[PATCH 2/2] fs: add super_operations infrastructure

[Yegor Yefremov]

On Mon, Feb 1, 2016 at 9:47 AM, Sascha Hauer <s.hauer at pengutronix.de> wrote:
> On Fri, Jan 29, 2016 at 03:06:21PM +0100, yegorslists at googlemail.com wrote:
>> From: Yegor Yefremov <yegorslists at googlemail.com>
>>
>> Signed-off-by: Yegor Yefremov <yegorslists at googlemail.com>
>> ---
>>  fs/Makefile        |  1 +
>>  fs/inode.c         | 34 ++++++++++++++++++++++++++++++++++
>>  fs/ubifs/super.c   | 26 +++++++++++---------------
>>  include/linux/fs.h |  6 ++++++
>>  4 files changed, 52 insertions(+), 15 deletions(-)
>>  create mode 100644 fs/inode.c
>>
>> diff --git a/fs/Makefile b/fs/Makefile
>> index 7896e38..320e6fa 100644
>> --- a/fs/Makefile
>> +++ b/fs/Makefile
>> @@ -5,6 +5,7 @@ obj-y                 += devfs-core.o
>>  obj-$(CONFIG_FS_DEVFS)       += devfs.o
>>  obj-$(CONFIG_FS_FAT) += fat/
>>  obj-y        += fs.o
>> +obj-y        += inode.o
>>  obj-$(CONFIG_FS_UBIFS)       += ubifs/
>>  obj-$(CONFIG_FS_TFTP)        += tftp.o
>>  obj-$(CONFIG_FS_OMAP4_USBBOOT)       += omap4_usbbootfs.o
>> diff --git a/fs/inode.c b/fs/inode.c
>> new file mode 100644
>> index 0000000..f9b1b63
>> --- /dev/null
>> +++ b/fs/inode.c
>> @@ -0,0 +1,34 @@
>> +/*
>> + * (C) 1997 Linus Torvalds
>> + * (C) 1999 Andrea Arcangeli <andrea at suse.de> (dynamic inode allocation)
>> + */
>> +
>> +#include <linux/mount.h>
>> +
>> +/**
>> + * iget_locked - obtain an inode from a mounted file system
>> + * @sb:              super block of file system
>> + * @ino:     inode number to get
>> + *
>> + * Search for the inode specified by @ino in the inode cache and if present
>> + * return it with an increased reference count. This is for file systems
>> + * where the inode number is sufficient for unique identification of an inode.
>> + *
>> + * If the inode is not in cache, allocate a new inode and return it locked,
>> + * hashed, and with the I_NEW flag set.  The file system gets to fill it in
>> + * before unlocking it via unlock_new_inode().
>> + */
>> +struct inode *iget_locked(struct super_block *sb, unsigned long ino)
>> +{
>> +     struct inode *inode;
>> +
>> +     inode = sb->s_op->alloc_inode(sb);
>> +     if (inode) {
>> +             inode->i_ino = ino;
>> +             inode->i_sb = sb;
>> +             list_add(&inode->i_sb_list, &sb->s_inodes);
>> +             inode->i_state = I_SYNC | I_NEW;
>> +     }
>> +
>> +     return inode;
>> +}
>
> Let's stop here, the border has been crossed. We can introduce no-op
> wrappers for functions which do not have any meaning in barebox
> (spinlocks and the like), we can introduce convenience wrappers for
> functions that have a different name in barebox (like kmalloc), but
> please let's not introduce functions which do more complicated things
> which depend on structures being ordered the way they are in the Linux
> kernel. I'm really afraid this way we merge more and more code from
> Linux that doesn't really make sense in the barebox context and that we
> have to keep code in the filesystem drivers just to make the parallel
> universe of the Linux wrappers happy. That's a can of worms I don't want
> to open.

OK. We'll see, what can be generalized between ubifs and squashfs
after I'm finished.

Yegor