Kmalloc while holding a spinlock

["Leonidas ." <leonidas137@xxxxxxxxx>]

Hi List,

I need to do lot of small allocations (around 70-80 bytes) from a critical region
while holding a spinlock. Total number of such allocation could go upto tens of 
thousands in few hours. So all these allocations use GFP_ATOMIC flag instead of
GFP_KERNEL. As per my understanding, GFP_ATOMIC can never fail, and I dont
want my allocations to fail. But this could be a bad idea right? Using GFP_ATOMIC
for all these allocations?

Is there any way I can avoid using GFP_ATOMIC without preallocating the buffers?

Something like,

ptr =  kmalloc(..., GFP_KERNEL | GFP_DONT_SLEEP );

if( !ptr ) {

 /*GFP_KERNEL failed, Use GFP_ATOMIC pool*/
 ptr =  kmalloc(..., GFP_ATOMIC );

}
/*Go ahead and do stuff*/


Is there any flag which fetches memory from GFP_KERNEL pool with a guarantee that 
it will not sleep? 

Is GFP_NOWAIT | GFP_KERNEL guarantee that?

Is ( GFP_NOWAIT | GFP_KERNEL ) == ( GFP_ATOMIC - dont access emergency pools)?

I think GFP_KERNEL | GFP_NOWAIT will solve my issue, what say?

-Leo.