Skip to content

KVM: xen: update shared_info and vcpu_info handling

Paul Durrant kvm linux-doc linux-kselftest linux-s390 lkml 14 versions – Applied

Status from this message: Applied to kvm-x86 xen, minus 18 and 19 (trylock stuff) and 21 (locking cleanup

Get series with b4
b4 am [email protected]

Versions newest first · linked by author and title

  1. v13 Latest21 patches 23 repliesApplied
  2. v1220 patches 37 repliesSuperseded
  3. v1119 patches 2 repliesSuperseded
Earlier versions · 11
VersionPostedPatchesRepliesStatus
v1019 patches7 repliesSuperseded
v915 patches3 repliesSuperseded
v815 patches20 repliesSuperseded
v711 patches22 repliesSuperseded
v6 posting 2/20 of 11 patches0 repliesSuperseded
v6 posting 1/211 patches1 replySuperseded
v510 patches8 repliesSuperseded
v413 patches7 repliesSuperseded
v313 patches11 repliesSuperseded
v212 patches19 repliesSuperseded
v18 patches17 repliesSuperseded

Patches in v13 Applied

#Subject and filesSizeRepliesApplied
1/21KVM: pfncache: Add a map helper functionvirt/kvm/pfncache.c+29 -180 replies
2/21KVM: pfncache: remove unnecessary exportsvirt/kvm/pfncache.c+0 -50 replies
3/21KVM: x86/xen: mark guest pages dirty with the pfncache lock heldarch/x86/kvm/xen.c+6 -70 replies
4/21KVM: pfncache: add a mark-dirty helperarch/x86/kvm/x86.c arch/x86/kvm/xen.c include/linux/kvm_host.h+17 -42 replies
5/21KVM: pfncache: remove KVM_GUEST_USES_PFN usagearch/x86/kvm/x86.c arch/x86/kvm/xen.c include/linux/kvm_host.h +2 more+16 -803 replies
6/21KVM: pfncache: stop open-coding offset_in_page()virt/kvm/pfncache.c+3 -30 replies
7/21KVM: pfncache: include page offset in uhva and use it consistentlyvirt/kvm/pfncache.c+21 -80 replies
8/21KVM: s390: Refactor kvm_is_error_gpa() into kvm_is_gpa_in_memslot()arch/s390/kvm/diag.c arch/s390/kvm/gaccess.c arch/s390/kvm/kvm-s390.c +3 more+15 -150 replies
9/21KVM: pfncache: allow a cache to be activated with a fixed (userspace) HVAinclude/linux/kvm_host.h virt/kvm/pfncache.c+89 -262 replies
10/21KVM: x86/xen: separate initialization of shared_info cache and contentarch/x86/kvm/xen.c+32 -230 replies
11/21KVM: x86/xen: re-initialize shared_info if guest (32/64-bit) mode is setarch/x86/kvm/xen.c+26 -30 replies
12/21KVM: x86/xen: allow shared_info to be mapped by fixed HVADocumentation/virt/kvm/api.rst arch/x86/kvm/xen.c include/uapi/linux/kvm.h+59 -162 replies
13/21KVM: x86/xen: allow vcpu_info to be mapped by fixed HVADocumentation/virt/kvm/api.rst arch/x86/kvm/xen.c include/uapi/linux/kvm.h+52 -110 replies
14/21KVM: selftests: map Xen's shared_info page using HVA rather than GFNtools/testing/selftests/kvm/x86_64/xen_shinfo_test.c+35 -90 replies
15/21KVM: selftests: re-map Xen's vcpu_info using HVA rather than GPAtools/testing/selftests/kvm/x86_64/xen_shinfo_test.c+15 -00 replies
16/21KVM: x86/xen: advertize the KVM_XEN_HVM_CONFIG_SHARED_INFO_HVA capabilityarch/x86/kvm/x86.c+2 -10 replies
17/21KVM: x86/xen: split up kvm_xen_set_evtchn_fast()arch/x86/kvm/xen.c+100 -750 replies
18/21KVM: x86/xen: don't block on pfncache locks in kvm_xen_set_evtchn_fast()arch/x86/kvm/xen.c+20 -102 replies
19/21KVM: pfncache: check the need for invalidation under read lock firstvirt/kvm/pfncache.c+19 -30 replies
20/21KVM: x86/xen: allow vcpu_info content to be 'safely' copiedarch/x86/kvm/xen.c+0 -30 replies
21/21KVM: pfncache: rework __kvm_gpc_refresh() to fix locking issuesvirt/kvm/pfncache.c+88 -964 replies

Reviews trailers per patch

20 of 21 patches have a Reviewed-by or Acked-by.

PatchDavid WoodhousePaul Durrant
1/21R-b
2/21R-b
3/21R-b
4/21R-b
5/21R-b
6/21R-b
7/21R-b
8/21
9/21R-b
10/21R-b
11/21R-b
12/21R-b
13/21R-b
14/21R-b
15/21R-b
16/21R-b
17/21R-b
18/21R-b
19/21R-b
20/21R-b
21/21R-b

Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.

Read the v13 thread

Cover letter [PATCH v13 00/21] KVM: xen: update shared_info and vcpu_info handling

From: Paul Durrant <[email protected]>

This series contains a new patch from Sean added since v12 [1]:

* KVM: s390: Refactor kvm_is_error_gpa() into kvm_is_gpa_in_memslot()

This frees up the function name kvm_is_error_gpa() such that it can then be re-defined in:

* KVM: pfncache: allow a cache to be activated with a fixed (userspace) HVA

to be used for a simple GPA validation helper function. The patch also now contains an either/or address check for GPA versus HVA in __kvm_gpc_refresh().

In:

* KVM: pfncache: add a mark-dirty helper

The function name has been changed from kvm_gpc_mark_dirty() to kvm_gpc_mark_dirty_in_slot().

In:

* KVM: x86/xen: allow shared_info to be mapped by fixed HVA

missing HVA validation checks have been added and the 'hva == 0' test has been changed to '!hva'. The KVM_XEN_ATTR_TYPE_SHARED_INFO and KVM_XEN_ATTR_TYPE_SHARED_INFO_HVA cases are still largely handled as one though as separation leads to duplicate calls to kvm_xen_shared_info_init() which looks messy.

Also, patches with a 'xen' prefix have now been modified to have a 'x86/xen' prefix and patches with a 'selftests / xen' prefix have been modified to have simply a 'selftests' prefix.

[1] https://lore.kernel.org/kvm/[email protected]/

David Woodhouse (1):
KVM: pfncache: rework __kvm_gpc_refresh() to fix locking issues

Paul Durrant (19):
KVM: pfncache: Add a map helper function
KVM: pfncache: remove unnecessary exports
KVM: x86/xen: mark guest pages dirty with the pfncache lock held
KVM: pfncache: add a mark-dirty helper
KVM: pfncache: remove KVM_GUEST_USES_PFN usage
KVM: pfncache: stop open-coding offset_in_page()
KVM: pfncache: include page offset in uhva and use it consistently
KVM: pfncache: allow a cache to be activated with a fixed (userspace)
HVA
KVM: x86/xen: separate initialization of shared_info cache and content
KVM: x86/xen: re-initialize shared_info if guest (32/64-bit) mode is
set
KVM: x86/xen: allow shared_info to be mapped by fixed HVA
KVM: x86/xen: allow vcpu_info to be mapped by fixed HVA
KVM: selftests: map Xen's shared_info page using HVA rather than GFN
KVM: selftests: re-map Xen's vcpu_info using HVA rather than GPA
KVM: x86/xen: advertize the KVM_XEN_HVM_CONFIG_SHARED_INFO_HVA
capability
KVM: x86/xen: split up kvm_xen_set_evtchn_fast()
KVM: x86/xen: don't block on pfncache locks in
kvm_xen_set_evtchn_fast()
KVM: pfncache: check the need for invalidation under read lock first
KVM: x86/xen: allow vcpu_info content to be 'safely' copied

Sean Christopherson (1):
KVM: s390: Refactor kvm_is_error_gpa() into kvm_is_gpa_in_memslot()

 Documentation/virt/kvm/api.rst                |  53 ++-
 arch/s390/kvm/diag.c                          |   2 +-
 arch/s390/kvm/gaccess.c                       |  14 +-
 arch/s390/kvm/kvm-s390.c                      |   4 +-
 arch/s390/kvm/priv.c                          |   4 +-
 arch/s390/kvm/sigp.c                          |   2 +-
 arch/x86/kvm/x86.c                            |   7 +-
 arch/x86/kvm/xen.c                            | 361 +++++++++++------
 include/linux/kvm_host.h                      |  49 ++-
 include/linux/kvm_types.h                     |   8 -
 include/uapi/linux/kvm.h                      |   9 +-
 .../selftests/kvm/x86_64/xen_shinfo_test.c    |  59 ++-
 virt/kvm/pfncache.c                           | 382 ++++++++++--------
 13 files changed, 591 insertions(+), 363 deletions(-)

base-commit: 7455665a3521aa7b56245c0a2810f748adc5fdd4

Signature · 1 line
2.39.2