Implemented real slab allocator. If you suspect it from being buggy,

just #define FAKE_SLAB at the top of lib/slab.c to bypass it.

TODO

@@ -1,7 +1,5 @@
 Core
 ~~~~
-- better memory allocators
-
 - static: check validity of route destination?
 - static: allow specifying a per-route filter program for setting route attributes?
 

lib/slab.c

@@ -1,7 +1,9 @@
 /*
- *	BIRD Resource Manager -- SLABs
+ *	BIRD Resource Manager -- A SLAB-like Memory Allocator
  *
- *	(c) 1998 Martin Mares <mj@ucw.cz>
+ *	Heavily inspired by the original SLAB paper by Jeff Bonwick.
+ *
+ *	(c) 1998--2000 Martin Mares <mj@ucw.cz>
  *
  *	Can be freely distributed and used under the terms of the GNU GPL.
  */
@@ -12,14 +14,16 @@
 #include "nest/bird.h"
 #include "lib/resource.h"
 
-/*
- *	These are only fake, soon to change...
- */
+#undef FAKE_SLAB	/* Turn on if you want to debug memory allocations */
 
-struct sl_obj {
-  node n;
-  byte data[0];
-};
+static void slab_free(resource *r);
+static void slab_dump(resource *r);
+
+#ifdef FAKE_SLAB
+
+/*
+ *  Fake version used for debugging.
+ */
 
 struct slab {
   resource r;
@@ -27,16 +31,18 @@ struct slab {
   list objs;
 };
 
-static void slab_free(resource *r);
-static void slab_dump(resource *r);
-
 static struct resclass sl_class = {
-  "Slab",
+  "FakeSlab",
   sizeof(struct slab),
   slab_free,
   slab_dump
 };
 
+struct sl_obj {
+  node n;
+  byte data[0];
+};
+
 slab *
 sl_new(pool *p, unsigned size)
 {
@@ -85,3 +91,182 @@ slab_dump(resource *r)
     cnt++;
   debug("(%d objects per %d bytes)\n", cnt, s->size);
 }
+
+#else
+
+/*
+ *  Real efficient version.
+ */
+
+#define SLAB_SIZE 4096
+#define MAX_EMPTY_HEADS 1
+
+struct slab {
+  resource r;
+  unsigned obj_size, head_size, objs_per_slab, num_empty_heads;
+  list empty_heads, partial_heads, full_heads;
+};
+
+static struct resclass sl_class = {
+  "Slab",
+  sizeof(struct slab),
+  slab_free,
+  slab_dump
+};
+
+struct sl_head {
+  node n;
+  struct sl_obj *first_free;
+  int num_full;
+};
+
+struct sl_obj {
+  struct sl_head *slab;
+  union {
+    struct sl_obj *next;
+    byte data[0];
+  } u;
+};
+
+struct sl_alignment {			/* Magic structure for testing of alignment */
+  byte data;
+  int x[0];
+};
+
+slab *
+sl_new(pool *p, unsigned size)
+{
+  slab *s = ralloc(p, &sl_class);
+  unsigned int align = sizeof(struct sl_alignment);
+  if (align < sizeof(int))
+    align = sizeof(int);
+  size += OFFSETOF(struct sl_obj, u.data);
+  if (size < sizeof(struct sl_obj))
+    size = sizeof(struct sl_obj);
+  size = (size + align - 1) / align * align;
+  s->obj_size = size;
+  s->head_size = (sizeof(struct sl_head) + align - 1) / align * align;
+  s->objs_per_slab = (SLAB_SIZE - s->head_size) / size;
+  if (!s->objs_per_slab)
+    bug("Slab: object too large");
+  s->num_empty_heads = 0;
+  init_list(&s->empty_heads);
+  init_list(&s->partial_heads);
+  init_list(&s->full_heads);
+  return s;
+}
+
+struct sl_head *
+sl_new_head(slab *s)
+{
+  struct sl_head *h = xmalloc(SLAB_SIZE);
+  struct sl_obj *o = (struct sl_obj *)((byte *)h+s->head_size);
+  struct sl_obj *no;
+  unsigned int n = s->objs_per_slab;
+
+  h->first_free = o;
+  h->num_full = 0;
+  while (n--)
+    {
+      o->slab = h;
+      no = (struct sl_obj *)((char *) o+s->obj_size);
+      o->u.next = n ? no : NULL;
+      o = no;
+    }
+  return h;
+}
+
+void *
+sl_alloc(slab *s)
+{
+  struct sl_head *h;
+  struct sl_obj *o;
+
+redo:
+  h = HEAD(s->partial_heads);
+  if (!h->n.next)
+    goto no_partial;
+okay:
+  o = h->first_free;
+  if (!o)
+    goto full_partial;
+  h->first_free = o->u.next;
+  h->num_full++;
+  return o->u.data;
+
+full_partial:
+  rem_node(&h->n);
+  add_tail(&s->full_heads, &h->n);
+  goto redo;
+
+no_partial:
+  h = HEAD(s->empty_heads);
+  if (h->n.next)
+    {
+      rem_node(&h->n);
+      add_head(&s->partial_heads, &h->n);
+      s->num_empty_heads--;
+      goto okay;
+    }
+  h = sl_new_head(s);
+  add_head(&s->partial_heads, &h->n);
+  goto okay;
+}
+
+void
+sl_free(slab *s, void *oo)
+{
+  struct sl_obj *o = SKIP_BACK(struct sl_obj, u.data, oo);
+  struct sl_head *h = o->slab;
+
+  o->u.next = h->first_free;
+  h->first_free = o;
+  if (!--h->num_full)
+    {
+      rem_node(&h->n);
+      if (s->num_empty_heads >= MAX_EMPTY_HEADS)
+	xfree(h);
+      else
+	{
+	  add_head(&s->empty_heads, &h->n);
+	  s->num_empty_heads++;
+	}
+    }
+  else if (!o->u.next)
+    {
+      rem_node(&h->n);
+      add_head(&s->partial_heads, &h->n);
+    }
+}
+
+static void
+slab_free(resource *r)
+{
+  slab *s = (slab *) r;
+  struct sl_head *h, *g;
+
+  WALK_LIST_DELSAFE(h, g, s->empty_heads)
+    xfree(h);
+  WALK_LIST_DELSAFE(h, g, s->partial_heads)
+    xfree(h);
+  WALK_LIST_DELSAFE(h, g, s->full_heads)
+    xfree(h);
+}
+
+static void
+slab_dump(resource *r)
+{
+  slab *s = (slab *) r;
+  int ec=0, pc=0, fc=0;
+  struct sl_head *h;
+
+  WALK_LIST(h, s->empty_heads)
+    ec++;
+  WALK_LIST(h, s->partial_heads)
+    pc++;
+  WALK_LIST(h, s->full_heads)
+    fc++;
+  debug("(%de+%dp+%df blocks per %d objs per %d bytes)\n", ec, pc, fc, s->objs_per_slab, s->obj_size);
+}
+
+#endif