On Thu, Apr 25, 2013 at 08:15:42PM +0800, Anand Jain wrote:
> --- a/disk-io.c
> +++ b/disk-io.c
> @@ -1290,8 +1290,10 @@ static int close_all_devices(struct btrfs_fs_info *fs_info)
> fprintf(stderr, "Warning, could not drop caches\n");
> }
> close(device->fd);
> - kfree(device->name);
> - kfree(device->label);
> + if (device->name)
> + kfree(device->name);
> + if (device->label)
> + kfree(device->label);
As we're going to use the kernel sources directly, we should also keep
the semantics of the helpers. The in-kernel kfree accepts NULL and does
nothing in that case, so should the userspace replacement which is now
implemented as:
#define kfree(x) free(x)
> kfree(device);
> }
> kfree(fs_info->fs_devices);
> diff --git a/utils.c b/utils.c
> index 7b028a7..8104465 100644
> --- a/utils.c
> +++ b/utils.c
> @@ -484,6 +484,8 @@ int btrfs_add_to_fsid(struct btrfs_trans_handle *trans,
> device = kmalloc(sizeof(*device), GFP_NOFS);
> if (!device)
> return -ENOMEM;
> + memset(device, 0, sizeof(*device));
Please replace the above kmalloc with kzalloc instead.
> +
> buf = kmalloc(sectorsize, GFP_NOFS);
> if (!buf) {
> kfree(device);
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
