Added logging to help troubleshoot common ballooning, hot add,
and versioning issues.
Signed-off-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If the guest does not support memory hotplugging, it should respond to
the host with zero pages added and successful result code. This signals
to the host that hotplugging is not supported and the host will avoid
sending future hot-add requests.
Signed-off-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We should intentionally declare the protocols to use for every known host
and default to using the latest protocol if the host is unknown or new.
Signed-off-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
I discovered that at least WS2016TP5 host has 60 seconds timeout for the
ICMSGTYPE_NEGOTIATE message so we need to lower guest's timeout a little
bit to make sure we always respond in time. Let's make it 55 seconds.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In commit 9a56e5d6a0ba ("Drivers: hv: make VMBus bus ids persistent")
the name of vmbus devices in sysfs changed to be (in 4.9-rc1):
/sys/bus/vmbus/vmbus-6aebe374-9ba0-11e6-933c-00259086b36b
The prefix ("vmbus-") is redundant and differs from how PCI is
represented in sysfs. Therefore simplify to:
/sys/bus/vmbus/6aebe374-9ba0-11e6-933c-00259086b36b
Please merge this before 4.9 is released and the old format
has to live forever.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The host keeps sending heartbeat packets independent of the
guest responding to them. Even though we respond to the heartbeat messages at
interrupt level, we can have situations where there maybe multiple heartbeat
messages pending that have not been responded to. For instance this occurs when the
VM is paused and the host continues to send the heartbeat messages.
Address this issue by draining and responding to all
the heartbeat messages that maybe pending.
Signed-off-by: Long Li <longli@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
CC: Stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The auto incremented counter is not being used anymore, get rid of it.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Some tools use bus ids to identify devices and they count on the fact
that these ids are persistent across reboot. This may be not true for
VMBus as we use auto incremented counter from alloc_channel() as such
id. Switch to using if_instance from channel offer, this id is supposed
to be persistent.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Under stress, we have seen allocation failure in time synch code. Avoid
this dynamic allocation.
Signed-off-by: Vivek Yadav <vyadav@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This enables support for more accurate TimeSync v4 samples when hosted
under Windows Server 2016 and newer hosts.
The new time samples include a "vmreferencetime" field that represents
the guest's TSC value when the host generated its time sample. This value
lets the guest calculate the latency in receiving the time sample. The
latency is added to the sample host time prior to updating the clock.
Signed-off-by: Alex Ng <alexng@messages.microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Only the first 50 samples after boot were being used to discipline the
clock. After the first 50 samples, any samples from the host were ignored
and the guest clock would eventually drift from the host clock.
This patch allows TimeSync-enabled guests to continuously synchronize the
clock with the host clock, even after the first 50 samples.
Signed-off-by: Alex Ng <alexng@messages.microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Different Windows host versions may reuse the same protocol version when
negotiating the TimeSync, Shutdown, and Heartbeat protocols. We should only
refer to the protocol version to avoid conflating the two concepts.
Signed-off-by: Alex Ng <alexng@messages.microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This fixes a sparse warning because hyperv_mmio resources
are only used in this one file and should be static.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Hyper-V host will send a VSS_OP_HOT_BACKUP request to check if guest is
ready for a live backup/snapshot. The driver should respond to the check
only if the daemon is running and listening to requests. This allows the
host to fallback to standard snapshots in case the VSS daemon is not
running.
Signed-off-by: Alex Ng <alexng@messages.microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Multiple VSS_OP_HOT_BACKUP requests may arrive in quick succession, even
though the host only signals once. The driver wass handling the first
request while ignoring the others in the ring buffer. We should poll the
VSS channel after handling a request to continue processing other requests.
Signed-off-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Introduce a mechanism to control how channels will be affinitized. We will
support two policies:
1. HV_BALANCED: All performance critical channels will be dstributed
evenly amongst all the available NUMA nodes. Once the Node is assigned,
we will assign the CPU based on a simple round robin scheme.
2. HV_LOCALIZED: Only the primary channels are distributed across all
NUMA nodes. Sub-channels will be in the same NUMA node as the primary
channel. This is the current behaviour.
The default policy will be the HV_BALANCED as it can minimize the remote
memory access on NUMA machines with applications that span NUMA nodes.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
With wrap around mappings for ring buffers we can always use a single
memcpy() to do the job.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Tested-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Make it possible to always use a single memcpy() or to provide a direct
link to a packet on the ring buffer by creating virtual mapping for two
copies of the ring buffer with vmap(). Utilize currently empty
hv_ringbuffer_cleanup() to do the unmap.
While on it, replace sizeof(struct hv_ring_buffer) check
in hv_ringbuffer_init() with BUILD_BUG_ON() as it is a compile time check.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In preparation for doing wrap around mappings for ring buffers cleanup
vmbus_open() function:
- check that ring sizes are PAGE_SIZE aligned (they are for all in-kernel
drivers now);
- kfree(open_info) on error only after we kzalloc() it (not an issue as it
is valid to call kfree(NULL);
- rename poorly named labels;
- use alloc_pages() instead of __get_free_pages() as we need struct page
pointer for future.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Reports for available memory should use the si_mem_available() value.
The previous freeram value does not include available page cache memory.
Signed-off-by: Alex Ng <alexng@messages.microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
lockdep reports possible circular locking dependency when udev is used
for memory onlining:
systemd-udevd/3996 is trying to acquire lock:
((memory_chain).rwsem){++++.+}, at: [<ffffffff810d137e>] __blocking_notifier_call_chain+0x4e/0xc0
but task is already holding lock:
(&dm_device.ha_region_mutex){+.+.+.}, at: [<ffffffffa015382e>] hv_memory_notifier+0x5e/0xc0 [hv_balloon]
...
which is probably a false positive because we take and release
ha_region_mutex from memory notifier chain depending on the arg. No real
deadlocks were reported so far (though I'm not really sure about
preemptible kernels...) but we don't really need to hold the mutex
for so long. We use it to protect ha_region_list (and its members) and the
num_pages_onlined counter. None of these operations require us to sleep
and nothing is slow, switch to using spinlock with interrupts disabled.
While on it, replace list_for_each -> list_for_each_entry as we actually
need entries in all these cases, drop meaningless list_empty() checks.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
With the recently introduced in-kernel memory onlining
(MEMORY_HOTPLUG_DEFAULT_ONLINE) these is no point in waiting for pages
to come online in the driver and we can get rid of the waiting.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
I'm observing the following hot add requests from the WS2012 host:
hot_add_req: start_pfn = 0x108200 count = 330752
hot_add_req: start_pfn = 0x158e00 count = 193536
hot_add_req: start_pfn = 0x188400 count = 239616
As the host doesn't specify hot add regions we're trying to create
128Mb-aligned region covering the first request, we create the 0x108000 -
0x160000 region and we add 0x108000 - 0x158e00 memory. The second request
passes the pfn_covered() check, we enlarge the region to 0x108000 -
0x190000 and add 0x158e00 - 0x188200 memory. The problem emerges with the
third request as it starts at 0x188400 so there is a 0x200 gap which is
not covered. As the end of our region is 0x190000 now it again passes the
pfn_covered() check were we just adjust the covered_end_pfn and make it
0x188400 instead of 0x188200 which means that we'll try to online
0x188200-0x188400 pages but these pages were never assigned to us and we
crash.
We can't react to such requests by creating new hot add regions as it may
happen that the whole suggested range falls into the previously identified
128Mb-aligned area so we'll end up adding nothing or create intersecting
regions and our current logic doesn't allow that. Instead, create a list of
such 'gaps' and check for them in the page online callback.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Windows 2012 (non-R2) does not specify hot add region in hot add requests
and the logic in hot_add_req() is trying to find a 128Mb-aligned region
covering the request. It may also happen that host's requests are not 128Mb
aligned and the created ha_region will start before the first specified
PFN. We can't online these non-present pages but we don't remember the real
start of the region.
This is a regression introduced by the commit 5abbbb75d7 ("Drivers: hv:
hv_balloon: don't lose memory when onlining order is not natural"). While
the idea of keeping the 'moving window' was wrong (as there is no guarantee
that hot add requests come ordered) we should still keep track of
covered_start_pfn. This is not a revert, the logic is different.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On Hyper-V, performance critical channels use the monitor
mechanism to signal the host when the guest posts mesages
for the host. This mechanism minimizes the hypervisor intercepts
and also makes the host more efficient in that each time the
host is woken up, it processes a batch of messages as opposed to
just one. The goal here is improve the throughput and this is at
the expense of increased latency.
Implement a mechanism to let the client driver decide if latency
is important.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The current delay between retries is unnecessarily high and is negatively
affecting the time it takes to boot the system.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
For synthetic NIC channels, enable explicit signaling policy as netvsc wants to
explicitly control when the host is to be signaled.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
There is a rare race when we remove an entry from the global list
hv_context.percpu_list[cpu] in hv_process_channel_removal() ->
percpu_channel_deq() -> list_del(): at this time, if vmbus_on_event() ->
process_chn_event() -> pcpu_relid2channel() is trying to query the list,
we can get the kernel fault.
Similarly, we also have the issue in the code path: vmbus_process_offer() ->
percpu_channel_enq().
We can resolve the issue by disabling the tasklet when updating the list.
The patch also moves vmbus_release_relid() to a later place where
the channel has been removed from the per-cpu and the global lists.
Reported-by: Rolf Neugebauer <rolf.neugebauer@docker.com>
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Background: userspace daemons registration protocol for Hyper-V utilities
drivers has two steps:
1) daemon writes its own version to kernel
2) kernel reads it and replies with module version
at this point we consider the handshake procedure being completed and we
do hv_poll_channel() transitioning the utility device to HVUTIL_READY
state. At this point we're ready to handle messages from kernel.
When hvutil_transport is in HVUTIL_TRANSPORT_CHARDEV mode we have a
single buffer for outgoing message. hvutil_transport_send() puts to this
buffer and till the buffer is cleared with hvt_op_read() returns -EFAULT
to all consequent calls. Host<->guest protocol guarantees there is no more
than one request at a time and we will not get new requests till we reply
to the previous one so this single message buffer is enough.
Now to the race. When we finish negotiation procedure and send kernel
module version to userspace with hvutil_transport_send() it goes into the
above mentioned buffer and if the daemon is slow enough to read it from
there we can get a collision when a request from the host comes, we won't
be able to put anything to the buffer so the request will be lost. To
solve the issue we need to know when the negotiation is really done (when
the version message is read by the daemon) and transition to HVUTIL_READY
state after this happens. Implement a callback on read to support this.
Old style netlink communication is not affected by the change, we don't
really know when these messages are delivered but we don't have a single
message buffer there.
Reported-by: Barry Davis <barry_davis@stormagic.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
vmbus_teardown_gpadl() can result in infinite wait when it is called on 5
second timeout in vmbus_open(). The issue is caused by the fact that gpadl
teardown operation won't ever succeed for an opened channel and the timeout
isn't always enough. As a guest, we can always trust the host to respond to
our request (and there is nothing we can do if it doesn't).
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In some cases create_gpadl_header() allocates submessages but we never
free them.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We use messagecount only once in vmbus_establish_gpadl() to check if
it is safe to iterate through the submsglist. We can just initialize
the list header in all cases in create_gpadl_header() instead.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When we crash from NMI context (e.g. after NMI injection from host when
'sysctl -w kernel.unknown_nmi_panic=1' is set) we hit
kernel BUG at mm/vmalloc.c:1530!
as vfree() is denied. While the issue could be solved with in_nmi() check
instead I opted for skipping vfree on all sorts of crashes to reduce the
amount of work which can cause consequent crashes. We don't really need to
free anything on crash.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Hyper-V Linux Integration Services use the VMBus implementation for
communication with the Hypervisor. VMBus registers its own interrupt
handler that completely bypasses the common Linux interrupt handling.
This implies that the interrupt entropy collector is not triggered.
This patch adds the interrupt entropy collection callback into the VMBus
interrupt handler function.
Cc: stable@kernel.org
Signed-off-by: Stephan Mueller <stephan.mueller@atsec.com>
Signed-off-by: Stephan Mueller <smueller@chronox.de>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
We set host_specified_ha_region = true on certain request but this is a
global state which stays 'true' forever. We need to reset it when we
receive a request where ha_region is not specified. I did not see any
real issues, the bug was found by code inspection.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When we iterate through all HA regions in handle_pg_range() we have an
assumption that all these regions are sorted in the list and the
'start_pfn >= has->end_pfn' check is enough to find the proper region.
Unfortunately it's not the case with WS2016 where host can hot-add regions
in a different order. We end up modifying the wrong HA region and crashing
later on pages online. Modify the check to make sure we found the region
we were searching for while iterating. Fix the same check in pfn_covered()
as well.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Kdump keeps biting. Turns out CHANNELMSG_UNLOAD_RESPONSE is always
delivered to the CPU which was used for initial contact or to CPU0
depending on host version. vmbus_wait_for_unload() doesn't account for
the fact that in case we're crashing on some other CPU we won't get the
CHANNELMSG_UNLOAD_RESPONSE message and our wait on the current CPU will
never end.
Do the following:
1) Check for completion_done() in the loop. In case interrupt handler is
still alive we'll get the confirmation we need.
2) Read message pages for all CPUs message page as we're unsure where
CHANNELMSG_UNLOAD_RESPONSE is going to be delivered to. We can race with
still-alive interrupt handler doing the same, add cmpxchg() to
vmbus_signal_eom() to not lose CHANNELMSG_UNLOAD_RESPONSE message.
3) Cleanup message pages on all CPUs. This is required (at least for the
current CPU as we're clearing CPU0 messages now but we may want to bring
up additional CPUs on crash) as new messages won't be delivered till we
consume what's pending. On boot we'll place message pages somewhere else
and we won't be able to read stale messages.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Hyper-V VMs can be replicated to another hosts and there is a feature to
set different IP for replicas, it is called 'Failover TCP/IP'. When
such guest starts Hyper-V host sends it KVP_OP_SET_IP_INFO message as soon
as we finish negotiation procedure. The problem is that it can happen (and
it actually happens) before userspace daemon connects and we reply with
HV_E_FAIL to the message. As there are no repetitions we fail to set the
requested IP.
Solve the issue by postponing our reply to the negotiation message till
userspace daemon is connected. We can't wait too long as there is a
host-side timeout (cca. 75 seconds) and if we fail to reply in this time
frame the whole KVP service will become inactive. The solution is not
ideal - if it takes userspace daemon more than 60 seconds to connect
IP Failover will still fail but I don't see a solution with our current
separation between kernel and userspace parts.
Other two modules (VSS and FCOPY) don't require such delay, leave them
untouched.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Simplify the logic that picks MMIO ranges by pulling out the
logic related to trying to lay frame buffer claim on top of where
the firmware placed the frame buffer.
Signed-off-by: Jake Oshins <jakeo@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Later in the boot sequence, we need to figure out which memory
ranges can be given out to various paravirtual drivers. The
hyperv_fb driver should, ideally, be placed right on top of
the frame buffer, without some other device getting plopped on
top of this range in the meantime. Recording this now allows
that to be guaranteed.
Signed-off-by: Jake Oshins <jakeo@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch changes vmbus_allocate_mmio() and vmbus_free_mmio() so
that when child paravirtual devices allocate memory-mapped I/O
space, they allocate it privately from a resource tree pointed
at by hyperv_mmio and also by the public resource tree
iomem_resource. This allows the region to be marked as "busy"
in the private tree, but a "bridge window" in the public tree,
guaranteeing that no two bridge windows will overlap each other
but while also allowing the PCI device children of the bridge
windows to overlap that window.
One might conclude that this belongs in the pnp layer, rather
than in this driver. Rafael Wysocki, the maintainter of the
pnp layer, has previously asked that we not modify the pnp layer
as it is considered deprecated. This patch is thus essentially
a workaround.
Signed-off-by: Jake Oshins <jakeo@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A patch later in this series allocates child nodes
in this resource tree. For that to work, this tree
needs to be sorted in ascending order.
Signed-off-by: Jake Oshins <jakeo@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch introduces a function that reverses everything
done by vmbus_allocate_mmio(). Existing code just called
release_mem_region(). Future patches in this series
require a more complex sequence of actions, so this function
is introduced to wrap those actions.
Signed-off-by: Jake Oshins <jakeo@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In existing code, this tree of resources is created
in single-threaded code and never modified after it is
created, and thus needs no locking. This patch introduces
a semaphore for tree access, as other patches in this
series introduce run-time modifications of this resource
tree which can happen on multiple threads.
Signed-off-by: Jake Oshins <jakeo@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Implement APIs for in-place consumption of vmbus packets. Currently, each
packet is copied and processed one at a time and as part of processing
each packet we potentially may signal the host (if it is waiting for
room to produce a packet).
These APIs help batched in-place processing of vmbus packets.
We also optimize host signaling by having a separate API to signal
the end of in-place consumption. With netvsc using these APIs,
on an iperf run on average I see about 20X reduction in checks to
signal the host.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In preparation for implementing APIs for in-place consumption of VMBUS
packets, movve some ring buffer functionality into hyperv.h
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In preparation for moving some ring buffer functionality out of the
vmbus driver, export the API for signaling the host.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Use the virt_xx barriers that have been defined for use in virtual machines.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Use the READ_ONCE macro to access variabes that can change asynchronously.
This is the recommended mechanism for dealing with "unsafe" compiler
optimizations.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Introduce separate functions for estimating how much can be read from
and written to the ring buffer.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On the consumer side, we have interrupt driven flow management of the
producer. It is sufficient to base the signaling decision on the
amount of space that is available to write after the read is complete.
The current code samples the previous available space and uses this
in making the signaling decision. This state can be stale and is
unnecessary. Since the state can be stale, we end up not signaling
the host (when we should) and this can result in a hang. Fix this
problem by removing the unnecessary check. I would like to thank
Arseney Romanenko <arseneyr@microsoft.com> for pointing out this issue.
Also, issue a full memory barrier before making the signaling descision
to correctly deal with potential reordering of the write (read index)
followed by the read of pending_sz.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Tested-by: Dexuan Cui <decui@microsoft.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Here is the big char/misc driver update for 4.6-rc1.
The majority of the patches here is hwtracing and some new mic drivers,
but there's a lot of other driver updates as well. Full details in the
shortlog.
All have been in linux-next for a while with no reported issues.
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2
iEYEABECAAYFAlbp9IcACgkQMUfUDdst+ykyJgCeLTC2QNGrh51kiJglkVJ0yD36
q4MAn0NkvSX2+iv5Jq8MaX6UQoRa4Nun
=MNjR
-----END PGP SIGNATURE-----
Merge tag 'char-misc-4.6-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc
Pull char/misc updates from Greg KH:
"Here is the big char/misc driver update for 4.6-rc1.
The majority of the patches here is hwtracing and some new mic
drivers, but there's a lot of other driver updates as well. Full
details in the shortlog.
All have been in linux-next for a while with no reported issues"
* tag 'char-misc-4.6-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc: (238 commits)
goldfish: Fix build error of missing ioremap on UM
nvmem: mediatek: Fix later provider initialization
nvmem: imx-ocotp: Fix return value of imx_ocotp_read
nvmem: Fix dependencies for !HAS_IOMEM archs
char: genrtc: replace blacklist with whitelist
drivers/hwtracing: make coresight-etm-perf.c explicitly non-modular
drivers: char: mem: fix IS_ERROR_VALUE usage
char: xillybus: Fix internal data structure initialization
pch_phub: return -ENODATA if ROM can't be mapped
Drivers: hv: vmbus: Support kexec on ws2012 r2 and above
Drivers: hv: vmbus: Support handling messages on multiple CPUs
Drivers: hv: utils: Remove util transport handler from list if registration fails
Drivers: hv: util: Pass the channel information during the init call
Drivers: hv: vmbus: avoid unneeded compiler optimizations in vmbus_wait_for_unload()
Drivers: hv: vmbus: remove code duplication in message handling
Drivers: hv: vmbus: avoid wait_for_completion() on crash
Drivers: hv: vmbus: don't loose HVMSG_TIMER_EXPIRED messages
misc: at24: replace memory_accessor with nvmem_device_read
eeprom: 93xx46: extend driver to plug into the NVMEM framework
eeprom: at25: extend driver to plug into the NVMEM framework
...
WS2012 R2 and above hosts can support kexec in that thay can support
reconnecting to the host (as would be needed in the kexec path)
on any CPU. Enable this. Pre ws2012 r2 hosts don't have this ability
and consequently cannot support kexec.
Signed-off-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Starting with Windows 2012 R2, message inteerupts can be delivered
on any VCPU in the guest. Support this functionality.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If util transport fails to initialize for any reason, the list of transport
handlers may become corrupted due to freeing the transport handler without
removing it from the list. Fix this by cleaning it up from the list.
Signed-off-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Pass the channel information to the util drivers that need to defer
reading the channel while they are processing a request. This would address
the following issue reported by Vitaly:
Commit 3cace4a616 ("Drivers: hv: utils: run polling callback always in
interrupt context") removed direct *_transaction.state = HVUTIL_READY
assignments from *_handle_handshake() functions introducing the following
race: if a userspace daemon connects before we get first non-negotiation
request from the server hv_poll_channel() won't set transaction state to
HVUTIL_READY as (!channel) condition will fail, we set it to non-NULL on
the first real request from the server.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Reported-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Message header is modified by the hypervisor and we read it in a loop,
we need to prevent compilers from optimizing accesses. There are no such
optimizations at this moment, this is just a future proof.
Suggested-by: Radim Krcmar <rkrcmar@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: Radim Kr.má<rkrcmar@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We have 3 functions dealing with messages and they all implement
the same logic to finalize reads, move it to vmbus_signal_eom().
Suggested-by: Radim Krcmar <rkrcmar@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: Radim Kr.má<rkrcmar@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
wait_for_completion() may sleep, it enables interrupts and this
is something we really want to avoid on crashes because interrupt
handlers can cause other crashes. Switch to the recently introduced
vmbus_wait_for_unload() doing busy wait instead.
Reported-by: Radim Krcmar <rkrcmar@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: Radim Kr.má<rkrcmar@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We must handle HVMSG_TIMER_EXPIRED messages in the interrupt context
and we offload all the rest to vmbus_on_msg_dpc() tasklet. This functions
loops to see if there are new messages pending. In case we'll ever see
HVMSG_TIMER_EXPIRED message there we're going to lose it as we can't
handle it from there. Avoid looping in vmbus_on_msg_dpc(), we're OK
with handling one message per interrupt.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: Radim Kr.má<rkrcmar@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
VMBus hypercall codes inside Hyper-V UAPI header will
be used by QEMU to implement VMBus host devices support.
Signed-off-by: Andrey Smetanin <asmetanin@virtuozzo.com>
Acked-by: K. Y. Srinivasan <kys@microsoft.com>
Reviewed-by: Roman Kagan <rkagan@virtuozzo.com>
CC: Gleb Natapov <gleb@kernel.org>
CC: Paolo Bonzini <pbonzini@redhat.com>
CC: Joerg Roedel <joro@8bytes.org>
CC: "K. Y. Srinivasan" <kys@microsoft.com>
CC: Haiyang Zhang <haiyangz@microsoft.com>
CC: Roman Kagan <rkagan@virtuozzo.com>
CC: Denis V. Lunev <den@openvz.org>
CC: qemu-devel@nongnu.org
[Do not rename the constant at the same time as moving it, as that
would cause semantic conflicts with the Hyper-V tree. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
On the channel send side, many of the VMBUS
device drivers explicity serialize access to the
outgoing ring buffer. Give more control to the
VMBUS device drivers in terms how to serialize
accesss to the outgoing ring buffer.
The default behavior will be to aquire the
ring lock to preserve the current behavior.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The function hv_ringbuffer_read() is called always on a pre-assigned
CPU. Each chnnel is bound to a specific CPU and this function is
always called on the CPU the channel is bound. There is no need to
acquire the spin lock; get rid of this overhead.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The hvsock driver needs this API to release all the resources related
to the channel.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This will be used by the coming hv_sock driver.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Only the coming hv_sock driver has a "true" value for this flag.
We treat the hvsock offers/channels as special VMBus devices.
Since the hv_sock driver handles all the hvsock offers/channels, we need to
tweak vmbus_match() for hv_sock driver, so we introduce this flag.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A function to send the type of message is also added.
The coming net/hvsock driver will use this function to proactively request
the host to offer a VMBus channel for a new hvsock connection.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When the hvsock channel's outbound ringbuffer is full (i.e.,
hv_ringbuffer_write() returns -EAGAIN), we should avoid the unnecessary
signaling the host.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
clocksource_change_rating() involves mutex usage and can't be called
in interrupt context. It also makes sense to avoid doing redundant work
on crash.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We have to call vmbus_initiate_unload() on crash to make kdump work but
the crash can also be happening in interrupt (e.g. Sysrq + c results in
such) where we can't schedule or the following will happen:
[ 314.905786] bad: scheduling from the idle thread!
Just skipping the wait (and even adding some random wait here) won't help:
to make host-side magic working we're supposed to receive CHANNELMSG_UNLOAD
(and actually confirm the fact that we received it) but we can't use
interrupt-base path (vmbus_isr()-> vmbus_on_msg_dpc()). Implement a simple
busy wait ignoring all the other messages and use it if we're in an
interrupt context.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When we pick a CPU to use for a new subchannel we try find a non-used one
on the appropriate NUMA node, we keep track of them with the
primary->alloced_cpus_in_node mask. Under normal circumstances we don't run
out of available CPUs but it is possible when we we don't initialize some
cpus in Linux, e.g. when we boot with 'nr_cpus=' limitation.
Avoid the infinite loop in init_vp_index() by checking that we still have
non-used CPUs in the alloced_cpus_in_node mask and resetting it in case
we don't.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add vendor and device attributes to VMBUS devices. These will be used
by Hyper-V tools as well user-level RDMA libraries that will use the
vendor/device tuple to discover the RDMA device.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cleanup vmbus_set_event() by inlining the hypercall to post
the event and since the return value of vmbus_set_event() is not checked,
make it void. As part of this cleanup, get rid of the function
hv_signal_event() as it is only callled from vmbus_set_event().
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Here's the big set of char/misc patches for 4.5-rc1.
Nothing major, lots of different driver subsystem updates, full details
in the shortlog. All of these have been in linux-next for a while.
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2
iEYEABECAAYFAlaV0GsACgkQMUfUDdst+ymlPgCg07GSc4SOWlUL2V36vQ0kucoO
YjAAoMfeUEhsf/NJ7iaAMGQVUQKuYVqr
=BlqH
-----END PGP SIGNATURE-----
Merge tag 'char-misc-4.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc
Pull char/misc updates from Greg KH:
"Here's the big set of char/misc patches for 4.5-rc1.
Nothing major, lots of different driver subsystem updates, full
details in the shortlog. All of these have been in linux-next for a
while"
* tag 'char-misc-4.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc: (71 commits)
mei: fix fasync return value on error
parport: avoid assignment in if
parport: remove unneeded space
parport: change style of NULL comparison
parport: remove unnecessary out of memory message
parport: remove braces
parport: quoted strings should not be split
parport: code indent should use tabs
parport: fix coding style
parport: EXPORT_SYMBOL should follow function
parport: remove trailing white space
parport: fix a trivial typo
coresight: Fix a typo in Kconfig
coresight: checking for NULL string in coresight_name_match()
Drivers: hv: vmbus: Treat Fibre Channel devices as performance critical
Drivers: hv: utils: fix hvt_op_poll() return value on transport destroy
Drivers: hv: vmbus: fix the building warning with hyperv-keyboard
extcon: add Maxim MAX3355 driver
Drivers: hv: ring_buffer: eliminate hv_ringbuffer_peek()
Drivers: hv: remove code duplication between vmbus_recvpacket()/vmbus_recvpacket_raw()
...
For performance critical devices, we distribute the incoming
channel interrupt load across available CPUs in the guest.
Include Fibre channel devices in the set of devices for which
we would distribute the interrupt load.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The return type of hvt_op_poll() is unsigned int and -EBADF is
inappropriate, poll functions return POLL* statuses.
Reported-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This struct is required for Hyper-V SynIC timers implementation inside KVM
and for upcoming Hyper-V VMBus support by userspace(QEMU). So place it into
Hyper-V UAPI header.
Signed-off-by: Andrey Smetanin <asmetanin@virtuozzo.com>
CC: Gleb Natapov <gleb@kernel.org>
CC: Paolo Bonzini <pbonzini@redhat.com>
CC: "K. Y. Srinivasan" <kys@microsoft.com>
CC: Haiyang Zhang <haiyangz@microsoft.com>
CC: Vitaly Kuznetsov <vkuznets@redhat.com>
CC: Roman Kagan <rkagan@virtuozzo.com>
CC: Denis V. Lunev <den@openvz.org>
CC: qemu-devel@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This struct is required for Hyper-V SynIC timers implementation inside KVM
and for upcoming Hyper-V VMBus support by userspace(QEMU). So place it into
Hyper-V UAPI header.
Signed-off-by: Andrey Smetanin <asmetanin@virtuozzo.com>
Acked-by: K. Y. Srinivasan <kys@microsoft.com>
Reviewed-by: Roman Kagan <rkagan@virtuozzo.com>
CC: Gleb Natapov <gleb@kernel.org>
CC: Paolo Bonzini <pbonzini@redhat.com>
CC: "K. Y. Srinivasan" <kys@microsoft.com>
CC: Haiyang Zhang <haiyangz@microsoft.com>
CC: Vitaly Kuznetsov <vkuznets@redhat.com>
CC: Roman Kagan <rkagan@virtuozzo.com>
CC: Denis V. Lunev <den@openvz.org>
CC: qemu-devel@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This constant is required for Hyper-V SynIC timers MSR's
support by userspace(QEMU).
Signed-off-by: Andrey Smetanin <asmetanin@virtuozzo.com>
Acked-by: K. Y. Srinivasan <kys@microsoft.com>
Reviewed-by: Roman Kagan <rkagan@virtuozzo.com>
CC: Gleb Natapov <gleb@kernel.org>
CC: Paolo Bonzini <pbonzini@redhat.com>
CC: "K. Y. Srinivasan" <kys@microsoft.com>
CC: Haiyang Zhang <haiyangz@microsoft.com>
CC: Vitaly Kuznetsov <vkuznets@redhat.com>
CC: Roman Kagan <rkagan@virtuozzo.com>
CC: Denis V. Lunev <den@openvz.org>
CC: qemu-devel@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Currently, there is only one user for hv_ringbuffer_read()/
hv_ringbuffer_peak() functions and the usage of these functions is:
- insecure as we drop ring_lock between them, someone else (in theory
only) can acquire it in between;
- non-optimal as we do a number of things (acquire/release the above
mentioned lock, calculate available space on the ring, ...) twice and
this path is performance-critical.
Remove hv_ringbuffer_peek() moving the logic from __vmbus_recvpacket() to
hv_ringbuffer_read().
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
vmbus_recvpacket() and vmbus_recvpacket_raw() are almost identical but
there are two discrepancies:
1) vmbus_recvpacket() doesn't propagate errors from hv_ringbuffer_read()
which looks like it is not desired.
2) There is an error message printed in packetlen > bufferlen case in
vmbus_recvpacket(). I'm removing it as it is usless for users to see
such messages and /vmbus_recvpacket_raw() doesn't have it.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
hv_ringbuffer_peek() does the same as hv_ringbuffer_read() without
advancing the read index. The only functional change this patch brings
is moving hv_need_to_signal_on_read() call under the ring_lock but this
function is just a couple of comparisons.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Convert 6+-string comments repeating function names to normal kernel-style
comments and fix a couple of other comment style issues. No textual or
functional changes intended.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The crash is observed when a service is being disabled host side while
userspace daemon is connected to the device:
[ 90.244859] general protection fault: 0000 [#1] SMP
...
[ 90.800082] Call Trace:
[ 90.800082] [<ffffffff81187008>] __fput+0xc8/0x1f0
[ 90.800082] [<ffffffff8118716e>] ____fput+0xe/0x10
...
[ 90.800082] [<ffffffff81015278>] do_signal+0x28/0x580
[ 90.800082] [<ffffffff81086656>] ? finish_task_switch+0xa6/0x180
[ 90.800082] [<ffffffff81443ebf>] ? __schedule+0x28f/0x870
[ 90.800082] [<ffffffffa01ebbaa>] ? hvt_op_read+0x12a/0x140 [hv_utils]
...
The problem is that hvutil_transport_destroy() which does misc_deregister()
freeing the appropriate device is reachable by two paths: module unload
and from util_remove(). While module unload path is protected by .owner in
struct file_operations util_remove() path is not. Freeing the device while
someone holds an open fd for it is a show stopper.
In general, it is not possible to revoke an fd from all users so the only
way to solve the issue is to defer freeing the hvutil_transport structure.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When Hyper-V host asks us to remove some util driver by closing the
appropriate channel there is no easy way to force the current file
descriptor holder to hang up but we can start to respond -EBADF to all
operations asking it to exit gracefully.
As we're setting hvt->mode from two separate contexts now we need to use
a proper locking.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
As a preparation to reusing outmsg_lock to protect test-and-set openrations
on 'mode' rename it the more general 'lock'.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
inmsg should be freed in case of on_msg() failure to avoid memory leak.
Preserve the error code from on_msg().
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When the handshake with daemon is complete, we should poll the channel since
during the handshake, we will not be processing any messages. This is a
potential bug if the host is waiting for a response from the guest.
I would like to thank Dexuan for pointing this out.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Force all channel messages to be delivered on CPU0. These messages are not
performance critical and are used during the setup and teardown of the
channel.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Hypervisor Top Level Functional Specification v3/4 says
that TSC page sequence value = -1(0xFFFFFFFF) is used to
indicate that TSC page no longer reliable source of reference
timer. Unfortunately, we found that Windows Hyper-V guest
side implementation uses sequence value = 0 to indicate
that Tsc page no longer valid. This is clearly visible
inside Windows 2012R2 ntoskrnl.exe HvlGetReferenceTime()
function dissassembly:
HvlGetReferenceTime proc near
xchg ax, ax
loc_1401C3132:
mov rax, cs:HvlpReferenceTscPage
mov r9d, [rax]
test r9d, r9d
jz short loc_1401C3176
rdtsc
mov rcx, cs:HvlpReferenceTscPage
shl rdx, 20h
or rdx, rax
mov rax, [rcx+8]
mov rcx, cs:HvlpReferenceTscPage
mov r8, [rcx+10h]
mul rdx
mov rax, cs:HvlpReferenceTscPage
add rdx, r8
mov ecx, [rax]
cmp ecx, r9d
jnz short loc_1401C3132
jmp short loc_1401C3184
loc_1401C3176:
mov ecx, 40000020h
rdmsr
shl rdx, 20h
or rdx, rax
loc_1401C3184:
mov rax, rdx
retn
HvlGetReferenceTime endp
This patch aligns Tsc page invalid sequence value with
Windows Hyper-V guest implementation which is more
compatible with both Hyper-V hypervisor and KVM hypervisor.
Signed-off-by: Andrey Smetanin <asmetanin@virtuozzo.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
CC: "K. Y. Srinivasan" <kys@microsoft.com>
CC: Haiyang Zhang <haiyangz@microsoft.com>
CC: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently we have two policies for deciding when to signal the host:
One based on the ring buffer state and the other based on what the
VMBUS client driver wants to do. Consider the case when the client
wants to explicitly control when to signal the host. In this case,
if the client were to defer signaling, we will not be able to signal
the host subsequently when the client does want to signal since the
ring buffer state will prevent the signaling. Implement logic to
have only one signaling policy in force for a given channel.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
Tested-by: Haiyang Zhang <haiyangz@microsoft.com>
Cc: <stable@vger.kernel.org> # v4.2+
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch makes 16GB GPUs work in Hyper-V VMs, since, for
compatibility reasons, the Hyper-V BIOS lists MMIO ranges in 2GB
chunks in its root bus's _CRS object.
Signed-off-by: Jake Oshins <jakeo@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
spinlock is unnecessary here.
mutex is enough.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We want to simplify vmbus_onoffer_rescind() by not invoking
hv_process_channel_removal(NULL, ...).
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In the path vmbus_onoffer_rescind() -> vmbus_device_unregister() ->
device_unregister() -> ... -> __device_release_driver(), we can see for a
device without a driver loaded: dev->driver is NULL, so
dev->bus->remove(dev), namely vmbus_remove(), isn't invoked.
As a result, vmbus_remove() -> hv_process_channel_removal() isn't invoked
and some cleanups(like sending a CHANNELMSG_RELID_RELEASED message to the
host) aren't done.
We can demo the issue this way:
1. rmmod hv_utils;
2. disable the Heartbeat Integration Service in Hyper-V Manager and lsvmbus
shows the device disappears.
3. re-enable the Heartbeat in Hyper-V Manager and modprobe hv_utils, but
lsvmbus shows the device can't appear again.
This is because, the host thinks the VM hasn't released the relid, so can't
re-offer the device to the VM.
We can fix the issue by moving hv_process_channel_removal()
from vmbus_close_internal() to vmbus_device_release(), since the latter is
always invoked on device_unregister(), whether or not the dev has a driver
loaded.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This fixes an incorrect assumption of channel state in the function.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
process_chn_event(), running in the tasklet, can race with
vmbus_close_internal() in the case of SMP guest, e.g., when the former is
accessing channel->inbound.ring_buffer, the latter could be freeing the
ring_buffer pages.
To resolve the race, we can serialize them by disabling the tasklet when
the latter is running here.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The irq we extract from ACPI is not used - we deliver hypervisor
interrupts on a special vector. Make the necessary adjustments.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Use uuid_le_cmp() for comparing GUIDs.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Consistently use uuid_le type in the Hyper-V driver code.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Backup integration service on WS2012 has appearently trouble to
negotiate with a guest which does not support the provided util version.
Currently the VSS driver supports only version 5/0. A WS2012 offers only
version 1/x and 3/x, and vmbus_prep_negotiate_resp correctly returns an
empty icframe_vercnt/icmsg_vercnt. But the host ignores that and
continues to send ICMSGTYPE_NEGOTIATE messages. The result are weird
errors during boot and general misbehaviour.
Check the Windows version to work around the host bug, skip hv_vss_init
on WS2012 and older.
Signed-off-by: Olaf Hering <olaf@aepfle.de>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This defines the channel type for PCI front-ends in Hyper-V VMs.
Signed-off-by: Jake Oshins <jakeo@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch exposes the function that hv_vmbus.ko uses to make hypercalls. This
is necessary for retargeting an interrupt when it is given a new affinity.
Since we are exporting this API, rename the API as it will be visible outside
the hv.c file.
Signed-off-by: Jake Oshins <jakeo@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch exposes the mapping between Linux CPU number and Hyper-V virtual
processor number. This is necessary because the hypervisor needs to know which
virtual processors to target when making a mapping in the Interrupt Redirection
Table in the I/O MMU.
Signed-off-by: Jake Oshins <jakeo@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Before vmbus_connect() synic is setup per vcpu - this means
hypervisor receives writes at synic msr's and probably allocate
hypervisor resources per synic setup.
If vmbus_connect() failed for some reason it's neccessary to cleanup
synic setup by call hv_synic_cleanup() at each vcpu to get a chance
to free allocated resources by hypervisor per synic.
This patch does appropriate cleanup in case of vmbus_connect() failure.
Signed-off-by: Andrey Smetanin <asmetanin@virtuozzo.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
CC: "K. Y. Srinivasan" <kys@microsoft.com>
CC: Haiyang Zhang <haiyangz@microsoft.com>
CC: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
All channel interrupts are bound to specific VCPUs in the guest
at the point channel is created. While currently, we invoke the
polling function on the correct CPU (the CPU to which the channel
is bound to) in some cases we may run the polling function in
a non-interrupt context. This potentially can cause an issue as the
polling function can be interrupted by the channel callback function.
Fix the issue by running the polling function on the appropriate CPU
at interrupt level. Additional details of the issue being addressed by
this patch are given below:
Currently hv_fcopy_onchannelcallback is called from interrupts and also
via the ->write function of hv_utils. Since the used global variables to
maintain state are not thread safe the state can get out of sync.
This affects the variable state as well as the channel inbound buffer.
As suggested by KY adjust hv_poll_channel to always run the given
callback on the cpu which the channel is bound to. This avoids the need
for locking because all the util services are single threaded and only
one transaction is active at any given point in time.
Additionally, remove the context variable, they will always be the same as
recv_channel.
Signed-off-by: Olaf Hering <olaf@aepfle.de>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Util services such as KVP and FCOPY need assistance from daemon's running
in user space. Increase the timeout so we don't prematurely terminate
the transaction in the kernel. Host sets up a 60 second timeout for
all util driver transactions. The host will retry the transaction if it
times out. Set the guest timeout at 30 seconds.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We were getting build warning about unused variable "tsc_msr" and
"va_tsc" while building for i386 allmodconfig.
Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This fixes the recent commit 3b71107d73b16074afa7658f3f0fcf837aabfe24:
Drivers: hv: vmbus: Further improve CPU affiliation logic
Without the fix, reloading hv_netvsc hangs the guest.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit e513229b4c ("Drivers: hv: vmbus: prevent cpu offlining on newer
hypervisors") was altering smp_ops.cpu_disable to prevent CPU offlining.
We can bo better by using cpu_hotplug_enable/disable functions instead of
such hard-coding.
Reported-by: Radim Kr.má <rkrcmar@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This is useful to analyze performance issue.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The current Hyper-V clock source is based on the per-partition reference counter
and this counter is being accessed via s synthetic MSR - HV_X64_MSR_TIME_REF_COUNT.
Hyper-V has a more efficient way of computing the per-partition reference
counter value that does not involve reading a synthetic MSR. We implement
a time source based on this mechanism.
Tested-by: Vivek Yadav <vyadav@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Migrate hv driver to the new 'set-state' interface provided by
clockevents core, the earlier 'set-mode' interface is marked obsolete
now.
This also enables us to implement callbacks for new states of clockevent
devices, for example: ONESHOT_STOPPED.
Cc: "K. Y. Srinivasan" <kys@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Cc: devel@linuxdriverproject.org
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fixes a bug where previously hv_ringbuffer_read would pass in the old
number of bytes available to read instead of the expected old read index
when calculating when to signal to the host that the ringbuffer is empty.
Since the previous write size is already saved, also changes the
hv_need_to_signal_on_read to use the previously read value rather than
recalculating it.
Signed-off-by: Christopher Oo <t-chriso@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Keep track of CPU affiliations of sub-channels within the scope of the primary
channel. This will allow us to better distribute the load amongst available
CPUs.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The current code tracks the assigned CPUs within a NUMA node in the context of
the primary channel. So, if we have a VM with a single NUMA node with 8 VCPUs, we may
end up unevenly distributing the channel load. Fix the issue by tracking affiliations
globally.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch deletes the logic from hyperv_fb which picked a range of MMIO space
for the frame buffer and adds new logic to hv_vmbus which picks ranges for
child drivers. The new logic isn't quite the same as the old, as it considers
more possible ranges.
Signed-off-by: Jake Oshins <jakeo@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch changes the logic in hv_vmbus to record all of the ranges in the
VM's firmware (BIOS or UEFI) that offer regions of memory-mapped I/O space for
use by paravirtual front-end drivers. The old logic just found one range
above 4GB and called it good. This logic will find any ranges above 1MB.
It would have been possible with this patch to just use existing resource
allocation functions, rather than keep track of the entire set of Hyper-V
related MMIO regions in VMBus. This strategy, however, is not sufficient
when the resource allocator needs to be aware of the constraints of a
Hyper-V virtual machine, which is what happens in the next patch in the series.
So this first patch exists to show the first steps in reworking the MMIO
allocation paths for Hyper-V front-end drivers.
Signed-off-by: Jake Oshins <jakeo@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We cycle through all the "high performance" channels to distribute
load across the available CPUs. Process the NetworkDirect as a
high performance device.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Hypervisor Top Level Functional Specification v3.1/4.0 notes that cpuid
(0x40000003) EDX's 10th bit should be used to check that Hyper-V guest
crash MSR's functionality available.
This patch should fix this recognition. Currently the code checks EAX
register instead of EDX.
Signed-off-by: Andrey Smetanin <asmetanin@virtuozzo.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Pre-Win2012R2 hosts don't properly handle CHANNELMSG_UNLOAD and
wait_for_completion() hangs. Avoid sending such request on old hosts.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This fixes a typo: base_flag_bumber to base_flag_number
Signed-off-by: Nik Nyby <nikolas@gnu.org>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We don't catch this allocation failure because there is a typo and we
check the wrong variable.
Fixes: 14b50f80c3 ('Drivers: hv: util: introduce hv_utils_transport abstraction')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The guest may have to send a completion packet back to the host.
To support this usage, permit sending a packet without a payload -
we would be only sending the descriptor in this case.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Support Win10 protocol for Dynamic Memory. Thia patch allows guests on Win10 hosts
to hot-add memory even when dynamic memory is not enabled on the guest.
Signed-off-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
struct hv_start_fcopy is too big to be on stack on i386, the following
warning is reported:
>> drivers/hv/hv_fcopy.c:159:1: warning: the frame size of 1088 bytes is larger than 1024 bytes [-Wframe-larger-than=]
Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
kzalloc() return value check was accidentally lost in 11bc3a5fa91f:
"Drivers: hv: kvp: convert to hv_utils_transport" commit.
We don't need to reset kvp_transaction.state here as we have the
kvp_timeout_func() timeout function and in case we're in OOM situation
it is preferable to wait.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
current_pt_regs() sometimes returns regs of the userspace process and in
case of a kernel crash this is not what we need to report. E.g. when we
trigger crash with sysrq we see the following:
...
RIP: 0010:[<ffffffff815b8696>] [<ffffffff815b8696>] sysrq_handle_crash+0x16/0x20
RSP: 0018:ffff8800db0a7d88 EFLAGS: 00010246
RAX: 000000000000000f RBX: ffffffff820a0660 RCX: 0000000000000000
...
at the same time current_pt_regs() give us:
ip=7f899ea7e9e0, ax=ffffffffffffffda, bx=26c81a0, cx=7f899ea7e9e0, ...
These registers come from the userspace process triggered the crash. As we
don't even know which process it was this information is rather useless.
When kernel crash happens through 'die' proper regs are being passed to
all receivers on the die_chain (and panic_notifier_list is being notified
with the string passed to panic() only). If panic() is called manually
(e.g. on BUG()) we won't get 'die' notification so keep the 'panic'
notification reporter as well but guard against double reporting.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Full kernel hang is observed when kdump kernel starts after a crash. This
hang happens in vmbus_negotiate_version() function on
wait_for_completion() as Hyper-V host (Win2012R2 in my testing) never
responds to CHANNELMSG_INITIATE_CONTACT as it thinks the connection is
already established. We need to perform some mandatory minimalistic
cleanup before we start new kernel.
Reported-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
At the very late stage of kexec a driver (which are not being unloaded) can
try to post a message or signal an event. This will crash the kernel as we
already did hv_cleanup() and the hypercall page is NULL.
Move all common (between 32 and 64 bit code) declarations to the beginning
of the do_hypercall() function. Unfortunately we have to write the
!hypercall_page check twice to not mix declarations and code.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When general-purpose kexec (not kdump) is being performed in Hyper-V guest
the newly booted kernel fails with an MCE error coming from the host. It
is the same error which was fixed in the "Drivers: hv: vmbus: Implement
the protocol for tearing down vmbus state" commit - monitor pages remain
special and when they're being written to (as the new kernel doesn't know
these pages are special) bad things happen. We need to perform some
minimalistic cleanup before booting a new kernel on kexec. To do so we
need to register a special machine_ops.shutdown handler to be executed
before the native_machine_shutdown(). Registering a shutdown notification
handler via the register_reboot_notifier() call is not sufficient as it
happens to early for our purposes. machine_ops is not being exported to
modules (and I don't think we want to export it) so let's do this in
mshyperv.c
The minimalistic cleanup consists of cleaning up clockevents, synic MSRs,
guest os id MSR, and hypercall MSR.
Kdump doesn't require all this stuff as it lives in a separate memory
space.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We already have hv_synic_free() which frees all per-cpu pages for all
CPUs, let's remove the hv_synic_free_cpu() call from hv_synic_cleanup()
so it will be possible to do separate cleanup (writing to MSRs) and final
freeing. This is going to be used to assist kexec.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Allocate ring buffer memory from the NUMA node assigned to the channel.
Since this is a performance and not a correctness issue, if the node specific
allocation were to fail, fall back and allocate without specifying the node.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Channels/sub-channels can be affinitized to VCPUs in the guest. Implement
this affinity in a way that is NUMA aware. The current protocol distributed
the primary channels uniformly across all available CPUs. The new protocol
is NUMA aware: primary channels are distributed across the available NUMA
nodes while the sub-channels within a primary channel are distributed amongst
CPUs within the NUMA node assigned to the primary channel.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Map target_cpu to target_vcpu using the mapping table.
We should use the mapping table to transform guest CPU ID to VP Index
as is done for the non-performance critical channels.
While the value CPU 0 is special and will
map to VP index 0, it is good to be consistent.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Memory notifiers are being executed in a sequential order and when one of
them fails returning something different from NOTIFY_OK the remainder of
the notification chain is not being executed. When a memory block is being
onlined in online_pages() we do memory_notify(MEM_GOING_ONLINE, ) and if
one of the notifiers in the chain fails we end up doing
memory_notify(MEM_CANCEL_ONLINE, ) so it is possible for a notifier to see
MEM_CANCEL_ONLINE without seeing the corresponding MEM_GOING_ONLINE event.
E.g. when CONFIG_KASAN is enabled the kasan_mem_notifier() is being used
to prevent memory hotplug, it returns NOTIFY_BAD for all MEM_GOING_ONLINE
events. As kasan_mem_notifier() comes before the hv_memory_notifier() in
the notification chain we don't see the MEM_GOING_ONLINE event and we do
not take the ha_region_mutex. We, however, see the MEM_CANCEL_ONLINE event
and unconditionally try to release the lock, the following is observed:
[ 110.850927] =====================================
[ 110.850927] [ BUG: bad unlock balance detected! ]
[ 110.850927] 4.1.0-rc3_bugxxxxxxx_test_xxxx #595 Not tainted
[ 110.850927] -------------------------------------
[ 110.850927] systemd-udevd/920 is trying to release lock
(&dm_device.ha_region_mutex) at:
[ 110.850927] [<ffffffff81acda0e>] mutex_unlock+0xe/0x10
[ 110.850927] but there are no more locks to release!
At the same time we can have the ha_region_mutex taken when we get the
MEM_CANCEL_ONLINE event in case one of the memory notifiers after the
hv_memory_notifier() in the notification chain failed so we need to add
the mutex_is_locked() check. In case of MEM_ONLINE we are always supposed
to have the mutex locked.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add support for Windows 10.
Signed-off-by: Keith Mange <keith.mange@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Primary channels are distributed evenly across all vcpus we have. When the host
asks us to create subchannels it usually makes us num_cpus-1 offers and we are
supposed to distribute the work evenly among the channel itself and all its
subchannels. Make sure they are all assigned to different vcpus.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We need to call init_vp_index() after we added the channel to the appropriate
list (global or subchannel) to be able to use this information when assigning
the channel to the particular vcpu. To do so we need to move a couple of
functions around. The only real change is the init_vp_index() call. This is a
small refactoring without a functional change.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It is unlikely that that host will ask us to close only one subchannel for a
device but let's be consistent. Do both num_sc++ and num_sc-- with
channel->lock to be on the safe side.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In case there was an error reported in the response to the CHANNELMSG_OPENCHANNEL
call we need to do the cleanup as a vmbus_open() user won't be doing it after
receiving an error. The cleanup should be done on all failure paths. We also need
to avoid returning open_info->response.open_result.status as the return value as
all other errors we return from vmbus_open() are -EXXX and vmbus_open() callers
are not supposed to analyze host error codes.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Implement the protocol for tearing down the monitor state established with
the host.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Tested-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
free_channel() has been invoked in
vmbus_remove() -> hv_process_channel_removal(), or vmbus_remove() ->
... -> vmbus_close_internal() -> hv_process_channel_removal().
We also change to use list_for_each_entry_safe(), because the entry
is removed in hv_process_channel_removal().
This patch fixes a bug in the vmbus unload path.
Thank Dan Carpenter for finding the issue!
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Cc: K. Y. Srinivasan <kys@microsoft.com>
Cc: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In case we do request_resource() in vmbus_acpi_add() we need to tear it down
to be able to load the driver again. Otherwise the following crash in observed
when hv_vmbus unload/load sequence is performed on a Generation2 instance:
[ 38.165701] BUG: unable to handle kernel paging request at ffffffffa00075a0
[ 38.166315] IP: [<ffffffff8107dc5f>] __request_resource+0x2f/0x50
[ 38.166315] PGD 1f34067 PUD 1f35063 PMD 3f723067 PTE 0
[ 38.166315] Oops: 0000 [#1] SMP
[ 38.166315] Modules linked in: hv_vmbus(+) [last unloaded: hv_vmbus]
[ 38.166315] CPU: 0 PID: 267 Comm: modprobe Not tainted 3.19.0-rc5_bug923184+ #486
[ 38.166315] Hardware name: Microsoft Corporation Virtual Machine/Virtual Machine, BIOS Hyper-V UEFI Release v1.0 11/26/2012
[ 38.166315] task: ffff88003f401cb0 ti: ffff88003f60c000 task.ti: ffff88003f60c000
[ 38.166315] RIP: 0010:[<ffffffff8107dc5f>] [<ffffffff8107dc5f>] __request_resource+0x2f/0x50
[ 38.166315] RSP: 0018:ffff88003f60fb58 EFLAGS: 00010286
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Unify driver registration reporting and move it to debug level as normally daemons write to syslog themselves
and these kernel messages are useless.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Introduce FCOPY_VERSION_1 to support kernel replying to the negotiation
message with its own version.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Introduce VSS_OP_REGISTER1 to support kernel replying to the negotiation
message with its own version.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Convert to hv_utils_transport to support both netlink and /dev/vmbus/hv_kvp communication methods.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Unify the code with the recently introduced hv_utils_transport. Netlink
communication is disabled for fcopy.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Convert to hv_utils_transport to support both netlink and /dev/vmbus/hv_vss communication methods.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The intention is to make KVP/VSS drivers work through misc char devices.
Introduce an abstraction for kernel/userspace communication to make the
migration smoother. Transport operational mode (netlink or char device)
is determined by the first received message. To support driver upgrades
the switch from netlink to chardev operational mode is supported.
Every hv_util daemon is supposed to register 2 callbacks:
1) on_msg() to get notified when the userspace daemon sent a message;
2) on_reset() to get notified when the userspace daemon drops the connection.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Get an additional reference otherwise a crash is observed when hv_utils module is being unloaded while
fcopy daemon is still running. .owner gives us an additional reference when
someone holds a descriptor for the device.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Switch to using the hvutil_device_state state machine from using 3 different state variables:
fcopy_transaction.active, opened, and in_hand_shake.
State transitions are:
-> HVUTIL_DEVICE_INIT when driver loads or on device release
-> HVUTIL_READY if the handshake was successful
-> HVUTIL_HOSTMSG_RECEIVED when there is a non-negotiation message from the host
-> HVUTIL_USERSPACE_REQ after userspace daemon read the message
-> HVUTIL_USERSPACE_RECV after/if userspace has replied
-> HVUTIL_READY after we respond to the host
-> HVUTIL_DEVICE_DYING on driver unload
In hv_fcopy_onchannelcallback() process ICMSGTYPE_NEGOTIATE messages even when
the userspace daemon is disconnected, otherwise we can make the host think
we don't support FCOPY and disable the service completely.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Switch to using the hvutil_device_state state machine from using kvp_transaction.active.
State transitions are:
-> HVUTIL_DEVICE_INIT when driver loads or on device release
-> HVUTIL_READY if the handshake was successful
-> HVUTIL_HOSTMSG_RECEIVED when there is a non-negotiation message from the host
-> HVUTIL_USERSPACE_REQ after we sent the message to the userspace daemon
-> HVUTIL_USERSPACE_RECV after/if the userspace daemon has replied
-> HVUTIL_READY after we respond to the host
-> HVUTIL_DEVICE_DYING on driver unload
In hv_vss_onchannelcallback() process ICMSGTYPE_NEGOTIATE messages even when
the userspace daemon is disconnected, otherwise we can make the host think
we don't support VSS and disable the service completely.
Unfortunately there is no good way we can figure out that the userspace daemon
has died (unless we start treating all timeouts as such), add a protection
against processing new VSS_OP_REGISTER messages while being in the middle of a
transaction (HVUTIL_USERSPACE_REQ or HVUTIL_USERSPACE_RECV state).
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Switch to using the hvutil_device_state state machine from using 2 different state variables: kvp_transaction.active and
in_hand_shake.
State transitions are:
-> HVUTIL_DEVICE_INIT when driver loads or on device release
-> HVUTIL_READY if the handshake was successful
-> HVUTIL_HOSTMSG_RECEIVED when there is a non-negotiation message from the host
-> HVUTIL_USERSPACE_REQ after we sent the message to the userspace daemon
-> HVUTIL_USERSPACE_RECV after/if the userspace daemon has replied
-> HVUTIL_READY after we respond to the host
-> HVUTIL_DEVICE_DYING on driver unload
In hv_kvp_onchannelcallback() process ICMSGTYPE_NEGOTIATE messages even when
the userspace daemon is disconnected, otherwise we can make the host think
we don't support KVP and disable the service completely.
Unfortunately there is no good way we can figure out that the userspace daemon
has died (unless we start treating all timeouts as such). In case the daemon
restarts we skip the negotiation procedure (so the daemon is supposed to has
the same version). This behavior is unchanged from in_handshake approach.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
KVP/VSS/FCOPY drivers work in fully serialized mode: we wait till userspace
daemon registers, wait for a message from the host, send this message to the
daemon, get the reply, send it back to host, wait for another message.
Introduce enum hvutil_device_state to represend this state in all 3 drivers.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
'fcopy_work' (and fcopy_work_func) is a misnomer as it sounds like we expect
this useful work to happen and in reality it is just an emergency escape when
timeout happens.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
'kvp_work' (and kvp_work_func) is a misnomer as it sounds like we expect
this useful work to happen and in reality it is just an emergency escape when
timeout happens.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In theory, the host is not supposed to issue any requests before be reply to
the previous one. In KVP we, however, support the following scenarios:
1) A message was received before userspace daemon registered;
2) A message was received while the previous one is still being processed.
In VSS we support only the former. Add support for the later, use
hv_poll_channel() to do the job.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In theory, the host is not supposed to issue any requests before be reply to
the previous one. In KVP we, however, support the following scenarios:
1) A message was received before userspace daemon registered;
2) A message was received while the previous one is still being processed.
In FCOPY we support only the former. Add support for the later, use
hv_poll_channel() to do the job.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Move poll_channel() to hyperv_vmbus.h and make it inline and rename it to hv_poll_channel() so it can be reused
in other hv_util modules.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We set kvp_context when we want to postpone receiving a packet from vmbus due
to the previous transaction being unfinished. We, however, never reset this
state, all consequent kvp_respond_to_host() calls will result in poll_channel()
calling hv_kvp_onchannelcallback(). This doesn't cause real issues as:
1) Host is supposed to serialize transactions as well
2) If no message is pending vmbus_recvpacket() will return 0 recvlen.
This is just a cleanup.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
These declarations are internal to hv_util module and hv_fcopy_* declarations
already reside there.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
balloon_wrk.num_pages is __u32 and it comes from host in struct dm_balloon
where it is also __u32. We, however, use 'int' in balloon_up() and in case
we happen to receive num_pages>INT_MAX request we'll end up allocating zero
pages as 'num_pages < alloc_unit' check in alloc_balloon_pages() will pass.
Change num_pages type to unsigned int.
In real life ballooning request come with num_pages in [512, 32768] range so
this is more a future-proof/cleanup.
Reported-by: Laszlo Ersek <lersek@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
'Drivers: hv: hv_balloon: refuse to balloon below the floor' fix does not
correctly handle the case when val.freeram < num_pages as val.freeram is
__kernel_ulong_t and the 'val.freeram - num_pages' value will be a huge
positive value instead of being negative.
Usually host doesn't ask us to balloon more than val.freeram but in case
he have a memory hog started after we post the last pressure report we
can get into troubles.
Suggested-by: Laszlo Ersek <lersek@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Most of the retries can be done within a millisecond successfully, so we
sleep 1ms before the first retry, then gradually increase the retry
interval to 2^n with max value of 2048ms. Doing so, we will have shorter
overall delay time, because most of the cases succeed within 1-2 attempts.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: K. Y. Srinivasan <kys@microsoft.com>
Reviewed-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
... and simplify alloc_balloon_pages() interface by removing redundant
alloc_error from it.
If we happen to enter balloon_up() with balloon_wrk.num_pages = 0 we will enter
infinite 'while (!done)' loop as alloc_balloon_pages() will be always returning
0 and not setting alloc_error. We will also be sending a meaningless message to
the host on every iteration.
The 'alloc_unit == 1 && alloc_error -> num_ballooned == 0' change and
alloc_error elimination requires a special comment. We do alloc_balloon_pages()
with 2 different alloc_unit values and there are 4 different
alloc_balloon_pages() results, let's check them all.
alloc_unit = 512:
1) num_ballooned = 0, alloc_error = 0: we do 'alloc_unit=1' and retry pre- and
post-patch.
2) num_ballooned > 0, alloc_error = 0: we check 'num_ballooned == num_pages'
and act accordingly, pre- and post-patch.
3) num_ballooned > 0, alloc_error > 0: we report this chunk and remain within
the loop, no changes here.
4) num_ballooned = 0, alloc_error > 0: we do 'alloc_unit=1' and retry pre- and
post-patch.
alloc_unit = 1:
1) num_ballooned = 0, alloc_error = 0: this can happen in two cases: when we
passed 'num_pages=0' to alloc_balloon_pages() or when there was no space in
bl_resp to place a single response. The second option is not possible as
bl_resp is of PAGE_SIZE size and single response 'union dm_mem_page_range' is
8 bytes, but the first one is (in theory, I think that Hyper-V host never
places such requests). Pre-patch code loops forever, post-patch code sends
a reply with more_pages = 0 and finishes.
2) num_ballooned > 0, alloc_error = 0: we ran out of space in bl_resp, we
report partial success and remain within the loop, no changes pre- and
post-patch.
3) num_ballooned > 0, alloc_error > 0: pre-patch code finishes, post-patch code
does one more try and if there is no progress (we finish with
'num_ballooned = 0') we finish. So we try a bit harder with this patch.
4) num_ballooned = 0, alloc_error > 0: both pre- and post-patch code enter
'more_pages = 0' branch and finish.
So this patch has two real effects:
1) We reply with an empty response to 'num_pages=0' request.
2) We try a bit harder on alloc_unit=1 allocations (and reply with an empty
tail reply in case we fail).
An empty reply should be supported by host as we were able to send it even with
pre-patch code when we were not able to allocate a single page.
Suggested-by: Laszlo Ersek <lersek@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 79208c57da ("Drivers: hv: hv_balloon: Make adjustments in computing
the floor") was inacurate as it introduced a jump in our piecewiese linear
'floor' function:
At 2048MB we have:
Left limit:
104 + 2048/8 = 360
Right limit:
256 + 2048/16 = 384 (so the right value is 232)
We now have to make an adjustment at 8192 boundary:
232 + 8192/16 = 744
512 + 8192/32 = 768 (so the right value is 488)
Suggested-by: Laszlo Ersek <lersek@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently we add memory in 128Mb blocks but the request from host can be
aligned differently. In such case we add a partially backed block and
when this block goes online we skip onlining pages which are not backed
(hv_online_page() callback serves this purpose). When we receive next
request for the same host add region we online pages which were not backed
before with hv_bring_pgs_online(). However, we don't check if the the block
in question was onlined and online this tail unconditionally. This is bad as
we avoid all online_pages() logic: these pages are not accounted, we don't
send notifications (and hv_balloon is not the only receiver of them),...
And, first of all, nobody asked as to online these pages. Solve the issue by
checking if the last previously backed page was onlined and onlining the tail
only in case it was.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It's not necessary any longer, since we can safely run the blocking
message handlers in vmbus_connection.work_queue now.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Cc: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Since the 2 fucntions can safely run in vmbus_connection.work_queue without
hang, we don't need to schedule new work items into the per-channel workqueue.
Actally we can even remove the per-channel workqueue now -- we'll do it
in the next patch.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Cc: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A work item in vmbus_connection.work_queue can sleep, waiting for a new
host message (usually it is some kind of "completion" message). Currently
the new message will be handled in the same workqueue, but since work items
in the workqueue is serialized, we actually have no chance to handle
the new message if the current work item is sleeping -- as as result, the
current work item will hang forever.
K. Y. has posted the below fix to resolve the issue:
Drivers: hv: vmbus: Perform device register in the per-channel work element
Actually we can simplify the fix by directly running non-blocking message
handlers in the dispatch tasklet (inspired by K. Y.).
This patch is the fundamental change. The following 2 patches will simplify
the message offering and rescind-offering handling a lot.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Cc: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Don't wait after sending request for offers to the host. This wait is
unnecessary and simply adds 5 seconds to the boot time.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Handle the case when the write to the ringbuffer fails. In this case,
unconditionally signal the host. Since we may have deferred signalling
the host based on the kick_q parameter, signalling the host
unconditionally in this case deals with the issue.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Export the vmbus_sendpacket_pagebuffer_ctl() interface. This export will be
used by the netvsc driver.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When a channel has been rescinded, the close operation is a noop.
Restructure the code so we deal with the rescind condition after
we properly cleanup the channel. I would like to thank
Dexuan Cui <decui@microsoft.com> for observing this problem.
The current code leaks memory when the channel is rescinded.
The current char-next branch is broken and this patch fixes
the bug.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The indenting makes it clear that there were curly braces intended here.
Fixes: 2dd37cb815 ('Drivers: hv: vmbus: Handle both rescind and offer messages in the same context')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
HV_CRASH_CTL_CRASH_NOTIFY is a 64 bit number. Depending on the usage context,
the value may be truncated. This patch is in response from the following
email from Wu Fengguang <fengguang.wu@intel.com>:
From: Wu Fengguang <fengguang.wu@intel.com>
Subject: [char-misc:char-misc-testing 25/45] drivers/hv/vmbus_drv.c:67:9: sparse:
constant 0x8000000000000000 is so big it is unsigned long
tree: git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc.git char-misc-testing
head: b3de8e3719
commit: 96c1d0581d [25/45] Drivers: hv: vmbus: Add support for VMBus panic notifier handler
reproduce:
# apt-get install sparse
git checkout 96c1d0581d
make ARCH=x86_64 allmodconfig
make C=1 CF=-D__CHECK_ENDIAN__
sparse warnings: (new ones prefixed by >>)
drivers/hv/vmbus_drv.c:67:9: sparse: constant 0x8000000000000000 is so big it is unsigned long
...
Signed-off-by: Nick Meier <nmeier@microsoft.com>
Reported-by: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Memory blocks can be onlined in random order. When this order is not natural
some memory pages are not onlined because of the redundant check in
hv_online_page().
Here is a real world scenario:
1) Host tries to hot-add the following (process_hot_add):
pg_start=rg_start=0x48000, pfn_cnt=111616, rg_size=262144
2) This results in adding 4 memory blocks:
[ 109.057866] init_memory_mapping: [mem 0x48000000-0x4fffffff]
[ 114.102698] init_memory_mapping: [mem 0x50000000-0x57ffffff]
[ 119.168039] init_memory_mapping: [mem 0x58000000-0x5fffffff]
[ 124.233053] init_memory_mapping: [mem 0x60000000-0x67ffffff]
The last one is incomplete but we have special has->covered_end_pfn counter to
avoid onlining non-backed frames and hv_bring_pgs_online() function to bring
them online later on.
3) Now we have 4 offline memory blocks: /sys/devices/system/memory/memory9-12
$ for f in /sys/devices/system/memory/memory*/state; do echo $f `cat $f`; done | grep -v onlin
/sys/devices/system/memory/memory10/state offline
/sys/devices/system/memory/memory11/state offline
/sys/devices/system/memory/memory12/state offline
/sys/devices/system/memory/memory9/state offline
4) We bring them online in non-natural order:
$grep MemTotal /proc/meminfo
MemTotal: 966348 kB
$echo online > /sys/devices/system/memory/memory12/state && grep MemTotal /proc/meminfo
MemTotal: 1019596 kB
$echo online > /sys/devices/system/memory/memory11/state && grep MemTotal /proc/meminfo
MemTotal: 1150668 kB
$echo online > /sys/devices/system/memory/memory9/state && grep MemTotal /proc/meminfo
MemTotal: 1150668 kB
As you can see memory9 block gives us zero additional memory. We can also
observe a huge discrepancy between host- and guest-reported memory sizes.
The root cause of the issue is the redundant pg >= covered_start_pfn check (and
covered_start_pfn advancing) in hv_online_page(). When upper memory block in
being onlined before the lower one (memory12 and memory11 in the above case) we
advance the covered_start_pfn pointer and all memory9 pages do not pass the
check. If the assumption that host always gives us requests in sequential order
and pg_start always equals rg_start when the first request for the new HA
region is received (that's the case in my testing) is correct than we can get
rid of covered_start_pfn and pg >= start_pfn check in hv_online_page() is
sufficient.
The current char-next branch is broken and this patch fixes
the bug.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When add_memory() fails the following BUG is observed:
[ 743.646107] hv_balloon: hot_add memory failed error is -17
[ 743.679973]
[ 743.680930] =====================================
[ 743.680930] [ BUG: bad unlock balance detected! ]
[ 743.680930] 3.19.0-rc5_bug1131426+ #552 Not tainted
[ 743.680930] -------------------------------------
[ 743.680930] kworker/0:2/255 is trying to release lock (&dm_device.ha_region_mutex) at:
[ 743.680930] [<ffffffff81aae5fe>] mutex_unlock+0xe/0x10
[ 743.680930] but there are no more locks to release!
This happens as we don't acquire ha_region_mutex and hot_add_req() expects us
to as it does unconditional mutex_unlock(). Acquire the lock on the error path.
The current char-next branch is broken and this patch fixes
the bug.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch is a continuation of the rescind handling cleanup work. We cannot
block in the global message handling work context especially if we are blocking
waiting for the host to wake us up. I would like to thank
Dexuan Cui <decui@microsoft.com> for observing this problem.
The current char-next branch is broken and this patch fixes
the bug.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/hv/vmbus_drv.c:51:5: sparse: symbol 'hyperv_panic_event' was not declared. Should it be static?
drivers/hv/vmbus_drv.c:51:5: sparse: symbol 'hyperv_panic_event' was not declared. Should it be static?
drivers/hv/vmbus_drv.c:51:5: sparse: symbol 'hyperv_panic_event' was not declared. Should it be static?
drivers/hv/vmbus_drv.c:51:5: sparse: symbol 'hyperv_panic_event' was not declared. Should it be static?
Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Implement an API that gives additional control on the what VMBUS flags will be
set as well as if the host needs to be signalled. This API will be
useful for clients that want to batch up requests to the host.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Implement an API for sending pagebuffers that gives more control to the client
in terms of setting the vmbus flags as well as deciding when to
notify the host. This will be useful for enabling batch processing.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The current algorithm for picking an outgoing channel was not distributing
the load well. Implement a simple round-robin scheme to ensure good
distribution of the outgoing traffic.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Reviewed-by: Long Li <longli@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Hyper-V allows a guest to notify the Hyper-V host that a panic
condition occured. This notification can include up to five 64
bit values. These 64 bit values are written into crash MSRs.
Once the data has been written into the crash MSRs, the host is
then notified by writing into a Crash Control MSR. On the Hyper-V
host, the panic notification data is captured in the Windows Event
log as a 18590 event.
Crash MSRs are defined in appendix H of the Hypervisor Top Level
Functional Specification. At the time of this patch, v4.0 is the
current functional spec. The URL for the v4.0 document is:
http://download.microsoft.com/download/A/B/4/AB43A34E-BDD0-4FA6-BDEF-79EEF16E880B/Hypervisor Top Level Functional Specification v4.0.docx
Signed-off-by: Nick Meier <nmeier@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When host asks us to balloon up we need to be sure we're not committing suicide
by overballooning. Use already existent 'floor' metric as our lowest possible
value for free ram.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When hot-added memory pages are not brought online or when some memory blocks
are sent offline the subsequent ballooning process kills the guest with OOM
killer. This happens as we don't report these pages as neither used nor free
and apparently host algorithm considers them as being unused. Keep track of
all online/offline operations and report all currently offline pages as being
used so host won't try to balloon them out.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When many memory regions are being added and automatically onlined the
following lockup is sometimes observed:
INFO: task udevd:1872 blocked for more than 120 seconds.
...
Call Trace:
[<ffffffff816ec0bc>] schedule_timeout+0x22c/0x350
[<ffffffff816eb98f>] wait_for_common+0x10f/0x160
[<ffffffff81067650>] ? default_wake_function+0x0/0x20
[<ffffffff816eb9fd>] wait_for_completion+0x1d/0x20
[<ffffffff8144cb9c>] hv_memory_notifier+0xdc/0x120
[<ffffffff816f298c>] notifier_call_chain+0x4c/0x70
...
When several memory blocks are going online simultaneously we got several
hv_memory_notifier() trying to acquire the ha_region_mutex. When this mutex is
being held by hot_add_req() all these competing acquire_region_mutex() do
mutex_trylock, fail, and queue themselves into wait_for_completion(..). However
when we do complete() from release_region_mutex() only one of them wakes up.
This could be solved by changing complete() -> complete_all() memory onlining
can be delayed as well, in that case we can still get several
hv_memory_notifier() runners at the same time trying to grab the mutex.
Only one of them will succeed and the others will hang for forever as
complete() is not being called. We don't see this issue often because we have
5sec onlining timeout in hv_mem_hot_add() and usually all udev events arrive
in this time frame.
Get rid of the trylock path, waiting on the mutex is supposed to provide the
required serialization.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently we log messages when either we are not able to map an ID to a
channel or when the channel does not have a callback associated
(in the channel interrupt handling path). These messages don't add
any value, get rid of them.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When the offer is rescinded, vmbus_close() can free up the channel;
deinitialize the service before closing the channel.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Properly rollback state in vmbus_pocess_offer() in the failure paths.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Alex Ng