<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux-toradex.git/arch/mips/include/asm/jump_label.h, branch v3.4.10</title>
<subtitle>Linux kernel for Apalis and Colibri modules</subtitle>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/'/>
<entry>
<title>static keys: Introduce 'struct static_key', static_key_true()/false() and static_key_slow_[inc|dec]()</title>
<updated>2012-02-24T09:05:59+00:00</updated>
<author>
<name>Ingo Molnar</name>
<email>mingo@elte.hu</email>
</author>
<published>2012-02-24T07:31:31+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=c5905afb0ee6550b42c49213da1c22d67316c194'/>
<id>c5905afb0ee6550b42c49213da1c22d67316c194</id>
<content type='text'>
So here's a boot tested patch on top of Jason's series that does
all the cleanups I talked about and turns jump labels into a
more intuitive to use facility. It should also address the
various misconceptions and confusions that surround jump labels.

Typical usage scenarios:

        #include &lt;linux/static_key.h&gt;

        struct static_key key = STATIC_KEY_INIT_TRUE;

        if (static_key_false(&amp;key))
                do unlikely code
        else
                do likely code

Or:

        if (static_key_true(&amp;key))
                do likely code
        else
                do unlikely code

The static key is modified via:

        static_key_slow_inc(&amp;key);
        ...
        static_key_slow_dec(&amp;key);

The 'slow' prefix makes it abundantly clear that this is an
expensive operation.

I've updated all in-kernel code to use this everywhere. Note
that I (intentionally) have not pushed through the rename
blindly through to the lowest levels: the actual jump-label
patching arch facility should be named like that, so we want to
decouple jump labels from the static-key facility a bit.

On non-jump-label enabled architectures static keys default to
likely()/unlikely() branches.

Signed-off-by: Ingo Molnar &lt;mingo@elte.hu&gt;
Acked-by: Jason Baron &lt;jbaron@redhat.com&gt;
Acked-by: Steven Rostedt &lt;rostedt@goodmis.org&gt;
Cc: a.p.zijlstra@chello.nl
Cc: mathieu.desnoyers@efficios.com
Cc: davem@davemloft.net
Cc: ddaney.cavm@gmail.com
Cc: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
Link: http://lkml.kernel.org/r/20120222085809.GA26397@elte.hu
Signed-off-by: Ingo Molnar &lt;mingo@elte.hu&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
So here's a boot tested patch on top of Jason's series that does
all the cleanups I talked about and turns jump labels into a
more intuitive to use facility. It should also address the
various misconceptions and confusions that surround jump labels.

Typical usage scenarios:

        #include &lt;linux/static_key.h&gt;

        struct static_key key = STATIC_KEY_INIT_TRUE;

        if (static_key_false(&amp;key))
                do unlikely code
        else
                do likely code

Or:

        if (static_key_true(&amp;key))
                do likely code
        else
                do unlikely code

The static key is modified via:

        static_key_slow_inc(&amp;key);
        ...
        static_key_slow_dec(&amp;key);

The 'slow' prefix makes it abundantly clear that this is an
expensive operation.

I've updated all in-kernel code to use this everywhere. Note
that I (intentionally) have not pushed through the rename
blindly through to the lowest levels: the actual jump-label
patching arch facility should be named like that, so we want to
decouple jump labels from the static-key facility a bit.

On non-jump-label enabled architectures static keys default to
likely()/unlikely() branches.

Signed-off-by: Ingo Molnar &lt;mingo@elte.hu&gt;
Acked-by: Jason Baron &lt;jbaron@redhat.com&gt;
Acked-by: Steven Rostedt &lt;rostedt@goodmis.org&gt;
Cc: a.p.zijlstra@chello.nl
Cc: mathieu.desnoyers@efficios.com
Cc: davem@davemloft.net
Cc: ddaney.cavm@gmail.com
Cc: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
Link: http://lkml.kernel.org/r/20120222085809.GA26397@elte.hu
Signed-off-by: Ingo Molnar &lt;mingo@elte.hu&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>jump label: Introduce static_branch() interface</title>
<updated>2011-04-04T16:48:08+00:00</updated>
<author>
<name>Jason Baron</name>
<email>jbaron@redhat.com</email>
</author>
<published>2011-03-16T21:29:47+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=d430d3d7e646eb1eac2bb4aa244a644312e67c76'/>
<id>d430d3d7e646eb1eac2bb4aa244a644312e67c76</id>
<content type='text'>
Introduce:

static __always_inline bool static_branch(struct jump_label_key *key);

instead of the old JUMP_LABEL(key, label) macro.

In this way, jump labels become really easy to use:

Define:

        struct jump_label_key jump_key;

Can be used as:

        if (static_branch(&amp;jump_key))
                do unlikely code

enable/disale via:

        jump_label_inc(&amp;jump_key);
        jump_label_dec(&amp;jump_key);

that's it!

For the jump labels disabled case, the static_branch() becomes an
atomic_read(), and jump_label_inc()/dec() are simply atomic_inc(),
atomic_dec() operations. We show testing results for this change below.

Thanks to H. Peter Anvin for suggesting the 'static_branch()' construct.

Since we now require a 'struct jump_label_key *key', we can store a pointer into
the jump table addresses. In this way, we can enable/disable jump labels, in
basically constant time. This change allows us to completely remove the previous
hashtable scheme. Thanks to Peter Zijlstra for this re-write.

Testing:

I ran a series of 'tbench 20' runs 5 times (with reboots) for 3
configurations, where tracepoints were disabled.

jump label configured in
avg: 815.6

jump label *not* configured in (using atomic reads)
avg: 800.1

jump label *not* configured in (regular reads)
avg: 803.4

Signed-off-by: Peter Zijlstra &lt;a.p.zijlstra@chello.nl&gt;
LKML-Reference: &lt;20110316212947.GA8792@redhat.com&gt;
Signed-off-by: Jason Baron &lt;jbaron@redhat.com&gt;
Suggested-by: H. Peter Anvin &lt;hpa@linux.intel.com&gt;
Tested-by: David Daney &lt;ddaney@caviumnetworks.com&gt;
Acked-by: Ralf Baechle &lt;ralf@linux-mips.org&gt;
Acked-by: David S. Miller &lt;davem@davemloft.net&gt;
Acked-by: Mathieu Desnoyers &lt;mathieu.desnoyers@efficios.com&gt;
Signed-off-by: Steven Rostedt &lt;rostedt@goodmis.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Introduce:

static __always_inline bool static_branch(struct jump_label_key *key);

instead of the old JUMP_LABEL(key, label) macro.

In this way, jump labels become really easy to use:

Define:

        struct jump_label_key jump_key;

Can be used as:

        if (static_branch(&amp;jump_key))
                do unlikely code

enable/disale via:

        jump_label_inc(&amp;jump_key);
        jump_label_dec(&amp;jump_key);

that's it!

For the jump labels disabled case, the static_branch() becomes an
atomic_read(), and jump_label_inc()/dec() are simply atomic_inc(),
atomic_dec() operations. We show testing results for this change below.

Thanks to H. Peter Anvin for suggesting the 'static_branch()' construct.

Since we now require a 'struct jump_label_key *key', we can store a pointer into
the jump table addresses. In this way, we can enable/disable jump labels, in
basically constant time. This change allows us to completely remove the previous
hashtable scheme. Thanks to Peter Zijlstra for this re-write.

Testing:

I ran a series of 'tbench 20' runs 5 times (with reboots) for 3
configurations, where tracepoints were disabled.

jump label configured in
avg: 815.6

jump label *not* configured in (using atomic reads)
avg: 800.1

jump label *not* configured in (regular reads)
avg: 803.4

Signed-off-by: Peter Zijlstra &lt;a.p.zijlstra@chello.nl&gt;
LKML-Reference: &lt;20110316212947.GA8792@redhat.com&gt;
Signed-off-by: Jason Baron &lt;jbaron@redhat.com&gt;
Suggested-by: H. Peter Anvin &lt;hpa@linux.intel.com&gt;
Tested-by: David Daney &lt;ddaney@caviumnetworks.com&gt;
Acked-by: Ralf Baechle &lt;ralf@linux-mips.org&gt;
Acked-by: David S. Miller &lt;davem@davemloft.net&gt;
Acked-by: Mathieu Desnoyers &lt;mathieu.desnoyers@efficios.com&gt;
Signed-off-by: Steven Rostedt &lt;rostedt@goodmis.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>MIPS: jump label: Add MIPS support.</title>
<updated>2011-01-18T18:30:24+00:00</updated>
<author>
<name>David Daney</name>
<email>ddaney@caviumnetworks.com</email>
</author>
<published>2010-12-28T21:26:23+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=94bb0c1ab293c298a8852e4f10c4215bad6daa9b'/>
<id>94bb0c1ab293c298a8852e4f10c4215bad6daa9b</id>
<content type='text'>
In order not to be left behind, we add jump label support for MIPS.

Tested on 64-bit big endian (Octeon), and 32-bit little endian
(malta/qemu).

Signed-off-by: David Daney &lt;ddaney@caviumnetworks.com&gt;
To: linux-mips@linux-mips.org
Cc: Steven Rostedt &lt;rostedt@goodmis.org&gt;
Cc: Jason Baron &lt;jbaron@redhat.com&gt;
Patchwork: https://patchwork.linux-mips.org/patch/1923/
Signed-off-by: Ralf Baechle &lt;ralf@linux-mips.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
In order not to be left behind, we add jump label support for MIPS.

Tested on 64-bit big endian (Octeon), and 32-bit little endian
(malta/qemu).

Signed-off-by: David Daney &lt;ddaney@caviumnetworks.com&gt;
To: linux-mips@linux-mips.org
Cc: Steven Rostedt &lt;rostedt@goodmis.org&gt;
Cc: Jason Baron &lt;jbaron@redhat.com&gt;
Patchwork: https://patchwork.linux-mips.org/patch/1923/
Signed-off-by: Ralf Baechle &lt;ralf@linux-mips.org&gt;
</pre>
</div>
</content>
</entry>
</feed>
