The patch titled
Subject: x86/hweight: force inlining of __arch_hweight{32,64}()
has been added to the -mm tree. Its filename is
x86-hweight-force-inlining-of-__arch_hweight3264.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/x86-hweight-force-inlining-of-__arch_hweight3264.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/x86-hweight-force-inlining-of-__arch_hweight3264.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/SubmitChecklist when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Denys Vlasenko <dvlasenk@xxxxxxxxxx>
Subject: x86/hweight: force inlining of __arch_hweight{32,64}()
With this config:
http://busybox.net/~vda/kernel_config_OPTIMIZE_INLINING_and_Os
gcc-4.7.2 generates many copies of these tiny functions:
__arch_hweight32 (35 copies):
55 push %rbp
e8 66 9b 4a 00 callq __sw_hweight32
48 89 e5 mov %rsp,%rbp
5d pop %rbp
c3 retq
__arch_hweight64 (8 copies):
55 push %rbp
e8 5e c2 8a 00 callq __sw_hweight64
48 89 e5 mov %rsp,%rbp
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/
To avoid touching 32-bit case where such change was not tested to be a win,
reformat __arch_hweight64() to have completely disjoint 64-bit and 32-bit
implementations. IOW: made #ifdef / #else / #endif blocks contain
complete function definitions for 32 bits and 64 bits
instead of having #ifdef / #else / #endif inside a single function body.
Only 64-bit __arch_hweight64() is __always_inline'd.
text data bss dec hex filename
86971120 17195912 36659200 140826232 864d678 vmlinux.before
86970954 17195912 36659200 140826066 864d5d2 vmlinux
Signed-off-by: Denys Vlasenko <dvlasenk@xxxxxxxxxx>
Cc: Ingo Molnar <mingo@xxxxxxxxxx>
Cc: Thomas Graf <tgraf@xxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Cc: David Rientjes <rientjes@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
arch/x86/include/asm/arch_hweight.h | 13 ++++++++-----
1 file changed, 8 insertions(+), 5 deletions(-)
diff -puN arch/x86/include/asm/arch_hweight.h~x86-hweight-force-inlining-of-__arch_hweight3264 arch/x86/include/asm/arch_hweight.h
--- a/arch/x86/include/asm/arch_hweight.h~x86-hweight-force-inlining-of-__arch_hweight3264
+++ a/arch/x86/include/asm/arch_hweight.h
@@ -21,7 +21,7 @@
* ARCH_HWEIGHT_CFLAGS in <arch/x86/Kconfig> for the respective
* compiler switches.
*/
-static inline unsigned int __arch_hweight32(unsigned int w)
+static __always_inline unsigned int __arch_hweight32(unsigned int w)
{
unsigned int res = 0;
@@ -42,20 +42,23 @@ static inline unsigned int __arch_hweigh
return __arch_hweight32(w & 0xff);
}
+#ifdef CONFIG_X86_32
static inline unsigned long __arch_hweight64(__u64 w)
{
- unsigned long res = 0;
-
-#ifdef CONFIG_X86_32
return __arch_hweight32((u32)w) +
__arch_hweight32((u32)(w >> 32));
+}
#else
+static __always_inline unsigned long __arch_hweight64(__u64 w)
+{
+ unsigned long res = 0;
+
asm (ALTERNATIVE("call __sw_hweight64", POPCNT64, X86_FEATURE_POPCNT)
: "="REG_OUT (res)
: REG_IN (w));
-#endif /* CONFIG_X86_32 */
return res;
}
+#endif /* CONFIG_X86_32 */
#endif
_
Patches currently in -mm which might be from dvlasenk@xxxxxxxxxx are
linux-bitmap-force-inlining-of-bitmap-weight-functions.patch
x86-hweight-force-inlining-of-__arch_hweight3264.patch
jiffies-force-inlining-of-mumsecs_to_jiffies.patch
linux-next.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
