Send Patchs with email

Commit 95158a89dd50 ("sched,rt: Use the full cpumask for balancing")
allow find_lock_lowest_rq to pick a task with migration disabled.
This commit is intended to push the current running task on this CPU
away.

There is a race scenario, which allows a migration disabled task to
be migrated to another CPU.

When there is a RT task with higher priority, rt sched class was
intended to migrate higher priority task to lowest rq via push_rt_tasks,
this WARNING will happen here.

With the system running on PREEMPT_RT, rt_spin_lock will disable
migration, this will make the problem easier to reproduce.

At the moment the following QEMU command line triggers an assertion
failure On xlnx-versal SOC:
qemu-system-aarch64 \
    -machine xlnx-versal-virt -nographic -smp 2 -m 128 \
    -fsdev local,id=shareid,path=${HOME}/work,security_model=none \
    -device virtio-9p-device,fsdev=shareid,mount_tag=share \
    -fsdev local,id=shareid1,path=${HOME}/Music,security_model=none \
    -device virtio-9p-device,fsdev=shareid1,mount_tag=share1

qemu-system-aarch64: ../migration/savevm.c:860:
vmstate_register_with_alias_id:
Assertion `!se->compat || se->instance_id == 0' failed.

This problem was fixed on arm virt platform in patch

This problem was fixed on arm virt platform in patch

https://lists.nongnu.org/archive/html/qemu-devel/2016-07/msg01119.html

It works perfectly on arm virt platform. but there is still there on
xlnx-versal SOC.

The main difference between arm virt and xlnx-versal is they use
different way to create virtio-mmio qdev. on arm virt, it calls
sysbus_create_simple("virtio-mmio", base, pic[irq]); which will call
sysbus_mmio_map internally and assign base address to subsys device
mmio correctly. but xlnx-versal's implements won't do this.

However, xlnx-versal can't switch to sysbus_create_simple() to create
virtio-mmio device. It's because xlnx-versal's cpu use
VersalVirt.soc.fpd.apu.mr as it's memory. which is subregion of
system_memory. sysbus_create_simple will add virtio to system_memory,
which can't be accessed by cpu.

     T21124                   p9_read_work
======================== second trans =================================
p9_client_walk
  p9_client_rpc
    p9_client_prepare_req
      p9_tag_alloc
        req = kmem_cache_alloc(p9_req_cache, GFP_NOFS);
        tag = idr_alloc
        << preempted >>
        req->tc.tag = tag;
                            /* req->[refcount/tag] == uninitialized */
                            m->rreq = p9_tag_lookup(m->client, m->rc.tag);
                              /* increments uninitalized refcount */

        refcount_set(&req->refcount, 2);
                            /* cb drops one ref */
                            p9_client_cb(req)
                            /* reader thread drops its ref:
                               request is incorrectly freed */
                            p9_req_put(req)
    /* use after free and ref underflow */
    p9_req_put(req)

We can solve this by simply assign mmio[0].addr directly. makes
virtio_mmio_bus_get_dev_path to produce correct unique device path.