Al Viro [Sat, 6 Apr 2013 00:44:08 +0000 (20:44 -0400)]
pxa3xx-gcu: stop embedding file_operations into device-private object
no need to do that anymore...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sat, 6 Apr 2013 00:39:36 +0000 (20:39 -0400)]
pxa3xx-gcu: quite playing silly buggers with ->f_op
misc device gets ->private_data pointing to struct miscdevice
on open(), so we can use that to get to per-device structure
instead of relying on file_operations being copied into it.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
David Howells [Thu, 4 Apr 2013 19:36:00 +0000 (20:36 +0100)]
silicom-bypass: Remove device_open/close()
Remove device_open/close() functions as they don't really do anything and
remove Device_Open as it isn't counted atomically and the value isn't used.
Signed-off-by: David Howells <dhowells@redhat.com>
David Howells [Thu, 4 Apr 2013 16:02:03 +0000 (17:02 +0100)]
proc: Kill create_proc_entry()
Kill create_proc_entry() in favour of create_proc_read_entry(), proc_create()
and proc_create_data().
Signed-off-by: David Howells <dhowells@redhat.com>
David Howells [Thu, 4 Apr 2013 15:49:26 +0000 (16:49 +0100)]
wlags49_h2: Don't use create_proc_entry()
create_proc_entry() shouldn't be used. Rather proc_create_data() should be
used. The proc_write() function is only used by #if'd out code, so delete it
for now.
Signed-off-by: David Howells <dhowells@redhat.com>
David Howells [Thu, 4 Apr 2013 15:44:51 +0000 (16:44 +0100)]
nubus: Don't use create_proc_entry()
Don't use create_proc_entry() in nubus_proc_subdir(). The files created aren't
given any way to use them, so for the moment use create_proc_read_entry() with
a NULL accessor and generate a compile-time warning.
Signed-off-by: David Howells <dhowells@redhat.com>
David Howells [Thu, 4 Apr 2013 15:58:25 +0000 (16:58 +0100)]
UM: Adjust printk in create_proc_mconsole()
Adjust printk in create_proc_mconsole() to reflect it is now using
proc_create() not create_proc_mconsole().
Signed-off-by: David Howells <dhowells@redhat.com>
Al Viro [Thu, 4 Apr 2013 23:09:41 +0000 (19:09 -0400)]
constify a bunch of struct file_operations instances
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Thu, 4 Apr 2013 23:18:04 +0000 (19:18 -0400)]
bluetooth: kill unused fops field in struct bt_sock_list
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Thu, 4 Apr 2013 23:16:06 +0000 (19:16 -0400)]
bluetooth: fix race in bt_procfs_init()
use proc_create_data() rather than set ->data after the file has
been created
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Thu, 4 Apr 2013 23:14:33 +0000 (19:14 -0400)]
bluetooth: kill unused 'module' argument of bt_procfs_init()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Thu, 4 Apr 2013 23:12:06 +0000 (19:12 -0400)]
bluetooth: don't bother with ->owner for procfs fops
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Wed, 3 Apr 2013 04:04:51 +0000 (00:04 -0400)]
x25: use proc_remove_subtree()
... and don't bother with ->owner, while we are at it - procfs fops
do not need it.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Tue, 2 Apr 2013 00:48:34 +0000 (20:48 -0400)]
create_proc_cpu_mask() doesn't need an argument...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Mon, 1 Apr 2013 03:21:50 +0000 (23:21 -0400)]
lparcfg: don't bother saving pointer to proc_dir_entry
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Mon, 1 Apr 2013 02:50:16 +0000 (22:50 -0400)]
prominfo_proc fixes
* check for proc_mkdir() failures
* use remove_proc_subtree()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Mon, 1 Apr 2013 01:45:09 +0000 (21:45 -0400)]
srm_env: don't bother with pointer to srm_env_t
Since the only thing in it the methods actually care about is
variable id, just store that directly.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 22:29:51 +0000 (18:29 -0400)]
srm_env: use proc_remove_subtree()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 22:16:14 +0000 (18:16 -0400)]
procfs: new helper - PDE_DATA(inode)
The only part of proc_dir_entry the code outside of fs/proc
really cares about is PDE(inode)->data. Provide a helper
for that; static inline for now, eventually will be moved
to fs/proc, along with the knowledge of struct proc_dir_entry
layout.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 19:41:41 +0000 (15:41 -0400)]
last_radio_log: switch to proc_create()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 19:30:40 +0000 (15:30 -0400)]
procfs: kill ->write_proc()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 19:29:21 +0000 (15:29 -0400)]
bfin_adv7393: switch to proc_create_data()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 19:18:19 +0000 (15:18 -0400)]
xtensa simdisk: switch to proc_create_data()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 19:07:00 +0000 (15:07 -0400)]
rtl8192u: switch to proc_create()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 17:50:52 +0000 (13:50 -0400)]
scsi_proc: make proc_scsi_host_open() preallocate a bigger buffer
Some of the ->show_info() instances really spew a lot; it's not a problem
wrt correctness (seq_read() will grow buffer and call the sucker again),
but in this case it makes sense to start with a somewhat bigger one -
they often do exceed one page worth of output.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 17:43:23 +0000 (13:43 -0400)]
new helper: single_open_size()
Same as single_open(), but preallocates the buffer of given size.
Doesn't make any sense for sizes up to PAGE_SIZE and doesn't make
sense if output of show() exceeds PAGE_SIZE only rarely - seq_read()
will take care of growing the buffer and redoing show(). If you
_know_ that it will be large, it might make more sense to look into
saner iterator, rather than go with single-shot one. If that's
impossible, single_open_size() might be for you.
Again, don't use that without a good reason; occasionally that's really
the best way to go, but very often there are better solutions.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 17:27:56 +0000 (13:27 -0400)]
scsi: bury ->proc_info()
all users converted to ->show_info()/->write_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 17:24:44 +0000 (13:24 -0400)]
atari_scsi: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 16:53:53 +0000 (12:53 -0400)]
aix7xxx_old: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 16:46:43 +0000 (12:46 -0400)]
sata_svw: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 16:17:09 +0000 (12:17 -0400)]
ips: fix format bugs exposed by previous commit
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 16:15:44 +0000 (12:15 -0400)]
ips: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 16:02:39 +0000 (12:02 -0400)]
acornscsi: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 08:42:22 +0000 (04:42 -0400)]
fas216: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 08:07:31 +0000 (04:07 -0400)]
aic7xxx: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 07:59:17 +0000 (03:59 -0400)]
aic79xx: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 07:32:45 +0000 (03:32 -0400)]
in2000: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 07:26:26 +0000 (03:26 -0400)]
atp870u: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 07:24:12 +0000 (03:24 -0400)]
dc395x: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 07:21:50 +0000 (03:21 -0400)]
dpt_i2o: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 07:18:35 +0000 (03:18 -0400)]
eata_pio: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 07:15:00 +0000 (03:15 -0400)]
aha1740: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 07:12:15 +0000 (03:12 -0400)]
aha152x: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 07:04:13 +0000 (03:04 -0400)]
advansys: fix buggered formats - unsigned long != u32 on 64bit
asc_prt_line() had been hiding several places where formats had not
matched the argument types. The previous commit has finally made them
visible...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 06:59:55 +0000 (02:59 -0400)]
advansys: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 06:14:18 +0000 (02:14 -0400)]
nsp_cs: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 06:11:39 +0000 (02:11 -0400)]
nsp32: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 06:08:31 +0000 (02:08 -0400)]
keucr: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 06:04:33 +0000 (02:04 -0400)]
rts5139: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 06:01:55 +0000 (02:01 -0400)]
tcm: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 06:00:06 +0000 (02:00 -0400)]
cciss: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 05:52:13 +0000 (01:52 -0400)]
BusLogic: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 05:46:06 +0000 (01:46 -0400)]
scsi_debug: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 05:42:38 +0000 (01:42 -0400)]
fusion: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 05:37:13 +0000 (01:37 -0400)]
usb-storage: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 05:15:54 +0000 (01:15 -0400)]
NCR5830: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 04:58:31 +0000 (00:58 -0400)]
ppa: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 04:55:40 +0000 (00:55 -0400)]
imm: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 04:52:08 +0000 (00:52 -0400)]
gdth: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 04:30:35 +0000 (00:30 -0400)]
switch wd33c93 to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 04:17:00 +0000 (00:17 -0400)]
wd7000: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 04:10:16 +0000 (00:10 -0400)]
sym53c8xx_2: switch to ->show_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 03:58:05 +0000 (23:58 -0400)]
scsi: saner replacements for ->proc_info()
It's still an obsolete interface; don't introduce those in new drivers.
However, it's saner than the ->proc_info() and commits after this one
will convert the existing ->proc_info() users to it.
The read side is ->show_info(seq_file *, struct Scsi_Host *); use
seq_... for generating contents.
The write side is ->write_info(struct Scsi_Host *, char *, int).
Again, this is driven by procfs needs; we are going to kill ->write_proc()
and ->read_proc() and this is the main obstacle to burying that piece of
shit.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 01:20:14 +0000 (21:20 -0400)]
procfs: don't allow to use proc_create, create_proc_entry, etc. for directories
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 00:34:21 +0000 (20:34 -0400)]
i2o: use proc_remove_subtree()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Fri, 5 Apr 2013 17:42:42 +0000 (13:42 -0400)]
silicom: bury bp_proc.c
It's a seriously rotten copy of parts of bp_mod.c; had been
ifdefed out all along, lacks a bunch of declarations that would
be needed if ifdef had been removed, all stuff in it is duplicated
in bp_mod.c anyway...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sat, 30 Mar 2013 23:11:39 +0000 (19:11 -0400)]
silicom: switch to create_proc_data(), clean procfs side of things up
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 00:16:20 +0000 (20:16 -0400)]
reiserfs: use proc_remove_subtree()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sat, 30 Mar 2013 21:20:54 +0000 (17:20 -0400)]
silicom: untangle module_init
Helper functions are often useful. So are local variables...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sat, 30 Mar 2013 20:44:11 +0000 (16:44 -0400)]
silicom: get_bypass_slave_pfs() open-codes lookup_port()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sat, 30 Mar 2013 20:07:02 +0000 (16:07 -0400)]
silicom: unobfuscate get_{status,master}_port_fn()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sat, 30 Mar 2013 19:41:56 +0000 (15:41 -0400)]
silicom: helper functions are often useful...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sat, 30 Mar 2013 17:26:37 +0000 (13:26 -0400)]
get rid of a bunch of open-coded create_proc_read_entry()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sat, 30 Mar 2013 17:15:27 +0000 (13:15 -0400)]
isp1362-hcd: don't reimplement proc_create_data()
... especially in a racy way
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sat, 30 Mar 2013 17:13:39 +0000 (13:13 -0400)]
rtl8192u: don't play with reassigning ->proc_fops, just use proc_create()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sat, 30 Mar 2013 05:07:56 +0000 (01:07 -0400)]
dgrp procfs fixes, part 6: just use proc_create{,_data}
->permission() of its own is a rudiment of sysctl imitation;
normal procfs logics will do just fine here, no need to
mess with ->proc_iops at all.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sat, 30 Mar 2013 05:03:53 +0000 (01:03 -0400)]
dgrp procfs fixes, part 5: per-node files
this "hooks" scheme is pointless - just make file_operations non-static
and consolidate initialiazation bits.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sat, 30 Mar 2013 04:36:23 +0000 (00:36 -0400)]
dgrp procfs fixes, part 4: get rid of sysctl-like machinery
racy and very overblown...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sat, 30 Mar 2013 00:49:43 +0000 (20:49 -0400)]
dgrp procfs fixes, part 3: kill dead code
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sat, 30 Mar 2013 00:45:38 +0000 (20:45 -0400)]
dgrp procfs fixes, part 2
All table entries either have non-NULL ->proc_file_fops or
non-NULL child.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sat, 30 Mar 2013 00:39:17 +0000 (20:39 -0400)]
dgrp procfs fixes, part 1
proc_create() has shat upon fops argument when mode is S_IFDIR.
Good thing, too, since fops passed to it is completely useless
for any directory. Just use proc_mkdir(), damnit.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Sean MacLennan [Tue, 9 Apr 2013 01:18:06 +0000 (21:18 -0400)]
The rtl8192e procfs-based debug interface seems very broken
The procfs debug code in rtl_debug.c is, ironically, very buggy: it lacks proper locking.
Since the most useful part of the code (the stats) are available through more
standard APIs, I think it is best to just delete the whole mess.
Signed-off-by: Sean MacLennan <seanm@seanm.ca>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 31 Mar 2013 19:02:51 +0000 (15:02 -0400)]
rtl8192e: switch to proc_create()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Fri, 29 Mar 2013 23:30:06 +0000 (19:30 -0400)]
rtl8192e: don't use create_proc_entry() for directories
proc_mkdir() is there for purpose...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Fri, 29 Mar 2013 23:27:05 +0000 (19:27 -0400)]
procfs: switch /proc/self away from proc_dir_entry
Just have it pinned in dcache all along and let procfs ->kill_sb()
drop it before kill_anon_super().
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Fri, 29 Mar 2013 16:23:28 +0000 (12:23 -0400)]
mode_t, whack-a-mole at 11...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Fri, 29 Mar 2013 03:01:34 +0000 (23:01 -0400)]
snd_info_register: switch to proc_create_data/proc_mkdir_mode
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Thu, 28 Mar 2013 22:56:21 +0000 (18:56 -0400)]
hysdn: stash pointer to card into proc_dir_entry->data
no need to search later - we know the card when we are
creating procfs entries
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Thu, 28 Mar 2013 22:11:13 +0000 (18:11 -0400)]
atags_proc: switch to proc_create_data()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Thu, 28 Mar 2013 17:12:32 +0000 (13:12 -0400)]
gadgetfs: don't bother with fops->owner
filesystem module as whole is pinned down by its superblock, no need
to have opened files on it to add anything to that.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Thu, 28 Mar 2013 17:00:31 +0000 (13:00 -0400)]
ccg: don't bother with fops->owner
filesystem module as whole is pinned down by its superblock, no need
to have opened files on it to add anything to that.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Thu, 28 Mar 2013 16:45:40 +0000 (12:45 -0400)]
spufs: don't bother with fops->owner
filesystem module as whole is pinned down by its superblock, no need
to have opened files on it to add anything to that.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Thu, 21 Mar 2013 15:06:46 +0000 (11:06 -0400)]
get rid of the last free_pipe_info() callers
and rename __free_pipe_info() to free_pipe_info()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Thu, 21 Mar 2013 15:04:15 +0000 (11:04 -0400)]
get rid of alloc_pipe_info() argument
not used anymore
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Thu, 21 Mar 2013 15:01:38 +0000 (11:01 -0400)]
get rid of pipe->inode
it's used only as a flag to distinguish normal pipes/FIFOs from the
internal per-task one used by file-to-file splice. And pipe->files
would work just as well for that purpose...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Thu, 21 Mar 2013 16:24:01 +0000 (12:24 -0400)]
introduce variants of pipe_lock/pipe_unlock for real pipes/FIFOs
fs/pipe.c file_operations methods *know* that pipe is not an internal one;
no need to check pipe->inode for those callers.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Thu, 21 Mar 2013 15:16:56 +0000 (11:16 -0400)]
pipe: set file->private_data to ->i_pipe
simplify get_pipe_info(), while we are at it
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Thu, 21 Mar 2013 06:32:24 +0000 (02:32 -0400)]
pipe: don't use ->i_mutex
now it can be done - put mutex into pipe_inode_info, use it instead
of ->i_mutex
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Thu, 21 Mar 2013 06:21:19 +0000 (02:21 -0400)]
pipe: take allocation and freeing of pipe_inode_info out of ->i_mutex
* new field - pipe->files; number of struct file over that pipe (all
sharing the same inode, of course); protected by inode->i_lock.
* pipe_release() decrements pipe->files, clears inode->i_pipe when
if the counter has reached 0 (all under ->i_lock) and, in that case,
frees pipe after having done pipe_unlock()
* fifo_open() starts with grabbing ->i_lock, and either bumps pipe->files
if ->i_pipe was non-NULL or allocates a new pipe (dropping and regaining
->i_lock) and rechecks ->i_pipe; if it's still NULL, inserts new pipe
there, otherwise bumps ->i_pipe->files and frees the one we'd allocated.
At that point we know that ->i_pipe is non-NULL and won't go away, so
we can do pipe_lock() on it and proceed as we used to. If we end up
failing, decrement pipe->files and if it reaches 0 clear ->i_pipe and
free the sucker after pipe_unlock().
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Thu, 21 Mar 2013 06:16:30 +0000 (02:16 -0400)]
pipe: preparation to new locking rules
* use the fact that file_inode(file)->i_pipe doesn't change
while the file is opened - no locks needed to access that.
* switch to pipe_lock/pipe_unlock where it's easy to do
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
projects / GitHub / LineageOS / G12 / android_kernel_amlogic_linux-4.9.git / log