btrfs: Document parameters of btrfs_reserve_extent
diff mbox

Message ID 1520936552-11586-1-git-send-email-nborisov@suse.com
State New
Headers show

Commit Message

Nikolay Borisov March 13, 2018, 10:22 a.m. UTC
This function is the entry to the extent allocator and as such has
quite a number of parameters. Some of those have subtle effects on the
allocation algorithm. Document the parameters.

Signed-off-by: Nikolay Borisov <nborisov@suse.com>
---
 fs/btrfs/extent-tree.c | 45 +++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 45 insertions(+)

Comments

Anand Jain March 13, 2018, 1:56 p.m. UTC | #1
On 03/13/2018 06:22 PM, Nikolay Borisov wrote:
> This function is the entry to the extent allocator and as such has
> quite a number of parameters. Some of those have subtle effects on the
> allocation algorithm. Document the parameters.
> 
> Signed-off-by: Nikolay Borisov <nborisov@suse.com>

  Reviewed-by: Anand Jain <anand.jain@oracle.com>

> ---
>   fs/btrfs/extent-tree.c | 45 +++++++++++++++++++++++++++++++++++++++++++++
>   1 file changed, 45 insertions(+)
> 
> diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
> index bb1861033071..60ece25101ee 100644
> --- a/fs/btrfs/extent-tree.c
> +++ b/fs/btrfs/extent-tree.c
> @@ -7961,6 +7961,51 @@ static void dump_space_info(struct btrfs_fs_info *fs_info,
>   	up_read(&info->groups_sem);
>   }
>   
> +/*
> + * btrfs_reserve_extent - entry point to the extent allocator. Tries to find a
> + *			  hole that is at least as big as @num_bytes.
> + *
> + * @root           -	The root that will contains this extent
> + *
> + * @ram_bytes      -	The amount of space in ram that @num_bytes take. This
> + *			is used for accounting purposes. This value differs
> + *			than @num_bytes only in the case of compressed extents.
> + *
> + * @num_bytes      -	Number of bytes to allocates on-disk.
> + *
> + * @min_alloc_size -	Indicates the minimum amount of memory that the
> + *			allocator should try to satisfy. In some cases
> + *			@num_bytes may be larger than what is required and if
> + *			the filesystem is fragmented then allocation fails.
> + *			However, the presence of @min_alloc_size gives a
> + *			chance to try and satisfy the smaller allocation.
> + *
> + * @empty_size     -	A hint that you plan on doing more cow. This is the
> + *			size in bytes the allocator should try to find free
> + *			next to the block it returns.  This is just a hint and
> + *			may be ignored by the allocator.
> + *
> + * @hint_byte      -	Hint to the allocator to start searching above the byte
> + *			address passed. It might be ignored.
> + *
> + * @ins            -	This key is modified to record the found hole. It will
> + *			have the following values:
> + *			ins->objectid == start position
> + *			ins->flags = BTRFS_EXTENT_ITEM_KEY
> + *			ins->offset == the size of the hole.
> + *
> + * @is_data        -	Boolean flag indicating whether an extent is
> + *			allocated for data (true) or metadata (false)
> + *
> + * @delalloc       -	Boolean flag indicating whether this allocation is for
> + *			delalloc or not. If 'true' data_rwsem of block groups
> + *			is going to be acquired.
> + *
> + *
> + * Returns 0 when an allocation succeeded or < 0 when an error occurred. In
> + * case -ENOSPC is returned then @ins->offset will contain the size of the
> + * largest available hole the allocator managed to find.
> + */
>   int btrfs_reserve_extent(struct btrfs_root *root, u64 ram_bytes,
>   			 u64 num_bytes, u64 min_alloc_size,
>   			 u64 empty_size, u64 hint_byte,
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
David Sterba March 14, 2018, 8:08 p.m. UTC | #2
On Tue, Mar 13, 2018 at 09:56:23PM +0800, Anand Jain wrote:
> 
> 
> On 03/13/2018 06:22 PM, Nikolay Borisov wrote:
> > This function is the entry to the extent allocator and as such has
> > quite a number of parameters. Some of those have subtle effects on the
> > allocation algorithm. Document the parameters.
> > 
> > Signed-off-by: Nikolay Borisov <nborisov@suse.com>
> 
>   Reviewed-by: Anand Jain <anand.jain@oracle.com>

Added to next, thanks.
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Liu Bo March 16, 2018, 6:43 p.m. UTC | #3
On Tue, Mar 13, 2018 at 3:22 AM, Nikolay Borisov <nborisov@suse.com> wrote:
> This function is the entry to the extent allocator and as such has
> quite a number of parameters. Some of those have subtle effects on the
> allocation algorithm. Document the parameters.
>
> Signed-off-by: Nikolay Borisov <nborisov@suse.com>
> ---
>  fs/btrfs/extent-tree.c | 45 +++++++++++++++++++++++++++++++++++++++++++++
>  1 file changed, 45 insertions(+)
>
> diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
> index bb1861033071..60ece25101ee 100644
> --- a/fs/btrfs/extent-tree.c
> +++ b/fs/btrfs/extent-tree.c
> @@ -7961,6 +7961,51 @@ static void dump_space_info(struct btrfs_fs_info *fs_info,
>         up_read(&info->groups_sem);
>  }
>
> +/*
> + * btrfs_reserve_extent - entry point to the extent allocator. Tries to find a
> + *                       hole that is at least as big as @num_bytes.
> + *
> + * @root           -   The root that will contains this extent
> + *
> + * @ram_bytes      -   The amount of space in ram that @num_bytes take. This
> + *                     is used for accounting purposes. This value differs
> + *                     than @num_bytes only in the case of compressed extents.
> + *
> + * @num_bytes      -   Number of bytes to allocates on-disk.
> + *
> + * @min_alloc_size -   Indicates the minimum amount of memory that the

Could you please pick another word instead of "memory", like hole or
extent or space?

Others look good to me.

Reviewed-by: Liu Bo <bo.liu@linux.alibaba.com>

thanks,
liubo

> + *                     allocator should try to satisfy. In some cases
> + *                     @num_bytes may be larger than what is required and if
> + *                     the filesystem is fragmented then allocation fails.
> + *                     However, the presence of @min_alloc_size gives a
> + *                     chance to try and satisfy the smaller allocation.
> + *
> + * @empty_size     -   A hint that you plan on doing more cow. This is the
> + *                     size in bytes the allocator should try to find free
> + *                     next to the block it returns.  This is just a hint and
> + *                     may be ignored by the allocator.
> + *
> + * @hint_byte      -   Hint to the allocator to start searching above the byte
> + *                     address passed. It might be ignored.
> + *
> + * @ins            -   This key is modified to record the found hole. It will
> + *                     have the following values:
> + *                     ins->objectid == start position
> + *                     ins->flags = BTRFS_EXTENT_ITEM_KEY
> + *                     ins->offset == the size of the hole.
> + *
> + * @is_data        -   Boolean flag indicating whether an extent is
> + *                     allocated for data (true) or metadata (false)
> + *
> + * @delalloc       -   Boolean flag indicating whether this allocation is for
> + *                     delalloc or not. If 'true' data_rwsem of block groups
> + *                     is going to be acquired.
> + *
> + *
> + * Returns 0 when an allocation succeeded or < 0 when an error occurred. In
> + * case -ENOSPC is returned then @ins->offset will contain the size of the
> + * largest available hole the allocator managed to find.
> + */
>  int btrfs_reserve_extent(struct btrfs_root *root, u64 ram_bytes,
>                          u64 num_bytes, u64 min_alloc_size,
>                          u64 empty_size, u64 hint_byte,
> --
> 2.7.4
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Nikolay Borisov March 16, 2018, 6:44 p.m. UTC | #4
On 16.03.2018 20:43, Liu Bo wrote:
> On Tue, Mar 13, 2018 at 3:22 AM, Nikolay Borisov <nborisov@suse.com> wrote:
>> This function is the entry to the extent allocator and as such has
>> quite a number of parameters. Some of those have subtle effects on the
>> allocation algorithm. Document the parameters.
>>
>> Signed-off-by: Nikolay Borisov <nborisov@suse.com>
>> ---
>>  fs/btrfs/extent-tree.c | 45 +++++++++++++++++++++++++++++++++++++++++++++
>>  1 file changed, 45 insertions(+)
>>
>> diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
>> index bb1861033071..60ece25101ee 100644
>> --- a/fs/btrfs/extent-tree.c
>> +++ b/fs/btrfs/extent-tree.c
>> @@ -7961,6 +7961,51 @@ static void dump_space_info(struct btrfs_fs_info *fs_info,
>>         up_read(&info->groups_sem);
>>  }
>>
>> +/*
>> + * btrfs_reserve_extent - entry point to the extent allocator. Tries to find a
>> + *                       hole that is at least as big as @num_bytes.
>> + *
>> + * @root           -   The root that will contains this extent
>> + *
>> + * @ram_bytes      -   The amount of space in ram that @num_bytes take. This
>> + *                     is used for accounting purposes. This value differs
>> + *                     than @num_bytes only in the case of compressed extents.
>> + *
>> + * @num_bytes      -   Number of bytes to allocates on-disk.
>> + *
>> + * @min_alloc_size -   Indicates the minimum amount of memory that the
> 
> Could you please pick another word instead of "memory", like hole or
> extent or space?

Good point, i guess it's a brain fair. I'd rather use "space". I see
David has already merged it. David do you mind fixing it in-place?

> 
> Others look good to me.
> 
> Reviewed-by: Liu Bo <bo.liu@linux.alibaba.com>
> 
> thanks,
> liubo
> 
>> + *                     allocator should try to satisfy. In some cases
>> + *                     @num_bytes may be larger than what is required and if
>> + *                     the filesystem is fragmented then allocation fails.
>> + *                     However, the presence of @min_alloc_size gives a
>> + *                     chance to try and satisfy the smaller allocation.
>> + *
>> + * @empty_size     -   A hint that you plan on doing more cow. This is the
>> + *                     size in bytes the allocator should try to find free
>> + *                     next to the block it returns.  This is just a hint and
>> + *                     may be ignored by the allocator.
>> + *
>> + * @hint_byte      -   Hint to the allocator to start searching above the byte
>> + *                     address passed. It might be ignored.
>> + *
>> + * @ins            -   This key is modified to record the found hole. It will
>> + *                     have the following values:
>> + *                     ins->objectid == start position
>> + *                     ins->flags = BTRFS_EXTENT_ITEM_KEY
>> + *                     ins->offset == the size of the hole.
>> + *
>> + * @is_data        -   Boolean flag indicating whether an extent is
>> + *                     allocated for data (true) or metadata (false)
>> + *
>> + * @delalloc       -   Boolean flag indicating whether this allocation is for
>> + *                     delalloc or not. If 'true' data_rwsem of block groups
>> + *                     is going to be acquired.
>> + *
>> + *
>> + * Returns 0 when an allocation succeeded or < 0 when an error occurred. In
>> + * case -ENOSPC is returned then @ins->offset will contain the size of the
>> + * largest available hole the allocator managed to find.
>> + */
>>  int btrfs_reserve_extent(struct btrfs_root *root, u64 ram_bytes,
>>                          u64 num_bytes, u64 min_alloc_size,
>>                          u64 empty_size, u64 hint_byte,
>> --
>> 2.7.4
>>
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> --
> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
David Sterba March 16, 2018, 7:41 p.m. UTC | #5
On Fri, Mar 16, 2018 at 08:44:56PM +0200, Nikolay Borisov wrote:
> >> + * @min_alloc_size -   Indicates the minimum amount of memory that the
> > 
> > Could you please pick another word instead of "memory", like hole or
> > extent or space?
> 
> Good point, i guess it's a brain fair. I'd rather use "space". I see
> David has already merged it. David do you mind fixing it in-place?

Yes, no problem.
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Patch
diff mbox

diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index bb1861033071..60ece25101ee 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -7961,6 +7961,51 @@  static void dump_space_info(struct btrfs_fs_info *fs_info,
 	up_read(&info->groups_sem);
 }
 
+/*
+ * btrfs_reserve_extent - entry point to the extent allocator. Tries to find a
+ *			  hole that is at least as big as @num_bytes.
+ *
+ * @root           -	The root that will contains this extent
+ *
+ * @ram_bytes      -	The amount of space in ram that @num_bytes take. This
+ *			is used for accounting purposes. This value differs
+ *			than @num_bytes only in the case of compressed extents.
+ *
+ * @num_bytes      -	Number of bytes to allocates on-disk.
+ *
+ * @min_alloc_size -	Indicates the minimum amount of memory that the
+ *			allocator should try to satisfy. In some cases
+ *			@num_bytes may be larger than what is required and if
+ *			the filesystem is fragmented then allocation fails.
+ *			However, the presence of @min_alloc_size gives a
+ *			chance to try and satisfy the smaller allocation.
+ *
+ * @empty_size     -	A hint that you plan on doing more cow. This is the
+ *			size in bytes the allocator should try to find free
+ *			next to the block it returns.  This is just a hint and
+ *			may be ignored by the allocator.
+ *
+ * @hint_byte      -	Hint to the allocator to start searching above the byte
+ *			address passed. It might be ignored.
+ *
+ * @ins            -	This key is modified to record the found hole. It will
+ *			have the following values:
+ *			ins->objectid == start position
+ *			ins->flags = BTRFS_EXTENT_ITEM_KEY
+ *			ins->offset == the size of the hole.
+ *
+ * @is_data        -	Boolean flag indicating whether an extent is
+ *			allocated for data (true) or metadata (false)
+ *
+ * @delalloc       -	Boolean flag indicating whether this allocation is for
+ *			delalloc or not. If 'true' data_rwsem of block groups
+ *			is going to be acquired.
+ *
+ *
+ * Returns 0 when an allocation succeeded or < 0 when an error occurred. In
+ * case -ENOSPC is returned then @ins->offset will contain the size of the
+ * largest available hole the allocator managed to find.
+ */
 int btrfs_reserve_extent(struct btrfs_root *root, u64 ram_bytes,
 			 u64 num_bytes, u64 min_alloc_size,
 			 u64 empty_size, u64 hint_byte,