Staging: hv: blk dev depends on SCSI

hv block driver uses scsi_*() interfaces so it should depend on SCSI.

ERROR: "scsi_normalize_sense" [drivers/staging/hv/hv_blkvsc.ko] undefined!
ERROR: "scsi_print_sense_hdr" [drivers/staging/hv/hv_blkvsc.ko] undefined!

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: Replace typedef SG_BUFFER_LIST by struct scatterlist

typedef SG_BUFFER_LIST is removed and its uses are replaced by the use of
struct scatterlist.

Signed-off-by: Nicolas Palix <npalix@diku.dk>
Cc: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove duplicated osd.o inclusions

Now that we have properly prefixed the osd.c functions, we don't need to
include it in each of the modules.  So only build it into the hv_vmbus
module.

Export the symbols that the other hv_ modules need, so that they can
properly find them.

Cc: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove timer wrapper functions

Use a real timer (there's only one in the code), no wrapper is needed,
it just increases the complexity for no reason.

Cc: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: osd: add osd_ prefix to global functions

Put a "osd_" prefix on the osd.c functions in order for us to play nicer
in the kernel namespace.

Cc: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: osd: remove physical address wrapper functions

Use the real functions the kernel provides, so that people can see what
is actually going on in the code easier.

Cc: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove custom rdmsrl and wrmsrl functions

Use the ones that the kernel provides, they do it correctly.

Cc: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove custom cpuid function

Use the one that the kernel provides, it does it correctly.

Cc: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove wrapper functions around kmap_

Remove PageMapVirtualAddress() and PageUnmapVirtualAddress() which
were wrappers around kmap_atomic() and kunmap_atomic()

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove wrapper function VirtualFree

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove wrapper functions for atomic operations

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove wrapper functions for bit operations

There were several Bit* functions that did nothing but call the kernel
functions with the parameters reversed.  Remove these and call the
functions directly.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove WaitEventClose()

All WaitEventClose() close did was call kfree(), so get rid of it and
replace it with a call to kfree()

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rework use of workqueues in osd

Change the usage of workqueues to be consistant with other parts of
the kernel.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: fix sparse NULL pointer warnings

This fixes up all of the sparse warnings where NULL should be used
instead of 0.

Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: fix sparse function warnings

This fixes up all of the sparse warnings about functions not being
properly declared.  Meaning, void functions need to say they are a void
function, otherwise the compiler assumes it is an integer here.

Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: fix sparse static warnings

This fixes up all of the sparse warnings about static functions.

Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: Remove typedef STORVSC_REQUEST and PSTORVSC_REQUEST

typedef STORVSC_REQUEST and PSTORVSC_REQUEST are removed and their
usages are replace by the use of struct hv_storvsc_request and
struct hv_storvsc_request * respectively.

Here is the semantic patch generated to perform this transformation:
(http://coccinelle.lip6.fr/)

//<smpl>
@rm_STORVSC_REQUEST@
@@
-typedef struct _STORVSC_REQUEST
+struct hv_storvsc_request
{...}
-STORVSC_REQUEST
;

@rm_PSTORVSC_REQUEST@
@@
-typedef struct _STORVSC_REQUEST *PSTORVSC_REQUEST;
+struct hv_storvsc_request;

@fixtypedef_STORVSC_REQUEST@
typedef STORVSC_REQUEST;
@@
-STORVSC_REQUEST
+struct hv_storvsc_request

@fixstruct__STORVSC_REQUEST@
@@
struct
-_STORVSC_REQUEST
+hv_storvsc_request

@fixtypedef_PSTORVSC_REQUEST@
typedef PSTORVSC_REQUEST;
@@
-PSTORVSC_REQUEST
+struct hv_storvsc_request*
//</smpl>

Signed-off-by: Nicolas Palix <npalix@diku.dk>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: Remove typedef NETVSC_PACKET and PNETVSC_PACKET

typedef NETVSC_PACKET and PNETVSC_PACKET are removed and their usages
are replace by the use of struct hv_netvsc_packet and
struct hv_netvsc_packet * respectively.

Here is the semantic patch generated to perform this transformation:
(http://coccinelle.lip6.fr/)

//<smpl>
@rm_NETVSC_PACKET@
@@
-typedef struct _NETVSC_PACKET
+struct hv_netvsc_packet
{...}
-NETVSC_PACKET
;

@rm_PNETVSC_PACKET@
@@
-typedef struct _NETVSC_PACKET *PNETVSC_PACKET;
+struct hv_netvsc_packet;

@fixtypedef_NETVSC_PACKET@
typedef NETVSC_PACKET;
@@
-NETVSC_PACKET
+struct hv_netvsc_packet

@fixstruct__NETVSC_PACKET@
@@
struct
-_NETVSC_PACKET
+hv_netvsc_packet

@fixtypedef_PNETVSC_PACKET@
typedef PNETVSC_PACKET;
@@
-PNETVSC_PACKET
+struct hv_netvsc_packet*
//</smpl>

Signed-off-by: Nicolas Palix <npalix@diku.dk>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: Remove typedef DRIVER_OBJECT and PDRIVER_OBJECT

typedef DRIVER_OBJECT and PDRIVER_OBJECT are removed and their usages
are replace by the use of struct hv_driver and struct hv_driver *
respectively.

Here is the semantic patch generated to perform this transformation:
(http://coccinelle.lip6.fr/)

//<smpl>
@rm_DRIVER_OBJECT@
@@
-typedef struct _DRIVER_OBJECT
+struct hv_driver
{...}
-DRIVER_OBJECT
;

@rm_PDRIVER_OBJECT@
@@
-typedef struct _DRIVER_OBJECT *PDRIVER_OBJECT;
+struct hv_driver;

@fixtypedef_DRIVER_OBJECT@
typedef DRIVER_OBJECT;
@@
-DRIVER_OBJECT
+struct hv_driver

@fixstruct__DRIVER_OBJECT@
@@
struct
-_DRIVER_OBJECT
+hv_driver

@fixtypedef_PDRIVER_OBJECT@
typedef PDRIVER_OBJECT;
@@
-PDRIVER_OBJECT
+struct hv_driver*
//</smpl>

Signed-off-by: Nicolas Palix <npalix@diku.dk>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: properly fix the printk() warnings

This fixes the printk() warnings on all platforms now (x86-64 and i386).

Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove more printk() warnings

This should fix up the rest of the printk() warnings on an i386 build

Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove HANDLE typedef

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv remove TIMER typedef

Remove the TIMER typedef and also replace HANDLE types that use
the timer calls.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove WAITEVENT typedef

Remove the WAITEVENT typedef and also replace HANDLE types that use
the WaitEvent calls with struct osd_waitevent.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: comment out blkdev variable in blkvsc_ioctl

This variable generated an unused variable warning due to other code
in the fuction being commented out.  This comments out the variable
defination so that the code compiles without warnings.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: check return value of driver_for_each_device()

The return value of driver_for_each_device() is now checked.  A
non-zero value simply generates a warning message, but it's better
than not checking at all.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: Transform PDEVICE_OBJECT and DEVICE_OBJECT typedefs into their corresponding structs

Remove typedef DEVICE_OBJECT and use a struct named hv_device instead.
Remove typedef PDEVICE_OBJECT which aliases a pointer and use
struct hv_device * instead.

Here is the semantic patch to perform this transformation:
(http://coccinelle.lip6.fr/)

//<smpl>
@rm_PDEVICE_OBJECT@
@@
-typedef struct _DEVICE_OBJECT *PDEVICE_OBJECT;

@rm_DEVICE_OBJECT@
@@
-typedef struct _DEVICE_OBJECT
+struct hv_device
{...}
-DEVICE_OBJECT
;

@fixtypedef_PDEVICE_OBJECT@
typedef PDEVICE_OBJECT;
@@
-PDEVICE_OBJECT
+struct hv_device*

@fixtypedef_DEVICE_OBJECT@
typedef DEVICE_OBJECT;
@@
-DEVICE_OBJECT
+struct hv_device

@fixstruct__DEVICE_OBJECT@
@@
struct
-_DEVICE_OBJECT
+hv_device
//</smpl>

Signed-off-by: Nicolas Palix <npalix@diku.dk>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: Remove WORKQUEUE typedef

WORKQUEUE was a wrapper around struct workqueue_struct so just use
that instead.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove NETVSC_DEVICE typedef

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove VMBUS_MSGINFO typedef

Note that this struct doesn't appear to be used anywhere and can
probably be removed.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove VMBUS_CONNECTION typedef

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove VMBUS_CONNECT_STATE typedef

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove VMBUS_CHANNEL_PACKET_MULITPAGE_BUFFER typedef

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove VMBUS_CHANNEL_PACKET_PAGE_BUFFER typedef

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: Hv.c: remove unused physAddr

physAddr was declared but never used in HvInit()

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: vmbus_drv.c: remove unused structs

vmbus_ctl_table_hdr, vmbus_dev_ctl_table, vmbus_ctl_table, and
vmus_root_ctl_table were never used.  This removes them.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: check return value of device_register()

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: check return value of bus_register()

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: vmbus_drv.c: fix print formatting

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: storvsc_drv.c: fix print formatting

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: RndisFilter.c: fix print formatting

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: NetVsc.c: fix print formatting

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: Remove X2V_LINUX check

Remove preprocessor check for X2V_LINUX in osd.c

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: fix blkvsc_ioctl() parameters

blkvsc_ioctl() had the wrong parameter list for  struct
block_device_operations

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: fix blkvsc_release() parameters

blkvsc_release() had the wrong parameter list for  struct
block_device_operations

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: fix blkvsc_open() parameters

blkvsc_open() had the wrong parameter list for  struct
block_device_operations

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: blkvsc_drv.c: fix print formatting

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: StorVsc.c: fix print formatting

There were a few places that used %lx when they should have used %x
and a few places that used %d when they should have used %ld

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: Remove C99 comments

Remove C99 // comments with traditional /* */ comments

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: Use %ld instead of %d for a long ints

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: force hyper-v drivers to be built as a module

Right now they can not be built into the kernel due to global
symbol name conflicts that this code is causing.

Thanks to Nicolas Palix <npalix@diku.dk> for pointing this out.

Reported-by: Nicolas Palix <npalix@diku.dk>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: Transform some kzalloc calls to kcalloc

Here is a short excerpt of the semantic patch performing
this transformation:
(http://www.emn.fr/x-info/coccinelle/)

//<smpl>
@@
expression E1,E2,E3;
@@
- kzalloc(E1*sizeof(E2),E3)
+ kcalloc(E1,sizeof(E2),E3)

@@
expression E1,E3;
type T;
@@
- kzalloc(E1*sizeof(T),E3)
+ kcalloc(E1,sizeof(T),E3)
//</smpl>

Signed-off-by: Nicolas Palix <npalix@diku.dk>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: Remove compatibility ifdefry

Signed-off-by: Moritz Muehlenhoff <jmm@debian.org>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: osd: remove GetTickCount and GetTimestamp wrappers

No one was even using them.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: fix up printk warnings

After LogMsg was converted to printk, lots of build warnings showed up
as no one was checking the arguments to LogMsg.  This patch fixes them
all.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: osd: remove PrintBytes wrapper

Use the "real" print_hex_dump_bytes call instead of a wrapper function.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: osd: remove LogMsg wrapper

Use the "real" printk call instead of a wrapper function.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: osd: remove MemoryFence wrapper

Use the "real" mb call instead of a wrapper function.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: osd: remove Sleep wrapper

Use the "real" udelay call instead of a wrapper function.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: osd: remove spinlock wrapper functions

Now that there are no users of the wrapper functions for spinlocks,
remove them.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: make gVmbusConnection.ChannelLock a real spinlock

Don't use the wrapper functions for this lock, make it a real
lock so that we know what is going on.

I don't think we really want to be doing a irqsave for this code, but I
left it alone to preserve the original codepath.  It should be reviewed
later.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: make gVmbusConnection.ChannelMsgLock a real spinlock

Don't use the wrapper functions for this lock, make it a real
lock so that we know what is going on.

I don't think we really want to be doing a irqsave for this code, but I
left it alone to preserve the original codepath.  It should be reviewed
later.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: make netDevice->ReceivePacketListLock a real spinlock

Don't use the wrapper functions for this lock, make it a real
lock so that we know what is going on.

I don't think we really want to be doing a irqsave for this code, but I
left it alone to preserve the original codepath.  It should be reviewed
later.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: make Device->RequestLock a real spinlock

Don't use the wrapper functions for this lock, make it a real
lock so that we know what is going on.

I don't think we really want to be doing a irqsave for this code, but I
left it alone to preserve the original codepath.  It should be reviewed
later.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: make RingInfo->RingLock a real spinlock

Don't use the wrapper functions for this lock, make it a real
lock so that we know what is going on.

I don't think we really want to be doing a irqsave for this code, but I
left it alone to preserve the original codepath.  It should be reviewed
later.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: make Channel->InboundLock a real spinlock

Don't use the wrapper functions for this lock, make it a real
lock so that we know what is going on.

I don't think we really want to be doing a irqsave for this code, but I
left it alone to preserve the original codepath.  It should be reviewed
later.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: osd: remove MemFree wrapper

Use the "real" kfree call instead of a wrapper function.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: osd: remove MemAllocAtomic wrapper

Use the "real" kmalloc call instead of a wrapper function.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: osd: remove MemAllocZeroed wrapper

Use the "real" kzalloc call instead of a wrapper function.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: osd: remove MemAlloc wrapper

Use the "real" kmalloc call instead of a wrapper function.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove TRUE, FALSE, and NULL usage

Don't define things that are either already provided (like NULL), or you
shouldn't use (like TRUE and FALSE).

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove FIELD_OFFSET usage

This isn't needed, or even used, at all.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove UNUSED_VAR usage

This isn't needed at all.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove STRUCT_PACKED and STRUCT_ALIGNED defines

Use the correct __attribute__((packed)) one if it's really needed.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove PAGE_SIZE and PAGE_SHIFT and __builtin functions

The kernel provides all of this, and actually gets it correct, so don't
try to redefine these types of things.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove MIN and MAX usages

The kernel has the "correct" min() and max() functions, so use those.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove #defines from osd.c

Remove the unneeded #defines from osd.c

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove BOOL and BOOLEAN typedefs

The BOOL and BOOLEAN typedefs are now removed from the Hyper-V driver
code.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove DWORD and BYTE typedefs

No one was even using them...

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove SIZE_T typedef

The SIZE_T typedef is now removed from the Hyper-V driver code.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove ULONG and LONG typedefs

The ULONG and LONG typedefs are now removed from the Hyper-V driver
code.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove ULONG_PTR typedef

The ULONG_PTR typedef is now removed from the Hyper-V driver code.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove ULONGLONG and LONGLONG typedefs

The ULONGLONG and LONGLONG typedefs are now removed from the Hyper-V
driver code.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove USHORT typedef

The USHORT typedef is now removed from the Hyper-V driver code.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove UINT64 and INT64 and UCHAR typedefs

The UINT64 and INT64 and UCHAR typedefs are now removed from the Hyper-V
driver code.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove UINT32 and INT32 typedefs

The UINT32 and INT32 typedefs are now removed from the Hyper-V driver
code.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove UINT16 and INT16 typedefs

The UINT16 and INT16 typedefs are now removed from the Hyper-V driver
code.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove UINT8 and INT8 typedefs

The UINT8 and INT8 typedefs are now removed from the Hyper-V driver
code.

Had to include <linux/kernel.h> in a few places to get the
build to work properly as well.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove VOID typedef

The VOID typedef is now removed from the Hyper-V driver code.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove PVOID typedef

The PVOID typedef is now removed from the Hyper-V driver code.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove INTERNAL typedef

The INTERNAL typedef is now removed from the Hyper-V driver code.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc: fix up driver_data usage

driver_data is gone now from struct device, so use the proper functions
to access it instead.

Thanks to Bill Pemberton for pointing out this build error.

Cc: Bill Pemberton <wfp5p@viridian.itc.Virginia.EDU>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: add the Hyper-V virtual network driver to the build

Add the Hyper-V virtual network driver to the kernel build system.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: make the Hyper-V virtual network driver build

The #define KERNEL_2_6_27 needs to be set, and I adjusted the include
directories a bit to get things to build properly.

The driver was changed to use net_device_ops, as that is needed to build
and operate properly now.

The hv_netvsc code should now build with no errors.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: blkvsc: fix up driver_data usage

driver_data is gone now from struct device, so use the proper functions
to access it instead.

Thanks to Bill Pemberton for pointing out this build error.

Cc: Bill Pemberton <wfp5p@viridian.itc.Virginia.EDU>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: add the Hyper-V virtual block driver to the build

Add the Hyper-V virtual block driver to the kernel build system.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: make the Hyper-V virtual block driver build

The #define KERNEL_2_6_27 needs to be set, and I adjusted the include
directories a bit to get things to build properly.

I also fixed up the direct access of bus_id, as that field is now gone.
Lots of block api changes were needed, and I don't think I got it
all correct.  It would be great of someone who knows the block api better
could review it.

The hv_blkvsc code should now build, with no errors.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: storvsc: fix up driver_data usage

driver_data is gone now from struct device, so use the proper functions
to access it instead.

Thanks to Bill Pemberton for pointing out this build error.

Cc: Bill Pemberton <wfp5p@viridian.itc.Virginia.EDU>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: add the Hyper-V virtual scsi driver to the build

Add the Hyper-V virtual scsi driver to the kernel build system.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>