* [PATCH] btrfs-progs: Use exclude_super_stripes instead of account_super_bytes
@ 2018-05-02 11:52 Nikolay Borisov
2018-05-02 12:29 ` Qu Wenruo
` (2 more replies)
0 siblings, 3 replies; 7+ messages in thread
From: Nikolay Borisov @ 2018-05-02 11:52 UTC (permalink / raw)
To: linux-btrfs; +Cc: wqu, osandov, Nikolay Borisov
Originally commit 2681e00f00fe ("btrfs-progs: check for matchingi
free space in cache") added the account_super_bytes function to prevent
false negative when running btrfs check. Turns out this function is
really copied exclude_super_stripes, excluding the calls to
exclude_super_stripes. Later commit e4797df6a9fa ("btrfs-progs: check
the free space tree in btrfsck") introduced proper version of
exclude_super_stripes. Instead of duplicating the function, just remove
account_super_bytes and use exclude_super_stripes instead of the former.
This also has the benefit of bringing the userspace code a bit closer
to the kernel counterpart.
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
---
extent-tree.c | 52 ++--------------------------------------------------
1 file changed, 2 insertions(+), 50 deletions(-)
diff --git a/extent-tree.c b/extent-tree.c
index ea205ccf4c30..391f0a784710 100644
--- a/extent-tree.c
+++ b/extent-tree.c
@@ -3164,54 +3164,6 @@ static int find_first_block_group(struct btrfs_root *root,
return ret;
}
-static void account_super_bytes(struct btrfs_fs_info *fs_info,
- struct btrfs_block_group_cache *cache)
-{
- u64 bytenr;
- u64 *logical;
- int stripe_len;
- int i, nr, ret;
-
- if (cache->key.objectid < BTRFS_SUPER_INFO_OFFSET) {
- stripe_len = BTRFS_SUPER_INFO_OFFSET - cache->key.objectid;
- cache->bytes_super += stripe_len;
- }
-
- for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) {
- bytenr = btrfs_sb_offset(i);
- ret = btrfs_rmap_block(fs_info,
- cache->key.objectid, bytenr,
- 0, &logical, &nr, &stripe_len);
- if (ret)
- return;
-
- while (nr--) {
- u64 start, len;
-
- if (logical[nr] > cache->key.objectid +
- cache->key.offset)
- continue;
-
- if (logical[nr] + stripe_len <= cache->key.objectid)
- continue;
-
- start = logical[nr];
- if (start < cache->key.objectid) {
- start = cache->key.objectid;
- len = (logical[nr] + stripe_len) - start;
- } else {
- len = min_t(u64, stripe_len,
- cache->key.objectid +
- cache->key.offset - start);
- }
-
- cache->bytes_super += len;
- }
-
- kfree(logical);
- }
-}
-
int btrfs_read_block_groups(struct btrfs_root *root)
{
struct btrfs_path *path;
@@ -3287,7 +3239,7 @@ int btrfs_read_block_groups(struct btrfs_root *root)
if (btrfs_chunk_readonly(info, cache->key.objectid))
cache->ro = 1;
- account_super_bytes(info, cache);
+ exclude_super_stripes(root, cache);
ret = update_space_info(info, cache->flags, found_key.offset,
btrfs_block_group_used(&cache->item),
@@ -3331,7 +3283,7 @@ btrfs_add_block_group(struct btrfs_fs_info *fs_info, u64 bytes_used, u64 type,
cache->flags = type;
btrfs_set_block_group_flags(&cache->item, type);
- account_super_bytes(fs_info, cache);
+ exclude_super_stripes(fs_info->extent_root, cache);
ret = update_space_info(fs_info, cache->flags, size, bytes_used,
&cache->space_info);
BUG_ON(ret);
--
2.7.4
^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: [PATCH] btrfs-progs: Use exclude_super_stripes instead of account_super_bytes
2018-05-02 11:52 [PATCH] btrfs-progs: Use exclude_super_stripes instead of account_super_bytes Nikolay Borisov
@ 2018-05-02 12:29 ` Qu Wenruo
2018-05-02 12:49 ` Nikolay Borisov
2018-05-04 1:53 ` Su Yue
2018-05-08 17:34 ` David Sterba
2 siblings, 1 reply; 7+ messages in thread
From: Qu Wenruo @ 2018-05-02 12:29 UTC (permalink / raw)
To: Nikolay Borisov, linux-btrfs; +Cc: wqu, osandov
On 2018年05月02日 19:52, Nikolay Borisov wrote:
> Originally commit 2681e00f00fe ("btrfs-progs: check for matchingi
> free space in cache") added the account_super_bytes function to prevent
> false negative when running btrfs check. Turns out this function is
> really copied exclude_super_stripes, excluding the calls to
> exclude_super_stripes. Later commit e4797df6a9fa ("btrfs-progs: check
> the free space tree in btrfsck") introduced proper version of
> exclude_super_stripes. Instead of duplicating the function, just remove
> account_super_bytes and use exclude_super_stripes instead of the former.
> This also has the benefit of bringing the userspace code a bit closer
> to the kernel counterpart.
>
> Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Since these two functions are the same, it's completely fine to remove one.
Reviewed-by: Qu Wenruo <wqu@suse.com>
Thanks,
Qu
> ---
> extent-tree.c | 52 ++--------------------------------------------------
> 1 file changed, 2 insertions(+), 50 deletions(-)
>
> diff --git a/extent-tree.c b/extent-tree.c
> index ea205ccf4c30..391f0a784710 100644
> --- a/extent-tree.c
> +++ b/extent-tree.c
> @@ -3164,54 +3164,6 @@ static int find_first_block_group(struct btrfs_root *root,
> return ret;
> }
>
> -static void account_super_bytes(struct btrfs_fs_info *fs_info,
> - struct btrfs_block_group_cache *cache)
> -{
> - u64 bytenr;
> - u64 *logical;
> - int stripe_len;
> - int i, nr, ret;
> -
> - if (cache->key.objectid < BTRFS_SUPER_INFO_OFFSET) {
> - stripe_len = BTRFS_SUPER_INFO_OFFSET - cache->key.objectid;
> - cache->bytes_super += stripe_len;
> - }
> -
> - for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) {
> - bytenr = btrfs_sb_offset(i);
> - ret = btrfs_rmap_block(fs_info,
> - cache->key.objectid, bytenr,
> - 0, &logical, &nr, &stripe_len);
> - if (ret)
> - return;
> -
> - while (nr--) {
> - u64 start, len;
> -
> - if (logical[nr] > cache->key.objectid +
> - cache->key.offset)
> - continue;
> -
> - if (logical[nr] + stripe_len <= cache->key.objectid)
> - continue;
> -
> - start = logical[nr];
> - if (start < cache->key.objectid) {
> - start = cache->key.objectid;
> - len = (logical[nr] + stripe_len) - start;
> - } else {
> - len = min_t(u64, stripe_len,
> - cache->key.objectid +
> - cache->key.offset - start);
> - }
> -
> - cache->bytes_super += len;
> - }
> -
> - kfree(logical);
> - }
> -}
> -
> int btrfs_read_block_groups(struct btrfs_root *root)
> {
> struct btrfs_path *path;
> @@ -3287,7 +3239,7 @@ int btrfs_read_block_groups(struct btrfs_root *root)
> if (btrfs_chunk_readonly(info, cache->key.objectid))
> cache->ro = 1;
>
> - account_super_bytes(info, cache);
> + exclude_super_stripes(root, cache);
>
> ret = update_space_info(info, cache->flags, found_key.offset,
> btrfs_block_group_used(&cache->item),
> @@ -3331,7 +3283,7 @@ btrfs_add_block_group(struct btrfs_fs_info *fs_info, u64 bytes_used, u64 type,
> cache->flags = type;
> btrfs_set_block_group_flags(&cache->item, type);
>
> - account_super_bytes(fs_info, cache);
> + exclude_super_stripes(fs_info->extent_root, cache);
> ret = update_space_info(fs_info, cache->flags, size, bytes_used,
> &cache->space_info);
> BUG_ON(ret);
>
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] btrfs-progs: Use exclude_super_stripes instead of account_super_bytes
2018-05-02 12:29 ` Qu Wenruo
@ 2018-05-02 12:49 ` Nikolay Borisov
2018-05-02 13:13 ` Qu Wenruo
0 siblings, 1 reply; 7+ messages in thread
From: Nikolay Borisov @ 2018-05-02 12:49 UTC (permalink / raw)
To: Qu Wenruo, linux-btrfs; +Cc: wqu, osandov
On 2.05.2018 15:29, Qu Wenruo wrote:
>
>
> On 2018年05月02日 19:52, Nikolay Borisov wrote:
>> Originally commit 2681e00f00fe ("btrfs-progs: check for matchingi
>> free space in cache") added the account_super_bytes function to prevent
>> false negative when running btrfs check. Turns out this function is
>> really copied exclude_super_stripes, excluding the calls to
>> exclude_super_stripes. Later commit e4797df6a9fa ("btrfs-progs: check
>> the free space tree in btrfsck") introduced proper version of
>> exclude_super_stripes. Instead of duplicating the function, just remove
>> account_super_bytes and use exclude_super_stripes instead of the former.
>> This also has the benefit of bringing the userspace code a bit closer
>> to the kernel counterpart.
>>
>> Signed-off-by: Nikolay Borisov <nborisov@suse.com>
>
> Since these two functions are the same, it's completely fine to remove one.
As I mentioned, they are not "comlpetely" the same. The difference is
that exclude_super_stripes will mark the stripes as
EXTENT_UPTODATE in fs_info->pinned_extents via add_excluded_extent.
Dunno if this has any repercussions on functionality. I've run the progs
test suite and didn't observe any regressions. Also looking at the usage
of fs_info->pinned_extents didn't see anything conspicuous.
>
> Reviewed-by: Qu Wenruo <wqu@suse.com>
>
> Thanks,
> Qu
>
>> ---
>> extent-tree.c | 52 ++--------------------------------------------------
>> 1 file changed, 2 insertions(+), 50 deletions(-)
>>
>> diff --git a/extent-tree.c b/extent-tree.c
>> index ea205ccf4c30..391f0a784710 100644
>> --- a/extent-tree.c
>> +++ b/extent-tree.c
>> @@ -3164,54 +3164,6 @@ static int find_first_block_group(struct btrfs_root *root,
>> return ret;
>> }
>>
>> -static void account_super_bytes(struct btrfs_fs_info *fs_info,
>> - struct btrfs_block_group_cache *cache)
>> -{
>> - u64 bytenr;
>> - u64 *logical;
>> - int stripe_len;
>> - int i, nr, ret;
>> -
>> - if (cache->key.objectid < BTRFS_SUPER_INFO_OFFSET) {
>> - stripe_len = BTRFS_SUPER_INFO_OFFSET - cache->key.objectid;
>> - cache->bytes_super += stripe_len;
>> - }
>> -
>> - for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) {
>> - bytenr = btrfs_sb_offset(i);
>> - ret = btrfs_rmap_block(fs_info,
>> - cache->key.objectid, bytenr,
>> - 0, &logical, &nr, &stripe_len);
>> - if (ret)
>> - return;
>> -
>> - while (nr--) {
>> - u64 start, len;
>> -
>> - if (logical[nr] > cache->key.objectid +
>> - cache->key.offset)
>> - continue;
>> -
>> - if (logical[nr] + stripe_len <= cache->key.objectid)
>> - continue;
>> -
>> - start = logical[nr];
>> - if (start < cache->key.objectid) {
>> - start = cache->key.objectid;
>> - len = (logical[nr] + stripe_len) - start;
>> - } else {
>> - len = min_t(u64, stripe_len,
>> - cache->key.objectid +
>> - cache->key.offset - start);
>> - }
>> -
>> - cache->bytes_super += len;
>> - }
>> -
>> - kfree(logical);
>> - }
>> -}
>> -
>> int btrfs_read_block_groups(struct btrfs_root *root)
>> {
>> struct btrfs_path *path;
>> @@ -3287,7 +3239,7 @@ int btrfs_read_block_groups(struct btrfs_root *root)
>> if (btrfs_chunk_readonly(info, cache->key.objectid))
>> cache->ro = 1;
>>
>> - account_super_bytes(info, cache);
>> + exclude_super_stripes(root, cache);
>>
>> ret = update_space_info(info, cache->flags, found_key.offset,
>> btrfs_block_group_used(&cache->item),
>> @@ -3331,7 +3283,7 @@ btrfs_add_block_group(struct btrfs_fs_info *fs_info, u64 bytes_used, u64 type,
>> cache->flags = type;
>> btrfs_set_block_group_flags(&cache->item, type);
>>
>> - account_super_bytes(fs_info, cache);
>> + exclude_super_stripes(fs_info->extent_root, cache);
>> ret = update_space_info(fs_info, cache->flags, size, bytes_used,
>> &cache->space_info);
>> BUG_ON(ret);
>>
>
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] btrfs-progs: Use exclude_super_stripes instead of account_super_bytes
2018-05-02 12:49 ` Nikolay Borisov
@ 2018-05-02 13:13 ` Qu Wenruo
2018-05-04 1:49 ` Su Yue
0 siblings, 1 reply; 7+ messages in thread
From: Qu Wenruo @ 2018-05-02 13:13 UTC (permalink / raw)
To: Nikolay Borisov, linux-btrfs; +Cc: wqu, osandov, Su Yue
On 2018年05月02日 20:49, Nikolay Borisov wrote:
>
>
> On 2.05.2018 15:29, Qu Wenruo wrote:
>>
>>
>> On 2018年05月02日 19:52, Nikolay Borisov wrote:
>>> Originally commit 2681e00f00fe ("btrfs-progs: check for matchingi
>>> free space in cache") added the account_super_bytes function to prevent
>>> false negative when running btrfs check. Turns out this function is
>>> really copied exclude_super_stripes, excluding the calls to
>>> exclude_super_stripes. Later commit e4797df6a9fa ("btrfs-progs: check
>>> the free space tree in btrfsck") introduced proper version of
>>> exclude_super_stripes. Instead of duplicating the function, just remove
>>> account_super_bytes and use exclude_super_stripes instead of the former.
>>> This also has the benefit of bringing the userspace code a bit closer
>>> to the kernel counterpart.
>>>
>>> Signed-off-by: Nikolay Borisov <nborisov@suse.com>
>>
>> Since these two functions are the same, it's completely fine to remove one.
>
> As I mentioned, they are not "comlpetely" the same. The difference is
> that exclude_super_stripes will mark the stripes as
> EXTENT_UPTODATE in fs_info->pinned_extents via add_excluded_extent.
> Dunno if this has any repercussions on functionality. I've run the progs
> test suite and didn't observe any regressions. Also looking at the usage
> of fs_info->pinned_extents didn't see anything conspicuous.
Yeah, still same conclusion here.
All pinned_extents usage I found is either really for pinned extents of
current transaction (EXTENT_DIRTY) or this excluded usage (EXTENT_UPTODATE).
And unlike EXTENT_DIRTY, EXTENT_UPTODATE won't be removed after
transaction commit, so if I didn't miss anything important, it should be OK.
Just adding Su for this, as he worked on pinning down tree blocks for
lowmem mode extent init re-init, he may be more experienced in this field.
Despite that, such abuse of EXTENT_* bits in different trees at least
need extra comment for them later.
Thanks,
Qu
>
>>
>> Reviewed-by: Qu Wenruo <wqu@suse.com>
>>
>> Thanks,
>> Qu
>>
>>> ---
>>> extent-tree.c | 52 ++--------------------------------------------------
>>> 1 file changed, 2 insertions(+), 50 deletions(-)
>>>
>>> diff --git a/extent-tree.c b/extent-tree.c
>>> index ea205ccf4c30..391f0a784710 100644
>>> --- a/extent-tree.c
>>> +++ b/extent-tree.c
>>> @@ -3164,54 +3164,6 @@ static int find_first_block_group(struct btrfs_root *root,
>>> return ret;
>>> }
>>>
>>> -static void account_super_bytes(struct btrfs_fs_info *fs_info,
>>> - struct btrfs_block_group_cache *cache)
>>> -{
>>> - u64 bytenr;
>>> - u64 *logical;
>>> - int stripe_len;
>>> - int i, nr, ret;
>>> -
>>> - if (cache->key.objectid < BTRFS_SUPER_INFO_OFFSET) {
>>> - stripe_len = BTRFS_SUPER_INFO_OFFSET - cache->key.objectid;
>>> - cache->bytes_super += stripe_len;
>>> - }
>>> -
>>> - for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) {
>>> - bytenr = btrfs_sb_offset(i);
>>> - ret = btrfs_rmap_block(fs_info,
>>> - cache->key.objectid, bytenr,
>>> - 0, &logical, &nr, &stripe_len);
>>> - if (ret)
>>> - return;
>>> -
>>> - while (nr--) {
>>> - u64 start, len;
>>> -
>>> - if (logical[nr] > cache->key.objectid +
>>> - cache->key.offset)
>>> - continue;
>>> -
>>> - if (logical[nr] + stripe_len <= cache->key.objectid)
>>> - continue;
>>> -
>>> - start = logical[nr];
>>> - if (start < cache->key.objectid) {
>>> - start = cache->key.objectid;
>>> - len = (logical[nr] + stripe_len) - start;
>>> - } else {
>>> - len = min_t(u64, stripe_len,
>>> - cache->key.objectid +
>>> - cache->key.offset - start);
>>> - }
>>> -
>>> - cache->bytes_super += len;
>>> - }
>>> -
>>> - kfree(logical);
>>> - }
>>> -}
>>> -
>>> int btrfs_read_block_groups(struct btrfs_root *root)
>>> {
>>> struct btrfs_path *path;
>>> @@ -3287,7 +3239,7 @@ int btrfs_read_block_groups(struct btrfs_root *root)
>>> if (btrfs_chunk_readonly(info, cache->key.objectid))
>>> cache->ro = 1;
>>>
>>> - account_super_bytes(info, cache);
>>> + exclude_super_stripes(root, cache);
>>>
>>> ret = update_space_info(info, cache->flags, found_key.offset,
>>> btrfs_block_group_used(&cache->item),
>>> @@ -3331,7 +3283,7 @@ btrfs_add_block_group(struct btrfs_fs_info *fs_info, u64 bytes_used, u64 type,
>>> cache->flags = type;
>>> btrfs_set_block_group_flags(&cache->item, type);
>>>
>>> - account_super_bytes(fs_info, cache);
>>> + exclude_super_stripes(fs_info->extent_root, cache);
>>> ret = update_space_info(fs_info, cache->flags, size, bytes_used,
>>> &cache->space_info);
>>> BUG_ON(ret);
>>>
>>
> --
> 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
>
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] btrfs-progs: Use exclude_super_stripes instead of account_super_bytes
2018-05-02 13:13 ` Qu Wenruo
@ 2018-05-04 1:49 ` Su Yue
0 siblings, 0 replies; 7+ messages in thread
From: Su Yue @ 2018-05-04 1:49 UTC (permalink / raw)
To: quwenruo.btrfs; +Cc: nborisov, linux-btrfs, wqu, osandov, Su Yue
On Wed, May 2, 2018 at 9:15 PM Qu Wenruo <quwenruo.btrfs@gmx.com> wrote:
> On 2018年05月02日 20:49, Nikolay Borisov wrote:
> >
> >
> > On 2.05.2018 15:29, Qu Wenruo wrote:
> >>
> >>
> >> On 2018年05月02日 19:52, Nikolay Borisov wrote:
> >>> Originally commit 2681e00f00fe ("btrfs-progs: check for matchingi
> >>> free space in cache") added the account_super_bytes function to
prevent
> >>> false negative when running btrfs check. Turns out this function is
> >>> really copied exclude_super_stripes, excluding the calls to
> >>> exclude_super_stripes. Later commit e4797df6a9fa ("btrfs-progs: check
> >>> the free space tree in btrfsck") introduced proper version of
> >>> exclude_super_stripes. Instead of duplicating the function, just
remove
> >>> account_super_bytes and use exclude_super_stripes instead of the
former.
> >>> This also has the benefit of bringing the userspace code a bit closer
> >>> to the kernel counterpart.
> >>>
> >>> Signed-off-by: Nikolay Borisov <nborisov@suse.com>
> >>
> >> Since these two functions are the same, it's completely fine to remove
one.
> >
> > As I mentioned, they are not "comlpetely" the same. The difference is
> > that exclude_super_stripes will mark the stripes as
> > EXTENT_UPTODATE in fs_info->pinned_extents via add_excluded_extent.
> > Dunno if this has any repercussions on functionality. I've run the progs
> > test suite and didn't observe any regressions. Also looking at the usage
> > of fs_info->pinned_extents didn't see anything conspicuous.
> Yeah, still same conclusion here.
> All pinned_extents usage I found is either really for pinned extents of
> current transaction (EXTENT_DIRTY) or this excluded usage
(EXTENT_UPTODATE).
> And unlike EXTENT_DIRTY, EXTENT_UPTODATE won't be removed after
> transaction commit, so if I didn't miss anything important, it should be
OK.
It seems related to extents of super block.
In btrfs-progs, every block group is cached by function cache_block_group().
This function calls remove_sb_from_cache() to exclude extents of SB in block
group from free space cache.
So, I think it should be OK too.
> Just adding Su for this, as he worked on pinning down tree blocks for
> lowmem mode extent init re-init, he may be more experienced in this field.
> Despite that, such abuse of EXTENT_* bits in different trees at least
> need extra comment for them later.
Agreed. Kernel part has comments for those codes even they are simple.
Thanks,
Su
> Thanks,
> Qu
> >
> >>
> >> Reviewed-by: Qu Wenruo <wqu@suse.com>
> >>
> >> Thanks,
> >> Qu
> >>
> >>> ---
> >>> extent-tree.c | 52
++--------------------------------------------------
> >>> 1 file changed, 2 insertions(+), 50 deletions(-)
> >>>
> >>> diff --git a/extent-tree.c b/extent-tree.c
> >>> index ea205ccf4c30..391f0a784710 100644
> >>> --- a/extent-tree.c
> >>> +++ b/extent-tree.c
> >>> @@ -3164,54 +3164,6 @@ static int find_first_block_group(struct
btrfs_root *root,
> >>> return ret;
> >>> }
> >>>
> >>> -static void account_super_bytes(struct btrfs_fs_info *fs_info,
> >>> - struct btrfs_block_group_cache *cache)
> >>> -{
> >>> - u64 bytenr;
> >>> - u64 *logical;
> >>> - int stripe_len;
> >>> - int i, nr, ret;
> >>> -
> >>> - if (cache->key.objectid < BTRFS_SUPER_INFO_OFFSET) {
> >>> - stripe_len = BTRFS_SUPER_INFO_OFFSET -
cache->key.objectid;
> >>> - cache->bytes_super += stripe_len;
> >>> - }
> >>> -
> >>> - for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) {
> >>> - bytenr = btrfs_sb_offset(i);
> >>> - ret = btrfs_rmap_block(fs_info,
> >>> - cache->key.objectid, bytenr,
> >>> - 0, &logical, &nr, &stripe_len);
> >>> - if (ret)
> >>> - return;
> >>> -
> >>> - while (nr--) {
> >>> - u64 start, len;
> >>> -
> >>> - if (logical[nr] > cache->key.objectid +
> >>> - cache->key.offset)
> >>> - continue;
> >>> -
> >>> - if (logical[nr] + stripe_len <=
cache->key.objectid)
> >>> - continue;
> >>> -
> >>> - start = logical[nr];
> >>> - if (start < cache->key.objectid) {
> >>> - start = cache->key.objectid;
> >>> - len = (logical[nr] + stripe_len) - start;
> >>> - } else {
> >>> - len = min_t(u64, stripe_len,
> >>> - cache->key.objectid +
> >>> - cache->key.offset - start);
> >>> - }
> >>> -
> >>> - cache->bytes_super += len;
> >>> - }
> >>> -
> >>> - kfree(logical);
> >>> - }
> >>> -}
> >>> -
> >>> int btrfs_read_block_groups(struct btrfs_root *root)
> >>> {
> >>> struct btrfs_path *path;
> >>> @@ -3287,7 +3239,7 @@ int btrfs_read_block_groups(struct btrfs_root
*root)
> >>> if (btrfs_chunk_readonly(info, cache->key.objectid))
> >>> cache->ro = 1;
> >>>
> >>> - account_super_bytes(info, cache);
> >>> + exclude_super_stripes(root, cache);
> >>>
> >>> ret = update_space_info(info, cache->flags,
found_key.offset,
> >>>
btrfs_block_group_used(&cache->item),
> >>> @@ -3331,7 +3283,7 @@ btrfs_add_block_group(struct btrfs_fs_info
*fs_info, u64 bytes_used, u64 type,
> >>> cache->flags = type;
> >>> btrfs_set_block_group_flags(&cache->item, type);
> >>>
> >>> - account_super_bytes(fs_info, cache);
> >>> + exclude_super_stripes(fs_info->extent_root, cache);
> >>> ret = update_space_info(fs_info, cache->flags, size, bytes_used,
> >>> &cache->space_info);
> >>> BUG_ON(ret);
> >>>
> >>
> > --
> > 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
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] btrfs-progs: Use exclude_super_stripes instead of account_super_bytes
2018-05-02 11:52 [PATCH] btrfs-progs: Use exclude_super_stripes instead of account_super_bytes Nikolay Borisov
2018-05-02 12:29 ` Qu Wenruo
@ 2018-05-04 1:53 ` Su Yue
2018-05-08 17:34 ` David Sterba
2 siblings, 0 replies; 7+ messages in thread
From: Su Yue @ 2018-05-04 1:53 UTC (permalink / raw)
To: nborisov; +Cc: linux-btrfs, wqu, osandov
On Wed, May 2, 2018 at 7:55 PM Nikolay Borisov <nborisov@suse.com> wrote:
> Originally commit 2681e00f00fe ("btrfs-progs: check for matchingi
> free space in cache") added the account_super_bytes function to prevent
> false negative when running btrfs check. Turns out this function is
> really copied exclude_super_stripes, excluding the calls to
> exclude_super_stripes. Later commit e4797df6a9fa ("btrfs-progs: check
> the free space tree in btrfsck") introduced proper version of
> exclude_super_stripes. Instead of duplicating the function, just remove
> account_super_bytes and use exclude_super_stripes instead of the former.
> This also has the benefit of bringing the userspace code a bit closer
> to the kernel counterpart.
> Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: Su Yue <suy.fnst@cn.fujitsu.com>
> ---
> extent-tree.c | 52 ++--------------------------------------------------
> 1 file changed, 2 insertions(+), 50 deletions(-)
> diff --git a/extent-tree.c b/extent-tree.c
> index ea205ccf4c30..391f0a784710 100644
> --- a/extent-tree.c
> +++ b/extent-tree.c
> @@ -3164,54 +3164,6 @@ static int find_first_block_group(struct
btrfs_root *root,
> return ret;
> }
> -static void account_super_bytes(struct btrfs_fs_info *fs_info,
> - struct btrfs_block_group_cache *cache)
> -{
> - u64 bytenr;
> - u64 *logical;
> - int stripe_len;
> - int i, nr, ret;
> -
> - if (cache->key.objectid < BTRFS_SUPER_INFO_OFFSET) {
> - stripe_len = BTRFS_SUPER_INFO_OFFSET -
cache->key.objectid;
> - cache->bytes_super += stripe_len;
> - }
> -
> - for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) {
> - bytenr = btrfs_sb_offset(i);
> - ret = btrfs_rmap_block(fs_info,
> - cache->key.objectid, bytenr,
> - 0, &logical, &nr, &stripe_len);
> - if (ret)
> - return;
> -
> - while (nr--) {
> - u64 start, len;
> -
> - if (logical[nr] > cache->key.objectid +
> - cache->key.offset)
> - continue;
> -
> - if (logical[nr] + stripe_len <=
cache->key.objectid)
> - continue;
> -
> - start = logical[nr];
> - if (start < cache->key.objectid) {
> - start = cache->key.objectid;
> - len = (logical[nr] + stripe_len) - start;
> - } else {
> - len = min_t(u64, stripe_len,
> - cache->key.objectid +
> - cache->key.offset - start);
> - }
> -
> - cache->bytes_super += len;
> - }
> -
> - kfree(logical);
> - }
> -}
> -
> int btrfs_read_block_groups(struct btrfs_root *root)
> {
> struct btrfs_path *path;
> @@ -3287,7 +3239,7 @@ int btrfs_read_block_groups(struct btrfs_root *root)
> if (btrfs_chunk_readonly(info, cache->key.objectid))
> cache->ro = 1;
> - account_super_bytes(info, cache);
> + exclude_super_stripes(root, cache);
> ret = update_space_info(info, cache->flags,
found_key.offset,
btrfs_block_group_used(&cache->item),
> @@ -3331,7 +3283,7 @@ btrfs_add_block_group(struct btrfs_fs_info
*fs_info, u64 bytes_used, u64 type,
> cache->flags = type;
> btrfs_set_block_group_flags(&cache->item, type);
> - account_super_bytes(fs_info, cache);
> + exclude_super_stripes(fs_info->extent_root, cache);
> ret = update_space_info(fs_info, cache->flags, size, bytes_used,
> &cache->space_info);
> BUG_ON(ret);
> --
> 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
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] btrfs-progs: Use exclude_super_stripes instead of account_super_bytes
2018-05-02 11:52 [PATCH] btrfs-progs: Use exclude_super_stripes instead of account_super_bytes Nikolay Borisov
2018-05-02 12:29 ` Qu Wenruo
2018-05-04 1:53 ` Su Yue
@ 2018-05-08 17:34 ` David Sterba
2 siblings, 0 replies; 7+ messages in thread
From: David Sterba @ 2018-05-08 17:34 UTC (permalink / raw)
To: Nikolay Borisov; +Cc: linux-btrfs, wqu, osandov
On Wed, May 02, 2018 at 02:52:54PM +0300, Nikolay Borisov wrote:
> Originally commit 2681e00f00fe ("btrfs-progs: check for matchingi
> free space in cache") added the account_super_bytes function to prevent
> false negative when running btrfs check. Turns out this function is
> really copied exclude_super_stripes, excluding the calls to
> exclude_super_stripes. Later commit e4797df6a9fa ("btrfs-progs: check
> the free space tree in btrfsck") introduced proper version of
> exclude_super_stripes. Instead of duplicating the function, just remove
> account_super_bytes and use exclude_super_stripes instead of the former.
> This also has the benefit of bringing the userspace code a bit closer
> to the kernel counterpart.
>
> Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Applied, thanks.
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2018-05-08 17:37 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-05-02 11:52 [PATCH] btrfs-progs: Use exclude_super_stripes instead of account_super_bytes Nikolay Borisov
2018-05-02 12:29 ` Qu Wenruo
2018-05-02 12:49 ` Nikolay Borisov
2018-05-02 13:13 ` Qu Wenruo
2018-05-04 1:49 ` Su Yue
2018-05-04 1:53 ` Su Yue
2018-05-08 17:34 ` David Sterba
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.