linux.git/drivers/hv, branch v3.11-rc2

drivers: hv: allocate synic structures before hv_synic_init()

2013-06-24T23:24:17+00:00

We currently allocate synic structures in hv_sync_init(), but there's no way for
the driver to know about the allocation failure and it may continue to use the
uninitialized pointers. Solve this by introducing helpers for allocating and
freeing and doing the allocation before the on_each_cpu() call in
vmbus_bus_init().

Cc: Haiyang Zhang 
Signed-off-by: Jason Wang 
Signed-off-by: K. Y. Srinivasan 
Signed-off-by: Greg Kroah-Hartman

drivers: hv: check interrupt mask before read_index

2013-06-24T23:24:17+00:00

This patches add a read barriers to force the driver to check the interrupt mask
before read_index. Otherwise we may lost a kick to host.

Cc: K. Y. Srinivasan 
Cc: Haiyang Zhang 
Signed-off-by: Jason Wang 
Signed-off-by:  K. Y. Srinivasan 
Signed-off-by: Greg Kroah-Hartman

drivers: hv: switch to use mb() instead of smp_mb()

2013-06-18T18:14:18+00:00

Even if guest were compiled without SMP support, it could not assume that host
wasn't. So switch to use mb() instead of smp_mb() to force memory barriers for
UP guest.

Signed-off-by: Jason Wang 
Cc: Haiyang Zhang 
Cc: stable 
Signed-off-by: K. Y. Srinivasan 
Signed-off-by: Greg Kroah-Hartman

Drivers: hv: vmbus: Implement multi-channel support

2013-06-03T21:09:14+00:00

Starting with Win8, the host supports multiple sub-channels for a given
device. As in the past, the initial channel offer specifies the device and
is associated with both the type and the instance GUIDs. For performance
critical devices, the host may support multiple sub-channels. The sub-channels
share the same type and instance GUID as the primary channel. The number of
sub-channels offerrred to the guest depends on the number of virtual CPUs
assigned to the guest. The guest can request the creation of these sub-channels
and once created and opened, the guest can distribute the traffic across all
the channels (the primary and the sub-channels). A request sent on a sub-channel
will have the response delivered on the same sub-channel.

At channel (sub-channel) creation we bind the channel interrupt to a CPU and
with this sub-channel support we will be able to spread the interrupt load
of a given device across all available CPUs.

Signed-off-by: K. Y. Srinivasan 
Reviewed-by: Haiyang Zhang 
Signed-off-by: Greg Kroah-Hartman

Drivers: hv: Fix a bug in get_vp_index()

2013-05-21T16:56:55+00:00

Linux' notion of cpuid is different from the Host's notion of CPUID. In the
call to bind the channel interrupts, we should use the host's notion of
CPU Ids. Fix this bug.

Signed-off-by: K. Y. Srinivasan 
Cc: Stable  (V3.9)
Signed-off-by: Greg Kroah-Hartman

Merge branch 'x86-paravirt-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

2013-04-30T15:41:21+00:00

Pull x86 paravirt update from Ingo Molnar:
 "Various paravirtualization related changes - the biggest one makes
  guest support optional via CONFIG_HYPERVISOR_GUEST"

* 'x86-paravirt-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86, wakeup, sleep: Use pvops functions for changing GDT entries
  x86, xen, gdt: Remove the pvops variant of store_gdt.
  x86-32, gdt: Store/load GDT for ACPI S3 or hibernation/resume path is not needed
  x86-64, gdt: Store/load GDT for ACPI S3 or hibernate/resume path is not needed.
  x86: Make Linux guest support optional
  x86, Kconfig: Move PARAVIRT_DEBUG into the paravirt menu

Drivers: hv: vmbus: Fix a bug in hv_need_to_signal()

2013-04-01T19:29:52+00:00

As part of updating the vmbus protocol, the function hv_need_to_signal()
was introduced. This functions helps optimize signalling from guest to
host. The newly added memory barrier is needed to ensure that we correctly
decide when to signal the host.

Signed-off-by: K. Y. Srinivasan 
Reviewed-by: Haiyang Zhang 
Reported-by: Olaf Hering 
Cc: Stable   (V3.8+)
Signed-off-by: Greg Kroah-Hartman

Drivers: hv: Notify the host of permanent hot-add failures

2013-03-29T16:03:21+00:00

If memory hot-add fails with the error -EEXIST, then this is a permanent
failure. Notify the host of this information, so the host will not attempt
hot-add again. If the failure were a transient failure, host will attempt
a hot-add after some delay.

In this version of the patch, I have added some additional comments
to clarify how the host treats different failure conditions.

Signed-off-by: K. Y. Srinivasan 
Signed-off-by: Greg Kroah-Hartman

Drivers: hv: balloon: Support 2M page allocations for ballooning

2013-03-29T16:03:21+00:00

On Hyper-V it will be very efficient to use 2M allocations in the guest as this
makes the ballooning protocol with the host that much more efficient. Hyper-V
uses page ranges (start pfn : number of pages) to specify memory being moved
around and with 2M pages this encoding can be very efficient. However, when
memory is returned to the guest, the host does not guarantee any granularity.
To deal with this issue, split the page soon after a successful 2M allocation
so that this memory can potentially be freed as 4K pages.

If 2M allocations fail, we revert to 4K allocations.

In this version of the patch, based on the feedback from Michal Hocko
, I have added some additional commentary to the patch
description.

Cc: Michal Hocko 
Signed-off-by: K. Y. Srinivasan 
Signed-off-by: Greg Kroah-Hartman

Drivers: hv: balloon: Permit Linux to specify hot-add alignment requirements

2013-03-29T15:53:12+00:00

Some Windows hosts permit the guest to specify memory hot-add alignment
requirements (if any). Linux currently requires a 128MB alignment on memory
segments that can be hot-added. Specify this alignment requirement to the
host.

Signed-off-by: K. Y. Srinivasan 
Signed-off-by: Greg Kroah-Hartman