Skip to contentSkip to conversation

Documentation: update nfs option in filesystem/vfat.txt

PATCH In review Namjae Jeon 2 participants 3 messages Sep 16, 2012 – Sep 17, 2012 Documentation

lore
Share & tools
lore
#0 lore raw

PATCH v3 5/5 Documentation: update nfs option in filesystem/vfat.txt

From: Namjae Jeon <[email protected]>

update nfs option in filesystem/vfat.txt

---
 Documentation/filesystems/vfat.txt |   19 ++++++++++++++-----
 1 file changed, 14 insertions(+), 5 deletions(-)
Documentation/filesystems/vfat.txt +14 -5
index de1e6c4..3371b8e 100644
@@ -141,13 +141,22 @@ discard       -- If set, issues discard/TRIM commands to the block
 		 device when blocks are freed. This is useful for SSD devices
 		 and sparse/thinly-provisoned LUNs.
 
-nfs           -- This option maintains an index (cache) of directory
-		 inodes by i_logstart which is used by the nfs-related code to
-		 improve look-ups.
-
-		 Enable this only if you want to export the FAT filesystem
+nfs=full_unstable|limited_stable
+	      -- Enable this only if you want to export the FAT filesystem
 		 over NFS
 
+		 full_unstable:This option maintains an index (cache) of
+		 directory inodes by i_logstart which is used by
+		 the nfs-related code to improve look-ups.Full file operations
+		 (read/write) over NFS is supported but with cache eviction
+		 at NFS server, this could result in ESTALE issues.
+
+		 limited_stable:This options makes NFS support for FAT as
+		 read-only and provides stable inode based on the on-disk
+		 location of the MS-DOS directory entry.This helps in
+		 reconstruction of the inodes in case of lookup failure,
+		 making FAT behaviour stable with respect to ESTALE issues.
+
 <bool>: 0,1,yes,no,true,false
 
 TODO
Signature · 1 line
1.7.9.5
#1 lore raw
Quoted 16 lines · Namjae Jeon

From: Namjae Jeon <[email protected]>

update nfs option in filesystem/vfat.txt

---
 Documentation/filesystems/vfat.txt |   19 ++++++++++++++-----
 1 file changed, 14 insertions(+), 5 deletions(-)
Documentation/filesystems/vfat.txt +0 -0
index de1e6c4..3371b8e 100644

Documentation/filesystems/vfat.txt discard -- If set, issues discard/TRIM commands to the block · line 144

Quoted 9 lines
Documentation/filesystems/vfat.txt +1 -5
@@ -141,13 +141,22 @@ discard       -- If set, issues discard/TRIM commands to the block
 		 device when blocks are freed. This is useful for SSD devices
 		 and sparse/thinly-provisoned LUNs.
 
-nfs           -- This option maintains an index (cache) of directory
-		 inodes by i_logstart which is used by the nfs-related code to
-		 improve look-ups.
-
-		 Enable this only if you want to export the FAT filesystem
+nfs=full_unstable|limited_stable

You still want to accept just "nfs" (defaulting to "full_unstable" behavior), for backwards compatibility.

Also I don't find this full_unstable/limited_stable terminology very intuitive. "stable" and "limited" could mean too many different things.

Maybe "nostale_ro|stale_rw" ?

Quoted 4 lines
Documentation/filesystems/vfat.txt +2 -0
+	      -- Enable this only if you want to export the FAT filesystem
 		 over NFS
 
+		 full_unstable:This option maintains an index (cache) of

Documentation/filesystems/vfat.txt discard -- If set, issues discard/TRIM commands to the block · line 158

Quoted 10 lines
Documentation/filesystems/vfat.txt +10 -0
+		 directory inodes by i_logstart which is used by
+		 the nfs-related code to improve look-ups.Full file operations
+		 (read/write) over NFS is supported but with cache eviction
+		 at NFS server, this could result in ESTALE issues.
+
+		 limited_stable:This options makes NFS support for FAT as
+		 read-only and provides stable inode based on the on-disk
+		 location of the MS-DOS directory entry.This helps in
+		 reconstruction of the inodes in case of lookup failure,
+		 making FAT behaviour stable with respect to ESTALE issues.

This option still scares me a little. Maybe a little more explanation:

This option bases the inode number and filehandle on the location of an file in the MS-DOS directory entry. This ensures that ESTALE will not be returned after a file is evicted from the inode cache. However, it means that operations such as rename, create, and unlink could cause filehandles that previously pointed at one file to point at a different file, potentially causing data corruption. For this reason, this option also mounts the filesystem readonly.

--b.

Quoted 7 lines
Documentation/filesystems/vfat.txt +1 -0
+
 <bool>: 0,1,yes,no,true,false
 
 TODO
Signature · 1 line
1.7.9.5
#2 lore raw

2012/9/17, J. Bruce Fields <[email protected]>:

Quoted 22 lines · Namjae Jeon
Quoted 21 lines

From: Namjae Jeon <[email protected]>

update nfs option in filesystem/vfat.txt

---
 Documentation/filesystems/vfat.txt |   19 ++++++++++++++-----
 1 file changed, 14 insertions(+), 5 deletions(-)
Documentation/filesystems/vfat.txt +0 -0
index de1e6c4..3371b8e 100644
@@ -141,13 +141,22 @@ discard       -- If set, issues discard/TRIM

commands to the block

Documentation/filesystems/vfat.txt +0 -0
 		 device when blocks are freed. This is useful for SSD devices
 		 and sparse/thinly-provisoned LUNs.
Quoted 7 lines

Documentation/filesystems/vfat.txt discard -- If set, issues discard/TRIM commands to the block · line 144

Quoted 6 lines
Documentation/filesystems/vfat.txt +1 -5
-nfs           -- This option maintains an index (cache) of directory
-		 inodes by i_logstart which is used by the nfs-related code to
-		 improve look-ups.
-
-		 Enable this only if you want to export the FAT filesystem
+nfs=full_unstable|limited_stable

Hi Bruce.

Quoted 2 lines

You still want to accept just "nfs" (defaulting to "full_unstable" behavior), for backwards compatibility.

Agree. I will consider backwards compatibilty as your opinion.

Quoted 5 lines

Also I don't find this full_unstable/limited_stable terminology very intuitive. "stable" and "limited" could mean too many different things.

Maybe "nostale_ro|stale_rw" ?

Yes, It would be better!

Quoted 18 lines
Quoted 5 lines
Documentation/filesystems/vfat.txt +2 -0
+	      -- Enable this only if you want to export the FAT filesystem
 		 over NFS
 
+		 full_unstable:This option maintains an index (cache) of

Documentation/filesystems/vfat.txt discard -- If set, issues discard/TRIM commands to the block · line 158

Quoted 10 lines
Documentation/filesystems/vfat.txt +10 -0
+		 directory inodes by i_logstart which is used by
+		 the nfs-related code to improve look-ups.Full file operations
+		 (read/write) over NFS is supported but with cache eviction
+		 at NFS server, this could result in ESTALE issues.
+
+		 limited_stable:This options makes NFS support for FAT as
+		 read-only and provides stable inode based on the on-disk
+		 location of the MS-DOS directory entry.This helps in
+		 reconstruction of the inodes in case of lookup failure,
+		 making FAT behaviour stable with respect to ESTALE issues.

This option still scares me a little. Maybe a little more explanation:

Quoted 9 lines

This option bases the inode number and filehandle on the location of an file in the MS-DOS directory entry. This ensures that ESTALE will not be returned after a file is evicted from the inode cache. However, it means that operations such as rename, create, and unlink could cause filehandles that previously pointed at one file to point at a different file, potentially causing data corruption. For this reason, this option also mounts the filesystem readonly.

Okay, I will update.

Thanks for your review!

Quoted 11 lines

--b.

Quoted 7 lines

+
<bool>: 0,1,yes,no,true,false

TODO
--
1.7.9.5