linux-f2fs-devel — The f2fs filesystem development

Re: [f2fs-dev] [PATCH] f2fs: add error checking in do_write_page()

[Chao Yu <ch...@ke...>]

On 7/24/2025 11:59 PM, mason.zhang wrote:
> Otherwise, the filesystem may unaware of potential file corruption.
> 
> Signed-off-by: mason.zhang <mas...@gm...>
> ---
>   fs/f2fs/segment.c | 7 +++++--
>   1 file changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
> index ae1223ef648f..ce80ba226aed 100644
> --- a/fs/f2fs/segment.c
> +++ b/fs/f2fs/segment.c
> @@ -3936,12 +3936,15 @@ static void do_write_page(struct f2fs_summary *sum, struct f2fs_io_info *fio)
>   	int seg_type = log_type_to_seg_type(type);
>   	bool keep_order = (f2fs_lfs_mode(fio->sbi) &&
>   				seg_type == CURSEG_COLD_DATA);
> +	int err;
>   
>   	if (keep_order)
>   		f2fs_down_read(&fio->sbi->io_order_lock);
>   
> -	if (f2fs_allocate_data_block(fio->sbi, fio->page, fio->old_blkaddr,
> -			&fio->new_blkaddr, sum, type, fio)) {
> +	err = f2fs_allocate_data_block(fio->sbi, fio->page, fio->old_blkaddr,
> +			&fio->new_blkaddr, sum, type, fio);
> +	if (unlikely(err)) {
> +		f2fs_err(fio->sbi, "Failed to allocate data block(%d)", err);

How about using f2fs_err_ratelimited() to avoid too many logs?

Can you please dump more informations about inode, page, blkaddr...?

Thanks,

>   		if (fscrypt_inode_uses_fs_layer_crypto(folio->mapping->host))
>   			fscrypt_finalize_bounce_page(&fio->encrypted_page);
>   		folio_end_writeback(folio);

Re: [f2fs-dev] [PATCH v2 2/2] f2fs: directly add newly allocated pre-dirty nat entry to dirty set list

[wangzijie <wan...@ho...>]

> On 7/22/25 22:36, wangzijie wrote:
> > When we need to alloc nat entry and set it dirty, we can directly add it to
> > dirty set list(or initialize its list_head for new_ne) instead of adding it
> > to clean list and make a move. Introduce init_dirty flag to do it.
> > 
> > Signed-off-by: wangzijie <wan...@ho...>
> > ---
> >  fs/f2fs/node.c | 37 ++++++++++++++++++++++++++++++-------
> >  1 file changed, 30 insertions(+), 7 deletions(-)
> > 
> > diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
> > index a23db6238..20bcf8559 100644
> > --- a/fs/f2fs/node.c
> > +++ b/fs/f2fs/node.c
> > @@ -185,7 +185,7 @@ static void __free_nat_entry(struct nat_entry *e)
> >  
> >  /* must be locked by nat_tree_lock */
> >  static struct nat_entry *__init_nat_entry(struct f2fs_nm_info *nm_i,
> > -	struct nat_entry *ne, struct f2fs_nat_entry *raw_ne, bool no_fail)
> > +	struct nat_entry *ne, struct f2fs_nat_entry *raw_ne, bool no_fail, bool init_dirty)
> >  {
> >  	if (no_fail)
> >  		f2fs_radix_tree_insert(&nm_i->nat_root, nat_get_nid(ne), ne);
> > @@ -195,6 +195,11 @@ static struct nat_entry *__init_nat_entry(struct f2fs_nm_info *nm_i,
> >  	if (raw_ne)
> >  		node_info_from_raw_nat(&ne->ni, raw_ne);
> >  
> > +	if (init_dirty) {
> > +		nm_i->nat_cnt[TOTAL_NAT]++;
> > +		return ne;
> > +	}
> > +
> >  	spin_lock(&nm_i->nat_list_lock);
> >  	list_add_tail(&ne->list, &nm_i->nat_entries);
> >  	spin_unlock(&nm_i->nat_list_lock);
> > @@ -256,7 +261,7 @@ static struct nat_entry_set *__grab_nat_entry_set(struct f2fs_nm_info *nm_i,
> >  }
> >  
> >  static void __set_nat_cache_dirty(struct f2fs_nm_info *nm_i,
> > -						struct nat_entry *ne)
> > +					struct nat_entry *ne, bool init_dirty)
> >  {
> >  	struct nat_entry_set *head;
> >  	bool new_ne = nat_get_blkaddr(ne) == NEW_ADDR;
> > @@ -275,6 +280,18 @@ static void __set_nat_cache_dirty(struct f2fs_nm_info *nm_i,
> >  
> >  	set_nat_flag(ne, IS_PREALLOC, new_ne);
> >  
> > +	if (init_dirty) {
> > +		nm_i->nat_cnt[DIRTY_NAT]++;
> > +		set_nat_flag(ne, IS_DIRTY, true);
> > +		spin_lock(&nm_i->nat_list_lock);
> > +		if (new_ne)
> > +			INIT_LIST_HEAD(&ne->list);
> > +		else
> > +			list_add_tail(&ne->list, &head->entry_list);
> > +		spin_unlock(&nm_i->nat_list_lock);
> > +		return;
> > +    }
> 
> Nit issue, above blanks should be replaced w/ tab.

Ah...my bad :-(

> Can we clean up like this?
> 
> diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
> index de99b42437c6..60fc2c7b8e10 100644
> --- a/fs/f2fs/node.c
> +++ b/fs/f2fs/node.c
> @@ -280,30 +280,23 @@ static void __set_nat_cache_dirty(struct f2fs_nm_info *nm_i,
> 
>  	set_nat_flag(ne, IS_PREALLOC, new_ne);
> 
> -	if (init_dirty) {
> -		nm_i->nat_cnt[DIRTY_NAT]++;
> -		set_nat_flag(ne, IS_DIRTY, true);
> -		spin_lock(&nm_i->nat_list_lock);
> -		if (new_ne)
> -			INIT_LIST_HEAD(&ne->list);
> -		else
> -			list_add_tail(&ne->list, &head->entry_list);
> -		spin_unlock(&nm_i->nat_list_lock);
> -		return;
> -    }
> -
>  	if (get_nat_flag(ne, IS_DIRTY))
>  		goto refresh_list;
> 
>  	nm_i->nat_cnt[DIRTY_NAT]++;
> -	nm_i->nat_cnt[RECLAIMABLE_NAT]--;
> +	if (!init_dirty)
> +		nm_i->nat_cnt[RECLAIMABLE_NAT]--;
>  	set_nat_flag(ne, IS_DIRTY, true);
>  refresh_list:
>  	spin_lock(&nm_i->nat_list_lock);
> -	if (new_ne)
> -		list_del_init(&ne->list);
> -	else
> +	if (new_ne) {
> +		if (init_dirty)
> +			INIT_LIST_HEAD(&ne->list);
> +		else
> +			list_del_init(&ne->list);
> +	} else {
>  		list_move_tail(&ne->list, &head->entry_list);
> +	}
>  	spin_unlock(&nm_i->nat_list_lock);
>  }
> 
> Thanks,

We need to init list_head before using list_move_tail.
I think we can do more clean up like this, keep refresh_list part code.

diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
index 20bcf8559..ebb624fa1 100644
--- a/fs/f2fs/node.c
+++ b/fs/f2fs/node.c
@@ -196,6 +196,7 @@ static struct nat_entry *__init_nat_entry(struct f2fs_nm_info *nm_i,
 		node_info_from_raw_nat(&ne->ni, raw_ne);
 
 	if (init_dirty) {
+		INIT_LIST_HEAD(&ne->list);
 		nm_i->nat_cnt[TOTAL_NAT]++;
 		return ne;
 	}
@@ -280,23 +281,12 @@ static void __set_nat_cache_dirty(struct f2fs_nm_info *nm_i,
 
 	set_nat_flag(ne, IS_PREALLOC, new_ne);
 
-	if (init_dirty) {
-		nm_i->nat_cnt[DIRTY_NAT]++;
-		set_nat_flag(ne, IS_DIRTY, true);
-		spin_lock(&nm_i->nat_list_lock);
-		if (new_ne)
-			INIT_LIST_HEAD(&ne->list);
-		else
-			list_add_tail(&ne->list, &head->entry_list);
-		spin_unlock(&nm_i->nat_list_lock);
-		return;
-    }
-
 	if (get_nat_flag(ne, IS_DIRTY))
 		goto refresh_list;
 
 	nm_i->nat_cnt[DIRTY_NAT]++;
-	nm_i->nat_cnt[RECLAIMABLE_NAT]--;
+	if (!init_dirty)
+		nm_i->nat_cnt[RECLAIMABLE_NAT]--;
 	set_nat_flag(ne, IS_DIRTY, true);
 refresh_list:
 	spin_lock(&nm_i->nat_list_lock);

> > +
> >  	if (get_nat_flag(ne, IS_DIRTY))
> >  		goto refresh_list;
> >  
> > @@ -441,7 +458,7 @@ static void cache_nat_entry(struct f2fs_sb_info *sbi, nid_t nid,
> >  	f2fs_down_write(&nm_i->nat_tree_lock);
> >  	e = __lookup_nat_cache(nm_i, nid, false);
> >  	if (!e)
> > -		e = __init_nat_entry(nm_i, new, ne, false);
> > +		e = __init_nat_entry(nm_i, new, ne, false, false);
> >  	else
> >  		f2fs_bug_on(sbi, nat_get_ino(e) != le32_to_cpu(ne->ino) ||
> >  				nat_get_blkaddr(e) !=
> > @@ -458,11 +475,13 @@ static void set_node_addr(struct f2fs_sb_info *sbi, struct node_info *ni,
> >  	struct f2fs_nm_info *nm_i = NM_I(sbi);
> >  	struct nat_entry *e;
> >  	struct nat_entry *new = __alloc_nat_entry(sbi, ni->nid, true);
> > +	bool init_dirty = false;
> >  
> >  	f2fs_down_write(&nm_i->nat_tree_lock);
> >  	e = __lookup_nat_cache(nm_i, ni->nid, true);
> >  	if (!e) {
> > -		e = __init_nat_entry(nm_i, new, NULL, true);
> > +		init_dirty = true;
> > +		e = __init_nat_entry(nm_i, new, NULL, true, true);
> >  		copy_node_info(&e->ni, ni);
> >  		f2fs_bug_on(sbi, ni->blk_addr == NEW_ADDR);
> >  	} else if (new_blkaddr == NEW_ADDR) {
> > @@ -498,7 +517,7 @@ static void set_node_addr(struct f2fs_sb_info *sbi, struct node_info *ni,
> >  	nat_set_blkaddr(e, new_blkaddr);
> >  	if (!__is_valid_data_blkaddr(new_blkaddr))
> >  		set_nat_flag(e, IS_CHECKPOINTED, false);
> > -	__set_nat_cache_dirty(nm_i, e);
> > +	__set_nat_cache_dirty(nm_i, e, init_dirty);
> >  
> >  	/* update fsync_mark if its inode nat entry is still alive */
> >  	if (ni->nid != ni->ino)
> > @@ -2924,6 +2943,7 @@ static void remove_nats_in_journal(struct f2fs_sb_info *sbi)
> >  	struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
> >  	struct f2fs_journal *journal = curseg->journal;
> >  	int i;
> > +	bool init_dirty;
> >  
> >  	down_write(&curseg->journal_rwsem);
> >  	for (i = 0; i < nats_in_cursum(journal); i++) {
> > @@ -2934,12 +2954,15 @@ static void remove_nats_in_journal(struct f2fs_sb_info *sbi)
> >  		if (f2fs_check_nid_range(sbi, nid))
> >  			continue;
> >  
> > +		init_dirty = false;
> > +
> >  		raw_ne = nat_in_journal(journal, i);
> >  
> >  		ne = __lookup_nat_cache(nm_i, nid, true);
> >  		if (!ne) {
> > +			init_dirty = true;
> >  			ne = __alloc_nat_entry(sbi, nid, true);
> > -			__init_nat_entry(nm_i, ne, &raw_ne, true);
> > +			__init_nat_entry(nm_i, ne, &raw_ne, true, true);
> >  		}
> >  
> >  		/*
> > @@ -2954,7 +2977,7 @@ static void remove_nats_in_journal(struct f2fs_sb_info *sbi)
> >  			spin_unlock(&nm_i->nid_list_lock);
> >  		}
> >  
> > -		__set_nat_cache_dirty(nm_i, ne);
> > +		__set_nat_cache_dirty(nm_i, ne, init_dirty);
> >  	}
> >  	update_nats_in_cursum(journal, -i);
> >  	up_write(&curseg->journal_rwsem);

Re: [f2fs-dev] [PATCH v2] mkfs.f2fs: support -C [no]hashonly to control linear lookup fallback

[Zhiguo N. <niu...@gm...>]

Chao Yu via Linux-f2fs-devel <lin...@li...>
于2025年7月25日周五 09:03写道：
>
> It provides a way to disable linear lookup fallback during mkfs.
>
> Behavior summary:
>                         Android         Distro
> By default              disabled        enabled
> Tune w/ [no]hashonly    no              yes
>
> Android case:
>
> 1.1) Disable linear lookup:
> mkfs.f2fs -f -g android -O casefold -C utf8:hashonly /dev/vdb
> dump.f2fs -d3 /dev/vdb |grep s_encoding_flags
> s_encoding_flags                        [0x       2 : 2]
>
> 1.2) Enable linear lookup:
> mkfs.f2fs -f -g android -O casefold -C utf8:nohashonly /dev/vdb
> dump.f2fs -d3 /dev/vdb |grep s_encoding_flags
> s_encoding_flags                        [0x       2 : 2]
Hi Chao,
Seems like a typo here? should be:
s_encoding_flags                        [0x       0 : 0] ?
thanks!
>
> 1.3) By default:
> mkfs.f2fs -f -g android -O casefold -C utf8 /dev/vdb
> dump.f2fs -d3 /dev/vdb |grep s_encoding_flags
> s_encoding_flags                        [0x       2 : 2]
>
> Distro case:
>
> 2.1) Disable linear lookup:
> mkfs.f2fs -f -O casefold -C utf8:hashonly /dev/vdb
> dump.f2fs -d3 /dev/vdb |grep s_encoding_flags
> s_encoding_flags                        [0x       2 : 2]
>
> 2.2) Enable linear lookup:
> mkfs.f2fs -f -O casefold -C utf8:nohashonly /dev/vdb
> dump.f2fs -d3 /dev/vdb |grep s_encoding_flags
> s_encoding_flags                        [0x       0 : 0]
>
> 2.3) By default:
> mkfs.f2fs -f -O casefold -C utf8 /dev/vdb
> dump.f2fs -d3 /dev/vdb |grep s_encoding_flags
> s_encoding_flags                        [0x       0 : 0]
>
> Signed-off-by: Chao Yu <ch...@ke...>
> ---
> v2:
> - disable linear lookup by default for Android case
>  include/f2fs_fs.h       |  3 ++-
>  lib/libf2fs.c           |  1 +
>  man/mkfs.f2fs.8         | 10 ++++++++--
>  mkfs/f2fs_format.c      |  3 +++
>  mkfs/f2fs_format_main.c |  3 ++-
>  5 files changed, 16 insertions(+), 4 deletions(-)
>
> diff --git a/include/f2fs_fs.h b/include/f2fs_fs.h
> index f7268d1..a8da8fa 100644
> --- a/include/f2fs_fs.h
> +++ b/include/f2fs_fs.h
> @@ -1478,7 +1478,8 @@ enum {
>
>  /* feature list in Android */
>  enum {
> -       F2FS_FEATURE_NAT_BITS = 0x0001,
> +       F2FS_FEATURE_NAT_BITS           = 0x0001,
> +       F2FS_FEATURE_LINEAR_LOOKUP      = 0x0002,
>  };
>
>  /* nolinear lookup tune */
> diff --git a/lib/libf2fs.c b/lib/libf2fs.c
> index 2f012c8..0e3e62a 100644
> --- a/lib/libf2fs.c
> +++ b/lib/libf2fs.c
> @@ -1424,6 +1424,7 @@ static const struct enc_flags {
>         char *param;
>  } encoding_flags[] = {
>         { F2FS_ENC_STRICT_MODE_FL, "strict" },
> +       { F2FS_ENC_NO_COMPAT_FALLBACK_FL, "hashonly"}
>  };
>
>  /* Return a positive number < 0xff indicating the encoding magic number
> diff --git a/man/mkfs.f2fs.8 b/man/mkfs.f2fs.8
> index 8b3b0cc..8cb7d32 100644
> --- a/man/mkfs.f2fs.8
> +++ b/man/mkfs.f2fs.8
> @@ -232,9 +232,15 @@ Use UTF-8 for casefolding.
>  .I flags:
>  .RS 1.2i
>  .TP 1.2i
> -.B strict
> +.B [no]strict
>  This flag specifies that invalid strings should be rejected by the filesystem.
> -Default is disabled.
> +For android case, it will disable linear lookup by default.
> +.RE
> +.RS 1.2i
> +.TP 1.2i
> +.B [no]hashonly
> +This flag specifies that there is no linear lookup fallback during lookup.
> +By default, linear lookup fallback is enabled.
>  .RE
>  .RE
>  .TP
> diff --git a/mkfs/f2fs_format.c b/mkfs/f2fs_format.c
> index 2680bd3..a45bbcb 100644
> --- a/mkfs/f2fs_format.c
> +++ b/mkfs/f2fs_format.c
> @@ -671,6 +671,9 @@ static int f2fs_prepare_super_block(void)
>         memcpy(sb->init_version, c.version, VERSION_LEN);
>
>         if (c.feature & F2FS_FEATURE_CASEFOLD) {
> +               if (!(c.s_encoding_flags & F2FS_ENC_NO_COMPAT_FALLBACK_FL) &&
> +                       (c.disabled_feature & F2FS_FEATURE_LINEAR_LOOKUP))
> +                       c.s_encoding_flags |= F2FS_ENC_NO_COMPAT_FALLBACK_FL;
>                 set_sb(s_encoding, c.s_encoding);
>                 set_sb(s_encoding_flags, c.s_encoding_flags);
>         }
> diff --git a/mkfs/f2fs_format_main.c b/mkfs/f2fs_format_main.c
> index f0bec4f..8f8e975 100644
> --- a/mkfs/f2fs_format_main.c
> +++ b/mkfs/f2fs_format_main.c
> @@ -143,7 +143,8 @@ static void add_default_options(void)
>                 force_overwrite = 1;
>                 c.wanted_sector_size = F2FS_BLKSIZE;
>                 c.root_uid = c.root_gid = 0;
> -               c.disabled_feature |= F2FS_FEATURE_NAT_BITS;
> +               c.disabled_feature |= F2FS_FEATURE_NAT_BITS |
> +                                       F2FS_FEATURE_LINEAR_LOOKUP;
>
>                 /* RO doesn't need any other features */
>                 if (c.feature & F2FS_FEATURE_RO)
> --
> 2.49.0
>
>
>
> _______________________________________________
> Linux-f2fs-devel mailing list
> Lin...@li...
> https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

Re: [f2fs-dev] [PATCH] f2fs: f2fs supports uncached buffered I/O

[hanqi <ha...@vi...>]

在 2025/7/24 21:09, Chao Yu 写道:
> On 2025/7/16 16:27, hanqi wrote:
>>
>>
>> 在 2025/7/16 11:43, Jens Axboe 写道:
>>> On 7/15/25 9:34 PM, hanqi wrote:
>>>>
>>>> ? 2025/7/15 22:28, Jens Axboe ??:
>>>>> On 7/14/25 9:10 PM, Qi Han wrote:
>>>>>> Jens has already completed the development of uncached buffered I/O
>>>>>> in git [1], and in f2fs, the feature can be enabled simply by 
>>>>>> setting
>>>>>> the FOP_DONTCACHE flag in f2fs_file_operations.
>>>>> You need to ensure that for any DONTCACHE IO that the completion is
>>>>> routed via non-irq context, if applicable. I didn't verify that 
>>>>> this is
>>>>> the case for f2fs. Generally you can deduce this as well through
>>>>> testing, I'd say the following cases would be interesting to test:
>>>>>
>>>>> 1) Normal DONTCACHE buffered read
>>>>> 2) Overwrite DONTCACHE buffered write
>>>>> 3) Append DONTCACHE buffered write
>>>>>
>>>>> Test those with DEBUG_ATOMIC_SLEEP set in your config, and it that
>>>>> doesn't complain, that's a great start.
>>>>>
>>>>> For the above test cases as well, verify that page cache doesn't 
>>>>> grow as
>>>>> IO is performed. A bit is fine for things like meta data, but 
>>>>> generally
>>>>> you want to see it remain basically flat in terms of page cache 
>>>>> usage.
>>>>>
>>>>> Maybe this is all fine, like I said I didn't verify. Just 
>>>>> mentioning it
>>>>> for completeness sake.
>>>> Hi, Jens
>>>> Thanks for your suggestion. As I mentioned earlier in [1], in f2fs,
>>>> the regular buffered write path invokes folio_end_writeback from a
>>>> softirq context. Therefore, it seems that f2fs may not be suitable
>>>> for DONTCACHE I/O writes.
>>>>
>>>> I?d like to ask a question: why is DONTCACHE I/O write restricted to
>>>> non-interrupt context only? Is it because dropping the page might be
>>>> too time-consuming to be done safely in interrupt context? This might
>>>> be a naive question, but I?d really appreciate your clarification.
>>>> Thanks in advance.
>>> Because (as of right now, at least) the code doing the invalidation
>>> needs process context. There are various reasons for this, which you'll
>>> see if you follow the path off folio_end_writeback() ->
>>> filemap_end_dropbehind_write() -> filemap_end_dropbehind() ->
>>> folio_unmap_invalidate(). unmap_mapping_folio() is one case, and while
>>> that may be doable, the inode i_lock is not IRQ safe.
>>>
>>> Most file systems have a need to punt some writeback completions to
>>> non-irq context, eg for file extending etc. Hence for most file 
>>> systems,
>>> the dontcache case just becomes another case that needs to go through
>>> that path.
>>>
>>> It'd certainly be possible to improve upon this, for example by having
>>> an opportunistic dontcache unmap from IRQ/soft-irq context, and then
>>> punting to a workqueue if that doesn't pan out. But this doesn't exist
>>> as of yet, hence the need for the workqueue punt.
>
> Thanks Jens for the detailed explanation.
>
>>
>> Hi, Jens
>> Thank you for your response. I tested uncached buffer I/O reads with
>> a 50GB dataset on a local F2FS filesystem, and the page cache size
>> only increased slightly, which I believe aligns with expectations.
>> After clearing the page cache, the page cache size returned to its
>> initial state. The test results are as follows:
>>
>> stat 50G.txt
>>     File: 50G.txt
>>     Size: 53687091200      Blocks: 104960712       IO Blocks: 512  
>> regular file
>>
>> [read before]:
>> echo 3 > /proc/sys/vm/drop_caches
>> 01:48:17        kbmemfree kbavail     kbmemused  %memused 
>> kbbuffers kbcached   kbcommit     %commit   kbactive kbinact     kbdirty
>> 01:50:59      6404648   8149508   2719384   23.40     512 1898092   
>> 199384760    823.75   1846756    466832     44
>>
>> ./uncached_io_test 8192 1 1 50G.txt
>> Starting 1 threads
>> reading bs 8192, uncached 1
>>     1s: 754MB/sec, MB=754
>>     ...
>>    64s: 844MB/sec, MB=262144
>>
>> [read after]:
>> 01:52:33      6326664   8121240   2747968    23.65      728 1947656   
>> 199384788    823.75   1887896    502004     68
>> echo 3 > /proc/sys/vm/drop_caches
>> 01:53:11      6351136   8096936   2772400   23.86     512 1900500   
>> 199385216    823.75   1847252    533768      104
>>
>> Hi Chao,
>> Given that F2FS currently calls folio_end_writeback in the softirq
>> context for normal write scenarios, could we first support uncached
>> buffer I/O reads? For normal uncached buffer I/O writes, would it be
>> feasible for F2FS to introduce an asynchronous workqueue to handle the
>> page drop operation in the future? What are your thoughts on this?
>
> Qi,
>
> Sorry for the delay.
>
> I think it will be good to support uncached buffered I/O in read path
> first, and then let's take a look what we can do for write path, anyway,
> let's do this step by step.
>
> Can you please update the patch?
> - support read path only
> - include test data in commit message
Chao

I will re-submit a patch to first enable F2FS support for uncached
buffer I/O reads. Following that, I will work on implementing
asynchronous page dropping in F2FS.

Thank you!
>
>> Thank you!
>>
>>

[f2fs-dev] [PATCH v2] mkfs.f2fs: support -C [no]hashonly to control linear lookup fallback

[Chao Yu <ch...@ke...>]

It provides a way to disable linear lookup fallback during mkfs.

Behavior summary:
			Android		Distro
By default		disabled	enabled
Tune w/ [no]hashonly	no		yes

Android case:

1.1) Disable linear lookup:
mkfs.f2fs -f -g android -O casefold -C utf8:hashonly /dev/vdb
dump.f2fs -d3 /dev/vdb |grep s_encoding_flags
s_encoding_flags                        [0x       2 : 2]

1.2) Enable linear lookup:
mkfs.f2fs -f -g android -O casefold -C utf8:nohashonly /dev/vdb
dump.f2fs -d3 /dev/vdb |grep s_encoding_flags
s_encoding_flags                        [0x       2 : 2]

1.3) By default:
mkfs.f2fs -f -g android -O casefold -C utf8 /dev/vdb
dump.f2fs -d3 /dev/vdb |grep s_encoding_flags
s_encoding_flags                        [0x       2 : 2]

Distro case:

2.1) Disable linear lookup:
mkfs.f2fs -f -O casefold -C utf8:hashonly /dev/vdb
dump.f2fs -d3 /dev/vdb |grep s_encoding_flags
s_encoding_flags                        [0x       2 : 2]

2.2) Enable linear lookup:
mkfs.f2fs -f -O casefold -C utf8:nohashonly /dev/vdb
dump.f2fs -d3 /dev/vdb |grep s_encoding_flags
s_encoding_flags                        [0x       0 : 0]

2.3) By default:
mkfs.f2fs -f -O casefold -C utf8 /dev/vdb
dump.f2fs -d3 /dev/vdb |grep s_encoding_flags
s_encoding_flags                        [0x       0 : 0]

Signed-off-by: Chao Yu <ch...@ke...>
---
v2:
- disable linear lookup by default for Android case
 include/f2fs_fs.h       |  3 ++-
 lib/libf2fs.c           |  1 +
 man/mkfs.f2fs.8         | 10 ++++++++--
 mkfs/f2fs_format.c      |  3 +++
 mkfs/f2fs_format_main.c |  3 ++-
 5 files changed, 16 insertions(+), 4 deletions(-)

diff --git a/include/f2fs_fs.h b/include/f2fs_fs.h
index f7268d1..a8da8fa 100644
--- a/include/f2fs_fs.h
+++ b/include/f2fs_fs.h
@@ -1478,7 +1478,8 @@ enum {
 
 /* feature list in Android */
 enum {
-	F2FS_FEATURE_NAT_BITS = 0x0001,
+	F2FS_FEATURE_NAT_BITS		= 0x0001,
+	F2FS_FEATURE_LINEAR_LOOKUP	= 0x0002,
 };
 
 /* nolinear lookup tune */
diff --git a/lib/libf2fs.c b/lib/libf2fs.c
index 2f012c8..0e3e62a 100644
--- a/lib/libf2fs.c
+++ b/lib/libf2fs.c
@@ -1424,6 +1424,7 @@ static const struct enc_flags {
 	char *param;
 } encoding_flags[] = {
 	{ F2FS_ENC_STRICT_MODE_FL, "strict" },
+	{ F2FS_ENC_NO_COMPAT_FALLBACK_FL, "hashonly"}
 };
 
 /* Return a positive number < 0xff indicating the encoding magic number
diff --git a/man/mkfs.f2fs.8 b/man/mkfs.f2fs.8
index 8b3b0cc..8cb7d32 100644
--- a/man/mkfs.f2fs.8
+++ b/man/mkfs.f2fs.8
@@ -232,9 +232,15 @@ Use UTF-8 for casefolding.
 .I flags:
 .RS 1.2i
 .TP 1.2i
-.B strict
+.B [no]strict
 This flag specifies that invalid strings should be rejected by the filesystem.
-Default is disabled.
+For android case, it will disable linear lookup by default.
+.RE
+.RS 1.2i
+.TP 1.2i
+.B [no]hashonly
+This flag specifies that there is no linear lookup fallback during lookup.
+By default, linear lookup fallback is enabled.
 .RE
 .RE
 .TP
diff --git a/mkfs/f2fs_format.c b/mkfs/f2fs_format.c
index 2680bd3..a45bbcb 100644
--- a/mkfs/f2fs_format.c
+++ b/mkfs/f2fs_format.c
@@ -671,6 +671,9 @@ static int f2fs_prepare_super_block(void)
 	memcpy(sb->init_version, c.version, VERSION_LEN);
 
 	if (c.feature & F2FS_FEATURE_CASEFOLD) {
+		if (!(c.s_encoding_flags & F2FS_ENC_NO_COMPAT_FALLBACK_FL) &&
+			(c.disabled_feature & F2FS_FEATURE_LINEAR_LOOKUP))
+			c.s_encoding_flags |= F2FS_ENC_NO_COMPAT_FALLBACK_FL;
 		set_sb(s_encoding, c.s_encoding);
 		set_sb(s_encoding_flags, c.s_encoding_flags);
 	}
diff --git a/mkfs/f2fs_format_main.c b/mkfs/f2fs_format_main.c
index f0bec4f..8f8e975 100644
--- a/mkfs/f2fs_format_main.c
+++ b/mkfs/f2fs_format_main.c
@@ -143,7 +143,8 @@ static void add_default_options(void)
 		force_overwrite = 1;
 		c.wanted_sector_size = F2FS_BLKSIZE;
 		c.root_uid = c.root_gid = 0;
-		c.disabled_feature |= F2FS_FEATURE_NAT_BITS;
+		c.disabled_feature |= F2FS_FEATURE_NAT_BITS |
+					F2FS_FEATURE_LINEAR_LOOKUP;
 
 		/* RO doesn't need any other features */
 		if (c.feature & F2FS_FEATURE_RO)
-- 
2.49.0

[f2fs-dev] [PATCH] mkfs.f2fs: Fix zoned alignment check for multi-device setups

[Daniel L. <ch...@go...>]

Commit 84447ee7212e correctly relocated the zone alignment check to its
proper location. However, this revealed that the original check's
condition was incorrect for multi-device setups.

This patch corrects the logic to check the alignment relative to the start
of the segment0 and improves the error messages.

Signed-off-by: Daniel Lee <ch...@go...>
---
 mkfs/f2fs_format.c | 38 +++++++++++++++++++++++++++-----------
 1 file changed, 27 insertions(+), 11 deletions(-)

diff --git a/mkfs/f2fs_format.c b/mkfs/f2fs_format.c
index 2680bd3..18f6e34 100644
--- a/mkfs/f2fs_format.c
+++ b/mkfs/f2fs_format.c
@@ -339,17 +339,6 @@ static int f2fs_prepare_super_block(void)
 	MSG(0, "Info: zone aligned segment0 blkaddr: %u\n",
 					get_sb(segment0_blkaddr));
 
-	if (c.zoned_mode &&
-		((c.ndevs == 1 &&
-			(get_sb(segment0_blkaddr) + c.start_sector /
-			DEFAULT_SECTORS_PER_BLOCK) % c.zone_blocks) ||
-		(c.ndevs > 1 &&
-			c.devices[1].start_blkaddr % c.zone_blocks))) {
-		MSG(1, "\tError: Unaligned segment0 block address %u\n",
-				get_sb(segment0_blkaddr));
-		return -1;
-	}
-
 	for (i = 0; i < c.ndevs; i++) {
 		if (i == 0) {
 			c.devices[i].total_segments =
@@ -390,6 +379,33 @@ static int f2fs_prepare_super_block(void)
 
 		c.total_segments += c.devices[i].total_segments;
 	}
+
+	if (c.zoned_mode) {
+		if (c.ndevs == 1 &&
+			(get_sb(segment0_blkaddr) + c.start_sector /
+			DEFAULT_SECTORS_PER_BLOCK) % c.zone_blocks) {
+			/*
+			 * With a sole zoned LU, segment0 start should be
+			 * aligned at the zone.
+			 */
+			MSG(1, "\tError: Unaligned segment0 start (%u) for zoned LU (zone_blocks: %lu)\n",
+				get_sb(segment0_blkaddr), c.zone_blocks);
+			return -1;
+		} else if (c.ndevs > 1 &&
+			(c.devices[1].start_blkaddr - get_sb(segment0_blkaddr)) % c.zone_blocks) {
+			/*
+			 * With the first device as a conventional LU and the
+			 * second as a zoned LU, the start address for the zoned
+			 * LU should be aligned to the zone size, starting from
+			 * segment0.
+			 */
+			MSG(1, "\tError: Unaligned start (%lu) for zoned LU from segment0 (%u) (zone_blocks: %lu)\n",
+				c.devices[1].start_blkaddr,
+				get_sb(segment0_blkaddr), c.zone_blocks);
+			return -1;
+		}
+	}
+
 	set_sb(segment_count, c.total_segments);
 	set_sb(segment_count_ckpt, F2FS_NUMBER_OF_CHECKPOINT_PACK);
 
-- 
2.50.1.552.g942d659e1b-goog

Re: [f2fs-dev] [PATCH v2] f2fs: vm_unmap_ram() may be called from an invalid context

[<pat...@ke...>]

Hello:

This patch was applied to jaegeuk/f2fs.git (dev)
by Jaegeuk Kim <ja...@ke...>:

On Thu, 24 Jul 2025 17:31:15 +0200 you wrote:
> When testing F2FS with xfstests using UFS backed virtual disks the
> kernel complains sometimes that f2fs_release_decomp_mem() calls
> vm_unmap_ram() from an invalid context. Example trace from
> f2fs/007 test:
> 
> f2fs/007 5s ...  [12:59:38][    8.902525] run fstests f2fs/007
> [   11.468026] BUG: sleeping function called from invalid context at mm/vmalloc.c:2978
> [   11.471849] in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 68, name: irq/22-ufshcd
> [   11.475357] preempt_count: 1, expected: 0
> [   11.476970] RCU nest depth: 0, expected: 0
> [   11.478531] CPU: 0 UID: 0 PID: 68 Comm: irq/22-ufshcd Tainted: G        W           6.16.0-rc5-xfstests-ufs-g40f92e79b0aa #9 PREEMPT(none)
> [   11.478535] Tainted: [W]=WARN
> [   11.478536] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2 04/01/2014
> [   11.478537] Call Trace:
> [   11.478543]  <TASK>
> [   11.478545]  dump_stack_lvl+0x4e/0x70
> [   11.478554]  __might_resched.cold+0xaf/0xbe
> [   11.478557]  vm_unmap_ram+0x21/0xb0
> [   11.478560]  f2fs_release_decomp_mem+0x59/0x80
> [   11.478563]  f2fs_free_dic+0x18/0x1a0
> [   11.478565]  f2fs_finish_read_bio+0xd7/0x290
> [   11.478570]  blk_update_request+0xec/0x3b0
> [   11.478574]  ? sbitmap_queue_clear+0x3b/0x60
> [   11.478576]  scsi_end_request+0x27/0x1a0
> [   11.478582]  scsi_io_completion+0x40/0x300
> [   11.478583]  ufshcd_mcq_poll_cqe_lock+0xa3/0xe0
> [   11.478588]  ufshcd_sl_intr+0x194/0x1f0
> [   11.478592]  ufshcd_threaded_intr+0x68/0xb0
> [   11.478594]  ? __pfx_irq_thread_fn+0x10/0x10
> [   11.478599]  irq_thread_fn+0x20/0x60
> [   11.478602]  ? __pfx_irq_thread_fn+0x10/0x10
> [   11.478603]  irq_thread+0xb9/0x180
> [   11.478605]  ? __pfx_irq_thread_dtor+0x10/0x10
> [   11.478607]  ? __pfx_irq_thread+0x10/0x10
> [   11.478609]  kthread+0x10a/0x230
> [   11.478614]  ? __pfx_kthread+0x10/0x10
> [   11.478615]  ret_from_fork+0x7e/0xd0
> [   11.478619]  ? __pfx_kthread+0x10/0x10
> [   11.478621]  ret_from_fork_asm+0x1a/0x30
> [   11.478623]  </TASK>
> 
> [...]

Here is the summary with links:
  - [f2fs-dev,v2] f2fs: vm_unmap_ram() may be called from an invalid context
    https://git.kernel.org/jaegeuk/f2fs/c/08a7efc5b02a

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

Re: [f2fs-dev] [PATCH] f2fs: ignore valid ratio when free section count is low

[<pat...@ke...>]

Hello:

This patch was applied to jaegeuk/f2fs.git (dev)
by Jaegeuk Kim <ja...@ke...>:

On Fri, 18 Jul 2025 15:04:31 -0700 you wrote:
> From: Daeho Jeong <dae...@go...>
> 
> Otherwise F2FS will not do GC in background in low free section.
> 
> Signed-off-by: Daeho Jeong <dae...@go...>
> ---
>  fs/f2fs/gc.c | 18 ++++++++++++------
>  1 file changed, 12 insertions(+), 6 deletions(-)

Here is the summary with links:
  - [f2fs-dev] f2fs: ignore valid ratio when free section count is low
    https://git.kernel.org/jaegeuk/f2fs/c/e6d5e789c3b2

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

Re: [f2fs-dev] [PATCH] f2fs: merge the two conditions to avoid code duplication

[<pat...@ke...>]

Hello:

This patch was applied to jaegeuk/f2fs.git (dev)
by Jaegeuk Kim <ja...@ke...>:

On Wed, 23 Jul 2025 22:58:37 +0800 you wrote:
> No functional changes.
> 
> Signed-off-by: mason.zhang <mas...@gm...>
> ---
>  fs/f2fs/gc.c | 7 +------
>  1 file changed, 1 insertion(+), 6 deletions(-)

Here is the summary with links:
  - [f2fs-dev] f2fs: merge the two conditions to avoid code duplication
    https://git.kernel.org/jaegeuk/f2fs/c/b93bf64e349b

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

Re: [f2fs-dev] [PATCH] f2fs: remove unnecessary tracepoint enabled check

[<pat...@ke...>]

Hello:

This patch was applied to jaegeuk/f2fs.git (dev)
by Jaegeuk Kim <ja...@ke...>:

On Wed, 23 Jul 2025 22:24:56 +0800 you wrote:
> From: Sheng Yong <she...@xi...>
> 
> There is no extra work before trace_f2fs_[dataread|datawrite]_end(),
> so there is no need to check trace_<tracepoint>_enabled().
> 
> Signed-off-by: Sheng Yong <she...@xi...>
> 
> [...]

Here is the summary with links:
  - [f2fs-dev] f2fs: remove unnecessary tracepoint enabled check
    https://git.kernel.org/jaegeuk/f2fs/c/95d7c508b212

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

[f2fs-dev] Patchwork summary for: f2fs

[<pat...@ke...>]

Hello:

The following patches were marked "accepted", because they were applied to
jaegeuk/f2fs.git (dev):

Patch: [f2fs-dev] f2fs: remove unnecessary tracepoint enabled check
  Submitter: Sheng Yong <she...@gm...>
  Committer: Jaegeuk Kim <ja...@ke...>
  Patchwork: https://patchwork.kernel.org/project/f2fs/list/?series=985162
  Lore link: https://lore.kernel.org/r/202...@xi...

Patch: [f2fs-dev] f2fs: don't break allocation when crossing contiguous sections
  Submitter: Chao Yu <ch...@ke...>
  Committer: Jaegeuk Kim <ja...@ke...>
  Patchwork: https://patchwork.kernel.org/project/f2fs/list/?series=984143
  Lore link: https://lore.kernel.org/r/202...@ke...

Patch: [f2fs-dev] f2fs: merge the two conditions to avoid code duplication
  Submitter: mason.zhang <mas...@gm...>
  Committer: Jaegeuk Kim <ja...@ke...>
  Patchwork: https://patchwork.kernel.org/project/f2fs/list/?series=985175
  Lore link: https://lore.kernel.org/r/202...@gm...

Patch: [f2fs-dev,v3] f2fs: zone: wait for inflight dio completion, excluding pinned files read using dio
  Submitter: yohan.joung <yoh...@sk...>
  Committer: Jaegeuk Kim <ja...@ke...>
  Patchwork: https://patchwork.kernel.org/project/f2fs/list/?series=984548
  Lore link: https://lore.kernel.org/r/202...@sk...

Patch: [f2fs-dev] f2fs: ignore valid ratio when free section count is low
  Submitter: Daeho Jeong <da...@gm...>
  Committer: Jaegeuk Kim <ja...@ke...>
  Patchwork: https://patchwork.kernel.org/project/f2fs/list/?series=983893
  Lore link: https://lore.kernel.org/r/202...@gm...

Patch: [f2fs-dev,v2] f2fs: vm_unmap_ram() may be called from an invalid context
  Submitter: Jan Prusakowski <jpr...@go...>
  Committer: Jaegeuk Kim <ja...@ke...>
  Patchwork: https://patchwork.kernel.org/project/f2fs/list/?series=985638
  Lore link: https://lore.kernel.org/r/202...@go...


Total patches: 6

-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

Re: [f2fs-dev] [PATCH v3] f2fs: zone: wait for inflight dio completion, excluding pinned files read using dio

[<pat...@ke...>]

Hello:

This patch was applied to jaegeuk/f2fs.git (dev)
by Jaegeuk Kim <ja...@ke...>:

On Tue, 22 Jul 2025 15:02:40 +0900 you wrote:
> read for the pinfile using Direct I/O do not wait for dio write.
> 
> Signed-off-by: yohan.joung <yoh...@sk...>
> ---
>  fs/f2fs/file.c | 8 ++++++--
>  1 file changed, 6 insertions(+), 2 deletions(-)

Here is the summary with links:
  - [f2fs-dev,v3] f2fs: zone: wait for inflight dio completion, excluding pinned files read using dio
    https://git.kernel.org/jaegeuk/f2fs/c/3bf1bab503a5

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

Re: [f2fs-dev] [PATCH] f2fs: don't break allocation when crossing contiguous sections

[<pat...@ke...>]

Hello:

This patch was applied to jaegeuk/f2fs.git (dev)
by Jaegeuk Kim <ja...@ke...>:

On Mon, 21 Jul 2025 10:02:31 +0800 you wrote:
> Commit 0638a3197c19 ("f2fs: avoid unused block when dio write in LFS
> mode") has fixed unused block issue for dio write in lfs mode.
> 
> However, f2fs_map_blocks() may break and return smaller extent when
> last allocated block locates in the end of section, even allocator
> can allocate contiguous blocks across sections.
> 
> [...]

Here is the summary with links:
  - [f2fs-dev] f2fs: don't break allocation when crossing contiguous sections
    https://git.kernel.org/jaegeuk/f2fs/c/f0a7adfedcc8

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

[f2fs-dev] [PATCH] f2fs: add error checking in do_write_page()

[mason.zhang <mas...@gm...>]

Otherwise, the filesystem may unaware of potential file corruption.

Signed-off-by: mason.zhang <mas...@gm...>
---
 fs/f2fs/segment.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index ae1223ef648f..ce80ba226aed 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -3936,12 +3936,15 @@ static void do_write_page(struct f2fs_summary *sum, struct f2fs_io_info *fio)
 	int seg_type = log_type_to_seg_type(type);
 	bool keep_order = (f2fs_lfs_mode(fio->sbi) &&
 				seg_type == CURSEG_COLD_DATA);
+	int err;
 
 	if (keep_order)
 		f2fs_down_read(&fio->sbi->io_order_lock);
 
-	if (f2fs_allocate_data_block(fio->sbi, fio->page, fio->old_blkaddr,
-			&fio->new_blkaddr, sum, type, fio)) {
+	err = f2fs_allocate_data_block(fio->sbi, fio->page, fio->old_blkaddr,
+			&fio->new_blkaddr, sum, type, fio);
+	if (unlikely(err)) {
+		f2fs_err(fio->sbi, "Failed to allocate data block(%d)", err);
 		if (fscrypt_inode_uses_fs_layer_crypto(folio->mapping->host))
 			fscrypt_finalize_bounce_page(&fio->encrypted_page);
 		folio_end_writeback(folio);
-- 
2.50.0

Re: [f2fs-dev] [PATCH] mkfs.f2fs: support -C [no]hashonly to control linear lookup fallback

[Chao Yu <ch...@ke...>]

On 7/24/2025 11:06 PM, Jaegeuk Kim wrote:
> Can we disable linear lookup by default, but add an option to enable it?

Sure.

> 
> On 07/24, Chao Yu wrote:
>> It provides a way to disable linear lookup fallback during mkfs.
>>
>> Usage:
>>
>> 1. Disable linear lookup:
>> mkfs.f2fs -f -O casefold -C utf8:hashonly /dev/vdb
>> dump.f2fs -d3 /dev/vdb |grep s_encoding_flags
>> s_encoding_flags                        [0x       2 : 2]
>>
>> 2. Enable linear lookup:
>> mkfs.f2fs -f -O casefold -C utf8:nohashonly /dev/vdb
>> dump.f2fs -d3 /dev/vdb |grep s_encoding_flags
>> s_encoding_flags                        [0x       0 : 0]
>>
>> mkfs.f2fs -f -O casefold -C utf8 /dev/vdb
>> dump.f2fs -d3 /dev/vdb |grep s_encoding_flags
>> s_encoding_flags                        [0x       0 : 0]

Let me update the logic to disable linear lookup by default for case 2.2).

1)  mkfs.f2fs -f -O casefold -C utf8:hashonly /dev/vdb
2.1) mkfs.f2fs -f -O casefold -C utf8:nohashonly /dev/vdb
2.2) mkfs.f2fs -f -O casefold -C utf8 /dev/vdb

Thanks,

>>
>> Signed-off-by: Chao Yu <ch...@ke...>
>> ---
>>   lib/libf2fs.c   | 1 +
>>   man/mkfs.f2fs.8 | 8 +++++++-
>>   2 files changed, 8 insertions(+), 1 deletion(-)
>>
>> diff --git a/lib/libf2fs.c b/lib/libf2fs.c
>> index 2f012c8..0e3e62a 100644
>> --- a/lib/libf2fs.c
>> +++ b/lib/libf2fs.c
>> @@ -1424,6 +1424,7 @@ static const struct enc_flags {
>>   	char *param;
>>   } encoding_flags[] = {
>>   	{ F2FS_ENC_STRICT_MODE_FL, "strict" },
>> +	{ F2FS_ENC_NO_COMPAT_FALLBACK_FL, "hashonly"}
>>   };
>>   
>>   /* Return a positive number < 0xff indicating the encoding magic number
>> diff --git a/man/mkfs.f2fs.8 b/man/mkfs.f2fs.8
>> index 8b3b0cc..b57f033 100644
>> --- a/man/mkfs.f2fs.8
>> +++ b/man/mkfs.f2fs.8
>> @@ -232,10 +232,16 @@ Use UTF-8 for casefolding.
>>   .I flags:
>>   .RS 1.2i
>>   .TP 1.2i
>> -.B strict
>> +.B [no]strict
>>   This flag specifies that invalid strings should be rejected by the filesystem.
>>   Default is disabled.
>>   .RE
>> +.RS 1.2i
>> +.TP 1.2i
>> +.B [no]hashonly
>> +This flag specifies that there is no linear lookup fallback during lookup.
>> +By default, linear lookup fallback is enabled.
>> +.RE
>>   .RE
>>   .TP
>>   .BI \-q
>> -- 
>> 2.49.0

[f2fs-dev] [PATCH v2] f2fs: vm_unmap_ram() may be called from an invalid context

[Jan P. <jpr...@go...>]

When testing F2FS with xfstests using UFS backed virtual disks the
kernel complains sometimes that f2fs_release_decomp_mem() calls
vm_unmap_ram() from an invalid context. Example trace from
f2fs/007 test:

f2fs/007 5s ...  [12:59:38][    8.902525] run fstests f2fs/007
[   11.468026] BUG: sleeping function called from invalid context at mm/vmalloc.c:2978
[   11.471849] in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 68, name: irq/22-ufshcd
[   11.475357] preempt_count: 1, expected: 0
[   11.476970] RCU nest depth: 0, expected: 0
[   11.478531] CPU: 0 UID: 0 PID: 68 Comm: irq/22-ufshcd Tainted: G        W           6.16.0-rc5-xfstests-ufs-g40f92e79b0aa #9 PREEMPT(none)
[   11.478535] Tainted: [W]=WARN
[   11.478536] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2 04/01/2014
[   11.478537] Call Trace:
[   11.478543]  <TASK>
[   11.478545]  dump_stack_lvl+0x4e/0x70
[   11.478554]  __might_resched.cold+0xaf/0xbe
[   11.478557]  vm_unmap_ram+0x21/0xb0
[   11.478560]  f2fs_release_decomp_mem+0x59/0x80
[   11.478563]  f2fs_free_dic+0x18/0x1a0
[   11.478565]  f2fs_finish_read_bio+0xd7/0x290
[   11.478570]  blk_update_request+0xec/0x3b0
[   11.478574]  ? sbitmap_queue_clear+0x3b/0x60
[   11.478576]  scsi_end_request+0x27/0x1a0
[   11.478582]  scsi_io_completion+0x40/0x300
[   11.478583]  ufshcd_mcq_poll_cqe_lock+0xa3/0xe0
[   11.478588]  ufshcd_sl_intr+0x194/0x1f0
[   11.478592]  ufshcd_threaded_intr+0x68/0xb0
[   11.478594]  ? __pfx_irq_thread_fn+0x10/0x10
[   11.478599]  irq_thread_fn+0x20/0x60
[   11.478602]  ? __pfx_irq_thread_fn+0x10/0x10
[   11.478603]  irq_thread+0xb9/0x180
[   11.478605]  ? __pfx_irq_thread_dtor+0x10/0x10
[   11.478607]  ? __pfx_irq_thread+0x10/0x10
[   11.478609]  kthread+0x10a/0x230
[   11.478614]  ? __pfx_kthread+0x10/0x10
[   11.478615]  ret_from_fork+0x7e/0xd0
[   11.478619]  ? __pfx_kthread+0x10/0x10
[   11.478621]  ret_from_fork_asm+0x1a/0x30
[   11.478623]  </TASK>

This patch modifies in_task() check inside f2fs_read_end_io() to also
check if interrupts are disabled. This ensures that pages are unmapped
asynchronously in an interrupt handler.

Fixes: bff139b49d9f ("f2fs: handle decompress only post processing in softirq")

Signed-off-by: Jan Prusakowski <jpr...@go...>

Reviewed-by: Chao Yu <ch...@ke...>
---
 fs/f2fs/data.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index d1a2616d41be..0acc25f996b3 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -280,7 +280,7 @@ static void f2fs_read_end_io(struct bio *bio)
 {
 	struct f2fs_sb_info *sbi = F2FS_F_SB(bio_first_folio_all(bio));
 	struct bio_post_read_ctx *ctx;
-	bool intask = in_task();
+	bool intask = in_task() && !irqs_disabled();
 
 	iostat_update_and_unbind_ctx(bio);
 	ctx = bio->bi_private;
-- 
2.50.1.487.gc89ff58d15-goog

Re: [f2fs-dev] [PATCH] mkfs.f2fs: support -C [no]hashonly to control linear lookup fallback

[Jaegeuk K. <ja...@ke...>]

Can we disable linear lookup by default, but add an option to enable it?

On 07/24, Chao Yu wrote:
> It provides a way to disable linear lookup fallback during mkfs.
> 
> Usage:
> 
> 1. Disable linear lookup:
> mkfs.f2fs -f -O casefold -C utf8:hashonly /dev/vdb
> dump.f2fs -d3 /dev/vdb |grep s_encoding_flags
> s_encoding_flags                        [0x       2 : 2]
> 
> 2. Enable linear lookup:
> mkfs.f2fs -f -O casefold -C utf8:nohashonly /dev/vdb
> dump.f2fs -d3 /dev/vdb |grep s_encoding_flags
> s_encoding_flags                        [0x       0 : 0]
> 
> mkfs.f2fs -f -O casefold -C utf8 /dev/vdb
> dump.f2fs -d3 /dev/vdb |grep s_encoding_flags
> s_encoding_flags                        [0x       0 : 0]
> 
> Signed-off-by: Chao Yu <ch...@ke...>
> ---
>  lib/libf2fs.c   | 1 +
>  man/mkfs.f2fs.8 | 8 +++++++-
>  2 files changed, 8 insertions(+), 1 deletion(-)
> 
> diff --git a/lib/libf2fs.c b/lib/libf2fs.c
> index 2f012c8..0e3e62a 100644
> --- a/lib/libf2fs.c
> +++ b/lib/libf2fs.c
> @@ -1424,6 +1424,7 @@ static const struct enc_flags {
>  	char *param;
>  } encoding_flags[] = {
>  	{ F2FS_ENC_STRICT_MODE_FL, "strict" },
> +	{ F2FS_ENC_NO_COMPAT_FALLBACK_FL, "hashonly"}
>  };
>  
>  /* Return a positive number < 0xff indicating the encoding magic number
> diff --git a/man/mkfs.f2fs.8 b/man/mkfs.f2fs.8
> index 8b3b0cc..b57f033 100644
> --- a/man/mkfs.f2fs.8
> +++ b/man/mkfs.f2fs.8
> @@ -232,10 +232,16 @@ Use UTF-8 for casefolding.
>  .I flags:
>  .RS 1.2i
>  .TP 1.2i
> -.B strict
> +.B [no]strict
>  This flag specifies that invalid strings should be rejected by the filesystem.
>  Default is disabled.
>  .RE
> +.RS 1.2i
> +.TP 1.2i
> +.B [no]hashonly
> +This flag specifies that there is no linear lookup fallback during lookup.
> +By default, linear lookup fallback is enabled.
> +.RE
>  .RE
>  .TP
>  .BI \-q
> -- 
> 2.49.0

Re: [f2fs-dev] [PATCH] f2fs: f2fs supports uncached buffered I/O

[Chao Yu <ch...@ke...>]

On 2025/7/16 16:27, hanqi wrote:
> 
> 
> 在 2025/7/16 11:43, Jens Axboe 写道:
>> On 7/15/25 9:34 PM, hanqi wrote:
>>>
>>> ? 2025/7/15 22:28, Jens Axboe ??:
>>>> On 7/14/25 9:10 PM, Qi Han wrote:
>>>>> Jens has already completed the development of uncached buffered I/O
>>>>> in git [1], and in f2fs, the feature can be enabled simply by setting
>>>>> the FOP_DONTCACHE flag in f2fs_file_operations.
>>>> You need to ensure that for any DONTCACHE IO that the completion is
>>>> routed via non-irq context, if applicable. I didn't verify that this is
>>>> the case for f2fs. Generally you can deduce this as well through
>>>> testing, I'd say the following cases would be interesting to test:
>>>>
>>>> 1) Normal DONTCACHE buffered read
>>>> 2) Overwrite DONTCACHE buffered write
>>>> 3) Append DONTCACHE buffered write
>>>>
>>>> Test those with DEBUG_ATOMIC_SLEEP set in your config, and it that
>>>> doesn't complain, that's a great start.
>>>>
>>>> For the above test cases as well, verify that page cache doesn't grow as
>>>> IO is performed. A bit is fine for things like meta data, but generally
>>>> you want to see it remain basically flat in terms of page cache usage.
>>>>
>>>> Maybe this is all fine, like I said I didn't verify. Just mentioning it
>>>> for completeness sake.
>>> Hi, Jens
>>> Thanks for your suggestion. As I mentioned earlier in [1], in f2fs,
>>> the regular buffered write path invokes folio_end_writeback from a
>>> softirq context. Therefore, it seems that f2fs may not be suitable
>>> for DONTCACHE I/O writes.
>>>
>>> I?d like to ask a question: why is DONTCACHE I/O write restricted to
>>> non-interrupt context only? Is it because dropping the page might be
>>> too time-consuming to be done safely in interrupt context? This might
>>> be a naive question, but I?d really appreciate your clarification.
>>> Thanks in advance.
>> Because (as of right now, at least) the code doing the invalidation
>> needs process context. There are various reasons for this, which you'll
>> see if you follow the path off folio_end_writeback() ->
>> filemap_end_dropbehind_write() -> filemap_end_dropbehind() ->
>> folio_unmap_invalidate(). unmap_mapping_folio() is one case, and while
>> that may be doable, the inode i_lock is not IRQ safe.
>>
>> Most file systems have a need to punt some writeback completions to
>> non-irq context, eg for file extending etc. Hence for most file systems,
>> the dontcache case just becomes another case that needs to go through
>> that path.
>>
>> It'd certainly be possible to improve upon this, for example by having
>> an opportunistic dontcache unmap from IRQ/soft-irq context, and then
>> punting to a workqueue if that doesn't pan out. But this doesn't exist
>> as of yet, hence the need for the workqueue punt.

Thanks Jens for the detailed explanation.

> 
> Hi, Jens
> Thank you for your response. I tested uncached buffer I/O reads with
> a 50GB dataset on a local F2FS filesystem, and the page cache size
> only increased slightly, which I believe aligns with expectations.
> After clearing the page cache, the page cache size returned to its
> initial state. The test results are as follows:
> 
> stat 50G.txt
>     File: 50G.txt
>     Size: 53687091200      Blocks: 104960712       IO Blocks: 512  regular file
> 
> [read before]:
> echo 3 > /proc/sys/vm/drop_caches
> 01:48:17        kbmemfree kbavail     kbmemused  %memused      kbbuffers kbcached   kbcommit     %commit   kbactive    kbinact     kbdirty
> 01:50:59      6404648   8149508   2719384   23.40     512     1898092   199384760    823.75   1846756    466832     44
> 
> ./uncached_io_test 8192 1 1 50G.txt
> Starting 1 threads
> reading bs 8192, uncached 1
>     1s: 754MB/sec, MB=754
>     ...
>    64s: 844MB/sec, MB=262144
> 
> [read after]:
> 01:52:33      6326664   8121240   2747968    23.65      728     1947656   199384788    823.75   1887896    502004     68
> echo 3 > /proc/sys/vm/drop_caches
> 01:53:11      6351136   8096936   2772400   23.86     512     1900500   199385216    823.75   1847252    533768      104
> 
> Hi Chao,
> Given that F2FS currently calls folio_end_writeback in the softirq
> context for normal write scenarios, could we first support uncached
> buffer I/O reads? For normal uncached buffer I/O writes, would it be
> feasible for F2FS to introduce an asynchronous workqueue to handle the
> page drop operation in the future? What are your thoughts on this?

Qi,

Sorry for the delay.

I think it will be good to support uncached buffered I/O in read path
first, and then let's take a look what we can do for write path, anyway,
let's do this step by step.

Can you please update the patch?
- support read path only
- include test data in commit message

> Thank you!
> 
> 

Re: [f2fs-dev] [PATCH] f2fs: vm_unmap_ram() may be called from an invalid context

[Chao Yu <ch...@ke...>]

Jan,

On 2025/7/24 18:30, Jan Prusakowski via Linux-f2fs-devel wrote:
> When testing F2FS with xfstests using UFS backed virtual disks the
> kernel complains sometimes that f2fs_release_decomp_mem() calls
> vm_unmap_ram() from an invalid context.

It will be better to include testcase & crash information in commit
message.

> 
> This patch modifies in_task() check inside f2fs_read_end_io() to also
> check if interrupts are disabled. This ensures that pages are unmapped
> asynchronously in an interrupt handler.
> 
> Fixes: bff139b (f2fs: handle decompress only post processing in softirq)

./scripts/checkpatch.pl will output below warning message:

WARNING: Please use correct Fixes: style 'Fixes: <12+ chars of sha1> ("<title line>")' - ie: 'Fixes: bff139b49d9f ("f2fs: handle decompress only post processing in softirq")'

Please use this instead:

Fixes: bff139b49d9f ("f2fs: handle decompress only post processing in softirq")

> 
> Signed-off-by: Jan Prusakowski <jpr...@go...>

Code part looks good to me, please feel free to add my RVB tag in next version:

Reviewed-by: Chao Yu <ch...@ke...>

Thanks,

> ---
>   fs/f2fs/data.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
> index d1a2616d41be..0acc25f996b3 100644
> --- a/fs/f2fs/data.c
> +++ b/fs/f2fs/data.c
> @@ -280,7 +280,7 @@ static void f2fs_read_end_io(struct bio *bio)
>   {
>   	struct f2fs_sb_info *sbi = F2FS_F_SB(bio_first_folio_all(bio));
>   	struct bio_post_read_ctx *ctx;
> -	bool intask = in_task();
> +	bool intask = in_task() && !irqs_disabled();
>   
>   	iostat_update_and_unbind_ctx(bio);
>   	ctx = bio->bi_private;

[f2fs-dev] [PATCH] f2fs: vm_unmap_ram() may be called from an invalid context

[Jan P. <jpr...@go...>]

When testing F2FS with xfstests using UFS backed virtual disks the
kernel complains sometimes that f2fs_release_decomp_mem() calls
vm_unmap_ram() from an invalid context.

This patch modifies in_task() check inside f2fs_read_end_io() to also
check if interrupts are disabled. This ensures that pages are unmapped
asynchronously in an interrupt handler.

Fixes: bff139b (f2fs: handle decompress only post processing in softirq)

Signed-off-by: Jan Prusakowski <jpr...@go...>
---
 fs/f2fs/data.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index d1a2616d41be..0acc25f996b3 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -280,7 +280,7 @@ static void f2fs_read_end_io(struct bio *bio)
 {
 	struct f2fs_sb_info *sbi = F2FS_F_SB(bio_first_folio_all(bio));
 	struct bio_post_read_ctx *ctx;
-	bool intask = in_task();
+	bool intask = in_task() && !irqs_disabled();
 
 	iostat_update_and_unbind_ctx(bio);
 	ctx = bio->bi_private;
-- 
2.50.0.727.gbf7dc18ff4-goog

Re: [f2fs-dev] [STABLE 5.15+] f2fs: sysfs: add encoding_flags entry

[Lee J. <le...@ke...>]

On Wed, 23 Jul 2025, Greg Kroah-Hartman wrote:

> On Wed, Jul 23, 2025 at 09:26:39AM +0100, Lee Jones wrote:
> > On Wed, 02 Jul 2025, Greg Kroah-Hartman wrote:
> > 
> > > On Tue, Jun 24, 2025 at 11:00:39AM +0100, Lee Jones wrote:
> > > > On Wed, 16 Apr 2025, Chao Yu wrote:
> > > > 
> > > > > This patch adds a new sysfs entry /sys/fs/f2fs/<disk>/encoding_flags,
> > > > > it is a read-only entry to show the value of sb.s_encoding_flags, the
> > > > > value is hexadecimal.
> > > > > 
> > > > > ===========================      ==========
> > > > > Flag_Name                        Flag_Value
> > > > > ===========================      ==========
> > > > > SB_ENC_STRICT_MODE_FL            0x00000001
> > > > > SB_ENC_NO_COMPAT_FALLBACK_FL     0x00000002
> > > > > ===========================      ==========
> > > > > 
> > > > > case#1
> > > > > mkfs.f2fs -f -O casefold -C utf8:strict /dev/vda
> > > > > mount /dev/vda /mnt/f2fs
> > > > > cat /sys/fs/f2fs/vda/encoding_flags
> > > > > 1
> > > > > 
> > > > > case#2
> > > > > mkfs.f2fs -f -O casefold -C utf8 /dev/vda
> > > > > fsck.f2fs --nolinear-lookup=1 /dev/vda
> > > > > mount /dev/vda /mnt/f2fs
> > > > > cat /sys/fs/f2fs/vda/encoding_flags
> > > > > 2
> > > > > 
> > > > > Signed-off-by: Chao Yu <ch...@ke...>
> > > > > ---
> > > > >  Documentation/ABI/testing/sysfs-fs-f2fs | 13 +++++++++++++
> > > > >  fs/f2fs/sysfs.c                         |  9 +++++++++
> > > > >  2 files changed, 22 insertions(+)
> > > > 
> > > > This patch, commit 617e0491abe4 ("f2fs: sysfs: export linear_lookup in
> > > > features directory") upstream, needs to find its way into all Stable
> > > > branches containing upstream commit 91b587ba79e1 ("f2fs: Introduce
> > > > linear search for dentries"), which is essentially linux-5.15.y and
> > > > newer.
> > > > 
> > > > stable/linux-5.4.y:
> > > > MISSING:     f2fs: Introduce linear search for dentries
> > > > MISSING:     f2fs: sysfs: export linear_lookup in features directory
> > > > 
> > > > stable/linux-5.10.y:
> > > > MISSING:     f2fs: Introduce linear search for dentries
> > > > MISSING:     f2fs: sysfs: export linear_lookup in features directory
> > > > 
> > > > stable/linux-5.15.y:
> > > > b0938ffd39ae f2fs: Introduce linear search for dentries [5.15.179]
> > > > MISSING:     f2fs: sysfs: export linear_lookup in features directory
> > > > 
> > > > stable/linux-6.1.y:
> > > > de605097eb17 f2fs: Introduce linear search for dentries [6.1.129]
> > > > MISSING:     f2fs: sysfs: export linear_lookup in features directory
> > > > 
> > > > stable/linux-6.6.y:
> > > > 0bf2adad03e1 f2fs: Introduce linear search for dentries [6.6.76]
> > > > MISSING:     f2fs: sysfs: export linear_lookup in features directory
> > > > 
> > > > stable/linux-6.12.y:
> > > > 00d1943fe46d f2fs: Introduce linear search for dentries [6.12.13]
> > > > MISSING:     f2fs: sysfs: export linear_lookup in features directory
> > > > 
> > > > mainline:
> > > > 91b587ba79e1 f2fs: Introduce linear search for dentries
> > > > 617e0491abe4 f2fs: sysfs: export linear_lookup in features directory
> > > 
> > > Great, then can someone submit these in a format we can apply them in?
> > > or do clean cherry-picks work properly?
> > 
> > Does this work:
> > 
> > Please backport upstream commit:
> > 
> >   617e0491abe4 f2fs: sysfs: export linear_lookup in features directory
> > 
> > ... to all stable branches up to and including linux-5.15.y.
> > 
> > If there are conflicts, I can do the backport myself and submit as patches.
> 
> There are conflicts, it doesn't apply to 6.1.y or 5.15.y :(

Okay, thanks for trying.  Leave it with me.

-- 
Lee Jones [李琼斯]

Re: [f2fs-dev] [PATCH v2 2/2] f2fs: directly add newly allocated pre-dirty nat entry to dirty set list

[Chao Yu <ch...@ke...>]

On 7/22/25 22:36, wangzijie wrote:
> When we need to alloc nat entry and set it dirty, we can directly add it to
> dirty set list(or initialize its list_head for new_ne) instead of adding it
> to clean list and make a move. Introduce init_dirty flag to do it.
> 
> Signed-off-by: wangzijie <wan...@ho...>
> ---
>  fs/f2fs/node.c | 37 ++++++++++++++++++++++++++++++-------
>  1 file changed, 30 insertions(+), 7 deletions(-)
> 
> diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
> index a23db6238..20bcf8559 100644
> --- a/fs/f2fs/node.c
> +++ b/fs/f2fs/node.c
> @@ -185,7 +185,7 @@ static void __free_nat_entry(struct nat_entry *e)
>  
>  /* must be locked by nat_tree_lock */
>  static struct nat_entry *__init_nat_entry(struct f2fs_nm_info *nm_i,
> -	struct nat_entry *ne, struct f2fs_nat_entry *raw_ne, bool no_fail)
> +	struct nat_entry *ne, struct f2fs_nat_entry *raw_ne, bool no_fail, bool init_dirty)
>  {
>  	if (no_fail)
>  		f2fs_radix_tree_insert(&nm_i->nat_root, nat_get_nid(ne), ne);
> @@ -195,6 +195,11 @@ static struct nat_entry *__init_nat_entry(struct f2fs_nm_info *nm_i,
>  	if (raw_ne)
>  		node_info_from_raw_nat(&ne->ni, raw_ne);
>  
> +	if (init_dirty) {
> +		nm_i->nat_cnt[TOTAL_NAT]++;
> +		return ne;
> +	}
> +
>  	spin_lock(&nm_i->nat_list_lock);
>  	list_add_tail(&ne->list, &nm_i->nat_entries);
>  	spin_unlock(&nm_i->nat_list_lock);
> @@ -256,7 +261,7 @@ static struct nat_entry_set *__grab_nat_entry_set(struct f2fs_nm_info *nm_i,
>  }
>  
>  static void __set_nat_cache_dirty(struct f2fs_nm_info *nm_i,
> -						struct nat_entry *ne)
> +					struct nat_entry *ne, bool init_dirty)
>  {
>  	struct nat_entry_set *head;
>  	bool new_ne = nat_get_blkaddr(ne) == NEW_ADDR;
> @@ -275,6 +280,18 @@ static void __set_nat_cache_dirty(struct f2fs_nm_info *nm_i,
>  
>  	set_nat_flag(ne, IS_PREALLOC, new_ne);
>  
> +	if (init_dirty) {
> +		nm_i->nat_cnt[DIRTY_NAT]++;
> +		set_nat_flag(ne, IS_DIRTY, true);
> +		spin_lock(&nm_i->nat_list_lock);
> +		if (new_ne)
> +			INIT_LIST_HEAD(&ne->list);
> +		else
> +			list_add_tail(&ne->list, &head->entry_list);
> +		spin_unlock(&nm_i->nat_list_lock);
> +		return;
> +    }

Nit issue, above blanks should be replaced w/ tab.

Can we clean up like this?

diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
index de99b42437c6..60fc2c7b8e10 100644
--- a/fs/f2fs/node.c
+++ b/fs/f2fs/node.c
@@ -280,30 +280,23 @@ static void __set_nat_cache_dirty(struct f2fs_nm_info *nm_i,

 	set_nat_flag(ne, IS_PREALLOC, new_ne);

-	if (init_dirty) {
-		nm_i->nat_cnt[DIRTY_NAT]++;
-		set_nat_flag(ne, IS_DIRTY, true);
-		spin_lock(&nm_i->nat_list_lock);
-		if (new_ne)
-			INIT_LIST_HEAD(&ne->list);
-		else
-			list_add_tail(&ne->list, &head->entry_list);
-		spin_unlock(&nm_i->nat_list_lock);
-		return;
-    }
-
 	if (get_nat_flag(ne, IS_DIRTY))
 		goto refresh_list;

 	nm_i->nat_cnt[DIRTY_NAT]++;
-	nm_i->nat_cnt[RECLAIMABLE_NAT]--;
+	if (!init_dirty)
+		nm_i->nat_cnt[RECLAIMABLE_NAT]--;
 	set_nat_flag(ne, IS_DIRTY, true);
 refresh_list:
 	spin_lock(&nm_i->nat_list_lock);
-	if (new_ne)
-		list_del_init(&ne->list);
-	else
+	if (new_ne) {
+		if (init_dirty)
+			INIT_LIST_HEAD(&ne->list);
+		else
+			list_del_init(&ne->list);
+	} else {
 		list_move_tail(&ne->list, &head->entry_list);
+	}
 	spin_unlock(&nm_i->nat_list_lock);
 }

Thanks,

> +
>  	if (get_nat_flag(ne, IS_DIRTY))
>  		goto refresh_list;
>  
> @@ -441,7 +458,7 @@ static void cache_nat_entry(struct f2fs_sb_info *sbi, nid_t nid,
>  	f2fs_down_write(&nm_i->nat_tree_lock);
>  	e = __lookup_nat_cache(nm_i, nid, false);
>  	if (!e)
> -		e = __init_nat_entry(nm_i, new, ne, false);
> +		e = __init_nat_entry(nm_i, new, ne, false, false);
>  	else
>  		f2fs_bug_on(sbi, nat_get_ino(e) != le32_to_cpu(ne->ino) ||
>  				nat_get_blkaddr(e) !=
> @@ -458,11 +475,13 @@ static void set_node_addr(struct f2fs_sb_info *sbi, struct node_info *ni,
>  	struct f2fs_nm_info *nm_i = NM_I(sbi);
>  	struct nat_entry *e;
>  	struct nat_entry *new = __alloc_nat_entry(sbi, ni->nid, true);
> +	bool init_dirty = false;
>  
>  	f2fs_down_write(&nm_i->nat_tree_lock);
>  	e = __lookup_nat_cache(nm_i, ni->nid, true);
>  	if (!e) {
> -		e = __init_nat_entry(nm_i, new, NULL, true);
> +		init_dirty = true;
> +		e = __init_nat_entry(nm_i, new, NULL, true, true);
>  		copy_node_info(&e->ni, ni);
>  		f2fs_bug_on(sbi, ni->blk_addr == NEW_ADDR);
>  	} else if (new_blkaddr == NEW_ADDR) {
> @@ -498,7 +517,7 @@ static void set_node_addr(struct f2fs_sb_info *sbi, struct node_info *ni,
>  	nat_set_blkaddr(e, new_blkaddr);
>  	if (!__is_valid_data_blkaddr(new_blkaddr))
>  		set_nat_flag(e, IS_CHECKPOINTED, false);
> -	__set_nat_cache_dirty(nm_i, e);
> +	__set_nat_cache_dirty(nm_i, e, init_dirty);
>  
>  	/* update fsync_mark if its inode nat entry is still alive */
>  	if (ni->nid != ni->ino)
> @@ -2924,6 +2943,7 @@ static void remove_nats_in_journal(struct f2fs_sb_info *sbi)
>  	struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
>  	struct f2fs_journal *journal = curseg->journal;
>  	int i;
> +	bool init_dirty;
>  
>  	down_write(&curseg->journal_rwsem);
>  	for (i = 0; i < nats_in_cursum(journal); i++) {
> @@ -2934,12 +2954,15 @@ static void remove_nats_in_journal(struct f2fs_sb_info *sbi)
>  		if (f2fs_check_nid_range(sbi, nid))
>  			continue;
>  
> +		init_dirty = false;
> +
>  		raw_ne = nat_in_journal(journal, i);
>  
>  		ne = __lookup_nat_cache(nm_i, nid, true);
>  		if (!ne) {
> +			init_dirty = true;
>  			ne = __alloc_nat_entry(sbi, nid, true);
> -			__init_nat_entry(nm_i, ne, &raw_ne, true);
> +			__init_nat_entry(nm_i, ne, &raw_ne, true, true);
>  		}
>  
>  		/*
> @@ -2954,7 +2977,7 @@ static void remove_nats_in_journal(struct f2fs_sb_info *sbi)
>  			spin_unlock(&nm_i->nid_list_lock);
>  		}
>  
> -		__set_nat_cache_dirty(nm_i, ne);
> +		__set_nat_cache_dirty(nm_i, ne, init_dirty);
>  	}
>  	update_nats_in_cursum(journal, -i);
>  	up_write(&curseg->journal_rwsem);

[f2fs-dev] [PATCH 3/3] f2fs: fix to trigger foreground gc during f2fs_map_blocks() in lfs mode

[Chao Yu <ch...@ke...>]

w/ "mode=lfs" mount option, generic/299 will cause system panic as below:

------------[ cut here ]------------
kernel BUG at fs/f2fs/segment.c:2835!
Call Trace:
 <TASK>
 f2fs_allocate_data_block+0x6f4/0xc50
 f2fs_map_blocks+0x970/0x1550
 f2fs_iomap_begin+0xb2/0x1e0
 iomap_iter+0x1d6/0x430
 __iomap_dio_rw+0x208/0x9a0
 f2fs_file_write_iter+0x6b3/0xfa0
 aio_write+0x15d/0x2e0
 io_submit_one+0x55e/0xab0
 __x64_sys_io_submit+0xa5/0x230
 do_syscall_64+0x84/0x2f0
 entry_SYSCALL_64_after_hwframe+0x76/0x7e
RIP: 0010:new_curseg+0x70f/0x720

The root cause of we run out-of-space is: in f2fs_map_blocks(), f2fs may
trigger foreground gc only if it allocates any physical block, it will be
a little bit later when there is multiple threads writing data w/
aio/dio/bufio method in parallel, since we always use OPU in lfs mode, so
f2fs_map_blocks() does block allocations aggressively.

In order to fix this issue, let's give a chance to trigger foreground
gc in prior to block allocation in f2fs_map_blocks().

Fixes: 36abef4e796d ("f2fs: introduce mode=lfs mount option")
Cc: Daeho Jeong <dae...@go...>
Signed-off-by: Chao Yu <ch...@ke...>
---
 fs/f2fs/data.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 4e62f7f00b70..ffcaf91c1703 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -1573,8 +1573,11 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, int flag)
 	end = pgofs + maxblocks;
 
 next_dnode:
-	if (map->m_may_create)
+	if (map->m_may_create) {
+		if (f2fs_lfs_mode(sbi))
+			f2fs_balance_fs(sbi, true);
 		f2fs_map_lock(sbi, flag);
+	}
 
 	/* When reading holes, we need its node page */
 	set_new_dnode(&dn, inode, NULL, NULL, 0);
-- 
2.49.0

[f2fs-dev] [PATCH 2/3] f2fs: fix to calculate dirty data during has_not_enough_free_secs()

[Chao Yu <ch...@ke...>]

In lfs mode, dirty data needs OPU, we'd better calculate lower_p and
upper_p w/ them during has_not_enough_free_secs(), otherwise we may
encounter out-of-space issue due to we missed to reclaim enough
free section w/ foreground gc.

Fixes: 36abef4e796d ("f2fs: introduce mode=lfs mount option")
Cc: Daeho Jeong <dae...@go...>
Signed-off-by: Chao Yu <ch...@ke...>
---
 fs/f2fs/segment.h | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h
index 2123645cf175..5e2ee5c686b1 100644
--- a/fs/f2fs/segment.h
+++ b/fs/f2fs/segment.h
@@ -668,8 +668,7 @@ static inline void __get_secs_required(struct f2fs_sb_info *sbi,
 	unsigned int dent_blocks = total_dent_blocks % CAP_BLKS_PER_SEC(sbi);
 	unsigned int data_blocks = 0;
 
-	if (f2fs_lfs_mode(sbi) &&
-		unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED))) {
+	if (f2fs_lfs_mode(sbi)) {
 		total_data_blocks = get_pages(sbi, F2FS_DIRTY_DATA);
 		data_secs = total_data_blocks / CAP_BLKS_PER_SEC(sbi);
 		data_blocks = total_data_blocks % CAP_BLKS_PER_SEC(sbi);
-- 
2.49.0

[f2fs-dev] [PATCH 1/3] f2fs: fix to update upper_p in __get_secs_required() correctly

[Chao Yu <ch...@ke...>]

Commit 1acd73edbbfe ("f2fs: fix to account dirty data in __get_secs_required()")
missed to calculate upper_p w/ data_secs, fix it.

Fixes: 1acd73edbbfe ("f2fs: fix to account dirty data in __get_secs_required()")
Cc: Daeho Jeong <dae...@go...>
Signed-off-by: Chao Yu <ch...@ke...>
---
 fs/f2fs/segment.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h
index d2c73f641134..2123645cf175 100644
--- a/fs/f2fs/segment.h
+++ b/fs/f2fs/segment.h
@@ -678,7 +678,7 @@ static inline void __get_secs_required(struct f2fs_sb_info *sbi,
 	if (lower_p)
 		*lower_p = node_secs + dent_secs + data_secs;
 	if (upper_p)
-		*upper_p = node_secs + dent_secs +
+		*upper_p = node_secs + dent_secs + data_secs +
 			(node_blocks ? 1 : 0) + (dent_blocks ? 1 : 0) +
 			(data_blocks ? 1 : 0);
 	if (curseg_p)
-- 
2.49.0