Re: new version of loop.c patch

[Rob McGee <rob0@runbox.com>]

On Tue, Sep 25, 2001 at 02:57:27PM +0300, Emil wrote:
> On 24 September 2001, beldridg@best.com <beldridg@best.com> wrote:
> > it looks like the loop.c patch hasn't been integrated into 2.4.10.
> 
> You may of course correct this little problem yourself (just remove the
> --dry-run option from the patch line and then apply the rejected file
> by hand.
> 
> For the lazy people here you have the new diff file

Lazy may be true. But in my case I'm also (over-)cautious and quite
clueless when it comes to reading code. This hunk:

> --- loop.c-2.4.original	Mon Sep 24 09:40:19 2001
> +++ patched-loop.c	Mon Sep 24 09:43:03 2001
> [snip]
> @@ -1034,6 +1034,7 @@
>  	for (i = 0; i < max_loop; i++)
>  		register_disk(NULL, MKDEV(MAJOR_NR, i), 1, &lo_fops, 0);
>  
> +	{ extern int init_module_aes(void); init_module_aes(); }
>  	printk(KERN_INFO "loop: loaded (max %d devices)\n", max_loop);
>  	return 0;
>  
> @@ -1047,6 +1048,7 @@
>  
>  void loop_exit(void) 
>  {
> +	{ extern void cleanup_module_aes(void); cleanup_module_aes(); }
>  	devfs_unregister(devfs_handle);
>  	if (devfs_unregister_blkdev(MAJOR_NR, "loop"))
>  		printk(KERN_WARNING "loop: cannot unregister blkdev\n");

... does not seem to be in HVR's loop.c patch. And the "_module_aes"
makes me think that this diff is actually for Loop-AES.

Here's the loop.c.rej I got:

***************
*** 474,482 ****
  	/*
  	 * piggy old buffer on original, and submit for I/O
  	 */
  	bh = loop_get_buffer(lo, rbh);
  	bh->b_private = rbh;
- 	IV = loop_get_iv(lo, bh->b_rsector);
  	if (rw == WRITE) {
  		set_bit(BH_Dirty, &bh->b_state);
  		if (lo_do_transfer(lo, WRITE, bh->b_data, rbh->b_data,
--- 492,502 ----
  	/*
  	 * piggy old buffer on original, and submit for I/O
  	 */
+ 	IV = loop_get_iv(lo, rbh->b_rsector);
+ 
  	bh = loop_get_buffer(lo, rbh);
  	bh->b_private = rbh;
+ 
  	if (rw == WRITE) {
  		set_bit(BH_Dirty, &bh->b_state);
  		if (lo_do_transfer(lo, WRITE, bh->b_data, rbh->b_data,
(EOF)

Looking in loop.c I see this:

*** 511,521 ***
	/*
	 * piggy old buffer on original, and submit for I/O
	 */
	bh = loop_get_buffer(lo, rbh);
	IV = loop_get_iv(lo, rbh->b_rsector);
	if (rw == WRITE) {
		set_bit(BH_Dirty, &bh->b_state);
		if (lo_do_transfer(lo, WRITE, bh->b_data, rbh->b_data,
				   bh->b_size, IV))
			goto err;
	}

It has the same "IV = ..." line as the "+" line in the .rej shows, but
in a slightly different order. And the "bh->b_private = rbh;" line isn't
there. However, I do find it up on line 452.

I'm sure this is trivially simple to many of you, but I really do not
know what to do with this loop.c.rej. My guess would be that it's not
needed at all.

Later I will try compiling it to see if it works, but that will take
awhile -- all I have is a P166. I will probably get a reply from the
list before this thing can finish the job. :)

Thank you,
    Rob - /dev/rob0

Linux-crypto:  cryptography in and on the Linux system
Archive:       http://mail.nl.linux.org/linux-crypto/