Skip to contentSkip to conversation

ext4: support xattr gnu.* namespace for the Hurd

PATCH In review Jan (janneke) Nieuwenhuizen 2 participants 4 messages May 25, 2020 – Jun 12, 2020 ext4 file system

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 ext4: support xattr gnu.* namespace for the Hurd

The Hurd gained[0] support for moving the translator and author fields out of the inode and into the "gnu.*" xattr namespace.

In anticipation of that, an xattr INDEX was reserved[1]. The Hurd has now been brought into compliance[2] with that.

This patch adds support for reading and writing such attributes from Linux; you can now do something like

    mkdir -p hurd-root/servers/socket
    touch hurd-root/servers/socket/1
    setfattr --name=gnu.translator --value='"/hurd/pflocal\0"' \
        hurd-root/servers/socket/1
    getfattr --name=gnu.translator hurd-root/servers/socket/1
    # file: 1
    gnu.translator="/hurd/pflocal"

to setup a pipe translator, which is being used to create[3] a vm-image for the Hurd from GNU Guix.

[0] https://summerofcode.withgoogle.com/projects/#5869799859027968
[1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=3980bd3b406addb327d858aebd19e229ea340b9a
[2] https://git.savannah.gnu.org/cgit/hurd/hurd.git/commit/?id=a04c7bf83172faa7cb080fbe3b6c04a8415ca645
[3] https://git.savannah.gnu.org/cgit/guix.git/log/?h=wip-hurd-vm

---
 fs/ext4/Makefile           |  3 ++-
 fs/ext4/xattr.c            |  2 ++
 fs/ext4/xattr.h            |  1 +
 fs/ext4/xattr_hurd.c       | 51 ++++++++++++++++++++++++++++++++++++++
 include/uapi/linux/xattr.h |  4 +++
 5 files changed, 60 insertions(+), 1 deletion(-)
 create mode 100644 fs/ext4/xattr_hurd.c
fs/ext4/Makefile +2 -1
index 4ccb3c9189d8..2e42f47a7f98 100644
@@ -9,7 +9,8 @@ ext4-y	:= balloc.o bitmap.o block_validity.o dir.o ext4_jbd2.o extents.o \
 		extents_status.o file.o fsmap.o fsync.o hash.o ialloc.o \
 		indirect.o inline.o inode.o ioctl.o mballoc.o migrate.o \
 		mmp.o move_extent.o namei.o page-io.o readpage.o resize.o \
-		super.o symlink.o sysfs.o xattr.o xattr_trusted.o xattr_user.o
+		super.o symlink.o sysfs.o xattr.o xattr_hurd.o xattr_trusted.o \
+		xattr_user.o
 
 ext4-$(CONFIG_EXT4_FS_POSIX_ACL)	+= acl.o
 ext4-$(CONFIG_EXT4_FS_SECURITY)		+= xattr_security.o
fs/ext4/xattr.c +2 -0
index 9b29a40738ac..7d2f6576d954 100644
@@ -93,6 +93,7 @@ static const struct xattr_handler * const ext4_xattr_handler_map[] = {
 #ifdef CONFIG_EXT4_FS_SECURITY
 	[EXT4_XATTR_INDEX_SECURITY]	     = &ext4_xattr_security_handler,
 #endif
+	[EXT4_XATTR_INDEX_HURD]		     = &ext4_xattr_hurd_handler,
 };
 
 const struct xattr_handler *ext4_xattr_handlers[] = {
@@ -105,6 +106,7 @@ const struct xattr_handler *ext4_xattr_handlers[] = {
 #ifdef CONFIG_EXT4_FS_SECURITY
 	&ext4_xattr_security_handler,
 #endif
+	&ext4_xattr_hurd_handler,
 	NULL
 };
 
fs/ext4/xattr.h +1 -0
index ffe21ac77f78..730b91fa0dd7 100644
@@ -124,6 +124,7 @@ struct ext4_xattr_inode_array {
 extern const struct xattr_handler ext4_xattr_user_handler;
 extern const struct xattr_handler ext4_xattr_trusted_handler;
 extern const struct xattr_handler ext4_xattr_security_handler;
+extern const struct xattr_handler ext4_xattr_hurd_handler;
 
 #define EXT4_XATTR_NAME_ENCRYPTION_CONTEXT "c"
 
fs/ext4/xattr_hurd.c new file +51 -0

51 lines not on this page · raw message

include/uapi/linux/xattr.h +4 -0
index c1395b5bd432..9463db2dfa9d 100644
@@ -7,6 +7,7 @@
   Copyright (C) 2001 by Andreas Gruenbacher <[email protected]>
   Copyright (c) 2001-2002 Silicon Graphics, Inc.  All Rights Reserved.
   Copyright (c) 2004 Red Hat, Inc., James Morris <[email protected]>
+  Copyright (c) 2020 Jan (janneke) Nieuwenhuizen <[email protected]>
 */
 
 #include <linux/libc-compat.h>
@@ -31,6 +32,9 @@
 #define XATTR_BTRFS_PREFIX "btrfs."
 #define XATTR_BTRFS_PREFIX_LEN (sizeof(XATTR_BTRFS_PREFIX) - 1)
 
+#define XATTR_HURD_PREFIX "gnu."
+#define XATTR_HURD_PREFIX_LEN (sizeof(XATTR_HURD_PREFIX) - 1)
+
 #define XATTR_SECURITY_PREFIX	"security."
 #define XATTR_SECURITY_PREFIX_LEN (sizeof(XATTR_SECURITY_PREFIX) - 1)
 
Signature · 1 line
2.26.2
maintainer
#1 lore raw
Quoted 17 lines · Jan (janneke) Nieuwenhuizen

The Hurd gained[0] support for moving the translator and author fields out of the inode and into the "gnu.*" xattr namespace.

In anticipation of that, an xattr INDEX was reserved[1]. The Hurd has now been brought into compliance[2] with that.

This patch adds support for reading and writing such attributes from Linux; you can now do something like

    mkdir -p hurd-root/servers/socket
    touch hurd-root/servers/socket/1
    setfattr --name=gnu.translator --value='"/hurd/pflocal\0"' \
        hurd-root/servers/socket/1
    getfattr --name=gnu.translator hurd-root/servers/socket/1
    # file: 1
    gnu.translator="/hurd/pflocal"
Quoted 7 lines

to setup a pipe translator, which is being used to create[3] a vm-image for the Hurd from GNU Guix.

[0] https://summerofcode.withgoogle.com/projects/#5869799859027968
[1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=3980bd3b406addb327d858aebd19e229ea340b9a
[2] https://git.savannah.gnu.org/cgit/hurd/hurd.git/commit/?id=a04c7bf83172faa7cb080fbe3b6c04a8415ca645
[3] https://git.savannah.gnu.org/cgit/guix.git/log/?h=wip-hurd-vm

This patch is missing a Signed-off-by. If you don't understand why this is really important, please read:

https://www.kernel.org/doc/html/latest/process/submitting-patches.html#sign-your-work-the-developer-s-certificate-of-origin

Can you resubmit with the DCO or confirm that it's OK for me to add your Signed-off-by?

- Ted

#2 lore raw

Theodore Y. Ts'o writes:

Hello!

Quoted 9 lines · Jan (janneke) Nieuwenhuizen
Quoted 8 lines

The Hurd gained[0] support for moving the translator and author fields out of the inode and into the "gnu.*" xattr namespace.

In anticipation of that, an xattr INDEX was reserved[1]. The Hurd has now been brought into compliance[2] with that.

This patch adds support for reading and writing such attributes from Linux; you can now do something like

[...]

Quoted 5 lines

This patch is missing a Signed-off-by. If you don't understand why this is really important, please read:

https://www.kernel.org/doc/html/latest/process/submitting-patches.html#sign-your-work-the-developer-s-certificate-of-origin

Ah, that makes sense. Sorry for omitting that/being not clear on it.

Quoted 2 lines

Can you resubmit with the DCO or confirm that it's OK for me to add your Signed-off-by?

Yes, that's OK, please do! I am the original author of this patch.

Greetings,
Janneke

Signature · 2 lines
Jan Nieuwenhuizen <[email protected]> | GNU LilyPond http://lilypond.org
Freelance IT http://JoyofSource.com | Avatar® http://AvatarAcademy.com
maintainer
#3 lore raw
Quoted 19 lines · Jan Nieuwenhuizen

Theodore Y. Ts'o writes:

Hello!

Quoted 9 lines · Jan (janneke) Nieuwenhuizen
Quoted 8 lines

The Hurd gained[0] support for moving the translator and author fields out of the inode and into the "gnu.*" xattr namespace.

In anticipation of that, an xattr INDEX was reserved[1]. The Hurd has now been brought into compliance[2] with that.

This patch adds support for reading and writing such attributes from Linux; you can now do something like

[...]

Quoted 4 lines

This patch is missing a Signed-off-by. If you don't understand why this is really important, please read:

Quoted 8 lines
Quoted 1 line

https://www.kernel.org/doc/html/latest/process/submitting-patches.html#sign-your-work-the-developer-s-certificate-of-origin

Ah, that makes sense. Sorry for omitting that/being not clear on it.

Quoted 2 lines

Can you resubmit with the DCO or confirm that it's OK for me to add your Signed-off-by?

Yes, that's OK, please do! I am the original author of this patch.

Thanks, I've applied this to the ext4.git tree.

- Ted