lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAL3q7H6pCGcT=xGX4h5Rcb2Ky1wzK=7s1iqA4XevsYE7KNc2EQ@mail.gmail.com>
Date: Tue, 13 Aug 2024 12:50:35 +0100
From: Filipe Manana <fdmanana@...nel.org>
To: Johannes Thumshirn <jth@...nel.org>
Cc: Chris Mason <clm@...com>, Josef Bacik <josef@...icpanda.com>, David Sterba <dsterba@...e.com>, 
	"open list:BTRFS FILE SYSTEM" <linux-btrfs@...r.kernel.org>, open list <linux-kernel@...r.kernel.org>, 
	Johannes Thumshirn <johannes.thumshirn@....com>, Qu Wenruo <wqu@...e.com>
Subject: Re: [PATCH v3] btrfs: reduce chunk_map lookups in btrfs_map_block

On Tue, Aug 13, 2024 at 12:37 PM Johannes Thumshirn <jth@...nel.org> wrote:
>
> From: Johannes Thumshirn <johannes.thumshirn@....com>
>
> Currently we're calling btrfs_num_copies() before btrfs_get_chunk_map() in
> btrfs_map_block(). But btrfs_num_copies() itself does a chunk map lookup
> to be able to calculate the number of copies.
>
> So split out the code getting the number of copies from btrfs_num_copies()
> into a helper called btrfs_chunk_map_num_copies() and directly call it
> from btrfs_map_block() and btrfs_num_copies().
>
> This saves us one rbtree lookup per btrfs_map_block() invocation.
>
> Reviewed-by: Qu Wenruo <wqu@...e.com>
> Signed-off-by: Johannes Thumshirn <johannes.thumshirn@....com>

Reviewed-by: Filipe Manana <fdmanana@...e.com>

Looks good, thanks.

> ---
> Changes in v2:
> - Added Reviewed-bys
> - Reflowed comments
> - Moved non RAID56 cases to the end without an if
> Link to v1:
> https://lore.kernel.org/all/20240812165931.9106-1-jth@kernel.org/
>
> Changes in v3:
> - constified btrfs_chunk_map_num_copies() parameter
> - fixed accidentially changed comment
> Link to v2:
> https://lore.kernel.org/all/bc73d318c7f24196cdc7305b6a6ce516fb4fc81d.1723546054.git.jth@kernel.org/
>
>  fs/btrfs/volumes.c | 49 +++++++++++++++++++++++++---------------------
>  1 file changed, 27 insertions(+), 22 deletions(-)
>
> diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
> index e07452207426..4b9b647a7e29 100644
> --- a/fs/btrfs/volumes.c
> +++ b/fs/btrfs/volumes.c
> @@ -5781,11 +5781,31 @@ void btrfs_mapping_tree_free(struct btrfs_fs_info *fs_info)
>         write_unlock(&fs_info->mapping_tree_lock);
>  }
>
> +static int btrfs_chunk_map_num_copies(const struct btrfs_chunk_map *map)
> +{
> +       enum btrfs_raid_types index = btrfs_bg_flags_to_raid_index(map->type);
> +
> +       if (map->type & BTRFS_BLOCK_GROUP_RAID5)
> +               return 2;
> +
> +       /*
> +        * There could be two corrupted data stripes, we need to loop
> +        * retry in order to rebuild the correct data.
> +        *
> +        * Fail a stripe at a time on every retry except the stripe
> +        * under reconstruction.
> +        */
> +       if (map->type & BTRFS_BLOCK_GROUP_RAID6)
> +               return map->num_stripes;
> +
> +       /* Non-RAID56, use their ncopies from btrfs_raid_array. */
> +       return btrfs_raid_array[index].ncopies;
> +}
> +
>  int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len)
>  {
>         struct btrfs_chunk_map *map;
> -       enum btrfs_raid_types index;
> -       int ret = 1;
> +       int ret;
>
>         map = btrfs_get_chunk_map(fs_info, logical, len);
>         if (IS_ERR(map))
> @@ -5797,22 +5817,7 @@ int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len)
>                  */
>                 return 1;
>
> -       index = btrfs_bg_flags_to_raid_index(map->type);
> -
> -       /* Non-RAID56, use their ncopies from btrfs_raid_array. */
> -       if (!(map->type & BTRFS_BLOCK_GROUP_RAID56_MASK))
> -               ret = btrfs_raid_array[index].ncopies;
> -       else if (map->type & BTRFS_BLOCK_GROUP_RAID5)
> -               ret = 2;
> -       else if (map->type & BTRFS_BLOCK_GROUP_RAID6)
> -               /*
> -                * There could be two corrupted data stripes, we need
> -                * to loop retry in order to rebuild the correct data.
> -                *
> -                * Fail a stripe at a time on every retry except the
> -                * stripe under reconstruction.
> -                */
> -               ret = map->num_stripes;
> +       ret = btrfs_chunk_map_num_copies(map);
>         btrfs_free_chunk_map(map);
>         return ret;
>  }
> @@ -6462,14 +6467,14 @@ int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
>         io_geom.stripe_index = 0;
>         io_geom.op = op;
>
> -       num_copies = btrfs_num_copies(fs_info, logical, fs_info->sectorsize);
> -       if (io_geom.mirror_num > num_copies)
> -               return -EINVAL;
> -
>         map = btrfs_get_chunk_map(fs_info, logical, *length);
>         if (IS_ERR(map))
>                 return PTR_ERR(map);
>
> +       num_copies = btrfs_chunk_map_num_copies(map);
> +       if (io_geom.mirror_num > num_copies)
> +               return -EINVAL;
> +
>         map_offset = logical - map->start;
>         io_geom.raid56_full_stripe_start = (u64)-1;
>         max_len = btrfs_max_io_len(map, map_offset, &io_geom);
> --
> 2.43.0
>
>

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ