<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux.git/fs/eventfd.c, branch v4.8</title>
<subtitle>Linux kernel source tree</subtitle>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/'/>
<entry>
<title>eventfd: document lockless access in eventfd_poll</title>
<updated>2016-03-22T22:36:02+00:00</updated>
<author>
<name>Paolo Bonzini</name>
<email>pbonzini@redhat.com</email>
</author>
<published>2016-03-22T21:27:14+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=a484c3dd9426e1f450c7ed8ad5a2002ea1b309ea'/>
<id>a484c3dd9426e1f450c7ed8ad5a2002ea1b309ea</id>
<content type='text'>
Since commit e22553e2a25e ("eventfd: don't take the spinlock in
eventfd_poll", 2015-02-17), eventfd is reading ctx-&gt;count outside
ctx-&gt;wqh.lock.

However, things aren't as simple as the read barrier in eventfd_poll
would suggest.  In fact, the read barrier, besides lacking a comment, is
not paired in any obvious manner with another read barrier, and it is
pointless because it is sitting between a write (deep in poll_wait) and
the read of ctx-&gt;count.  The read barrier is acting just as a compiler
barrier, for which we can use READ_ONCE instead.  This is what the code
change in this patch does.

The documentation change is just as important, however.  The question,
posed by Andrea Arcangeli, is then why the thing is safe on
architectures where spin_unlock does not imply a store-load memory
barrier.  The answer is that it's safe because writes of ctx-&gt;count use
the same lock as poll_wait, and hence an acquire barrier implicit in
poll_wait provides the necessary synchronization between eventfd_poll
and callers of wake_up_locked_poll.  This is sort of mentioned in the
commit message with respect to eventfd_ctx_read ("eventfd_read is
similar, it will do a single decrement with the lock held") but it
applies to all other callers too.  It's tricky enough that it should be
documented in the code.

Signed-off-by: Paolo Bonzini &lt;pbonzini@redhat.com&gt;
Reviewed-by: Andrea Arcangeli &lt;aarcange@redhat.com&gt;
Cc: Chris Mason &lt;clm@fb.com&gt;
Cc: Davide Libenzi &lt;davidel@xmailserver.org&gt;
Signed-off-by: Andrew Morton &lt;akpm@linux-foundation.org&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Since commit e22553e2a25e ("eventfd: don't take the spinlock in
eventfd_poll", 2015-02-17), eventfd is reading ctx-&gt;count outside
ctx-&gt;wqh.lock.

However, things aren't as simple as the read barrier in eventfd_poll
would suggest.  In fact, the read barrier, besides lacking a comment, is
not paired in any obvious manner with another read barrier, and it is
pointless because it is sitting between a write (deep in poll_wait) and
the read of ctx-&gt;count.  The read barrier is acting just as a compiler
barrier, for which we can use READ_ONCE instead.  This is what the code
change in this patch does.

The documentation change is just as important, however.  The question,
posed by Andrea Arcangeli, is then why the thing is safe on
architectures where spin_unlock does not imply a store-load memory
barrier.  The answer is that it's safe because writes of ctx-&gt;count use
the same lock as poll_wait, and hence an acquire barrier implicit in
poll_wait provides the necessary synchronization between eventfd_poll
and callers of wake_up_locked_poll.  This is sort of mentioned in the
commit message with respect to eventfd_ctx_read ("eventfd_read is
similar, it will do a single decrement with the lock held") but it
applies to all other callers too.  It's tricky enough that it should be
documented in the code.

Signed-off-by: Paolo Bonzini &lt;pbonzini@redhat.com&gt;
Reviewed-by: Andrea Arcangeli &lt;aarcange@redhat.com&gt;
Cc: Chris Mason &lt;clm@fb.com&gt;
Cc: Davide Libenzi &lt;davidel@xmailserver.org&gt;
Signed-off-by: Andrew Morton &lt;akpm@linux-foundation.org&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>Documentation: filesystem: Fix typo in fs/eventfd.c</title>
<updated>2015-12-08T13:52:03+00:00</updated>
<author>
<name>Masanari Iida</name>
<email>standby24x7@gmail.com</email>
</author>
<published>2015-09-22T03:04:17+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=20d5a865e1f1d98e8520c12fc647c9d4e47f46b3'/>
<id>20d5a865e1f1d98e8520c12fc647c9d4e47f46b3</id>
<content type='text'>
This patch fix typos found in Documentation/filesystems.xml,
DocBook/filesystems/API-eventfd-signal.html, and
DocBook/filesystems.aux.xml

These files are generated from comments within the source,
so I had to fix typos in fs/eventfd.c

Signed-off-by: Masanari Iida &lt;standby24x7@gmail.com&gt;
Signed-off-by: Jiri Kosina &lt;jkosina@suse.cz&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This patch fix typos found in Documentation/filesystems.xml,
DocBook/filesystems/API-eventfd-signal.html, and
DocBook/filesystems.aux.xml

These files are generated from comments within the source,
so I had to fix typos in fs/eventfd.c

Signed-off-by: Masanari Iida &lt;standby24x7@gmail.com&gt;
Signed-off-by: Jiri Kosina &lt;jkosina@suse.cz&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>eventfd: don't take the spinlock in eventfd_poll</title>
<updated>2015-02-17T22:34:52+00:00</updated>
<author>
<name>Chris Mason</name>
<email>clm@fb.com</email>
</author>
<published>2015-02-17T21:46:07+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=e22553e2a25ed3f2a9c874088e0f20cdcd97c7b0'/>
<id>e22553e2a25ed3f2a9c874088e0f20cdcd97c7b0</id>
<content type='text'>
The spinlock in eventfd_poll is trying to protect the count of events so
it can decide if it should return POLLIN, POLLERR, or POLLOUT.  But,
because of the way we drop the lock after calling poll_wait, and drop it
again before returning, we have the same pile of races with the lock as
we do with a single read of ctx-&gt;count().

This replaces the lock with a read barrier and single read.

eventfd_write does a single bump of ctx-&gt;count, so this should not add
new races with adding events.  eventfd_read is similar, it will do a
single decrement with the lock held, and so we're making the race with
concurrent readers slightly larger.

This spinlock is the top CPU user in kernel code during one of our
workloads.  Removing it gives us a ~2% boost.

[arnd@arndb.de: avoid unused variable warning]
[dan.carpenter@oracle.com: type bug in eventfd_poll()]
Signed-off-by: Chris Mason &lt;clm@fb.com&gt;
Cc: Davide Libenzi &lt;davidel@xmailserver.org&gt;
Signed-off-by: Arnd Bergmann &lt;arnd@arndb.de&gt;
Signed-off-by: Dan Carpenter &lt;dan.carpenter@oracle.com&gt;
Signed-off-by: Andrew Morton &lt;akpm@linux-foundation.org&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
The spinlock in eventfd_poll is trying to protect the count of events so
it can decide if it should return POLLIN, POLLERR, or POLLOUT.  But,
because of the way we drop the lock after calling poll_wait, and drop it
again before returning, we have the same pile of races with the lock as
we do with a single read of ctx-&gt;count().

This replaces the lock with a read barrier and single read.

eventfd_write does a single bump of ctx-&gt;count, so this should not add
new races with adding events.  eventfd_read is similar, it will do a
single decrement with the lock held, and so we're making the race with
concurrent readers slightly larger.

This spinlock is the top CPU user in kernel code during one of our
workloads.  Removing it gives us a ~2% boost.

[arnd@arndb.de: avoid unused variable warning]
[dan.carpenter@oracle.com: type bug in eventfd_poll()]
Signed-off-by: Chris Mason &lt;clm@fb.com&gt;
Cc: Davide Libenzi &lt;davidel@xmailserver.org&gt;
Signed-off-by: Arnd Bergmann &lt;arnd@arndb.de&gt;
Signed-off-by: Dan Carpenter &lt;dan.carpenter@oracle.com&gt;
Signed-off-by: Andrew Morton &lt;akpm@linux-foundation.org&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>fs: Convert show_fdinfo functions to void</title>
<updated>2014-11-05T19:13:23+00:00</updated>
<author>
<name>Joe Perches</name>
<email>joe@perches.com</email>
</author>
<published>2014-09-29T23:08:25+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=a3816ab0e8fe542a89a53b82506a8ddac063fbe3'/>
<id>a3816ab0e8fe542a89a53b82506a8ddac063fbe3</id>
<content type='text'>
seq_printf functions shouldn't really check the return value.
Checking seq_has_overflowed() occasionally is used instead.

Update vfs documentation.

Link: http://lkml.kernel.org/p/e37e6e7b76acbdcc3bb4ab2a57c8f8ca1ae11b9a.1412031505.git.joe@perches.com

Cc: David S. Miller &lt;davem@davemloft.net&gt;
Cc: Al Viro &lt;viro@zeniv.linux.org.uk&gt;
Signed-off-by: Joe Perches &lt;joe@perches.com&gt;
[ did a few clean ups ]
Signed-off-by: Steven Rostedt &lt;rostedt@goodmis.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
seq_printf functions shouldn't really check the return value.
Checking seq_has_overflowed() occasionally is used instead.

Update vfs documentation.

Link: http://lkml.kernel.org/p/e37e6e7b76acbdcc3bb4ab2a57c8f8ca1ae11b9a.1412031505.git.joe@perches.com

Cc: David S. Miller &lt;davem@davemloft.net&gt;
Cc: Al Viro &lt;viro@zeniv.linux.org.uk&gt;
Signed-off-by: Joe Perches &lt;joe@perches.com&gt;
[ did a few clean ups ]
Signed-off-by: Steven Rostedt &lt;rostedt@goodmis.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>eventfd_ctx_fdget(): use fdget() instead of fget()</title>
<updated>2014-01-25T08:13:04+00:00</updated>
<author>
<name>Al Viro</name>
<email>viro@zeniv.linux.org.uk</email>
</author>
<published>2013-12-23T21:51:33+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=36a7411724b1caf2fa92b5e4a41576ee8f16769e'/>
<id>36a7411724b1caf2fa92b5e4a41576ee8f16769e</id>
<content type='text'>
Signed-off-by: Al Viro &lt;viro@zeniv.linux.org.uk&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Signed-off-by: Al Viro &lt;viro@zeniv.linux.org.uk&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>fs, eventfd: add procfs fdinfo helper</title>
<updated>2012-12-18T01:15:27+00:00</updated>
<author>
<name>Cyrill Gorcunov</name>
<email>gorcunov@openvz.org</email>
</author>
<published>2012-12-18T00:04:57+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=cbac5542d48127b546a23d816380a7926eee1c25'/>
<id>cbac5542d48127b546a23d816380a7926eee1c25</id>
<content type='text'>
This allows us to print out raw counter value.  The /proc/pid/fdinfo/fd
output is

 | pos:	0
 | flags:	04002
 | eventfd-count:               5a

Signed-off-by: Cyrill Gorcunov &lt;gorcunov@openvz.org&gt;
Acked-by: Pavel Emelyanov &lt;xemul@parallels.com&gt;
Cc: Oleg Nesterov &lt;oleg@redhat.com&gt;
Cc: Andrey Vagin &lt;avagin@openvz.org&gt;
Cc: Al Viro &lt;viro@ZenIV.linux.org.uk&gt;
Cc: Alexey Dobriyan &lt;adobriyan@gmail.com&gt;
Cc: James Bottomley &lt;jbottomley@parallels.com&gt;
Cc: "Aneesh Kumar K.V" &lt;aneesh.kumar@linux.vnet.ibm.com&gt;
Cc: Alexey Dobriyan &lt;adobriyan@gmail.com&gt;
Cc: Matthew Helsley &lt;matt.helsley@gmail.com&gt;
Cc: "J. Bruce Fields" &lt;bfields@fieldses.org&gt;
Cc: "Aneesh Kumar K.V" &lt;aneesh.kumar@linux.vnet.ibm.com&gt;
Cc: Tvrtko Ursulin &lt;tvrtko.ursulin@onelan.co.uk&gt;
Signed-off-by: Andrew Morton &lt;akpm@linux-foundation.org&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This allows us to print out raw counter value.  The /proc/pid/fdinfo/fd
output is

 | pos:	0
 | flags:	04002
 | eventfd-count:               5a

Signed-off-by: Cyrill Gorcunov &lt;gorcunov@openvz.org&gt;
Acked-by: Pavel Emelyanov &lt;xemul@parallels.com&gt;
Cc: Oleg Nesterov &lt;oleg@redhat.com&gt;
Cc: Andrey Vagin &lt;avagin@openvz.org&gt;
Cc: Al Viro &lt;viro@ZenIV.linux.org.uk&gt;
Cc: Alexey Dobriyan &lt;adobriyan@gmail.com&gt;
Cc: James Bottomley &lt;jbottomley@parallels.com&gt;
Cc: "Aneesh Kumar K.V" &lt;aneesh.kumar@linux.vnet.ibm.com&gt;
Cc: Alexey Dobriyan &lt;adobriyan@gmail.com&gt;
Cc: Matthew Helsley &lt;matt.helsley@gmail.com&gt;
Cc: "J. Bruce Fields" &lt;bfields@fieldses.org&gt;
Cc: "Aneesh Kumar K.V" &lt;aneesh.kumar@linux.vnet.ibm.com&gt;
Cc: Tvrtko Ursulin &lt;tvrtko.ursulin@onelan.co.uk&gt;
Signed-off-by: Andrew Morton &lt;akpm@linux-foundation.org&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>eventfd: change int to __u64 in eventfd_signal()</title>
<updated>2012-06-01T00:49:32+00:00</updated>
<author>
<name>Sha Zhengju</name>
<email>handai.szj@taobao.com</email>
</author>
<published>2012-05-31T23:26:41+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=ee62c6b2dc93c09585b51fad18449dc5edb9977f'/>
<id>ee62c6b2dc93c09585b51fad18449dc5edb9977f</id>
<content type='text'>
eventfd_ctx-&gt;count is an __u64 counter which is allowed to reach
ULLONG_MAX.  eventfd_write() adds a __u64 value to "count", but the kernel
side eventfd_signal() only adds an int value to it.  Make them consistent.

[akpm@linux-foundation.org: update interface documentation]
Signed-off-by: Sha Zhengju &lt;handai.szj@taobao.com&gt;
Cc: Davide Libenzi &lt;davidel@xmailserver.org&gt;
Signed-off-by: Andrew Morton &lt;akpm@linux-foundation.org&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
eventfd_ctx-&gt;count is an __u64 counter which is allowed to reach
ULLONG_MAX.  eventfd_write() adds a __u64 value to "count", but the kernel
side eventfd_signal() only adds an int value to it.  Make them consistent.

[akpm@linux-foundation.org: update interface documentation]
Signed-off-by: Sha Zhengju &lt;handai.szj@taobao.com&gt;
Cc: Davide Libenzi &lt;davidel@xmailserver.org&gt;
Signed-off-by: Andrew Morton &lt;akpm@linux-foundation.org&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>fs: reduce the use of module.h wherever possible</title>
<updated>2012-02-29T00:31:58+00:00</updated>
<author>
<name>Paul Gortmaker</name>
<email>paul.gortmaker@windriver.com</email>
</author>
<published>2011-11-17T04:57:37+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=630d9c47274aa89bfa77fe6556d7818bdcb12992'/>
<id>630d9c47274aa89bfa77fe6556d7818bdcb12992</id>
<content type='text'>
For files only using THIS_MODULE and/or EXPORT_SYMBOL, map
them onto including export.h -- or if the file isn't even
using those, then just delete the include.  Fix up any implicit
include dependencies that were being masked by module.h along
the way.

Signed-off-by: Paul Gortmaker &lt;paul.gortmaker@windriver.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
For files only using THIS_MODULE and/or EXPORT_SYMBOL, map
them onto including export.h -- or if the file isn't even
using those, then just delete the include.  Fix up any implicit
include dependencies that were being masked by module.h along
the way.

Signed-off-by: Paul Gortmaker &lt;paul.gortmaker@windriver.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>Docbook: add fs/eventfd.c and fix typos in it</title>
<updated>2011-02-21T23:07:04+00:00</updated>
<author>
<name>Randy Dunlap</name>
<email>randy.dunlap@oracle.com</email>
</author>
<published>2011-02-21T04:08:35+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=361821854b71fc3a53c9e17701538247bddbd4ba'/>
<id>361821854b71fc3a53c9e17701538247bddbd4ba</id>
<content type='text'>
Add fs/eventfd.c to filesystems docbook.
Make typo corrections in fs/eventfd.c.

Signed-off-by: Randy Dunlap &lt;randy.dunlap@oracle.com&gt;
Cc: Davide Libenzi &lt;davidel@xmailserver.org&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Add fs/eventfd.c to filesystems docbook.
Make typo corrections in fs/eventfd.c.

Signed-off-by: Randy Dunlap &lt;randy.dunlap@oracle.com&gt;
Cc: Davide Libenzi &lt;davidel@xmailserver.org&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>llseek: automatically add .llseek fop</title>
<updated>2010-10-15T13:53:27+00:00</updated>
<author>
<name>Arnd Bergmann</name>
<email>arnd@arndb.de</email>
</author>
<published>2010-08-15T16:52:59+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=6038f373a3dc1f1c26496e60b6c40b164716f07e'/>
<id>6038f373a3dc1f1c26496e60b6c40b164716f07e</id>
<content type='text'>
All file_operations should get a .llseek operation so we can make
nonseekable_open the default for future file operations without a
.llseek pointer.

The three cases that we can automatically detect are no_llseek, seq_lseek
and default_llseek. For cases where we can we can automatically prove that
the file offset is always ignored, we use noop_llseek, which maintains
the current behavior of not returning an error from a seek.

New drivers should normally not use noop_llseek but instead use no_llseek
and call nonseekable_open at open time.  Existing drivers can be converted
to do the same when the maintainer knows for certain that no user code
relies on calling seek on the device file.

The generated code is often incorrectly indented and right now contains
comments that clarify for each added line why a specific variant was
chosen. In the version that gets submitted upstream, the comments will
be gone and I will manually fix the indentation, because there does not
seem to be a way to do that using coccinelle.

Some amount of new code is currently sitting in linux-next that should get
the same modifications, which I will do at the end of the merge window.

Many thanks to Julia Lawall for helping me learn to write a semantic
patch that does all this.

===== begin semantic patch =====
// This adds an llseek= method to all file operations,
// as a preparation for making no_llseek the default.
//
// The rules are
// - use no_llseek explicitly if we do nonseekable_open
// - use seq_lseek for sequential files
// - use default_llseek if we know we access f_pos
// - use noop_llseek if we know we don't access f_pos,
//   but we still want to allow users to call lseek
//
@ open1 exists @
identifier nested_open;
@@
nested_open(...)
{
&lt;+...
nonseekable_open(...)
...+&gt;
}

@ open exists@
identifier open_f;
identifier i, f;
identifier open1.nested_open;
@@
int open_f(struct inode *i, struct file *f)
{
&lt;+...
(
nonseekable_open(...)
|
nested_open(...)
)
...+&gt;
}

@ read disable optional_qualifier exists @
identifier read_f;
identifier f, p, s, off;
type ssize_t, size_t, loff_t;
expression E;
identifier func;
@@
ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off)
{
&lt;+...
(
   *off = E
|
   *off += E
|
   func(..., off, ...)
|
   E = *off
)
...+&gt;
}

@ read_no_fpos disable optional_qualifier exists @
identifier read_f;
identifier f, p, s, off;
type ssize_t, size_t, loff_t;
@@
ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off)
{
... when != off
}

@ write @
identifier write_f;
identifier f, p, s, off;
type ssize_t, size_t, loff_t;
expression E;
identifier func;
@@
ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off)
{
&lt;+...
(
  *off = E
|
  *off += E
|
  func(..., off, ...)
|
  E = *off
)
...+&gt;
}

@ write_no_fpos @
identifier write_f;
identifier f, p, s, off;
type ssize_t, size_t, loff_t;
@@
ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off)
{
... when != off
}

@ fops0 @
identifier fops;
@@
struct file_operations fops = {
 ...
};

@ has_llseek depends on fops0 @
identifier fops0.fops;
identifier llseek_f;
@@
struct file_operations fops = {
...
 .llseek = llseek_f,
...
};

@ has_read depends on fops0 @
identifier fops0.fops;
identifier read_f;
@@
struct file_operations fops = {
...
 .read = read_f,
...
};

@ has_write depends on fops0 @
identifier fops0.fops;
identifier write_f;
@@
struct file_operations fops = {
...
 .write = write_f,
...
};

@ has_open depends on fops0 @
identifier fops0.fops;
identifier open_f;
@@
struct file_operations fops = {
...
 .open = open_f,
...
};

// use no_llseek if we call nonseekable_open
////////////////////////////////////////////
@ nonseekable1 depends on !has_llseek &amp;&amp; has_open @
identifier fops0.fops;
identifier nso ~= "nonseekable_open";
@@
struct file_operations fops = {
...  .open = nso, ...
+.llseek = no_llseek, /* nonseekable */
};

@ nonseekable2 depends on !has_llseek @
identifier fops0.fops;
identifier open.open_f;
@@
struct file_operations fops = {
...  .open = open_f, ...
+.llseek = no_llseek, /* open uses nonseekable */
};

// use seq_lseek for sequential files
/////////////////////////////////////
@ seq depends on !has_llseek @
identifier fops0.fops;
identifier sr ~= "seq_read";
@@
struct file_operations fops = {
...  .read = sr, ...
+.llseek = seq_lseek, /* we have seq_read */
};

// use default_llseek if there is a readdir
///////////////////////////////////////////
@ fops1 depends on !has_llseek &amp;&amp; !nonseekable1 &amp;&amp; !nonseekable2 &amp;&amp; !seq @
identifier fops0.fops;
identifier readdir_e;
@@
// any other fop is used that changes pos
struct file_operations fops = {
... .readdir = readdir_e, ...
+.llseek = default_llseek, /* readdir is present */
};

// use default_llseek if at least one of read/write touches f_pos
/////////////////////////////////////////////////////////////////
@ fops2 depends on !fops1 &amp;&amp; !has_llseek &amp;&amp; !nonseekable1 &amp;&amp; !nonseekable2 &amp;&amp; !seq @
identifier fops0.fops;
identifier read.read_f;
@@
// read fops use offset
struct file_operations fops = {
... .read = read_f, ...
+.llseek = default_llseek, /* read accesses f_pos */
};

@ fops3 depends on !fops1 &amp;&amp; !fops2 &amp;&amp; !has_llseek &amp;&amp; !nonseekable1 &amp;&amp; !nonseekable2 &amp;&amp; !seq @
identifier fops0.fops;
identifier write.write_f;
@@
// write fops use offset
struct file_operations fops = {
... .write = write_f, ...
+	.llseek = default_llseek, /* write accesses f_pos */
};

// Use noop_llseek if neither read nor write accesses f_pos
///////////////////////////////////////////////////////////

@ fops4 depends on !fops1 &amp;&amp; !fops2 &amp;&amp; !fops3 &amp;&amp; !has_llseek &amp;&amp; !nonseekable1 &amp;&amp; !nonseekable2 &amp;&amp; !seq @
identifier fops0.fops;
identifier read_no_fpos.read_f;
identifier write_no_fpos.write_f;
@@
// write fops use offset
struct file_operations fops = {
...
 .write = write_f,
 .read = read_f,
...
+.llseek = noop_llseek, /* read and write both use no f_pos */
};

@ depends on has_write &amp;&amp; !has_read &amp;&amp; !fops1 &amp;&amp; !fops2 &amp;&amp; !has_llseek &amp;&amp; !nonseekable1 &amp;&amp; !nonseekable2 &amp;&amp; !seq @
identifier fops0.fops;
identifier write_no_fpos.write_f;
@@
struct file_operations fops = {
... .write = write_f, ...
+.llseek = noop_llseek, /* write uses no f_pos */
};

@ depends on has_read &amp;&amp; !has_write &amp;&amp; !fops1 &amp;&amp; !fops2 &amp;&amp; !has_llseek &amp;&amp; !nonseekable1 &amp;&amp; !nonseekable2 &amp;&amp; !seq @
identifier fops0.fops;
identifier read_no_fpos.read_f;
@@
struct file_operations fops = {
... .read = read_f, ...
+.llseek = noop_llseek, /* read uses no f_pos */
};

@ depends on !has_read &amp;&amp; !has_write &amp;&amp; !fops1 &amp;&amp; !fops2 &amp;&amp; !has_llseek &amp;&amp; !nonseekable1 &amp;&amp; !nonseekable2 &amp;&amp; !seq @
identifier fops0.fops;
@@
struct file_operations fops = {
...
+.llseek = noop_llseek, /* no read or write fn */
};
===== End semantic patch =====

Signed-off-by: Arnd Bergmann &lt;arnd@arndb.de&gt;
Cc: Julia Lawall &lt;julia@diku.dk&gt;
Cc: Christoph Hellwig &lt;hch@infradead.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
All file_operations should get a .llseek operation so we can make
nonseekable_open the default for future file operations without a
.llseek pointer.

The three cases that we can automatically detect are no_llseek, seq_lseek
and default_llseek. For cases where we can we can automatically prove that
the file offset is always ignored, we use noop_llseek, which maintains
the current behavior of not returning an error from a seek.

New drivers should normally not use noop_llseek but instead use no_llseek
and call nonseekable_open at open time.  Existing drivers can be converted
to do the same when the maintainer knows for certain that no user code
relies on calling seek on the device file.

The generated code is often incorrectly indented and right now contains
comments that clarify for each added line why a specific variant was
chosen. In the version that gets submitted upstream, the comments will
be gone and I will manually fix the indentation, because there does not
seem to be a way to do that using coccinelle.

Some amount of new code is currently sitting in linux-next that should get
the same modifications, which I will do at the end of the merge window.

Many thanks to Julia Lawall for helping me learn to write a semantic
patch that does all this.

===== begin semantic patch =====
// This adds an llseek= method to all file operations,
// as a preparation for making no_llseek the default.
//
// The rules are
// - use no_llseek explicitly if we do nonseekable_open
// - use seq_lseek for sequential files
// - use default_llseek if we know we access f_pos
// - use noop_llseek if we know we don't access f_pos,
//   but we still want to allow users to call lseek
//
@ open1 exists @
identifier nested_open;
@@
nested_open(...)
{
&lt;+...
nonseekable_open(...)
...+&gt;
}

@ open exists@
identifier open_f;
identifier i, f;
identifier open1.nested_open;
@@
int open_f(struct inode *i, struct file *f)
{
&lt;+...
(
nonseekable_open(...)
|
nested_open(...)
)
...+&gt;
}

@ read disable optional_qualifier exists @
identifier read_f;
identifier f, p, s, off;
type ssize_t, size_t, loff_t;
expression E;
identifier func;
@@
ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off)
{
&lt;+...
(
   *off = E
|
   *off += E
|
   func(..., off, ...)
|
   E = *off
)
...+&gt;
}

@ read_no_fpos disable optional_qualifier exists @
identifier read_f;
identifier f, p, s, off;
type ssize_t, size_t, loff_t;
@@
ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off)
{
... when != off
}

@ write @
identifier write_f;
identifier f, p, s, off;
type ssize_t, size_t, loff_t;
expression E;
identifier func;
@@
ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off)
{
&lt;+...
(
  *off = E
|
  *off += E
|
  func(..., off, ...)
|
  E = *off
)
...+&gt;
}

@ write_no_fpos @
identifier write_f;
identifier f, p, s, off;
type ssize_t, size_t, loff_t;
@@
ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off)
{
... when != off
}

@ fops0 @
identifier fops;
@@
struct file_operations fops = {
 ...
};

@ has_llseek depends on fops0 @
identifier fops0.fops;
identifier llseek_f;
@@
struct file_operations fops = {
...
 .llseek = llseek_f,
...
};

@ has_read depends on fops0 @
identifier fops0.fops;
identifier read_f;
@@
struct file_operations fops = {
...
 .read = read_f,
...
};

@ has_write depends on fops0 @
identifier fops0.fops;
identifier write_f;
@@
struct file_operations fops = {
...
 .write = write_f,
...
};

@ has_open depends on fops0 @
identifier fops0.fops;
identifier open_f;
@@
struct file_operations fops = {
...
 .open = open_f,
...
};

// use no_llseek if we call nonseekable_open
////////////////////////////////////////////
@ nonseekable1 depends on !has_llseek &amp;&amp; has_open @
identifier fops0.fops;
identifier nso ~= "nonseekable_open";
@@
struct file_operations fops = {
...  .open = nso, ...
+.llseek = no_llseek, /* nonseekable */
};

@ nonseekable2 depends on !has_llseek @
identifier fops0.fops;
identifier open.open_f;
@@
struct file_operations fops = {
...  .open = open_f, ...
+.llseek = no_llseek, /* open uses nonseekable */
};

// use seq_lseek for sequential files
/////////////////////////////////////
@ seq depends on !has_llseek @
identifier fops0.fops;
identifier sr ~= "seq_read";
@@
struct file_operations fops = {
...  .read = sr, ...
+.llseek = seq_lseek, /* we have seq_read */
};

// use default_llseek if there is a readdir
///////////////////////////////////////////
@ fops1 depends on !has_llseek &amp;&amp; !nonseekable1 &amp;&amp; !nonseekable2 &amp;&amp; !seq @
identifier fops0.fops;
identifier readdir_e;
@@
// any other fop is used that changes pos
struct file_operations fops = {
... .readdir = readdir_e, ...
+.llseek = default_llseek, /* readdir is present */
};

// use default_llseek if at least one of read/write touches f_pos
/////////////////////////////////////////////////////////////////
@ fops2 depends on !fops1 &amp;&amp; !has_llseek &amp;&amp; !nonseekable1 &amp;&amp; !nonseekable2 &amp;&amp; !seq @
identifier fops0.fops;
identifier read.read_f;
@@
// read fops use offset
struct file_operations fops = {
... .read = read_f, ...
+.llseek = default_llseek, /* read accesses f_pos */
};

@ fops3 depends on !fops1 &amp;&amp; !fops2 &amp;&amp; !has_llseek &amp;&amp; !nonseekable1 &amp;&amp; !nonseekable2 &amp;&amp; !seq @
identifier fops0.fops;
identifier write.write_f;
@@
// write fops use offset
struct file_operations fops = {
... .write = write_f, ...
+	.llseek = default_llseek, /* write accesses f_pos */
};

// Use noop_llseek if neither read nor write accesses f_pos
///////////////////////////////////////////////////////////

@ fops4 depends on !fops1 &amp;&amp; !fops2 &amp;&amp; !fops3 &amp;&amp; !has_llseek &amp;&amp; !nonseekable1 &amp;&amp; !nonseekable2 &amp;&amp; !seq @
identifier fops0.fops;
identifier read_no_fpos.read_f;
identifier write_no_fpos.write_f;
@@
// write fops use offset
struct file_operations fops = {
...
 .write = write_f,
 .read = read_f,
...
+.llseek = noop_llseek, /* read and write both use no f_pos */
};

@ depends on has_write &amp;&amp; !has_read &amp;&amp; !fops1 &amp;&amp; !fops2 &amp;&amp; !has_llseek &amp;&amp; !nonseekable1 &amp;&amp; !nonseekable2 &amp;&amp; !seq @
identifier fops0.fops;
identifier write_no_fpos.write_f;
@@
struct file_operations fops = {
... .write = write_f, ...
+.llseek = noop_llseek, /* write uses no f_pos */
};

@ depends on has_read &amp;&amp; !has_write &amp;&amp; !fops1 &amp;&amp; !fops2 &amp;&amp; !has_llseek &amp;&amp; !nonseekable1 &amp;&amp; !nonseekable2 &amp;&amp; !seq @
identifier fops0.fops;
identifier read_no_fpos.read_f;
@@
struct file_operations fops = {
... .read = read_f, ...
+.llseek = noop_llseek, /* read uses no f_pos */
};

@ depends on !has_read &amp;&amp; !has_write &amp;&amp; !fops1 &amp;&amp; !fops2 &amp;&amp; !has_llseek &amp;&amp; !nonseekable1 &amp;&amp; !nonseekable2 &amp;&amp; !seq @
identifier fops0.fops;
@@
struct file_operations fops = {
...
+.llseek = noop_llseek, /* no read or write fn */
};
===== End semantic patch =====

Signed-off-by: Arnd Bergmann &lt;arnd@arndb.de&gt;
Cc: Julia Lawall &lt;julia@diku.dk&gt;
Cc: Christoph Hellwig &lt;hch@infradead.org&gt;
</pre>
</div>
</content>
</entry>
</feed>
