[ofw] cl_spinlock_osd.h

[Smith, Stan]

Remove code which is duplicated in the implementation of MS code KeAcquireSpinLock(), as indicated by our Microsoft OFA members.

Signed-off-by: ***@intel.com

--- inc/kernel/complib/cl_spinlock_osd.h Wed Mar 06 10:12:19 2013
+++ inc/kernel/complib/cl_spinlock_osd.h Tue Feb 26 12:08:56 2013
@@ -91,15 +91,8 @@
cl_spinlock_acquire(
IN cl_spinlock_t* const p_spinlock )
{
- KIRQL irql = KeGetCurrentIrql();
CL_ASSERT( p_spinlock );
-
- if (irql == DISPATCH_LEVEL) {
- KeAcquireSpinLockAtDpcLevel( &p_spinlock->lock );
- p_spinlock->irql = irql;
- }
- else
- KeAcquireSpinLock( &p_spinlock->lock, &p_spinlock->irql );
+ KeAcquireSpinLock( &p_spinlock->lock, &p_spinlock->irql );
}

Basically move from

CL_INLINE void
cl_spinlock_acquire(
IN cl_spinlock_t* const p_spinlock )
{
KIRQL irql = KeGetCurrentIrql();
CL_ASSERT( p_spinlock );

if (irql == DISPATCH_LEVEL) {
KeAcquireSpinLockAtDpcLevel( &p_spinlock->lock );
p_spinlock->irql = irql;
}
else
KeAcquireSpinLock( &p_spinlock->lock, &p_spinlock->irql );
}

To

CL_INLINE void
cl_spinlock_acquire(
IN cl_spinlock_t* const p_spinlock )
{
CL_ASSERT( p_spinlock );
KeAcquireSpinLock( &p_spinlock->lock, &p_spinlock->irql );
}

[Fab Tillier]

Is there similar logic for the release case?

Anyway, this looks good to me.  It would be great to move away from complib altogether (at least in the kernel drivers).

Cheers,
-Fab


From: ofw-***@lists.openfabrics.org [mailto:ofw-***@lists.openfabrics.org] On Behalf Of Smith, Stan
Sent: Tuesday, April 9, 2013 4:25 PM
To: ***@lists.openfabrics.org
Subject: [ofw] cl_spinlock_osd.h

Remove code which is duplicated in the implementation of MS code KeAcquireSpinLock(), as indicated by our Microsoft OFA members.

Signed-off-by: ***@intel.com

--- inc/kernel/complib/cl_spinlock_osd.h             Wed Mar 06 10:12:19 2013
+++ inc/kernel/complib/cl_spinlock_osd.h          Tue Feb 26 12:08:56 2013
@@ -91,15 +91,8 @@
cl_spinlock_acquire(
                IN           cl_spinlock_t* const       p_spinlock )
{
-              KIRQL irql = KeGetCurrentIrql();
               CL_ASSERT( p_spinlock );
-
-              if (irql == DISPATCH_LEVEL) {
-                              KeAcquireSpinLockAtDpcLevel( &p_spinlock->lock );
-                              p_spinlock->irql = irql;
-              }
-              else
-                              KeAcquireSpinLock( &p_spinlock->lock, &p_spinlock->irql );
+             KeAcquireSpinLock( &p_spinlock->lock, &p_spinlock->irql );
}

Basically move from

CL_INLINE void
cl_spinlock_acquire(
                IN           cl_spinlock_t* const       p_spinlock )
{
                KIRQL irql = KeGetCurrentIrql();
                CL_ASSERT( p_spinlock );

                if (irql == DISPATCH_LEVEL) {
                                KeAcquireSpinLockAtDpcLevel( &p_spinlock->lock );
                                p_spinlock->irql = irql;
                }
                else
                                KeAcquireSpinLock( &p_spinlock->lock, &p_spinlock->irql );
}

To

CL_INLINE void
cl_spinlock_acquire(
                IN           cl_spinlock_t* const       p_spinlock )
{
                CL_ASSERT( p_spinlock );
                KeAcquireSpinLock( &p_spinlock->lock, &p_spinlock->irql );
}

[Smith, Stan]

Fab writes: Is there similar logic for the release case?

Yes - Similar logic can be applied to the release side - MPI/DAPL/IPOIB testing all passed.


--- C:/Users/scsmith/AppData/Local/Temp/cl_spinlock_osd.h-revBASE.svn000.tmp.h Wed May 16 14:38:16 2012

+++ C:/Users/scsmith/Documents/openIB-windows/ofw/gen1/branches/WOF3-2/inc/kernel/complib/cl_spinlock_osd.h Thu Apr 11 08:43:59 2013

@@ -91,15 +91,8 @@

cl_spinlock_acquire(

IN cl_spinlock_t* const p_spinlock )

{

- KIRQL irql = KeGetCurrentIrql();

CL_ASSERT( p_spinlock );

-

- if (irql == DISPATCH_LEVEL) {

- KeAcquireSpinLockAtDpcLevel( &p_spinlock->lock );

- p_spinlock->irql = irql;

- }

- else

- KeAcquireSpinLock( &p_spinlock->lock, &p_spinlock->irql );

+ KeAcquireSpinLock( &p_spinlock->lock, &p_spinlock->irql );

}


CL_INLINE void cl_spinlock_release(



#ifdef NTDDI_WIN8

@@ -114,10 +107,7 @@

{

CL_ASSERT( p_spinlock );

- if (p_spinlock->irql == DISPATCH_LEVEL)

- KeReleaseSpinLockFromDpcLevel( &p_spinlock->lock );

- else

- KeReleaseSpinLock( &p_spinlock->lock, p_spinlock->irql );

+ KeReleaseSpinLock( &p_spinlock->lock, p_spinlock->irql );

}


From: Fab Tillier [mailto:***@microsoft.com]
Sent: Wednesday, April 10, 2013 10:29 AM
To: Smith, Stan
Subject: RE: cl_spinlock_osd.h

Is there similar logic for the release case?

CL_INLINE void cl_spinlock_release(
IN cl_spinlock_t* const p_spinlock )
{
CL_ASSERT( p_spinlock );

if (p_spinlock->irql == DISPATCH_LEVEL)
KeReleaseSpinLockFromDpcLevel( &p_spinlock->lock );
else
KeReleaseSpinLock( &p_spinlock->lock, p_spinlock->irql );
}

?

Anyway, this looks good to me. It would be great to move away from complib altogether (at least in the kernel drivers).

Cheers,
-Fab

From: ofw-***@lists.openfabrics.org<mailto:ofw-***@lists.openfabrics.org> [mailto:ofw-***@lists.openfabrics.org] On Behalf Of Smith, Stan
Sent: Tuesday, April 9, 2013 4:25 PM
To: ***@lists.openfabrics.org<mailto:***@lists.openfabrics.org>
Subject: [ofw] cl_spinlock_osd.h

Remove code which is duplicated in the implementation of MS code KeAcquireSpinLock(), as indicated by our Microsoft OFA members.

Signed-off-by: ***@intel.com<mailto:***@intel.com>

--- inc/kernel/complib/cl_spinlock_osd.h Wed Mar 06 10:12:19 2013
+++ inc/kernel/complib/cl_spinlock_osd.h Tue Feb 26 12:08:56 2013
@@ -91,15 +91,8 @@
cl_spinlock_acquire(
IN cl_spinlock_t* const p_spinlock )
{
- KIRQL irql = KeGetCurrentIrql();
CL_ASSERT( p_spinlock );
-
- if (irql == DISPATCH_LEVEL) {
- KeAcquireSpinLockAtDpcLevel( &p_spinlock->lock );
- p_spinlock->irql = irql;
- }
- else
- KeAcquireSpinLock( &p_spinlock->lock, &p_spinlock->irql );
+ KeAcquireSpinLock( &p_spinlock->lock, &p_spinlock->irql );
}

Basically move from

CL_INLINE void
cl_spinlock_acquire(
IN cl_spinlock_t* const p_spinlock )
{
KIRQL irql = KeGetCurrentIrql();
CL_ASSERT( p_spinlock );

if (irql == DISPATCH_LEVEL) {
KeAcquireSpinLockAtDpcLevel( &p_spinlock->lock );
p_spinlock->irql = irql;
}
else
KeAcquireSpinLock( &p_spinlock->lock, &p_spinlock->irql );
}

To

CL_INLINE void
cl_spinlock_acquire(
IN cl_spinlock_t* const p_spinlock )
{
CL_ASSERT( p_spinlock );
KeAcquireSpinLock( &p_spinlock->lock, &p_spinlock->irql );
}