Implement NUMA support in uma(9) and malloc(9). Allocations from specific

domains can be done by the _domain() API variants.  UMA also supports a
first-touch policy via the NUMA zone flag.

The slab layer is now segregated by VM domains and is precise.  It handles
iteration for round-robin directly.  The per-cpu cache layer remains
a mix of domains according to where memory is allocated and freed.  Well
behaved clients can achieve perfect locality with no performance penalty.

The direct domain allocation functions have to visit the slab layer and
so require per-zone locks which come at some expense.

Reviewed by:	Attilio (a slightly older version)
Tested by:	pho
Sponsored by:	Netflix, Dell/EMC Isilon

1 files changed, 8 insertions, 0 deletions

diff --git a/sys/sys/malloc.h b/sys/sys/malloc.h
index 47ca13f5ebd6..57a06e8ccfd4 100644
--- a/sys/sys/malloc.h
+++ b/sys/sys/malloc.h
@@ -175,9 +175,17 @@ void	*contigmalloc(unsigned long size, struct malloc_type *type, int flags,
 	    vm_paddr_t low, vm_paddr_t high, unsigned long alignment,
 	    vm_paddr_t boundary) __malloc_like __result_use_check
 	    __alloc_size(1) __alloc_align(6);
+void	*contigmalloc_domain(unsigned long size, struct malloc_type *type,
+	    int domain, int flags, vm_paddr_t low, vm_paddr_t high,
+	    unsigned long alignment, vm_paddr_t boundary)
+	    __malloc_like __result_use_check __alloc_size(1) __alloc_align(6);
 void	free(void *addr, struct malloc_type *type);
+void	free_domain(void *addr, struct malloc_type *type);
 void	*malloc(unsigned long size, struct malloc_type *type, int flags)
 	    __malloc_like __result_use_check __alloc_size(1);
+void	*malloc_domain(unsigned long size, struct malloc_type *type,
+	    int domain, int flags)
+	    __malloc_like __result_use_check __alloc_size(1);
 void	*mallocarray(size_t nmemb, size_t size, struct malloc_type *type,
 	    int flags) __malloc_like __result_use_check
 	    __alloc_size(1) __alloc_size(2);