Re: [PATCH] ext4: initialize multi-block allocator before checking block descriptors

[Azat Khuzhin <a3at.mail@gmail.com>]

On Mon, Mar 17, 2014 at 6:00 AM,  <tytso@mit.edu> wrote:
> On Mon, Mar 17, 2014 at 12:56:59AM +0400, a3at.mail@gmail.com wrote:
>> After I tested ext4 dev branch (eb3e7abb161ad5), without any xfs-tests complaints,
>> I understand what goes wrong, you have not last version of
>> this patch, the latest is v3.
>> (Actually you have description from last patch, but not the latest changes.)
>
> What I did was to take your v3 version of the patch, and then since
> that patch removed the label failed_mount5, I changed instances of
> failed_mount4a to failed_mount5, just for aesthetic reasons.
>
> So there is no substantive difference between what is in the ext4
> patch queue and your v3 patch.   All I did was this:

Sorry I didn't look good at diff of two patches, my mistake.

>
> diff --git a/fs/ext4/super.c b/fs/ext4/super.c
> index d73f1d9..01c5088 100644
> --- a/fs/ext4/super.c
> +++ b/fs/ext4/super.c
> @@ -4100,14 +4100,14 @@ no_journal:
>         if (err) {
>                 ext4_msg(sb, KERN_ERR, "failed to reserve %llu clusters for "
>                          "reserved pool", ext4_calculate_resv_clusters(sb));
> -               goto failed_mount4a;
> +               goto failed_mount5;
>         }
>
>         err = ext4_setup_system_zone(sb);
>         if (err) {
>                 ext4_msg(sb, KERN_ERR, "failed to initialize system "
>                          "zone (%d)", err);
> -               goto failed_mount4a;
> +               goto failed_mount5;
>         }
>
>         err = ext4_register_li_request(sb, first_not_zeroed);
> @@ -4184,7 +4184,7 @@ failed_mount7:
>         ext4_unregister_li_request(sb);
>  failed_mount6:
>         ext4_release_system_zone(sb);
> -failed_mount4a:
> +failed_mount5:
>         dput(sb->s_root);
>         sb->s_root = NULL;
>  failed_mount4:
>
>
> Did you actually test your v3 patch on top of the dev branch?  Or did
> you just note that the patch in the ext4 patch queue was different,
> and assumed it was the v2 version of your patch?

Yes, I actually test v3 patch, but not on the top of dev branch,
instead the dev branch was on the top of v3 patch.
(but it changes nothing), with "-O bigalloc" and  "-o block_validity"

Also today I recheck this using kvm/qemu, with v3 patch on the top of
ext4 dev branch, and still nothing, here is last commits:
62f5f55 ext4: initialize multi-block allocator before checking block descriptors
eb3e7ab ext4: fix partial cluster handling for bigalloc file systems
97d3979 ext4: delete path dealloc code in ext4_ext_handle_uninitialized_extents

The only difference between your tests and my, is that you have i386,
while i have x86_64. But I really doubt that this is significantly.

>
> Regards,
>
>                                                 - Ted



-- 
Respectfully
Azat Khuzhin