linux-next: duplicate patches in the kvm-x86 tree

linux-next: duplicate patches in the kvm-x86 tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 598 bytes --]

Hi all,

The following commits are also in Linus Torvalds' tree as different
commits (but the same patches):

  afb2c7e27a7f ("KVM: selftests: Force load all supported XSAVE state in state test")
  d7b8762ec4a3 ("KVM: selftests: Load XSAVE state into untouched vCPU during state test")
  ff0654c71fb6 ("KVM: selftests: Touch relevant XSAVE state in guest for state test")
  27526efb5cff ("KVM: x86: Constrain guest-supported xfeatures only at KVM_GET_XSAVE{2}")
  2d287ec65e79 ("x86/fpu: Allow caller to constrain xfeatures when copying to uabi buffer")

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: linux-next: duplicate patches in the kvm-x86 tree

[Sean Christopherson]

On Tue, Oct 17, 2023, Stephen Rothwell wrote:
> Hi all,
> 
> The following commits are also in Linus Torvalds' tree as different
> commits (but the same patches):
> 
>   afb2c7e27a7f ("KVM: selftests: Force load all supported XSAVE state in state test")
>   d7b8762ec4a3 ("KVM: selftests: Load XSAVE state into untouched vCPU during state test")
>   ff0654c71fb6 ("KVM: selftests: Touch relevant XSAVE state in guest for state test")
>   27526efb5cff ("KVM: x86: Constrain guest-supported xfeatures only at KVM_GET_XSAVE{2}")
>   2d287ec65e79 ("x86/fpu: Allow caller to constrain xfeatures when copying to uabi buffer")

Should be resolved, dropped my kvm-x86/fpu branch.

Re: linux-next: duplicate patches in the kvm-x86 tree

[Sean Christopherson]

On Fri, Sep 08, 2023, Stephen Rothwell wrote:
> Hi all,
> 
> The following commits are also in Linus Torvalds' tree as different
> commits (but the same patches):

I'll force push, Paolo squashed a minor fixup and I haven't updated. 

linux-next: duplicate patches in the kvm-x86 tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 3905 bytes --]

Hi all,

The following commits are also in Linus Torvalds' tree as different
commits (but the same patches):

  004c297c327f ("KVM: x86/mmu: Harden TDP MMU iteration against root w/o shadow page")
  064a5ab685bd ("KVM: x86/mmu: Cleanup sanity check of SPTEs at SP free")
  1265fb534fa1 ("KVM: x86: Remove the unused page-track hook track_flush_slot()")
  21baf78ef845 ("KVM: x86/mmu: BUG() in rmap helpers iff CONFIG_BUG_ON_DATA_CORRUPTION=y")
  21e0e1efd880 ("KVM: x86/mmu: Assert that correct locks are held for page write-tracking")
  2431c9ab231a ("KVM: x86/mmu: Drop infrastructure for multiple page-track modes")
  2d28b1230c1c ("drm/i915/gvt: Drop final dependencies on KVM internal details")
  2f502998b046 ("KVM: x86/mmu: Move kvm_arch_flush_shadow_{all,memslot}() to mmu.c")
  3e90c27b4209 ("KVM: x86/mmu: Guard against collision with KVM-defined PFERR_IMPLICIT_ACCESS")
  3f8eb1d7d3ee ("KVM: drm/i915/gvt: Drop @vcpu from KVM's ->track_write() hook")
  472ba3231883 ("KVM: x86/mmu: Plumb "struct kvm" all the way to pte_list_remove()")
  4879a4370304 ("drm/i915/gvt: Use an "unsigned long" to iterate over memslot gfns")
  49a83e190b5b ("drm/i915/gvt: Protect gfn hash table with vgpu_lock")
  4b42f39917c1 ("KVM: x86/mmu: Rename page-track APIs to reflect the new reality")
  4f121b5d2228 ("KVM: x86/mmu: Bug the VM if a vCPU ends up in long mode without PAE enabled")
  537eef32e720 ("drm/i915/gvt: Explicitly check that vGPU is attached before shadowing")
  5a9481e69942 ("KVM: x86/mmu: Delete pgprintk() and all its usage")
  6a718c54c2ee ("drm/i915/gvt: Don't try to unpin an empty page range")
  6d6ff9e6db15 ("KVM: x86/mmu: Bug the VM if write-tracking is used but not enabled")
  732f57612d5c ("KVM: x86/mmu: Add helper to convert root hpa to shadow page")
  81d4621b7d9f ("KVM: x86/mmu: Disallow guest from using !visible slots for page tables")
  823ab2ea8429 ("drm/i915/gvt: remove interface intel_gvt_is_valid_gfn")
  865327865164 ("drm/i915/gvt: Verify pfn is "valid" before dereferencing "struct page"")
  982758f88bb9 ("KVM: x86/mmu: Replace MMU_DEBUG with proper KVM_PROVE_MMU Kconfig")
  9e3f832edfca ("KVM: x86/mmu: Harden new PGD against roots without shadow pages")
  a328a359d99b ("KVM: x86/mmu: Use dummy root, backed by zero page, for !visible guest roots")
  a41e34b05da7 ("drm/i915/gvt: Don't bother removing write-protection on to-be-deleted slot")
  ac5e77621712 ("drm/i915/gvt: Drop unused helper intel_vgpu_reset_gtt()")
  b9ae8a09f357 ("drm/i915/gvt: switch from ->track_flush_slot() to ->track_remove_region()")
  bc90c971dcb7 ("KVM: x86/mmu: Use BUILD_BUG_ON_INVALID() for KVM_MMU_WARN_ON() stub")
  bfd926291c58 ("KVM: x86/mmu: Include mmu.h in spte.h")
  c7784ee6cc06 ("KVM: x86/mmu: Avoid pointer arithmetic when iterating over SPTEs")
  c94811471997 ("drm/i915/gvt: Error out on an attempt to shadowing an unknown GTT entry type")
  ca181aa50724 ("KVM: x86/mmu: Drop @slot param from exported/external page-track APIs")
  e19314998321 ("KVM: x86/mmu: Move the lockdep_assert of mmu_lock to inside clear_dirty_pt_masked()")
  e1cb291d66ac ("KVM: x86/mmu: Rename MMU_WARN_ON() to KVM_MMU_WARN_ON()")
  e27395fb1b87 ("drm/i915/gvt: Verify hugepages are contiguous in physical address space")
  e2fe84fb5eae ("KVM: x86/mmu: Don't rely on page-track mechanism to flush on memslot change")
  eae2d71635a1 ("KVM: x86/mmu: Handle KVM bookkeeping in page-track APIs, not callers")
  ed501863ae54 ("KVM: x86/mmu: Delete rmap_printk() and all its usage")
  f018c319cc2f ("drm/i915/gvt: Don't rely on KVM's gfn_to_pfn() to query possible 2M GTT")
  f01ebf874adb ("KVM: x86/mmu: Delete the "dbg" module param")
  f1c58cdb8e04 ("KVM: x86/mmu: Don't bounce through page-track mechanism for guest PTEs")
  f969ecabe30b ("drm/i915/gvt: Put the page reference obtained by KVM's gfn_to_pfn()")

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: linux-next: duplicate patches in the kvm-x86 tree

[Sean Christopherson]

On Tue, Jan 24, 2023, Paolo Bonzini wrote:
> On Tue, Jan 24, 2023 at 9:47 AM Marc Zyngier <maz@kernel.org> wrote:
> >
> > Hi Stephen,
> >
> > On Tue, 24 Jan 2023 01:55:15 +0000,
> > Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> > >
> > > Hi all,
> > >
> > > The following commits are also in other tree(s?) as different
> > > commits (but the same patches):
> > >
> > >   0b6639e8ed87 ("KVM: s390: Move hardware setup/unsetup to init/exit")
> > >   0c2be59e0b53 ("KVM: x86: Use KBUILD_MODNAME to specify vendor module name")
> > >   1334f214d19f ("KVM: s390: Unwind kvm_arch_init() piece-by-piece() if a step fails")
> >
> > [...]
> >
> > > I guess someone has rebased one of the kvm trees and it had already been
> > > merged into another (like the kvm or kvm-arm trees).
> >
> > Huh, that's worrying. I'm carrying the kvm-hw-enable-refactor branch
> > from the KVM tree, which I understood to be a stable branch[1], and
> > which I merged to avoid conflicts to be propagated everywhere.
> 
> It wasn't 100% guaranteed to be stable because it was meant to be
> tested and have fixes squashed in. But since I had no issues reported
> from either maintainers or bots,

There's one issue, but I didn't explicitly call out that it could be squashed.

https://lore.kernel.org/all/20230119182158.4026656-1-seanjc@google.com

> I will indeed merge commit 9f1a4c004869 aka kvm/kvm-hw-enable-refactor into
> kvm/next. Sean, please rebase to drop the duplicate commits.

Will do.

Re: linux-next: duplicate patches in the kvm-x86 tree

[Paolo Bonzini]

On Tue, Jan 24, 2023 at 9:47 AM Marc Zyngier <maz@kernel.org> wrote:
>
> Hi Stephen,
>
> On Tue, 24 Jan 2023 01:55:15 +0000,
> Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> >
> > Hi all,
> >
> > The following commits are also in other tree(s?) as different
> > commits (but the same patches):
> >
> >   0b6639e8ed87 ("KVM: s390: Move hardware setup/unsetup to init/exit")
> >   0c2be59e0b53 ("KVM: x86: Use KBUILD_MODNAME to specify vendor module name")
> >   1334f214d19f ("KVM: s390: Unwind kvm_arch_init() piece-by-piece() if a step fails")
>
> [...]
>
> > I guess someone has rebased one of the kvm trees and it had already been
> > merged into another (like the kvm or kvm-arm trees).
>
> Huh, that's worrying. I'm carrying the kvm-hw-enable-refactor branch
> from the KVM tree, which I understood to be a stable branch[1], and
> which I merged to avoid conflicts to be propagated everywhere.

It wasn't 100% guaranteed to be stable because it was meant to be
tested and have fixes squashed in. But since I had no issues reported
from either maintainers or bots, I will indeed merge commit
9f1a4c004869 aka kvm/kvm-hw-enable-refactor into kvm/next. Sean,
please rebase to drop the duplicate commits.

Paolo

>
> Paolo, Sean: what is the *real* status of this branch?
>
>         M.
>
> [1] https://lore.kernel.org/r/4d73d1b9-2c28-ab6a-2963-579bcc7a9e67@redhat.com
>
> --
> Without deviation from the norm, progress is not possible.
>

Re: linux-next: duplicate patches in the kvm-x86 tree

[Marc Zyngier]

Hi Stephen,

On Tue, 24 Jan 2023 01:55:15 +0000,
Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> 
> Hi all,
> 
> The following commits are also in other tree(s?) as different
> commits (but the same patches):
> 
>   0b6639e8ed87 ("KVM: s390: Move hardware setup/unsetup to init/exit")
>   0c2be59e0b53 ("KVM: x86: Use KBUILD_MODNAME to specify vendor module name")
>   1334f214d19f ("KVM: s390: Unwind kvm_arch_init() piece-by-piece() if a step fails")

[...]

> I guess someone has rebased one of the kvm trees and it had already been
> merged into another (like the kvm or kvm-arm trees).

Huh, that's worrying. I'm carrying the kvm-hw-enable-refactor branch
from the KVM tree, which I understood to be a stable branch[1], and
which I merged to avoid conflicts to be propagated everywhere.

Paolo, Sean: what is the *real* status of this branch?

	M.

[1] https://lore.kernel.org/r/4d73d1b9-2c28-ab6a-2963-579bcc7a9e67@redhat.com

-- 
Without deviation from the norm, progress is not possible.

linux-next: duplicate patches in the kvm-x86 tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 4339 bytes --]

Hi all,

The following commits are also in other tree(s?) as different
commits (but the same patches):

  0b6639e8ed87 ("KVM: s390: Move hardware setup/unsetup to init/exit")
  0c2be59e0b53 ("KVM: x86: Use KBUILD_MODNAME to specify vendor module name")
  1334f214d19f ("KVM: s390: Unwind kvm_arch_init() piece-by-piece() if a step fails")
  145648f3d255 ("KVM: x86: Serialize vendor module initialization (hardware setup)")
  1b2d6f07a3bf ("KVM: Drop arch hardware (un)setup hooks")
  1f930fb59c72 ("KVM: x86: Do VMX/SVM support checks directly in vendor code")
  260a8c61527e ("KVM: x86: Do compatibility checks when onlining CPU")
  27c87909f037 ("KVM: arm64: Simplify the CPUHP logic")
  3f1e7970df4a ("KVM: VMX: Reset eVMCS controls in VP assist page during hardware disabling")
  474856bad921 ("KVM: PPC: Move processor compatibility check to module init")
  4e27b7e44713 ("KVM: VMX: Don't bother disabling eVMCS static key on module exit")
  546df64b3d07 ("KVM: Ensure CPU is stable during low level hardware enable/disable")
  6391320c531e ("KVM: VMX: Use current CPU's info to perform "disabled by BIOS?" checks")
  65d06a60543d ("KVM: Clean up error labels in kvm_init()")
  6705b876091e ("KVM: VMX: Shuffle support checks and hardware enabling code around")
  6c5d2dbcc71c ("KVM: Rename and move CPUHP_AP_KVM_STARTING to ONLINE section")
  6e9f5ebba4c4 ("KVM: arm64: Mark kvm_arm_init() and its unique descendants as __init")
  714d938385ce ("KVM: x86: Move guts of kvm_arch_init() to standalone helper")
  71875bd668c7 ("KVM: VMX: Move Hyper-V eVMCS initialization to helper")
  772bb3ffe6e0 ("KVM: Opt out of generic hardware enabling on s390 and PPC")
  784277da2097 ("KVM: Teardown VFIO ops earlier in kvm_exit()")
  79e49a0fa367 ("KVM: s390: Do s390 specific init without bouncing through kvm_init()")
  81566442e4bf ("KVM: Make hardware_enable_failed a local variable in the "enable all" path")
  8d040a03e829 ("KVM: x86: Move CPU compat checks hook to kvm_x86_ops (from kvm_x86_init_ops)")
  8e868404d176 ("KVM: MIPS: Hardcode callbacks to hardware virtualization extensions")
  9316983f4605 ("KVM: Drop kvm_count_lock and instead protect kvm_usage_count with kvm_lock")
  aa899f4725e8 ("KVM: Drop kvm_arch_{init,exit}() hooks")
  af5b59ac929b ("KVM: Remove on_each_cpu(hardware_disable_nolock) in kvm_exit()")
  b23df1859468 ("KVM: MIPS: Setup VZ emulation? directly from kvm_mips_init()")
  b38c794ec895 ("KVM: x86: Move hardware setup/unsetup to init/exit")
  b89f7faedfa3 ("KVM: x86: Unify pr_fmt to use module name for all KVM modules")
  ba3ad92110a6 ("KVM: s390: Mark __kvm_s390_init() and its descendants as __init")
  c017865c7cbc ("KVM: x86: Do CPU compatibility checks in x86 code")
  c2330114638a ("KVM: VMX: Do _all_ initialization before exposing /dev/kvm to userspace")
  c7e997d0e9f3 ("KVM: MIPS: Register die notifier prior to kvm_init()")
  c9d8dd2372a3 ("KVM: Register syscore (suspend/resume) ops early in kvm_init()")
  cc565465bb27 ("KVM: Drop kvm_arch_check_processor_compat() hook")
  d3174cf0ec90 ("KVM: arm64: Free hypervisor allocations if vector slot init fails")
  d4739c7e93de ("KVM: SVM: Check for SVM support in CPU compatibility checks")
  d64252c198aa ("KVM: Register /dev/kvm as the _very_ last thing during initialization")
  d80071eb35d8 ("KVM: x86: Do timer initialization after XCR0 configuration")
  da8de651943a ("KVM: Allocate cpus_hardware_enabled after arch hardware setup")
  da8efe2b928d ("KVM: Use a per-CPU variable to track which CPUs have enabled virtualization")
  dacd86171174 ("KVM: arm64: Do arm/arch initialization without bouncing through kvm_init()")
  de2646ea615e ("KVM: RISC-V: Do arch init directly in riscv_kvm_init()")
  dff01c0d47f9 ("KVM: Initialize IRQ FD after arch hardware setup")
  e3785ed68b99 ("KVM: arm64: Unregister perf callbacks if hypervisor finalization fails")
  e9c576b0f8bc ("KVM: Disable CPU hotplug during hardware enabling/disabling")
  ebb10d254041 ("KVM: VMX: Make VMCS configuration/capabilities structs read-only after init")
  ef307e18c456 ("KVM: RISC-V: Tag init functions and data with __init, __ro_after_init")

I guess someone has rebased one of the kvm trees and it had already been
merged into another (like the kvm or kvm-arm trees).

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]