kmem slab fixes

- Default SPL_KMEM_CACHE_DELAY changed to 15 to match Solaris.
- Aged out slab checking occurs every SPL_KMEM_CACHE_DELAY / 3.
- skc->skc_reap tunable added whichs allows callers of
  spl_slab_reclaim() to cap the number of slabs reclaimed.
  On Solaris all eligible slabs are always reclaimed, and this
  is still the default behavior.  However, I suspect that is
  not always wise for reasons such as in the next comment.
- spl_slab_reclaim() added cond_resched() while walking the
  slab/object free lists.  Soft lockups were observed when
  freeing large numbers of vmalloc'd slabs/objets.
- spl_slab_reclaim() 'sks->sks_ref > 0' check changes from
  incorrect 'break' to 'continue' to ensure all slabs are
  checked.
- spl_cache_age() reworked to avoid a deadlock with
  do_flush_tlb_all() which occured because we slept waiting
  for completion in spl_cache_age().  To waiting for magazine
  reclamation to finish is not required so we no longer wait.
- spl_magazine_create() and spl_magazine_destroy() shifted
  back to using for_each_online_cpu() instead of the
  spl_on_each_cpu() approach which was of course a bad idea
  due to memory allocations which Ricardo pointed out.

include/sys/kmem.h

@@ -239,7 +239,8 @@ extern struct rw_semaphore spl_kmem_cache_sem;
 #define SKS_MAGIC			0x22222222
 #define SKC_MAGIC			0x2c2c2c2c
 
-#define SPL_KMEM_CACHE_DELAY		5	/* Minimum slab release age */
+#define SPL_KMEM_CACHE_DELAY		15	/* Minimum slab release age */
+#define SPL_KMEM_CACHE_REAP		0	/* Default reap everything */
 #define SPL_KMEM_CACHE_OBJ_PER_SLAB	32	/* Target objects per slab */
 #define SPL_KMEM_CACHE_OBJ_PER_SLAB_MIN	8	/* Minimum objects per slab */
 #define SPL_KMEM_CACHE_ALIGN		8	/* Default object alignment */
@@ -292,6 +293,7 @@ typedef struct spl_kmem_cache {
 	uint32_t		skc_slab_objs;	/* Objects per slab */
 	uint32_t		skc_slab_size;	/* Slab size */
 	uint32_t		skc_delay;	/* Slab reclaim interval */
+	uint32_t		skc_reap;	/* Slab reclaim count */
 	atomic_t		skc_ref;	/* Ref count callers */
 	struct delayed_work	skc_work;	/* Slab reclaim work */
         struct work_struct work;