[PATCH 3/3][reiser4] dont get radix-tree dirty tagging out of sync

[PATCH 3/3][reiser4] dont get radix-tree dirty tagging out of sync

[Ryan Hope]

This was item #14 on the todo list for reiser4 inclusion in mainline:

diff --git a/fs/reiser4/page_cache.c b/fs/reiser4/page_cache.c
index fe71368..a662c25 100644
--- a/fs/reiser4/page_cache.c
+++ b/fs/reiser4/page_cache.c
@@ -467,15 +467,14 @@ int reiser4_set_page_dirty_internal(struct page *page)
  	BUG_ON(mapping == NULL);

  	if (!TestSetPageDirty(page)) {
+		spin_lock_irq(&mapping->tree_lock);
  		if (mapping_cap_account_dirty(mapping))
  			inc_zone_page_state(page, NR_FILE_DIRTY);
-
+		radix_tree_tag_set(&mapping->page_tree,
+			page_index(page), PAGECACHE_TAG_DIRTY);
  		__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
+		spin_unlock_irq(&mapping->tree_lock);
  	}
-
-	/* znode must be dirty ? */
-	if (mapping->host == reiser4_get_super_fake(mapping->host->i_sb))
-		assert("", JF_ISSET(jprivate(page), JNODE_DIRTY));
  	return 0;
  }

-Ryan

Re: [PATCH 3/3][reiser4] dont get radix-tree dirty tagging out of sync

[Nick Piggin]

On Tuesday 12 August 2008 08:40, Ryan Hope wrote:
> This was item #14 on the todo list for reiser4 inclusion in mainline:
>
> diff --git a/fs/reiser4/page_cache.c b/fs/reiser4/page_cache.c
> index fe71368..a662c25 100644
> --- a/fs/reiser4/page_cache.c
> +++ b/fs/reiser4/page_cache.c
> @@ -467,15 +467,14 @@ int reiser4_set_page_dirty_internal(struct page
> *page) BUG_ON(mapping == NULL);
>
>   	if (!TestSetPageDirty(page)) {
> +		spin_lock_irq(&mapping->tree_lock);
>   		if (mapping_cap_account_dirty(mapping))
>   			inc_zone_page_state(page, NR_FILE_DIRTY);
> -
> +		radix_tree_tag_set(&mapping->page_tree,
> +			page_index(page), PAGECACHE_TAG_DIRTY);
>   		__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
> +		spin_unlock_irq(&mapping->tree_lock);
>   	}
> -
> -	/* znode must be dirty ? */
> -	if (mapping->host == reiser4_get_super_fake(mapping->host->i_sb))
> -		assert("", JF_ISSET(jprivate(page), JNODE_DIRTY));
>   	return 0;
>   }

Any reason why this can't use a generic function such as
__set_page_dirty_nobuffers? There are accounting changes gone in
there now which I suspetc may be wrong now in reiser4 (eg. task
io accounting).

Actually every site that does a radix_tree_operation there should
be reviewed to try to use core functoins.

Re: [PATCH 3/3][reiser4] dont get radix-tree dirty tagging out of sync

[Ryan Hope]

I tried just using __set_page_dirty_nobuffers but that caused issues
with ext3/ext4 (i can send a bug trace later if needed).

                                __inc_bdi_stat(mapping->backing_dev_info,
                                                BDI_RECLAIMABLE);
                                task_io_account_write(PAGE_CACHE_SIZE);

^^^ the above code in __set_page_dirty_nobuffers causes an issue with
do_writepages in ext3/4 when I use something like the code below:

int reiser4_set_page_dirty_internal(struct page *page)
{
     return __set_page_dirty_nobuffers(page);
}

On Mon, Aug 11, 2008 at 10:36 PM, Nick Piggin <nickpiggin@yahoo.com.au> wrote:
> On Tuesday 12 August 2008 08:40, Ryan Hope wrote:
>> This was item #14 on the todo list for reiser4 inclusion in mainline:
>>
>> diff --git a/fs/reiser4/page_cache.c b/fs/reiser4/page_cache.c
>> index fe71368..a662c25 100644
>> --- a/fs/reiser4/page_cache.c
>> +++ b/fs/reiser4/page_cache.c
>> @@ -467,15 +467,14 @@ int reiser4_set_page_dirty_internal(struct page
>> *page) BUG_ON(mapping == NULL);
>>
>>       if (!TestSetPageDirty(page)) {
>> +             spin_lock_irq(&mapping->tree_lock);
>>               if (mapping_cap_account_dirty(mapping))
>>                       inc_zone_page_state(page, NR_FILE_DIRTY);
>> -
>> +             radix_tree_tag_set(&mapping->page_tree,
>> +                     page_index(page), PAGECACHE_TAG_DIRTY);
>>               __mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
>> +             spin_unlock_irq(&mapping->tree_lock);
>>       }
>> -
>> -     /* znode must be dirty ? */
>> -     if (mapping->host == reiser4_get_super_fake(mapping->host->i_sb))
>> -             assert("", JF_ISSET(jprivate(page), JNODE_DIRTY));
>>       return 0;
>>   }
>
> Any reason why this can't use a generic function such as
> __set_page_dirty_nobuffers? There are accounting changes gone in
> there now which I suspetc may be wrong now in reiser4 (eg. task
> io accounting).
>
> Actually every site that does a radix_tree_operation there should
> be reviewed to try to use core functoins.
>

Re: [PATCH 3/3][reiser4] dont get radix-tree dirty tagging out of sync

[Nick Piggin]

On Tuesday 12 August 2008 14:12, Ryan Hope wrote:
> I tried just using __set_page_dirty_nobuffers but that caused issues
> with ext3/ext4 (i can send a bug trace later if needed).
>
>                                 __inc_bdi_stat(mapping->backing_dev_info,
>                                                 BDI_RECLAIMABLE);
>                                 task_io_account_write(PAGE_CACHE_SIZE);
>
> ^^^ the above code in __set_page_dirty_nobuffers causes an issue with
> do_writepages in ext3/4 when I use something like the code below:
>
> int reiser4_set_page_dirty_internal(struct page *page)
> {
>      return __set_page_dirty_nobuffers(page);
> }

Hmm, it causes issues in ext3/4 even though it doesn't change any
code executed by ext3/4? Yes, if you could send a trace...

Re: [PATCH 3/3][reiser4] dont get radix-tree dirty tagging out of sync

[Ryan Hope]

On Tue, Aug 12, 2008 at 2:00 AM, Nick Piggin <nickpiggin@yahoo.com.au> wrote:
> On Tuesday 12 August 2008 14:12, Ryan Hope wrote:
>> I tried just using __set_page_dirty_nobuffers but that caused issues
>> with ext3/ext4 (i can send a bug trace later if needed).
>>
>>                                 __inc_bdi_stat(mapping->backing_dev_info,
>>                                                 BDI_RECLAIMABLE);
>>                                 task_io_account_write(PAGE_CACHE_SIZE);
>>
>> ^^^ the above code in __set_page_dirty_nobuffers causes an issue with
>> do_writepages in ext3/4 when I use something like the code below:
>>
>> int reiser4_set_page_dirty_internal(struct page *page)
>> {
>>      return __set_page_dirty_nobuffers(page);
>> }
>
> Hmm, it causes issues in ext3/4 even though it doesn't change any
> code executed by ext3/4? Yes, if you could send a trace...
>

Pid: 2620, comm: bonnie Not tainted 2.6.27-rc2-zen1 #1
 [<c026d729>] do_writepages+0x49/0x50
 [<c02dc830>] ext3_write_inode+0x0/0x40
 [<c02dc860>] ext3_write_inode+0x30/0x40
 [<c02ad3a2>] __writeback_single_inode+0x282/0x390
 [<c02ad9d4>] generic_sync_sb_inodes+0x304/0x3a0
 [<f9b9b234>] reiser4_sync_inodes+0x54/0x130 [reiser4]
 [<c02ad940>] generic_sync_sb_inodes+0x270/0x3a0
 [<c02adc64>] writeback_inodes+0x44/0xd0
 [<c026d0d0>] balance_dirty_pages_ratelimited_nr+0x230/0x370
 [<f9b79a5c>] reiser4_txn_end+0x4ec/0xae0 [reiser4]
 [<f9bc95ca>] balance_dirty_page_cluster+0x10a/0x1b0 [reiser4]
 [<f9bd17a0>] write_cryptcompress+0x610/0xdf0 [reiser4]
 [<f9bc2e76>] reiser4_write_careful+0xb6/0x880 [reiser4]
 [<c021c3ae>] update_curr+0x4e/0x70
 [<c02214c2>] task_tick_fair+0x32/0x90
 [<c023cf57>] hrtimer_forward+0xf7/0x150
 [<c0240523>] getnstimeofday+0x43/0xf0
 [<c0243885>] clockevents_program_event+0xa5/0x160
 [<c022f746>] run_timer_softirq+0x146/0x1c0
 [<c02449ca>] tick_program_event+0x3a/0x70
 [<c034a74c>] security_file_permission+0xc/0x10
 [<c028dcca>] rw_verify_area+0x4a/0xc0
 [<f9bc2dc0>] reiser4_write_careful+0x0/0x880 [reiser4]
 [<c028e1a0>] vfs_write+0xa0/0x140
 [<c028e311>] sys_write+0x41/0x80
 [<c020330d>] sysenter_do_call+0x12/0x25
 =======================

Re: [PATCH 3/3][reiser4] dont get radix-tree dirty tagging out of sync

[Edward Shishkin]

Nick Piggin wrote:
> On Tuesday 12 August 2008 08:40, Ryan Hope wrote:
>   
>> This was item #14 on the todo list for reiser4 inclusion in mainline:
>>     

No. This patch is a nonsense.
Where did you see  radix-tree dirty tagging here?


>> diff --git a/fs/reiser4/page_cache.c b/fs/reiser4/page_cache.c
>> index fe71368..a662c25 100644
>> --- a/fs/reiser4/page_cache.c
>> +++ b/fs/reiser4/page_cache.c
>> @@ -467,15 +467,14 @@ int reiser4_set_page_dirty_internal(struct page
>> *page) BUG_ON(mapping == NULL);
>>
>>   	if (!TestSetPageDirty(page)) {
>> +		spin_lock_irq(&mapping->tree_lock);
>>   		if (mapping_cap_account_dirty(mapping))
>>   			inc_zone_page_state(page, NR_FILE_DIRTY);
>> -
>> +		radix_tree_tag_set(&mapping->page_tree,
>> +			page_index(page), PAGECACHE_TAG_DIRTY);
>>   		__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
>> +		spin_unlock_irq(&mapping->tree_lock);
>>   	}
>> -
>> -	/* znode must be dirty ? */
>> -	if (mapping->host == reiser4_get_super_fake(mapping->host->i_sb))
>> -		assert("", JF_ISSET(jprivate(page), JNODE_DIRTY));
>>   	return 0;
>>   }
>>     
>
> Any reason why this can't use a generic function such as
> __set_page_dirty_nobuffers? 

Currently reiser4 is working around "anonymous" pages dirtied
outside of reiser4 context (e.g. via mmap), where some reiser4
specific work (jnode creation, capturing by an atom) can not
be done.

> There are accounting changes gone in
> there now which I suspetc may be wrong now in reiser4 (eg. task
> io accounting).
>   

Yes, this is definitely wrong. Thanks for pointing this out.
Such poking around vfs internals should be fixed, otherwise we'll
have permanent problems.

I have cc-ed Vladimir: maybe he has some hints. At least, I know
that he looked at this problem..

Thanks,
Edward.

> Actually every site that does a radix_tree_operation there should
> be reviewed to try to use core functoins.
>
>