email-ssengar@linux.microsoft.com> List of files: /linux/drivers/hv/Kconfig

d01b9a9f2d0131e1e249177a70e6b80d146d16d2 - Drivers: hv: Kconfig: Add HYPERV_VTL_MODE

Tue, 11 Apr 2023 07:55:31 +0200

Drivers: hv: Kconfig: Add HYPERV_VTL_MODEAdd HYPERV_VTL_MODE Kconfig flag for VTL mode.Signed-off-by: Saurabh Sengar Reviewed-by: Michael Kelley Link: https://lore.kernel.org/r/1681192532-15460-5-git-send-email-ssengar@linux.microsoft.comSigned-off-by: Wei Liu List of files: /linux/drivers/hv/Kconfig

25727aaed6514b88f98a18862c6f2d65a0b0ec3b - hv_netvsc: Remove second mapping of send and recv buffers

Sun, 26 Mar 2023 15:52:05 +0200

hv_netvsc: Remove second mapping of send and recv buffersWith changes to how Hyper-V guest VMs flip memory between private(encrypted) and shared (decrypted), creating a second kernel virtualmapping for shared memory is no longer necessary. Everything neededfor the transition to shared is handled by set_memory_decrypted().As such, remove the code to create and manage the secondmapping for the pre-allocated send and recv buffers. This mappingis the last user of hv_map_memory()/hv_unmap_memory(), so deletethese functions as well. Finally, hv_map_memory() is the lastuser of vmap_pfn() in Hyper-V guest code, so remove the Kconfigselection of VMAP_PFN.Signed-off-by: Michael Kelley Reviewed-by: Tianyu Lan Link: https://lore.kernel.org/r/1679838727-87310-11-git-send-email-mikelley@microsoft.comSigned-off-by: Wei Liu List of files: /linux/drivers/hv/Kconfig

f83705a51275ed29117d46e1d68e8b16dcb40507 - Driver: VMBus: Add Devicetree support

Mon, 20 Mar 2023 08:47:40 +0100

Driver: VMBus: Add Devicetree supportUpdate the driver to support Devicetree boot as well along with ACPI.At present the Devicetree parsing only provides the mmio region infoand is not the exact copy of ACPI parsing. This is sufficient to caterall the current Devicetree usecases for VMBus.Currently Devicetree is supported only for x86 systems.Signed-off-by: Saurabh Sengar Reviewed-by: Michael Kelley Link: https://lore.kernel.org/r/1679298460-11855-6-git-send-email-ssengar@linux.microsoft.comSigned-off-by: Wei Liu List of files: /linux/drivers/hv/Kconfig

1dc2f2b81a6a9895da59f3915760f6c0c3074492 - hv: utils: add PTP_1588_CLOCK to Kconfig to fix build

Fri, 26 Nov 2021 03:33:16 +0100

hv: utils: add PTP_1588_CLOCK to Kconfig to fix buildThe hyperv utilities use PTP clock interfaces and should depend aa kconfig symbol such that they will be built as a loadable module orbuiltin so that linker errors do not happen.Prevents these build errors:ld: drivers/hv/hv_util.o: in function `hv_timesync_deinit':hv_util.c:(.text+0x37d): undefined reference to `ptp_clock_unregister'ld: drivers/hv/hv_util.o: in function `hv_timesync_init':hv_util.c:(.text+0x738): undefined reference to `ptp_clock_register'Fixes: 3716a49a81ba ("hv_utils: implement Hyper-V PTP source")Signed-off-by: Randy Dunlap Reported-by: kernel test robot Cc: Arnd Bergmann Cc: "K. Y. Srinivasan" Cc: Haiyang Zhang Cc: Stephen Hemminger Cc: Wei Liu Cc: Dexuan Cui Cc: linux-hyperv@vger.kernel.orgCc: Greg Kroah-Hartman Reviewed-by: Michael Kelley Link: https://lore.kernel.org/r/20211126023316.25184-1-rdunlap@infradead.orgSigned-off-by: Wei Liu List of files: /linux/drivers/hv/Kconfig

9a8797722e4239242d0cb4cc4baa805df6ac979e - Drivers: hv: vmbus: Initialize VMbus ring buffer for Isolation VM

Mon, 25 Oct 2021 14:21:14 +0200

Drivers: hv: vmbus: Initialize VMbus ring buffer for Isolation VMVMbus ring buffer are shared with host and it's need tobe accessed via extra address space of Isolation VM withAMD SNP support. This patch is to map the ring bufferaddress in extra address space via vmap_pfn(). Hyperv setmemory host visibility hvcall smears data in the ring bufferand so reset the ring buffer memory to zero after mapping.Reviewed-by: Michael Kelley Signed-off-by: Tianyu Lan Link: https://lore.kernel.org/r/20211025122116.264793-10-ltykernel@gmail.comSigned-off-by: Wei Liu List of files: /linux/drivers/hv/Kconfig

7aff79e297ee1aa0126924921fd87a4ae59d2467 - Drivers: hv: Enable Hyper-V code to be built on ARM64

Wed, 04 Aug 2021 17:52:39 +0200

Drivers: hv: Enable Hyper-V code to be built on ARM64Update drivers/hv/Kconfig so CONFIG_HYPERV can be selected onARM64, causing the Hyper-V specific code to be built. Exclude theHyper-V enlightened clocks/timers code from being built for ARM64.Signed-off-by: Michael Kelley Reviewed-by: Boqun Feng Acked-by: Marc Zyngier Acked-by: Mark Rutland Acked-by: Catalin Marinas Link: https://lore.kernel.org/r/1628092359-61351-6-git-send-email-mikelley@microsoft.comSigned-off-by: Wei Liu List of files: /linux/drivers/hv/Kconfig