On Sun, Oct 28, 2012 at 3:55 PM, Zheng Liu [off-list ref] wrote:

Hi Zhiyong,

On Thu, Oct 25, 2012 at 11:08:55PM +0800, zwu.kernel@gmail.com wrote:
[snip]

quoted

@@ -199,6 +342,54 @@ err:
 }

 /*
+ * Main function to update access frequency from read/writepage(s) hooks
+ */
+inline void hot_update_freqs(struct inode *inode, u64 start,
+                     u64 len, int rw)

This function seems too big.  So we really need to inline this function?

As Dave said in his comments, it will add a function call
overhead even when tracking is not enabled. a static inline function
will just result in no extra overhead other than the if
statement....

Regards,
Zheng

quoted

+{
+     struct hot_info *root = inode->i_sb->s_hot_root;
+     struct hot_inode_item *he;
+     struct hot_range_item *hr;
+     u32 cur, end;
+
+     if (!root || (len == 0))
+             return;
+
+     he = hot_inode_item_find(root, inode->i_ino);
+     if (IS_ERR(he)) {
+             WARN_ON(1);
+             return;
+     }
+
+     spin_lock(&he->hot_inode.lock);
+     hot_freq_data_update(&he->hot_inode.hot_freq_data, rw);
+     spin_unlock(&he->hot_inode.lock);
+
+     /*
+      * Align ranges on RANGE_SIZE boundary
+      * to prevent proliferation of range structs
+      */
+     end = (start + len + RANGE_SIZE - 1) >> RANGE_BITS;
+     for (cur = (start >> RANGE_BITS); cur < end; cur++) {
+             hr = hot_range_item_find(he, cur);
+             if (IS_ERR(hr)) {
+                     WARN_ON(1);
+                     hot_inode_item_put(he);
+                     return;
+             }
+
+             spin_lock(&hr->hot_range.lock);
+             hot_freq_data_update(&hr->hot_range.hot_freq_data, rw);
+             spin_unlock(&hr->hot_range.lock);
+
+             hot_range_item_put(hr);
+     }
+
+     hot_inode_item_put(he);
+}
+EXPORT_SYMBOL_GPL(hot_update_freqs);

-- 
Regards,

Zhi Yong Wu