Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1199899

[PATCH 3/3] jiffies: Force inlining of {m,u}msecs_to_jiffies()

From Denys Vlasenko <dvlasenk@redhat.com>
Newsgroups linux.kernel
Subject [PATCH 3/3] jiffies: Force inlining of {m,u}msecs_to_jiffies()
Date 2015-08-04 16:20 +0200
Message-ID <pTLep-4wJ-13@gated-at.bofh.it> (permalink)
References <pTLep-4wJ-11@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


With this config:
http://busybox.net/~vda/kernel_config_OPTIMIZE_INLINING_and_Os
gcc-4.7.2 generates many copies of these tiny functions:

msecs_to_jiffies (45 copies):
55                      push   %rbp
48 89 e5                mov    %rsp,%rbp
e8 59 ec 03 00          callq  __msecs_to_jiffies
5d                      pop    %rbp
c3                      retq

usecs_to_jiffies (10 copies):
55                      push   %rbp
48 89 e5                mov    %rsp,%rbp
e8 5d 54 5e ff          callq  __usecs_to_jiffies
5d                      pop    %rbp
c3                      retq

See https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66122

This patch fixes this via s/inline/__always_inline/

    text     data      bss       dec     hex filename
86970954 17195912 36659200 140826066 864d5d2 vmlinux.before
86966150 17195912 36659200 140821262 864c30e vmlinux

Signed-off-by: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Thomas Graf <tgraf@suug.ch>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org
---
 include/linux/jiffies.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/linux/jiffies.h b/include/linux/jiffies.h
index 535fd3b..1ba48a1 100644
--- a/include/linux/jiffies.h
+++ b/include/linux/jiffies.h
@@ -351,7 +351,7 @@ static inline unsigned long _msecs_to_jiffies(const unsigned int m)
  * directly here and from __msecs_to_jiffies() in the case where
  * constant folding is not possible.
  */
-static inline unsigned long msecs_to_jiffies(const unsigned int m)
+static __always_inline unsigned long msecs_to_jiffies(const unsigned int m)
 {
 	if (__builtin_constant_p(m)) {
 		if ((int)m < 0)
@@ -405,7 +405,7 @@ static inline unsigned long _usecs_to_jiffies(const unsigned int u)
  * directly here and from __msecs_to_jiffies() in the case where
  * constant folding is not possible.
  */
-static inline unsigned long usecs_to_jiffies(const unsigned int u)
+static __always_inline unsigned long usecs_to_jiffies(const unsigned int u)
 {
 	if (__builtin_constant_p(u)) {
 		if (u > jiffies_to_usecs(MAX_JIFFY_OFFSET))
-- 
1.8.1.4

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Back to linux.kernel | Previous | NextNext in thread | Find similar | Unroll thread


Thread

[PATCH 3/3] jiffies: Force inlining of {m,u}msecs_to_jiffies() Denys Vlasenko <dvlasenk@redhat.com> - 2015-08-04 16:20 +0200
  [tip:core/types] jiffies: Force inlining of {m,u}msecs_to_jiffies  () tip-bot for Denys Vlasenko <tipbot@zytor.com> - 2015-08-05 22:20 +0200

csiph-web