Re: [PATCH] Permit inode & dentry hash tables to be allocated >MAX_ORDER size [#2]

[Andrew Morton]

David Howells <dhowells@xxxxxxxxxx> wrote:
>
> Here's an update to my patch.

-static void __init dcache_init(unsigned long mempages)
+static void __init dcache_init_early(void)
 {
-	struct hlist_head *d;
-	unsigned long order;
-	unsigned int nr_hash;
-	int i;
+	struct hlist_head *p;
+	int loop;
+
+	dentry_hashtable =
+		alloc_large_system_hash("Dentry cache",
+					sizeof(struct hlist_head),
+					dhash_entries,
+					13,
+					0,
+					&d_hash_shift,
+					&d_hash_mask);
+
+	p = dentry_hashtable;
+	loop = 1 << d_hash_shift;
+	do {
+		INIT_HLIST_HEAD(p);
+		p++;
+		loop--;
+	} while (loop);
+}

We have an opportunity to make this loop less baroque.

	for (i = 0; i < (1 << d_hash_shift); i++)
		INIT_HLIST_HEAD(p[i]);




+void __init vfs_caches_init_early(void)

This function was declared

+extern void vfs_caches_init_early(void);

whereas in other places you _have_ made the definition of __init functions
include the "__init".  I'm not sure which is best, really, but it seems
better to include the __init in the declaration.

+static inline int log2(unsigned long x) __attribute__((pure));

What's the attribute((pure)) for?

It generates a warning on older gcc - please use __attribute_pure__.

+static inline int log2(unsigned long x)

This is just asking for namespace collisions.

+{
+	int r = 0;
+	for (x >>= 1; x > 0; x >>= 1)
+		r++;
+	return r;
+}

The other four or five implementations of log2() use ffx(~n).

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/