btrfs: send: Proactively round up to kmalloc bucket size

Instead of discovering the kmalloc bucket size _after_ allocation, round
up proactively so the allocation is explicitly made for the full size,
allowing the compiler to correctly reason about the resulting size of
the buffer through the existing __alloc_size() hint.

Cc: Chris Mason <clm@fb.com>
Cc: Josef Bacik <josef@toxicpanda.com>
Cc: linux-btrfs@vger.kernel.org
Acked-by: David Sterba <dsterba@suse.com>
Link: https://lore.kernel.org/lkml/20220922133014.GI32411@suse.cz
Signed-off-by: Kees Cook <keescook@chromium.org>
Link: https://lore.kernel.org/r/20220923202822.2667581-8-keescook@chromium.org

diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index 4ef4167..f53e804 100644 (file)
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -439,6 +439,11 @@ static int fs_path_ensure_buf(struct fs_path *p, int len)
        old_buf_len = p->buf_len;
 
        /*
+        * Allocate to the next largest kmalloc bucket size, to let
+        * the fast path happen most of the time.
+        */
+       len = kmalloc_size_roundup(len);
+       /*
         * First time the inline_buf does not suffice
         */
        if (p->buf == p->inline_buf) {
@@ -451,11 +456,7 @@ static int fs_path_ensure_buf(struct fs_path *p, int len)
        if (!tmp_buf)
                return -ENOMEM;
        p->buf = tmp_buf;
-       /*
-        * The real size of the buffer is bigger, this will let the fast path
-        * happen most of the time
-        */
-       p->buf_len = ksize(p->buf);
+       p->buf_len = len;
 
        if (p->reversed) {
                tmp_buf = p->buf + old_buf_len - path_len - 1;