Skip to content
  1. Jul 17, 2021
    • Linus Torvalds's avatar
      Merge tag 'io_uring-5.14-2021-07-16' of git://git.kernel.dk/linux-block · 13fdaf04
      Linus Torvalds authored 
      Pull io_uring fixes from Jens Axboe:
 "Two small fixes: one fixing the process target of a check, and the
  other a minor issue with the drain error handling"

* tag 'io_uring-5.14-2021-07-16' of git://git.kernel.dk/linux-block:
  io_uring: fix io_drain_req()
  io_uring: use right task for exiting checks
      13fdaf04
    • Linus Torvalds's avatar
      Merge tag 'zonefs-5.14-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/dlemoal/zonefs · 45312bd7
      Linus Torvalds authored 
      Pull zonefs fix from Damien Le Moal:
 "A single patch to remove an unnecessary NULL bio check (from
  Xianting)"

* tag 'zonefs-5.14-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/dlemoal/zonefs:
  zonefs: remove redundant null bio check
      45312bd7
    • Linus Torvalds's avatar
      Merge tag 'drm-fixes-2021-07-16' of git://anongit.freedesktop.org/drm/drm · 786cb0a2
      Linus Torvalds authored 
      Pull drm fixes from Dave Airlie:
 "Regular rc2 fixes though a bit more than usual at rc2 stage, people
  must have been testing early or else some fixes from last week got a
  bit laggy.

  There is one larger change in the amd fixes to amalgamate some power
  management code on the newer chips with the code from the older chips,
  it should only affects chips where support was introduced in rc1 and
  it should make future fixes easier to maintain probably a good idea to
  merge it now.

  Otherwise it's mostly fixes across the board.

  dma-buf:
   - Fix fence leak in sync_file_merge() error code

  drm/panel:
   - nt35510: Don't fail on DSI reads

  fbdev:
   - Avoid use-after-free by not deleting current video mode

  ttm:
   - Avoid NULL-ptr deref in ttm_range_man_fini()

  vmwgfx:
   - Fix a merge commit

  qxl:
   - fix a TTM regression

  amdgpu:
   - SR-IOV fixes
   - RAS fixes
   - eDP fixes
   - SMU13 code unification to facilitate fixes in the future
   - Add new renoir DID
   - Yellow Carp fixes
   - Beige Goby fixes
   - Revert a bunch of TLB fixes that caused regressions
   - Revert an LTTPR display regression

  amdkfd
   - Fix VRAM access regression
   - SVM fixes

  i915:
   - Fix -EDEADLK handling regression
   - Drop the page table optimisation"

* tag 'drm-fixes-2021-07-16' of git://anongit.freedesktop.org/drm/drm: (29 commits)
  drm/amdgpu: add another Renoir DID
  drm/ttm: add a check against null pointer dereference
  drm/i915/gtt: drop the page table optimisation
  drm/i915/gt: Fix -EDEADLK handling regression
  drm/amd/pm: Add waiting for response of mode-reset message for yellow carp
  Revert "drm/amdkfd: Add heavy-weight TLB flush after unmapping"
  Revert "drm/amdgpu: Add table_freed parameter to amdgpu_vm_bo_update"
  Revert "drm/amdkfd: Make TLB flush conditional on mapping"
  Revert "drm/amdgpu: Fix warning of Function parameter or member not described"
  Revert "drm/amdkfd: Add memory sync before TLB flush on unmap"
  drm/amd/pm: Fix BACO state setting for Beige_Goby
  drm/amdgpu: Restore msix after FLR
  drm/amdkfd: Allow CPU access for all VRAM BOs
  drm/amdgpu/display - only update eDP's backlight level when necessary
  drm/amdkfd: handle fault counters on invalid address
  drm/amdgpu: Correct the irq numbers for virtual crtc
  drm/amd/display: update header file name
  drm/amd/pm: drop smu_v13_0_1.c|h files for yellow carp
  drm/amd/display: remove faulty assert
  Revert "drm/amd/display: Always write repeater mode regardless of LTTPR"
  ...
      786cb0a2
    • Linus Torvalds's avatar
      Merge branch 'urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu · 6e442d06
      Linus Torvalds authored 
      Pull RCU fixes from Paul McKenney:

 - fix regressions induced by a merge-window change in scheduler
   semantics, which means that smp_processor_id() can no longer be used
   in kthreads using simple affinity to bind themselves to a specific
   CPU.

 - fix a bug in Tasks Trace RCU that was thought to be strictly
   theoretical. However, production workloads have started hitting this,
   so these fixes need to be merged sooner rather than later.

 - fix a minor printk()-format-mismatch issue introduced during the
   merge window.

* 'urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu:
  rcu: Fix pr_info() formats and values in show_rcu_gp_kthreads()
  rcu-tasks: Don't delete holdouts within trc_wait_for_one_reader()
  rcu-tasks: Don't delete holdouts within trc_inspect_reader()
  refscale: Avoid false-positive warnings in ref_scale_reader()
  scftorture: Avoid false-positive warnings in scftorture_invoker()
      6e442d06
  3. Jul 16, 2021
  5. Jul 15, 2021
    • Vitaly Kuznetsov's avatar
      KVM: selftests: smm_test: Test SMM enter from L2 · d951b221
      Vitaly Kuznetsov authored 
      

Two additional tests are added:
- SMM triggered from L2 does not currupt L1 host state.
- Save/restore during SMM triggered from L2 does not corrupt guest/host
  state.

Signed-off-by: default avatarVitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210628104425.391276-7-vkuznets@redhat.com>
Reviewed-by: default avatarMaxim Levitsky <mlevitsk@redhat.com>
Signed-off-by: default avatarPaolo Bonzini <pbonzini@redhat.com>
      d951b221
    • Vitaly Kuznetsov's avatar
      KVM: nSVM: Restore nested control upon leaving SMM · bb00bd9c
      Vitaly Kuznetsov authored 
      

If the VM was migrated while in SMM, no nested state was saved/restored,
and therefore svm_leave_smm has to load both save and control area
of the vmcb12. Save area is already loaded from HSAVE area,
so now load the control area as well from the vmcb12.

Signed-off-by: default avatarVitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210628104425.391276-6-vkuznets@redhat.com>
Signed-off-by: default avatarPaolo Bonzini <pbonzini@redhat.com>
      bb00bd9c
    • Vitaly Kuznetsov's avatar
      KVM: nSVM: Fix L1 state corruption upon return from SMM · 37be407b
      Vitaly Kuznetsov authored 
      VMCB split commit 4995a368 ("KVM: SVM: Use a separate vmcb for the
nested L2 guest") broke return from SMM when we entered there from guest
(L2) mode. Gen2 WS2016/Hyper-V is known to do this on boot. The problem
manifests itself like this:

  kvm_exit:             reason EXIT_RSM rip 0x7ffbb280 info 0 0
  kvm_emulate_insn:     0:7ffbb280: 0f aa
  kvm_smm_transition:   vcpu 0: leaving SMM, smbase 0x7ffb3000
  kvm_nested_vmrun:     rip: 0x000000007ffbb280 vmcb: 0x0000000008224000
    nrip: 0xffffffffffbbe119 int_ctl: 0x01020000 event_inj: 0x00000000
    npt: on
  kvm_nested_intercepts: cr_read: 0000 cr_write: 0010 excp: 40060002
    intercepts: fd44bfeb 0000217f 00000000
  kvm_entry:            vcpu 0, rip 0xffffffffffbbe119
  kvm_exit:             reason EXIT_NPF rip 0xffffffffffbbe119 info
    200000006 1ab000
  kvm_nested_vmexit:    vcpu 0 reason npf rip 0xffffffffffbbe119 info1
    0x0000000200000006 info2 0x00000000001ab000 intr_info 0x00000000
    error_code 0x00000000
  kvm_page_fault:       address 1ab000 error_code 6
  kvm_nested_vmexit_inject: reason EXIT_NPF info1 200000006 info2 1ab000
    int_info 0 int_info_err 0
  kvm_entry:            vcpu 0, rip 0x7ffbb280
  kvm_exit:             reason EXIT_EXCP_GP rip 0x7ffbb280 info 0 0
  kvm_emulate_insn:     0:7ffbb280: 0f aa
  kvm_inj_exception:    #GP (0x0)

Note: return to L2 succeeded but upon first exit to L1 its RIP points to
'RSM' instruction but we're not in SMM.

The problem appears to be that VMCB01 gets irreversibly destroyed during
SMM execution. Previously, we used to have 'hsave' VMCB where regular
(pre-SMM) L1's state was saved upon nested_svm_vmexit() but now we just
switch to VMCB01 from VMCB02.

Pre-split (working) flow looked like:
- SMM is triggered during L2's execution
- L2's state is pushed to SMRAM
- nested_svm_vmexit() restores L1's state from 'hsave'
- SMM -> RSM
- enter_svm_guest_mode() switches to L2 but keeps 'hsave' intact so we have
  pre-SMM (and pre L2 VMRUN) L1's state there
- L2's state is restored from SMRAM
- upon first exit L1's state is restored from L1.

This was always broken with regards to svm_get_nested_state()/
svm_set_nested_state(): 'hsave' was never a part of what's being
save and restored so migration happening during SMM triggered from L2 would
never restore L1's state correctly.

Post-split flow (broken) looks like:
- SMM is triggered during L2's execution
- L2's state is pushed to SMRAM
- nested_svm_vmexit() switches to VMCB01 from VMCB02
- SMM -> RSM
- enter_svm_guest_mode() switches from VMCB01 to VMCB02 but pre-SMM VMCB01
  is already lost.
- L2's state is restored from SMRAM
- upon first exit L1's state is restored from VMCB01 but it is corrupted
 (reflects the state during 'RSM' execution).

VMX doesn't have this problem because unlike VMCB, VMCS keeps both guest
and host state so when we switch back to VMCS02 L1's state is intact there.

To resolve the issue we need to save L1's state somewhere. We could've
created a third VMCB for SMM but that would require us to modify saved
state format. L1's architectural HSAVE area (pointed by MSR_VM_HSAVE_PA)
seems appropriate: L0 is free to save any (or none) of L1's state there.
Currently, KVM does 'none'.

Note, for nested state migration to succeed, both source and destination
hypervisors must have the fix. We, however, don't need to create a new
flag indicating the fact that HSAVE area is now populated as migration
during SMM triggered from L2 was always broken.

Fixes: 4995a368

 ("KVM: SVM: Use a separate vmcb for the nested L2 guest")
Signed-off-by: default avatarVitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: default avatarPaolo Bonzini <pbonzini@redhat.com>
      37be407b
    • Vitaly Kuznetsov's avatar
      KVM: nSVM: Introduce svm_copy_vmrun_state() · 0a758290
      Vitaly Kuznetsov authored 
      

Separate the code setting non-VMLOAD-VMSAVE state from
svm_set_nested_state() into its own function. This is going to be
re-used from svm_enter_smm()/svm_leave_smm().

Signed-off-by: default avatarVitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210628104425.391276-4-vkuznets@redhat.com>
Signed-off-by: default avatarPaolo Bonzini <pbonzini@redhat.com>
      0a758290
    • Vitaly Kuznetsov's avatar
      KVM: nSVM: Check that VM_HSAVE_PA MSR was set before VMRUN · fb79f566
      Vitaly Kuznetsov authored 
      

APM states that "The address written to the VM_HSAVE_PA MSR, which holds
the address of the page used to save the host state on a VMRUN, must point
to a hypervisor-owned page. If this check fails, the WRMSR will fail with
a #GP(0) exception. Note that a value of 0 is not considered valid for the
VM_HSAVE_PA MSR and a VMRUN that is attempted while the HSAVE_PA is 0 will
fail with a #GP(0) exception."

svm_set_msr() already checks that the supplied address is valid, so only
check for '0' is missing. Add it to nested_svm_vmrun().

Signed-off-by: default avatarVitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210628104425.391276-3-vkuznets@redhat.com>
Reviewed-by: default avatarMaxim Levitsky <mlevitsk@redhat.com>
Signed-off-by: default avatarPaolo Bonzini <pbonzini@redhat.com>
      fb79f566
    • Vitaly Kuznetsov's avatar
      KVM: nSVM: Check the value written to MSR_VM_HSAVE_PA · fce7e152
      Vitaly Kuznetsov authored 
      

APM states that #GP is raised upon write to MSR_VM_HSAVE_PA when
the supplied address is not page-aligned or is outside of "maximum
supported physical address for this implementation".
page_address_valid() check seems suitable. Also, forcefully page-align
the address when it's written from VMM.

Signed-off-by: default avatarVitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210628104425.391276-2-vkuznets@redhat.com>
Cc: stable@vger.kernel.org
Reviewed-by: default avatarMaxim Levitsky <mlevitsk@redhat.com>
[Add comment about behavior for host-provided values. - Paolo]
Signed-off-by: default avatarPaolo Bonzini <pbonzini@redhat.com>
      fce7e152
    • Sean Christopherson's avatar
      KVM: SVM: Fix sev_pin_memory() error checks in SEV migration utilities · c7a1b2b6
      Sean Christopherson authored 
      

Use IS_ERR() instead of checking for a NULL pointer when querying for
sev_pin_memory() failures.  sev_pin_memory() always returns an error code
cast to a pointer, or a valid pointer; it never returns NULL.

Reported-by: default avatarDan Carpenter <dan.carpenter@oracle.com>
Cc: Steve Rutherford <srutherford@google.com>
Cc: Brijesh Singh <brijesh.singh@amd.com>
Cc: Ashish Kalra <ashish.kalra@amd.com>
Fixes: d3d1af85 ("KVM: SVM: Add KVM_SEND_UPDATE_DATA command")
Fixes: 15fb7de1

 ("KVM: SVM: Add KVM_SEV_RECEIVE_UPDATE_DATA command")
Signed-off-by: default avatarSean Christopherson <seanjc@google.com>
Message-Id: <20210506175826.2166383-3-seanjc@google.com>
Signed-off-by: default avatarPaolo Bonzini <pbonzini@redhat.com>
      c7a1b2b6
    • Sean Christopherson's avatar
      KVM: SVM: Return -EFAULT if copy_to_user() for SEV mig packet header fails · b4a69392
      Sean Christopherson authored 
      

Return -EFAULT if copy_to_user() fails; if accessing user memory faults,
copy_to_user() returns the number of bytes remaining, not an error code.

Reported-by: default avatarDan Carpenter <dan.carpenter@oracle.com>
Cc: Steve Rutherford <srutherford@google.com>
Cc: Brijesh Singh <brijesh.singh@amd.com>
Cc: Ashish Kalra <ashish.kalra@amd.com>
Fixes: d3d1af85

 ("KVM: SVM: Add KVM_SEND_UPDATE_DATA command")
Signed-off-by: default avatarSean Christopherson <seanjc@google.com>
Message-Id: <20210506175826.2166383-2-seanjc@google.com>
Signed-off-by: default avatarPaolo Bonzini <pbonzini@redhat.com>
      b4a69392
    • Maxim Levitsky's avatar
      KVM: SVM: add module param to control the #SMI interception · 4b639a9f
      Maxim Levitsky authored 
      

In theory there are no side effects of not intercepting #SMI,
because then #SMI becomes transparent to the OS and the KVM.

Plus an observation on recent Zen2 CPUs reveals that these
CPUs ignore #SMI interception and never deliver #SMI VMexits.

This is also useful to test nested KVM to see that L1
handles #SMIs correctly in case when L1 doesn't intercept #SMI.

Finally the default remains the same, the SMI are intercepted
by default thus this patch doesn't have any effect unless
non default module param value is used.

Signed-off-by: default avatarMaxim Levitsky <mlevitsk@redhat.com>
Message-Id: <20210707125100.677203-4-mlevitsk@redhat.com>
Signed-off-by: default avatarPaolo Bonzini <pbonzini@redhat.com>
      4b639a9f
    • Maxim Levitsky's avatar
      KVM: SVM: remove INIT intercept handler · 896707c2
      Maxim Levitsky authored 
      Kernel never sends real INIT even to CPUs, other than on boot.

Thus INIT interception is an error which should be caught
by a check for an unknown VMexit reason.

On top of that, the current INIT VM exit handler skips
the current instruction which is wrong.
That was added in commit 5ff3a351 ("KVM: x86: Move trivial
instruction-based exit handlers to common code").

Fixes: 5ff3a351

 ("KVM: x86: Move trivial instruction-based exit handlers to common code")
Signed-off-by: default avatarMaxim Levitsky <mlevitsk@redhat.com>
Message-Id: <20210707125100.677203-3-mlevitsk@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: default avatarPaolo Bonzini <pbonzini@redhat.com>
      896707c2
    • Maxim Levitsky's avatar
      KVM: SVM: #SMI interception must not skip the instruction · 991afbbe
      Maxim Levitsky authored 
      Commit 5ff3a351 ("KVM: x86: Move trivial instruction-based
exit handlers to common code"), unfortunately made a mistake of
treating nop_on_interception and nop_interception in the same way.

Former does truly nothing while the latter skips the instruction.

SMI VM exit handler should do nothing.
(SMI itself is handled by the host when we do STGI)

Fixes: 5ff3a351

 ("KVM: x86: Move trivial instruction-based exit handlers to common code")
Signed-off-by: default avatarMaxim Levitsky <mlevitsk@redhat.com>
Message-Id: <20210707125100.677203-2-mlevitsk@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: default avatarPaolo Bonzini <pbonzini@redhat.com>
      991afbbe