diff mbox series

[f2fs-dev,v3,1/5] zonefs: pass GFP_KERNEL to blkdev_zone_mgmt() call

Message ID 20240128-zonefs_nofs-v3-1-ae3b7c8def61@wdc.com (mailing list archive)
State Accepted
Commit 9105ce591b424771b1502ef9836ca7953c3e0af4
Headers show
Series block: remove gfp_mask for blkdev_zone_mgmt() | expand

Commit Message

Johannes Thumshirn Jan. 29, 2024, 7:52 a.m. UTC
Pass GFP_KERNEL instead of GFP_NOFS to the blkdev_zone_mgmt() call in
zonefs_zone_mgmt().

As as zonefs_zone_mgmt() and zonefs_inode_zone_mgmt() are never called
from a place that can recurse back into the filesystem on memory reclaim,
it is save to call blkdev_zone_mgmt() with GFP_KERNEL.

Link: https://lore.kernel.org/all/ZZcgXI46AinlcBDP@casper.infradead.org/
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
---
 fs/zonefs/super.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Comments

Damien Le Moal Jan. 29, 2024, 11:25 p.m. UTC | #1
On 1/29/24 16:52, Johannes Thumshirn wrote:
> Pass GFP_KERNEL instead of GFP_NOFS to the blkdev_zone_mgmt() call in
> zonefs_zone_mgmt().
> 
> As as zonefs_zone_mgmt() and zonefs_inode_zone_mgmt() are never called
> from a place that can recurse back into the filesystem on memory reclaim,
> it is save to call blkdev_zone_mgmt() with GFP_KERNEL.

s/save/safe

> Link: https://lore.kernel.org/all/ZZcgXI46AinlcBDP@casper.infradead.org/
> Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
> ---
>  fs/zonefs/super.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/fs/zonefs/super.c b/fs/zonefs/super.c
> index 93971742613a..63fbac018c04 100644
> --- a/fs/zonefs/super.c
> +++ b/fs/zonefs/super.c
> @@ -113,7 +113,7 @@ static int zonefs_zone_mgmt(struct super_block *sb,
>  
>  	trace_zonefs_zone_mgmt(sb, z, op);
>  	ret = blkdev_zone_mgmt(sb->s_bdev, op, z->z_sector,
> -			       z->z_size >> SECTOR_SHIFT, GFP_NOFS);
> +			       z->z_size >> SECTOR_SHIFT, GFP_KERNEL);
>  	if (ret) {
>  		zonefs_err(sb,
>  			   "Zone management operation %s at %llu failed %d\n",

I think this is OK but I will need a little more time to fully convince myself.
So let me look again at the code to check all the calls contexts.
Damien Le Moal Feb. 9, 2024, 7:44 a.m. UTC | #2
On 1/29/24 16:52, Johannes Thumshirn wrote:
> Pass GFP_KERNEL instead of GFP_NOFS to the blkdev_zone_mgmt() call in
> zonefs_zone_mgmt().
> 
> As as zonefs_zone_mgmt() and zonefs_inode_zone_mgmt() are never called
> from a place that can recurse back into the filesystem on memory reclaim,
> it is save to call blkdev_zone_mgmt() with GFP_KERNEL.
> 
> Link: https://lore.kernel.org/all/ZZcgXI46AinlcBDP@casper.infradead.org/
> Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
> ---
>  fs/zonefs/super.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/fs/zonefs/super.c b/fs/zonefs/super.c
> index 93971742613a..63fbac018c04 100644
> --- a/fs/zonefs/super.c
> +++ b/fs/zonefs/super.c
> @@ -113,7 +113,7 @@ static int zonefs_zone_mgmt(struct super_block *sb,
>  
>  	trace_zonefs_zone_mgmt(sb, z, op);
>  	ret = blkdev_zone_mgmt(sb->s_bdev, op, z->z_sector,
> -			       z->z_size >> SECTOR_SHIFT, GFP_NOFS);
> +			       z->z_size >> SECTOR_SHIFT, GFP_KERNEL);
>  	if (ret) {
>  		zonefs_err(sb,
>  			   "Zone management operation %s at %llu failed %d\n",
> 

Given that zonefs_inode_zone_mgmt() which calls zonefs_zone_mgmt() is only used
for sequential zone inodes, and that these inodes cannot be written with
buffered IOs, I think this is safe as we will never have dirty pages to
writeback for reclaim. So there should be no risk of re-entering the FS on
reclaim with GFP_KERNEL.

So:

Acked-by: Damien Le Moal <dlemoal@kernel.org>
diff mbox series

Patch

diff --git a/fs/zonefs/super.c b/fs/zonefs/super.c
index 93971742613a..63fbac018c04 100644
--- a/fs/zonefs/super.c
+++ b/fs/zonefs/super.c
@@ -113,7 +113,7 @@  static int zonefs_zone_mgmt(struct super_block *sb,
 
 	trace_zonefs_zone_mgmt(sb, z, op);
 	ret = blkdev_zone_mgmt(sb->s_bdev, op, z->z_sector,
-			       z->z_size >> SECTOR_SHIFT, GFP_NOFS);
+			       z->z_size >> SECTOR_SHIFT, GFP_KERNEL);
 	if (ret) {
 		zonefs_err(sb,
 			   "Zone management operation %s at %llu failed %d\n",