drm/amd/amdgpu: fixing read wrong pf2vf data in SRIOV

stable inclusion
from stable-v5.10.146
commit fda04a0bab7fdc3ac17508f00584580f858ade76
category: bugfix
bugzilla: https://gitee.com/openeuler/kernel/issues/I6D0VX

Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=fda04a0bab7fdc3ac17508f00584580f858ade76



--------------------------------

commit 9a458402 upstream.

[Why]
This fixes 892deb48 ("drm/amdgpu: Separate vf2pf work item init from virt data exchange").
we should read pf2vf data based at mman.fw_vram_usage_va after gmc
sw_init. commit 892deb48 breaks this logic.

[How]
calling amdgpu_virt_exchange_data in amdgpu_virt_init_data_exchange to
set the right base in the right sequence.

v2:
call amdgpu_virt_init_data_exchange after gmc sw_init to make data
exchange workqueue run

v3:
clean up the code logic

v4:
add some comment and make the code more readable

Fixes: 892deb48 ("drm/amdgpu: Separate vf2pf work item init from virt data exchange")
Signed-off-by: Jingwen Chen <Jingwen.Chen2@amd.com>
Reviewed-by: Horace Chen <horace.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Jialin Zhang <zhangjialin11@huawei.com>
Reviewed-by: Zheng Zengkai <zhengzengkai@huawei.com>