When a guest communicate with the hypervisor, it must use HV_HYP_PAGE to
calculate PFN, so introduce a few hvpfn helper functions as the
counterpart of the page helper functions. This is the preparation for
supporting guest whose PAGE_SIZE is not 4k.

Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Link: https://lore.kernel.org/r/20200916034817.30282-7-boqun.feng@gmail.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>

There will be more places other than vmbus where we need to calculate
the Hyper-V page PFN from a virtual address, so move virt_to_hvpfn() to
hyperv generic header.

Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Link: https://lore.kernel.org/r/20200916034817.30282-6-boqun.feng@gmail.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>

This patch introduces two types of GPADL: HV_GPADL_{BUFFER, RING}. The
types of GPADL are purely the concept in the guest, IOW the hypervisor
treat them as the same.

The reason of introducing the types for GPADL is to support guests whose
page size is not 4k (the page size of Hyper-V hypervisor). In these
guests, both the headers and the data parts of the ringbuffers need to
be aligned to the PAGE_SIZE, because 1) some of the ringbuffers will be
mapped into userspace and 2) we use "double mapping" mechanism to
support fast wrap-around, and "double mapping" relies on ringbuffers
being page-aligned. However, the Hyper-V hypervisor only uses 4k
(HV_HYP_PAGE_SIZE) headers. Our solution to this is that we always make
the headers of ringbuffers take one guest page and when GPADL is
established between the guest and hypervisor, the only first 4k of
header is used. To handle this special case, we need the types of GPADL
to differ different guest memory usage for GPADL.

Type enum is introduced along with several general interfaces to
describe the differences between normal buffer GPADL and ringbuffer
GPADL.

Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Link: https://lore.kernel.org/r/20200916034817.30282-4-boqun.feng@gmail.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>

The spinlock is (now) *not used to protect test-and-set accesses
to attributes of the structure or sc_list operations.

There is, AFAICT, a distinct lack of {WRITE,READ}_ONCE()s in the
handling of channel->state, but the changes below do not seem to
make things "worse".  ;-)

Signed-off-by: Andrea Parri (Microsoft) <parri.andrea@gmail.com>
Link: https://lore.kernel.org/r/20200617164642.37393-9-parri.andrea@gmail.com
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Signed-off-by: Wei Liu <wei.liu@kernel.org>

The field is read only in numa_node_show() and it is already stored twice
(after a call to cpu_to_node()) in target_cpu_store() and init_vp_index();
there is no need to "cache" its value in the channel data structure.

Signed-off-by: Andrea Parri (Microsoft) <parri.andrea@gmail.com>
Link: https://lore.kernel.org/r/20200617164642.37393-3-parri.andrea@gmail.com
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Signed-off-by: Wei Liu <wei.liu@kernel.org>

The field is read only in __vmbus_open() and it is already stored twice
(after a call to hv_cpu_number_to_vp_number()) in target_cpu_store() and
init_vp_index(); there is no need to "cache" its value in the channel
data structure.

Suggested-by: Michael Kelley <mikelley@microsoft.com>
Signed-off-by: Andrea Parri (Microsoft) <parri.andrea@gmail.com>
Link: https://lore.kernel.org/r/20200617164642.37393-2-parri.andrea@gmail.com
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Signed-off-by: Wei Liu <wei.liu@kernel.org>

init_vp_index() uses the (per-node) hv_numa_map[] masks to record the
CPUs allocated for channel interrupts at a given time, and distribute
the performance-critical channels across the available CPUs: in part.,
the mask of "candidate" target CPUs in a given NUMA node, for a newly
offered channel, is determined by XOR-ing the node's CPU mask and the
node's hv_numa_map.  This operation/mechanism assumes that no offline
CPUs is set in the hv_numa_map mask, an assumption that does not hold
since such mask is currently not updated when a channel is removed or
assigned to a different CPU.

To address the issues described above, this adds hooks in the channel
removal path (hv_process_channel_removal()) and in target_cpu_store()
in order to clear, resp. to update, the hv_numa_map[] masks as needed.
This also adds a (missed) update of the masks in init_vp_index() (cf.,
e.g., the memory-allocation failure path in this function).

Like in the case of init_vp_index(), such hooks require to determine
if the given channel is performance critical.  init_vp_index() does
this by parsing the channel's offer, it can not rely on the device
data structure (device_obj) to retrieve such information because the
device data structure has not been allocated/linked with the channel
by the time that init_vp_index() executes.  A similar situation may
hold in hv_is_alloced_cpu() (defined below); the adopted approach is
to "cache" the device type of the channel, as computed by parsing the
channel's offer, in the channel structure itself.

Fixes: 7527810573436f ("Drivers: hv: vmbus: Introduce the CHANNELMSG_MODIFYCHANNEL message type")
Signed-off-by: Andrea Parri (Microsoft) <parri.andrea@gmail.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Link: https://lore.kernel.org/r/20200522171901.204127-3-parri.andrea@gmail.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>

The current codebase makes use of the zero-length array language
extension to the C90 standard, but the preferred mechanism to declare
variable-length types such as these ones is a flexible array member[1][2],
introduced in C99:

struct foo {
        int stuff;
        struct boo array[];
};

By making use of the mechanism above, we will get a compiler warning
in case the flexible array does not occur last in the structure, which
will help us prevent some kind of undefined behavior bugs from being
inadvertently introduced[3] to the codebase from now on.

Also, notice that, dynamic memory allocations won't be affected by
this change:

"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]

sizeof(flexible-array-member) triggers a warning because flexible array
members have incomplete type[1]. There are some instances of code in
which the sizeof operator is being incorrectly/erroneously applied to
zero-length arrays and the result is zero. Such instances may be hiding
some bugs. So, this work (flexible-array member conversions) will also
help to get completely rid of those sorts of issues.

This issue was found with the help of Coccinelle.

[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit 76497732932f ("cxgb3/l2t: Fix undefined behaviour")

Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Link: https://lore.kernel.org/r/20200507185323.GA14416@embeddedor
Signed-off-by: Wei Liu <wei.liu@kernel.org>

For each storvsc_device, storvsc keeps track of the channel target CPUs
associated to the device (alloced_cpus) and it uses this information to
fill a "cache" (stor_chns) mapping CPU->channel according to a certain
heuristic.  Update the alloced_cpus mask and the stor_chns array when a
channel of the storvsc device is re-assigned to a different CPU.

Signed-off-by: Andrea Parri (Microsoft) <parri.andrea@gmail.com>
Cc: "James E.J. Bottomley" <jejb@linux.ibm.com>
Cc: "Martin K. Petersen" <martin.petersen@oracle.com>
Cc: <linux-scsi@vger.kernel.org>
Link: https://lore.kernel.org/r/20200406001514.19876-12-parri.andrea@gmail.com
Reviewed-by; Long Li <longli@microsoft.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
[ wei: fix a small issue reported by kbuild test robot <lkp@intel.com> ]
Signed-off-by: Wei Liu <wei.liu@kernel.org>

VMBus version 4.1 and later support the CHANNELMSG_MODIFYCHANNEL(22)
message type which can be used to request Hyper-V to change the vCPU
that a channel will interrupt.

Introduce the CHANNELMSG_MODIFYCHANNEL message type, and define the
vmbus_send_modifychannel() function to send CHANNELMSG_MODIFYCHANNEL
requests to the host via a hypercall.  The function is then used to
define a sysfs "store" operation, which allows to change the (v)CPU
the channel will interrupt by using the sysfs interface.  The feature
can be used for load balancing or other purposes.

One interesting catch here is that Hyper-V can *not* currently ACK
CHANNELMSG_MODIFYCHANNEL messages with the promise that (after the ACK
is sent) the channel won't send any more interrupts to the "old" CPU.

The peculiarity of the CHANNELMSG_MODIFYCHANNEL messages is problematic
if the user want to take a CPU offline, since we don't want to take a
CPU offline (and, potentially, "lose" channel interrupts on such CPU)
if the host is still processing a CHANNELMSG_MODIFYCHANNEL message
associated to that CPU.

It is worth mentioning, however, that we have been unable to observe
the above mentioned "race": in all our tests, CHANNELMSG_MODIFYCHANNEL
requests appeared *as if* they were processed synchronously by the host.

Suggested-by: Michael Kelley <mikelley@microsoft.com>
Signed-off-by: Andrea Parri (Microsoft) <parri.andrea@gmail.com>
Link: https://lore.kernel.org/r/20200406001514.19876-11-parri.andrea@gmail.com
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
[ wei: fix conflict in channel_mgmt.c ]
Signed-off-by: Wei Liu <wei.liu@kernel.org>