[PATCH v2 15/16] hyper-v: trace vmbus_release_relid()

Add tracepoint to CHANNELMSG_RELID_RELEASED sender. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/channel_mgmt.c | 7 +-- drivers/hv/hv_trace.h | 16 2 files changed, 21 insertions(+), 2 deletions(-) diff --git a/drivers/hv/channel_mg

[PATCH v2 02/16] hyper-v: trace vmbus_on_message()

Add tracepoint to vmbus_on_message() which is called when we start processing a blocking from work context. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/channel_mgmt.c | 2 ++ drivers/hv/hv_trace.h | 5 + 2 files changed, 7 insertions(+) diff --git a/driv

[PATCH v2 03/16] hyper-v: trace vmbus_onoffer()

Add tracepoint to CHANNELMSG_OFFERCHANNEL handler. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/channel_mgmt.c | 2 ++ drivers/hv/hv_trace.h | 37 + 2 files changed, 39 insertions(+) diff --git a/drivers/hv/channel_mg

[PATCH v2 07/16] hyper-v: trace vmbus_ongpadl_torndown()

Add tracepoint to CHANNELMSG_GPADL_TORNDOWN handler. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/channel_mgmt.c | 2 ++ drivers/hv/hv_trace.h | 8 2 files changed, 10 insertions(+) diff --git a/drivers/hv/channel_mgmt.c b/drivers/hv/channel_mgmt.c

[PATCH v2 00/16] Hyper-V: add tracing to VMBus module and trace all messages

ANNELMSG_UNLOAD/CHANNELMSG_UNLOAD_RESPONSE are not traced as these are mostly used on crash. Vitaly Kuznetsov (16): hyper-v: trace vmbus_on_msg_dpc() hyper-v: trace vmbus_on_message() hyper-v: trace vmbus_onoffer() hyper-v: trace vmbus_onoffer_rescind() hyper-v: trace vmbus_onopen_result()

Re: [PATCH 02/16] hyper-v: trace vmbus_on_message()

Steven Rostedt <rost...@goodmis.org> writes: > On Wed, 20 Sep 2017 19:21:53 +0200 > Vitaly Kuznetsov <vkuzn...@redhat.com> wrote: > >> diff --git a/drivers/hv/hv_trace.h b/drivers/hv/hv_trace.h >> index 9a29ef55477d..72911dfc9682 100644 >> --- a/dr

[PATCH 12/16] hyper-v: trace vmbus_establish_gpadl()

Add tracepoint to CHANNELMSG_GPADL_HEADER/CHANNELMSG_GPADL_BODY sender. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/channel.c | 6 ++ drivers/hv/hv_trace.h | 42 ++ 2 files changed, 48 insertions(+) diff --git a/driv

[PATCH 11/16] hyper-v: trace vmbus_close_internal()

Add tracepoint to CHANNELMSG_CLOSECHANNEL sender. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/channel.c | 2 ++ drivers/hv/hv_trace.h | 15 +++ 2 files changed, 17 insertions(+) diff --git a/drivers/hv/channel.c b/drivers/hv/channel.c index 7b114b

[PATCH 08/16] hyper-v: trace vmbus_onversion_response()

Add tracepoint to CHANNELMSG_VERSION_RESPONSE handler. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/channel_mgmt.c | 3 +++ drivers/hv/hv_trace.h | 11 +++ 2 files changed, 14 insertions(+) diff --git a/drivers/hv/channel_mgmt.c b/drivers/hv/channel_

[PATCH 14/16] hyper-v: trace vmbus_negotiate_version()

Add tracepoint to CHANNELMSG_INITIATE_CONTACT sender. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/connection.c | 3 +++ drivers/hv/hv_trace.h | 26 ++ 2 files changed, 29 insertions(+) diff --git a/drivers/hv/connection.c b/driv

[PATCH 13/16] hyper-v: trace vmbus_teardown_gpadl()

Add tracepoint to CHANNELMSG_GPADL_TEARDOWN sender. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/channel.c | 2 ++ drivers/hv/hv_trace.h | 18 ++ 2 files changed, 20 insertions(+) diff --git a/drivers/hv/channel.c b/drivers/hv/channel.c

[PATCH 15/16] hyper-v: trace vmbus_release_relid()

Add tracepoint to CHANNELMSG_RELID_RELEASED sender. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/channel_mgmt.c | 7 +-- drivers/hv/hv_trace.h | 16 2 files changed, 21 insertions(+), 2 deletions(-) diff --git a/drivers/hv/channel_mg

[PATCH 16/16] hyper-v: trace vmbus_send_tl_connect_request()

Add tracepoint to CHANNELMSG_TL_CONNECT_REQUEST sender. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/channel.c | 7 ++- drivers/hv/hv_trace.h | 20 2 files changed, 26 insertions(+), 1 deletion(-) diff --git a/drivers/hv/channel.c b/driv

[PATCH 04/16] hyper-v: trace vmbus_onoffer_rescind()

Add tracepoint to CHANNELMSG_RESCIND_CHANNELOFFER handler. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/channel_mgmt.c | 2 ++ drivers/hv/hv_trace.h | 8 2 files changed, 10 insertions(+) diff --git a/drivers/hv/channel_mgmt.c b/drivers/hv/channel_

[PATCH 09/16] hyper-v: trace vmbus_request_offers()

Add tracepoint to CHANNELMSG_REQUESTOFFERS sender. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/channel_mgmt.c | 4 +++- drivers/hv/hv_trace.h | 8 2 files changed, 11 insertions(+), 1 deletion(-) diff --git a/drivers/hv/channel_mgmt.c b/driv

[PATCH 10/16] hyper-v: trace vmbus_open()

Add tracepoint to CHANNELMSG_OPENCHANNEL sender. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/channel.c | 2 ++ drivers/hv/hv_trace.h | 27 +++ 2 files changed, 29 insertions(+) diff --git a/drivers/hv/channel.c b/drivers/hv/channel.c

[PATCH 03/16] hyper-v: trace vmbus_onoffer()

Add tracepoint to CHANNELMSG_OFFERCHANNEL handler. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/channel_mgmt.c | 2 ++ drivers/hv/hv_trace.h | 37 + 2 files changed, 39 insertions(+) diff --git a/drivers/hv/channel_mg

[PATCH 07/16] hyper-v: trace vmbus_ongpadl_torndown()

Add tracepoint to CHANNELMSG_GPADL_TORNDOWN handler. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/channel_mgmt.c | 2 ++ drivers/hv/hv_trace.h | 8 2 files changed, 10 insertions(+) diff --git a/drivers/hv/channel_mgmt.c b/drivers/hv/channel_mgmt.c

[PATCH 06/16] hyper-v: trace vmbus_ongpadl_created()

Add tracepoint to CHANNELMSG_GPADL_CREATED handler. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/channel_mgmt.c | 2 ++ drivers/hv/hv_trace.h | 17 + 2 files changed, 19 insertions(+) diff --git a/drivers/hv/channel_mgmt.c b/drivers/hv/channel_

[PATCH 05/16] hyper-v: trace vmbus_onopen_result()

Add tracepoint to CHANNELMSG_OPENCHANNEL_RESULT handler. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/channel_mgmt.c | 2 ++ drivers/hv/hv_trace.h | 17 + 2 files changed, 19 insertions(+) diff --git a/drivers/hv/channel_mgmt.c b/driv

[PATCH 02/16] hyper-v: trace vmbus_on_message()

Add tracepoint to vmbus_on_message() which is called when we start processing a blocking from work context. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/channel_mgmt.c | 2 ++ drivers/hv/hv_trace.h | 8 2 files changed, 10 insertions(+) diff --git a/d

[PATCH 00/16] Hyper-V: add tracing to VMBus module and trace all messages

.. 465.932709: vmbus_close_internal: sending child_relid 0x10, ret 0 kworker/5:4-2535 [005] 465.986417: vmbus_release_relid: sending child_relid 0x10, ret 0 CHANNELMSG_UNLOAD/CHANNELMSG_UNLOAD_RESPONSE are not traced as these are mostly used on crash. Vitaly Kuznetsov (16): hyper-v: trace v

[PATCH 01/16] hyper-v: trace vmbus_on_msg_dpc()

Add tracing subsystem to Hyper-V VMBus module and add tracepoint to vmbus_on_msg_dpc() which is called when we receive a message from host. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- drivers/hv/Makefile | 4 +++- drivers/hv/hv_trace.c | 4 drivers/hv/hv_t

Re: [PATCH 1/1] Drivers: hv: vmbus: Fix rescind handling issues

Vitaly Kuznetsov <vkuzn...@redhat.com> writes: > > Reverting 6f3d791f300618caf82a2be0c27456edd76d5164 still helps. In addition to the above I got the following crash while playing with 4.14-rc1 (unmodified): [ 55.810080] kernel tried to execute NX-protected page - exploit attem

Re: [PATCH 1/1] Drivers: hv: vmbus: Fix rescind handling issues

Stephen Hemminger <step...@networkplumber.org> writes: > On Sep 15, 2017 11:01 AM, "KY Srinivasan" <k...@microsoft.com> wrote: > > > Vitaly Kuznetsov <vkuzn...@redhat.com> writes: > > > > > > > > I'm see

Re: [PATCH 1/1] Drivers: hv: vmbus: Fix rescind handling issues

Vitaly Kuznetsov <vkuzn...@redhat.com> writes: > Stephen Hemminger <step...@networkplumber.org> writes: > >> On Fri, 11 Aug 2017 10:03:59 -0700 >> k...@exchange.microsoft.com wrote: >> >>> From: K. Y. Srinivasan <k...@microsoft.com> >>

Re: [PATCH 1/1] Drivers: hv: vmbus: Fix rescind handling issues

Stephen Hemminger writes: > On Fri, 11 Aug 2017 10:03:59 -0700 > k...@exchange.microsoft.com wrote: > >> From: K. Y. Srinivasan >> >> This patch handles the following issues that were observed when we are >> handling racing channel offer message

[PATCH] x86/hyper-v: Remove doubled HV_X64_EX_PROCESSOR_MASKS_RECOMMENDED definition

Commit 7dcf90e9e032 ("PCI: hv: Use vPCI protocol version 1.2") and commit 628f54cc6451 ("x86/hyper-v: Support extended CPU ranges for TLB flush hypercalls") adding the same definition came through different trees. Fix the glitch. Signed-off-by: Vitaly Kuznetsov <vkuzn...@re

Re: [PATCH v10 0/9] Hyper-V: paravirtualized remote TLB flushing and hypercall improvements

Ingo Molnar <mi...@kernel.org> writes: > * Vitaly Kuznetsov <vkuzn...@redhat.com> wrote: > >> > Changes since v9: >> > - Rebase to 4.13-rc3. >> > - Drop PATCH1 as it was already taken by Greg to char-misc tree. There're >> > no >> &g

Re: [PATCH v10 0/9] Hyper-V: paravirtualized remote TLB flushing and hypercall improvements

Vitaly Kuznetsov <vkuzn...@redhat.com> writes: > Changes since v9: > - Rebase to 4.13-rc3. > - Drop PATCH1 as it was already taken by Greg to char-misc tree. There're no > functional dependencies on this patch so the series can go through a > different tree > (and it

Re: [PATCH] tools: hv: handle EINTR in hv_fcopy_daemon

Olaf Hering writes: > If strace attaches to the daemon pread returns with EINTR, and the > process exits. Catch this case and continue with the next iteration. > > Signed-off-by: Olaf Hering > --- > tools/hv/hv_fcopy_daemon.c | 2 ++ > 1 file changed, 2

Re: [PATCH v10 0/9] Hyper-V: paravirtualized remote TLB flushing and hypercall improvements

Ingo Molnar writes: > I'm getting this build failure with this series: > > arch/x86/hyperv/mmu.c: In function ‘hyperv_setup_mmu_ops’: > arch/x86/hyperv/mmu.c:256:3: error: ‘pv_mmu_ops’ undeclared (first use in > this > function) >pv_mmu_ops.flush_tlb_others =

Re: [PATCH v10 0/9] Hyper-V: paravirtualized remote TLB flushing and hypercall improvements

Ingo Molnar <mi...@kernel.org> writes: > * Vitaly Kuznetsov <vkuzn...@redhat.com> wrote: > >> Vitaly Kuznetsov <vkuzn...@redhat.com> writes: >> >> > Changes since v9: >> > - Rebase to 4.13-rc3. >> > - Drop PAT

Re: [PATCH v10 0/9] Hyper-V: paravirtualized remote TLB flushing and hypercall improvements

Vitaly Kuznetsov <vkuzn...@redhat.com> writes: > Changes since v9: > - Rebase to 4.13-rc3. > - Drop PATCH1 as it was already taken by Greg to char-misc tree. There're no > functional dependencies on this patch so the series can go through a > different tree > (and it

Re: [PATCH net-next 0/1] netvsc: another VF datapath fix

Stephen Hemminger <step...@networkplumber.org> writes: > On Tue, 08 Aug 2017 17:24:03 +0200 > Vitaly Kuznetsov <vkuzn...@redhat.com> wrote: > >> Stephen Hemminger <step...@networkplumber.org> writes: >> >> > On Tue, 08 Aug 2017 16:03:56 +0200 &

Re: [PATCH net-next 0/1] netvsc: another VF datapath fix

Stephen Hemminger <step...@networkplumber.org> writes: > On Tue, 08 Aug 2017 16:03:56 +0200 > Vitaly Kuznetsov <vkuzn...@redhat.com> wrote: > >> Stephen Hemminger <step...@networkplumber.org> writes: >> >> > Previous fix was incomplete. >&

Re: [PATCH net-next 0/1] netvsc: another VF datapath fix

Stephen Hemminger writes: > Previous fix was incomplete. > Not really related to this patch series (which btw fixes my issue, thanks!), but I found one addition issue. Systemd fails to rename VF interface: kernel: mlx4_core 0002:00:02.0 eth2: joined to eth1

Re: [PATCH net-next 1/1] netvsc: fix rtnl deadlock on unregister of vf

Vitaly Kuznetsov <vkuzn...@redhat.com> writes: > Vitaly Kuznetsov <vkuzn...@redhat.com> writes: > >> Stephen Hemminger <step...@networkplumber.org> writes: >> >>> With new transparent VF support, it is possible to get a deadlock >>> when some

Re: [PATCH net-next 1/1] netvsc: fix rtnl deadlock on unregister of vf

Vitaly Kuznetsov <vkuzn...@redhat.com> writes: > Stephen Hemminger <step...@networkplumber.org> writes: > >> With new transparent VF support, it is possible to get a deadlock >> when some of the deferred work is running and the unregister_vf >> is trying to c

Re: [PATCH net-next 1/1] netvsc: fix rtnl deadlock on unregister of vf

similar to bonding and team device). > > Reported-by: Vitaly Kuznetsov <vkuzn...@redhat.com> > Fixes: 0c195567a8f6 ("netvsc: transparent VF management") > Signed-off-by: Stephen Hemminger <sthem...@microsoft.com> > --- > drivers/net/hyperv/netvsc_drv.

[PATCH v10 9/9] tracing/hyper-v: trace hyperv_mmu_flush_tlb_others()

Add Hyper-V tracing subsystem and trace hyperv_mmu_flush_tlb_others(). Tracing is done the same way we do xen_mmu_flush_tlb_others(). Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> Reviewed-by: Stephen Hemm

[PATCH v10 8/9] x86/hyper-v: support extended CPU ranges for TLB flush hypercalls

Hyper-V hosts may support more than 64 vCPUs, we need to use HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE_EX/LIST_EX hypercalls in this case. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> Reviewed-by: Stephen Hemminger <sthem..

[PATCH v10 7/9] x86/hyper-v: use hypercall for remote TLB flush

for optimization on the host side as both hypercall and native tlb flush will result in vmexit. The hint is also not set on modern Hyper-V versions. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> Reviewed-by: Stephen Hemm

[PATCH v10 3/9] x86/hyper-v: fast hypercall implementation

Hyper-V supports 'fast' hypercalls when all parameters are passed through registers. Implement an inline version of a simpliest of these calls: hypercall with one 8-byte input and no output. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <and

[PATCH v10 4/9] hyper-v: use fast hypercall for HVCALL_SIGNAL_EVENT

We need to pass only 8 bytes of input for HvSignalEvent which makes it a perfect fit for fast hypercall. hv_input_signal_event_buffer is not needed any more and hv_input_signal_event is converted to union for convenience. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by

[PATCH v10 6/9] hyper-v: globalize vp_index

To support implementing remote TLB flushing on Hyper-V with a hypercall we need to make vp_index available outside of vmbus module. Rename and globalize. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> Reviewed-by: Steph

[PATCH v10 5/9] x86/hyper-v: implement rep hypercalls

for HvFlushVirtualAddressList hypercall for remote TLB flushing. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> Reviewed-by: Stephen Hemminger <sthem...@microsoft.com> --- arch/x86/include/asm/mshyperv.h | 39

[PATCH v10 2/9] x86/hyper-v: make hv_do_hypercall() inline

hypercall and with inlining this is somewhat important, add the clobbers. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> Reviewed-by: Stephen Hemminger <sthem...@microsoft.com> --- arch/x86/hyperv/hv

[PATCH v10 1/9] x86/hyper-v: include hyperv/ only when CONFIG_HYPERV is set

Code is arch/x86/hyperv/ is only needed when CONFIG_HYPERV is set, the 'basic' support and detection lives in arch/x86/kernel/cpu/mshyperv.c which is included when CONFIG_HYPERVISOR_GUEST is set. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <and

[PATCH v10 0/9] Hyper-V: paravirtualized remote TLB flushing and hypercall improvements

hypercalls implementation, hyperv tracing subsystem (which only traces the newly added remote TLB flush for now). Vitaly Kuznetsov (9): x86/hyper-v: include hyperv/ only when CONFIG_HYPERV is set x86/hyper-v: make hv_do_hypercall() inline x86/hyper-v: fast hypercall implementation hyper-v: use fast

[PATCH v9 10/10] tracing/hyper-v: trace hyperv_mmu_flush_tlb_others()

Add Hyper-V tracing subsystem and trace hyperv_mmu_flush_tlb_others(). Tracing is done the same way we do xen_mmu_flush_tlb_others(). Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> Reviewed-by: Stephen Hemm

[PATCH v9 05/10] hyper-v: use fast hypercall for HVCALL_SIGNAL_EVENT

We need to pass only 8 bytes of input for HvSignalEvent which makes it a perfect fit for fast hypercall. hv_input_signal_event_buffer is not needed any more and hv_input_signal_event is converted to union for convenience. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by

[PATCH v9 09/10] x86/hyper-v: support extended CPU ranges for TLB flush hypercalls

Hyper-V hosts may support more than 64 vCPUs, we need to use HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE_EX/LIST_EX hypercalls in this case. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> Reviewed-by: Stephen Hemminger <sthem..

[PATCH v9 06/10] x86/hyper-v: implement rep hypercalls

for HvFlushVirtualAddressList hypercall for remote TLB flushing. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> Reviewed-by: Stephen Hemminger <sthem...@microsoft.com> --- arch/x86/include/asm/mshyperv.h | 39

[PATCH v9 04/10] x86/hyper-v: fast hypercall implementation

Hyper-V supports 'fast' hypercalls when all parameters are passed through registers. Implement an inline version of a simpliest of these calls: hypercall with one 8-byte input and no output. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <and

[PATCH v9 08/10] x86/hyper-v: use hypercall for remote TLB flush

for optimization on the host side as both hypercall and native tlb flush will result in vmexit. The hint is also not set on modern Hyper-V versions. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> Reviewed-by: Stephen Hemm

[PATCH v9 07/10] hyper-v: globalize vp_index

To support implementing remote TLB flushing on Hyper-V with a hypercall we need to make vp_index available outside of vmbus module. Rename and globalize. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> Reviewed-by: Steph

[PATCH v9 03/10] x86/hyper-v: make hv_do_hypercall() inline

hypercall and with inlining this is somewhat important, add the clobbers. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> Reviewed-by: Stephen Hemminger <sthem...@microsoft.com> --- arch/x86/hyperv/hv

[PATCH v9 00/10] Hyper-V: paravirtualized remote TLB flushing and hypercall improvements

2m9.948s This series brings a number of small improvements along the way: fast hypercall implementation and using it for event signaling, rep hypercalls implementation, hyperv tracing subsystem (which only traces the newly added remote TLB flush for now). Vitaly Kuznetsov (10): x86/hyper-v

[PATCH v9 01/10] x86/hyper-v: include hyperv/ only when CONFIG_HYPERV is set

Code is arch/x86/hyperv/ is only needed when CONFIG_HYPERV is set, the 'basic' support and detection lives in arch/x86/kernel/cpu/mshyperv.c which is included when CONFIG_HYPERVISOR_GUEST is set. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <and

[PATCH v9 02/10] x86/hyper-v: stash the max number of virtual/logical processor

Max virtual processor will be needed for 'extended' hypercalls supporting more than 64 vCPUs. While on it, unify on 'Hyper-V' in mshyperv.c as we currently have a mix, report acquired misc features as well. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shev

Re: [PATCH v3 08/10] x86/hyper-v: use hypercall for remote TLB flush

Andy Lutomirski <l...@kernel.org> writes: > On Tue, May 23, 2017 at 5:36 AM, Vitaly Kuznetsov <vkuzn...@redhat.com> wrote: >> Andy Lutomirski <l...@kernel.org> writes: >> >>> >>> Also, can you share the benchmark you used for these patches

Re: [PATCH 08/10] x86/hyper-v: use hypercall for remote TLB flush

kbuild test robot writes: > Hi Vitaly, > > [auto build test ERROR on next-20170619] > [cannot apply to tip/x86/core linus/master linux/master v4.12-rc6 v4.12-rc5 > v4.12-rc4 v4.12-rc7] > [if your patch is applied to the wrong git tree, please drop us a note to > help improve

[PATCH v2 2/2] hyper-v: read TSC frequency from a synthetic MSR

a error prone calibration. Reported-by: Ladi Prosek <lpro...@redhat.com> Tested-by: Ladi Prosek <lpro...@redhat.com> Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- arch/x86/kernel/cpu/mshyperv.c | 15 +++ 1 file changed, 15 insertions(+) diff --git a/a

[PATCH v2 1/2] hyper-v: check frequency MSRs presence according to the specification

accordingly. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- arch/x86/include/uapi/asm/hyperv.h | 15 ++- arch/x86/kernel/cpu/mshyperv.c | 3 ++- 2 files changed, 8 insertions(+), 10 deletions(-) diff --git a/arch/x86/include/uapi/asm/hyperv.h b/arch/x86/includ

[PATCH v2 0/2] hyper-v: avoid tsc calibration

information we can easily avoid calibration. This is already done for other hypervisors (KVM, Vmware). Vitaly Kuznetsov (2): hyper-v: check frequency MSRs presence according to the specification hyper-v: read TSC frequency from a synthetic MSR arch/x86/include/uapi/asm/hyperv.h | 15

[PATCH 2/2] hyper-v: read TSC frequency from a synthetic MSR

a error prone calibration. Reported-by: Ladi Prosek <lpro...@redhat.com> Tested-by: Ladi Prosek <lpro...@redhat.com> Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- arch/x86/kernel/cpu/mshyperv.c | 15 +++ 1 file changed, 15 insertions(+) diff --git a/a

[PATCH 1/2] hyper-v: check frequency MSRs presence according to the specification

accordingly. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- arch/x86/include/uapi/asm/hyperv.h | 15 ++- arch/x86/kernel/cpu/mshyperv.c | 3 ++- 2 files changed, 8 insertions(+), 10 deletions(-) diff --git a/arch/x86/include/uapi/asm/hyperv.h b/arch/x86/includ

[PATCH 0/2] hyper-v: avoid tsc calibration

TSC calibration on virtual machines is always error prone. It was found that in nested environments Gen2 instances may get stuck on boot. As Hyper-V hosts provide us with all the required information we can easily avoid calibration. This is already done for other hypervisors (KVM, Vmware). Vitaly

Re: [PATCH v8 07/10] hyper-v: globalize vp_index

Stephen Hemminger <step...@networkplumber.org> writes: > On Wed, 14 Jun 2017 04:31:32 + > Jork Loeser <jork.loe...@microsoft.com> wrote: > >> > From: Vitaly Kuznetsov [mailto:vkuzn...@redhat.com] >> > Sent: Tuesday, June 13, 2017 19:29 >> > &g

Re: [PATCH v8 07/10] hyper-v: globalize vp_index

Stephen Hemminger <step...@networkplumber.org> writes: > On Fri, 9 Jun 2017 15:27:33 +0200 > Vitaly Kuznetsov <vkuzn...@redhat.com> wrote: > >> To support implementing remote TLB flushing on Hyper-V with a hypercall >> we need to make vp_index availabl

Re: [PATCH v8 10/10] tracing/hyper-v: trace hyperv_mmu_flush_tlb_others()

Steven Rostedt writes: > On Fri, 09 Jun 2017 20:53:53 +0200 > Paul Bolle wrote: > >> On Fri, 2017-06-09 at 14:32 -0400, Steven Rostedt wrote: >> > I'm sure it works, but it just adds one more way of doing the same >> > thing. I thought that was what perl

[PATCH v8 10/10] tracing/hyper-v: trace hyperv_mmu_flush_tlb_others()

Add Hyper-V tracing subsystem and trace hyperv_mmu_flush_tlb_others(). Tracing is done the same way we do xen_mmu_flush_tlb_others(). Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- MAINTAINERS | 1 + arch/x86/hyperv/mmu.c | 7 +++ ar

[PATCH v8 08/10] x86/hyper-v: use hypercall for remote TLB flush

for optimization on the host side as both hypercall and native tlb flush will result in vmexit. The hint is also not set on modern Hyper-V versions. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- arch/x86/hyperv/Makefile | 2 +- arch/x86/hyperv/hv_init.c | 2 + arch/x86/

[PATCH v8 09/10] x86/hyper-v: support extended CPU ranges for TLB flush hypercalls

Hyper-V hosts may support more than 64 vCPUs, we need to use HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE_EX/LIST_EX hypercalls in this case. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- Changes since v7: - Drop explicit casting from virt_to_phys() [Andy Shevchenko] - Re-format ma

[PATCH v8 07/10] hyper-v: globalize vp_index

To support implementing remote TLB flushing on Hyper-V with a hypercall we need to make vp_index available outside of vmbus module. Rename and globalize. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> --- arch/x86/hype

[PATCH v8 06/10] x86/hyper-v: implement rep hypercalls

for HvFlushVirtualAddressList hypercall for remote TLB flushing. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> --- arch/x86/include/asm/mshyperv.h | 39 +++ 1 file changed, 39 insertions(+) diff --gi

[PATCH v8 04/10] x86/hyper-v: fast hypercall implementation

Hyper-V supports 'fast' hypercalls when all parameters are passed through registers. Implement an inline version of a simpliest of these calls: hypercall with one 8-byte input and no output. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <and

[PATCH v8 05/10] hyper-v: use fast hypercall for HVCALL_SIGNAL_EVENT

We need to pass only 8 bytes of input for HvSignalEvent which makes it a perfect fit for fast hypercall. hv_input_signal_event_buffer is not needed any more and hv_input_signal_event is converted to union for convenience. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by

[PATCH v8 03/10] x86/hyper-v: make hv_do_hypercall() inline

hypercall and with inlining this is somewhat important, add the clobbers. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> --- Changes since v7: - Use U64_MAX instead of (u64)ULLONG_MAX [Andy Shevchenko] --- arch/x86/hype

[PATCH v8 00/10] Hyper-V: paravirtualized remote TLB flushing and hypercall improvements

along the way: fast hypercall implementation and using it for event signaling, rep hypercalls implementation, hyperv tracing subsystem (which only traces the newly added remote TLB flush for now). Vitaly Kuznetsov (10): x86/hyper-v: include hyperv/ only when CONFIG_HYPERV is set x86/hyper-v

[PATCH v8 01/10] x86/hyper-v: include hyperv/ only when CONFIG_HYPERV is set

Code is arch/x86/hyperv/ is only needed when CONFIG_HYPERV is set, the 'basic' support and detection lives in arch/x86/kernel/cpu/mshyperv.c which is included when CONFIG_HYPERVISOR_GUEST is set. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <and

[PATCH v8 02/10] x86/hyper-v: stash the max number of virtual/logical processor

Max virtual processor will be needed for 'extended' hypercalls supporting more than 64 vCPUs. While on it, unify on 'Hyper-V' in mshyperv.c as we currently have a mix, report acquired misc features as well. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shev

Re: [PATCH v4 10/10] tracing/hyper-v: trace hyperv_mmu_flush_tlb_others()

Steven Rostedt writes: >> >> I'm pretty sure I tested tracing this even before sending v2 of this >> series, I'll retest before sending v7. > > Even if it does work, it's still fragile as it uses an > no-longer-supported framework. > I hope I fixed it in v7 of my patchset,

[PATCH v7 10/10] tracing/hyper-v: trace hyperv_mmu_flush_tlb_others()

Add Hyper-V tracing subsystem and trace hyperv_mmu_flush_tlb_others(). Tracing is done the same way we do xen_mmu_flush_tlb_others(). Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- Changes since v6: - Drop DEFINE_TRACE() from mmu.c, change tracing magic a bit to keep things com

[PATCH v7 08/10] x86/hyper-v: use hypercall for remote TLB flush

for optimization on the host side as both hypercall and native tlb flush will result in vmexit. The hint is also not set on modern Hyper-V versions. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- arch/x86/hyperv/Makefile | 2 +- arch/x86/hyperv/hv_init.c | 2 + arch/x86/

[PATCH v7 06/10] x86/hyper-v: implement rep hypercalls

for HvFlushVirtualAddressList hypercall for remote TLB flushing. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> --- arch/x86/include/asm/mshyperv.h | 39 +++ 1 file changed, 39 insertions(+) diff --gi

[PATCH v7 09/10] x86/hyper-v: support extended CPU ranges for TLB flush hypercalls

Hyper-V hosts may support more than 64 vCPUs, we need to use HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE_EX/LIST_EX hypercalls in this case. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- Changes since v6: - Fix a bug in cpumask_to_vp_set() by setting valid_bank_mask right, add a

[PATCH v7 07/10] hyper-v: globalize vp_index

To support implementing remote TLB flushing on Hyper-V with a hypercall we need to make vp_index available outside of vmbus module. Rename and globalize. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> --- arch/x86/hype

[PATCH v7 02/10] x86/hyper-v: stash the max number of virtual/logical processor

Max virtual processor will be needed for 'extended' hypercalls supporting more than 64 vCPUs. While on it, unify on 'Hyper-V' in mshyperv.c as we currently have a mix, report acquired misc features as well. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shev

[PATCH v7 03/10] x86/hyper-v: make hv_do_hypercall() inline

hypercall and with inlining this is somewhat important, add the clobbers. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> --- arch/x86/hyperv/hv_init.c | 54 - arch/x86/include/as

[PATCH v7 05/10] hyper-v: use fast hypercall for HVCALL_SIGNAL_EVENT

We need to pass only 8 bytes of input for HvSignalEvent which makes it a perfect fit for fast hypercall. hv_input_signal_event_buffer is not needed any more and hv_input_signal_event is converted to union for convenience. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by

[PATCH v7 01/10] x86/hyper-v: include hyperv/ only when CONFIG_HYPERV is set

Code is arch/x86/hyperv/ is only needed when CONFIG_HYPERV is set, the 'basic' support and detection lives in arch/x86/kernel/cpu/mshyperv.c which is included when CONFIG_HYPERVISOR_GUEST is set. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <and

[PATCH v7 00/10] Hyper-V: paravirtualized remote TLB flushing and hypercall improvements

lls implementation, hyperv tracing subsystem (which only traces the newly added remote TLB flush for now). Vitaly Kuznetsov (10): x86/hyper-v: include hyperv/ only when CONFIG_HYPERV is set x86/hyper-v: stash the max number of virtual/logical processor x86/hyper-v: make hv_do_hypercall() inline

Re: [PATCH v4 10/10] tracing/hyper-v: trace hyperv_mmu_flush_tlb_others()

Steven Rostedt <rost...@goodmis.org> writes: > On Wed, 24 May 2017 14:04:05 +0200 > Vitaly Kuznetsov <vkuzn...@redhat.com> wrote: > >> Add Hyper-V tracing subsystem and trace hyperv_mmu_flush_tlb_others(). >> Tracing is done the same way we do xen_mmu_flus

[PATCH v6 09/10] x86/hyper-v: support extended CPU ranges for TLB flush hypercalls

Hyper-V hosts may support more than 64 vCPUs, we need to use HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE_EX/LIST_EX hypercalls in this case. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- Changes since v5: - Do single pass in cpumask_to_vp_set() as it turns our that empty banks are

[PATCH v6 07/10] hyper-v: globalize vp_index

To support implementing remote TLB flushing on Hyper-V with a hypercall we need to make vp_index available outside of vmbus module. Rename and globalize. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> --- Changes sinc

[PATCH v6 10/10] tracing/hyper-v: trace hyperv_mmu_flush_tlb_others()

Add Hyper-V tracing subsystem and trace hyperv_mmu_flush_tlb_others(). Tracing is done the same way we do xen_mmu_flush_tlb_others(). Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevche...@gmail.com> --- MAINTAINERS

[PATCH v6 08/10] x86/hyper-v: use hypercall for remote TLB flush

for optimization on the host side as both hypercall and native tlb flush will result in vmexit. The hint is also not set on modern Hyper-V versions. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> --- Changes since v5: - Use BIT() in asm/hyperv.h [Andy Shevchenko] - Split off fill_gva_list()

[PATCH v6 04/10] x86/hyper-v: fast hypercall implementation

Hyper-V supports 'fast' hypercalls when all parameters are passed through registers. Implement an inline version of a simpliest of these calls: hypercall with one 8-byte input and no output. Signed-off-by: Vitaly Kuznetsov <vkuzn...@redhat.com> Reviewed-by: Andy Shevchenko <and

<    1   2   3   4   5   6   7   8   9   10   >