vfs.git pile 2
Al Viro
maintainerNext pile (and there'll be one or two more). The large piece in this one is getting rid of /proc/*/ns/* weirdness; among other things, it allows to (finally) make nameidata completely opaque outside of fs/namei.c, making for easier further cleanups in there.
I have _not_ included removal of unshare_fs_struct() into that pile; I think it'll turn out to be OK, but it's clearly the next cycle fodder.
Please, pull from the usual place -git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs.git for-linus
Shortlog:
Al Viro (20):
common object embedded into various struct ....ns
make mntns ->get()/->put()/->install()/->inum() work with &mnt_ns->ns
netns: switch ->get()/->put()/->install()/->inum() to working with &net->ns switch the rest of proc_ns_operations to working with &...->ns make proc_ns_operations work with struct ns_common * instead of void * new helpers: ns_alloc_inum/ns_free_inum
copy address of proc_ns_ops into ns_common
bury struct proc_ns in fs/proc
take the targets of /proc/*/ns/* symlinks to separate fs
kill proc_ns completely
make nameidata completely opaque outside of fs/namei.c
make default ->i_fop have ->open() fail with ENXIO path_init(): store the "base" pointer to file in nameidata itself fs/namei.c: new helper (path_cleanup())
path_init(): don't bother with LOOKUP_PARENT in argument fs/namei.c: fold link_path_walk() call into path_init() coda_venus_readdir(): use file_inode()
spawn PID 1 without CLONE_FS, give kernel threads zero umask
lustre: get rid of playing with ->fs
btrfs: filp_open() returns ERR_PTR() on failure, not NULL...
Diffstat:
arch/ia64/kernel/perfmon.c | 10 -- .../lustre/lustre/include/linux/lustre_compat25.h | 24 --- drivers/staging/lustre/lustre/llite/dir.c | 2 +- drivers/staging/lustre/lustre/llite/llite_lib.c | 17 +-- fs/Makefile | 2 +- fs/btrfs/volumes.c | 2 +- fs/coda/dir.c | 4 +- fs/fs_struct.c | 2 +- fs/inode.c | 11 +- fs/internal.h | 5 + fs/mount.h | 3 +- fs/namei.c | 98 +++++++------ fs/namespace.c | 51 +++---- fs/nsfs.c | 161 +++++++++++++++++++++ fs/proc/inode.c | 10 +- fs/proc/internal.h | 2 +- fs/proc/namespaces.c | 153 ++------------------ include/linux/fs.h | 1 - include/linux/ipc_namespace.h | 3 +- include/linux/namei.h | 25 +--- include/linux/ns_common.h | 12 ++ include/linux/pid_namespace.h | 3 +- include/linux/proc_ns.h | 43 +++--- include/linux/user_namespace.h | 3 +- include/linux/utsname.h | 3 +- include/net/net_namespace.h | 3 +- include/uapi/linux/magic.h | 1 + init/main.c | 6 +- init/version.c | 5 +- ipc/msgutil.c | 5 +- ipc/namespace.c | 32 ++-- kernel/kmod.c | 2 + kernel/nsproxy.c | 10 +- kernel/pid.c | 5 +- kernel/pid_namespace.c | 29 ++-- kernel/user.c | 5 +- kernel/user_namespace.c | 29 ++-- kernel/utsname.c | 31 ++-- net/Makefile | 2 - net/core/net_namespace.c | 39 ++--- net/nonet.c | 26 ---- net/socket.c | 19 --- 42 files changed, 425 insertions(+), 474 deletions(-) create mode 100644 fs/nsfs.c create mode 100644 include/linux/ns_common.h delete mode 100644 net/nonet.c
Al Viro
maintainerQuoted 7 lines · Al Viro
Next pile (and there'll be one or two more). The large piece in this one is getting rid of
/proc/*/ns/*weirdness; among other things, it allows to (finally) make nameidata completely opaque outside offs/namei.c, making for easier further cleanups in there.I have _not_ included removal of
unshare_fs_struct()into that pile; I think it'll turn out to be OK, but it's clearly the next cycle fodder.
... but I have included the "obviously safe" patch separating PID 1 ->fs. Without letting it sit in -next long enough. And it turns out to be not safe at all - it might be salvagable, but not in this form.
Mea maxima culpa - I should've known better than that. Self-LART applied; please, drop this pull request, I'll send a saner one (for stuff that *did* sit in -next long enough) in a bit.
Again, my apologies ;-/
Al Viro
maintainerQuoted 13 lines · Al Viro
Quoted 7 lines · Al Viro
Next pile (and there'll be one or two more). The large piece in this one is getting rid of
/proc/*/ns/*weirdness; among other things, it allows to (finally) make nameidata completely opaque outside offs/namei.c, making for easier further cleanups in there.I have _not_ included removal of
unshare_fs_struct()into that pile; I think it'll turn out to be OK, but it's clearly the next cycle fodder.... but I have included the "obviously safe" patch separating PID 1 ->fs. Without letting it sit in -next long enough. And it turns out to be not safe at all - it might be salvagable, but not in this form.
Quoted 5 lines
Mea maxima culpa - I should've known better than that. Self-LART applied; please, drop this pull request, I'll send a saner one (for stuff that *did* sit in -next long enough) in a bit.
Again, my apologies ;-/
... and this time around - only the stuff that has sat in -next for a while, with no "obviously safe" additions thrown in. Please, pull fromgit://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs.git for-linus
Shortlog:
Al Viro (17):
common object embedded into various struct ....ns
make mntns ->get()/->put()/->install()/->inum() work with &mnt_ns->ns
netns: switch ->get()/->put()/->install()/->inum() to working with &net->ns switch the rest of proc_ns_operations to working with &...->ns make proc_ns_operations work with struct ns_common * instead of void * new helpers: ns_alloc_inum/ns_free_inum
copy address of proc_ns_ops into ns_common
bury struct proc_ns in fs/proc
take the targets of /proc/*/ns/* symlinks to separate fs
kill proc_ns completely
make nameidata completely opaque outside of fs/namei.c
make default ->i_fop have ->open() fail with ENXIO path_init(): store the "base" pointer to file in nameidata itself fs/namei.c: new helper (path_cleanup())
path_init(): don't bother with LOOKUP_PARENT in argument fs/namei.c: fold link_path_walk() call into path_init() coda_venus_readdir(): use file_inode()
Diffstat:
arch/ia64/kernel/perfmon.c | 10 --- fs/Makefile | 2 +- fs/coda/dir.c | 4 +- fs/inode.c | 11 ++- fs/internal.h | 5 ++ fs/mount.h | 3 +- fs/namei.c | 98 +++++++++++++------------ fs/namespace.c | 51 ++++++------- fs/nsfs.c | 161 +++++++++++++++++++++++++++++++++++++++++ fs/proc/inode.c | 10 +-- fs/proc/internal.h | 2 +- fs/proc/namespaces.c | 153 ++++----------------------------------- include/linux/fs.h | 1 - include/linux/ipc_namespace.h | 3 +- include/linux/namei.h | 25 +------ include/linux/ns_common.h | 12 +++ include/linux/pid_namespace.h | 3 +- include/linux/proc_ns.h | 43 ++++++----- include/linux/user_namespace.h | 3 +- include/linux/utsname.h | 3 +- include/net/net_namespace.h | 3 +- include/uapi/linux/magic.h | 1 + init/main.c | 2 + init/version.c | 5 +- ipc/msgutil.c | 5 +- ipc/namespace.c | 32 ++++---- kernel/nsproxy.c | 10 +-- kernel/pid.c | 5 +- kernel/pid_namespace.c | 29 ++++---- kernel/user.c | 5 +- kernel/user_namespace.c | 29 ++++---- kernel/utsname.c | 31 ++++---- net/Makefile | 2 - net/core/net_namespace.c | 39 +++++----- net/nonet.c | 26 ------- net/socket.c | 19 ----- 36 files changed, 416 insertions(+), 430 deletions(-) create mode 100644 fs/nsfs.c create mode 100644 include/linux/ns_common.h delete mode 100644 net/nonet.c