From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756530AbbHDOPW (ORCPT ); Tue, 4 Aug 2015 10:15:22 -0400 Received: from mx1.redhat.com ([209.132.183.28]:32795 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751425AbbHDOPT (ORCPT ); Tue, 4 Aug 2015 10:15:19 -0400 From: Denys Vlasenko To: Ingo Molnar Cc: Denys Vlasenko , Thomas Graf , Peter Zijlstra , David Rientjes , Andrew Morton , linux-kernel@vger.kernel.org Subject: [PATCH 1/3] linux/bitmap: Force inlining of bitmap weight functions Date: Tue, 4 Aug 2015 16:15:14 +0200 Message-Id: <1438697716-28121-1-git-send-email-dvlasenk@redhat.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org With this config: http://busybox.net/~vda/kernel_config_OPTIMIZE_INLINING_and_Os gcc-4.7.2 generates many copies of these tiny functions: bitmap_weight (55 copies): 55 push %rbp 48 89 e5 mov %rsp,%rbp e8 3f 3a 8b 00 callq __bitmap_weight 5d pop %rbp c3 retq hweight_long (23 copies): 55 push %rbp e8 b5 65 8e 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/ While at it, replaced two "__inline__" with usual "inline" (the rest of the source file uses the latter). text data bss dec hex filename 86971357 17195880 36659200 140826437 864d745 vmlinux.before 86971120 17195912 36659200 140826232 864d678 vmlinux Signed-off-by: Denys Vlasenko Cc: Ingo Molnar Cc: Thomas Graf Cc: Peter Zijlstra Cc: David Rientjes Cc: Andrew Morton Cc: linux-kernel@vger.kernel.org --- include/linux/bitmap.h | 2 +- include/linux/bitops.h | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/include/linux/bitmap.h b/include/linux/bitmap.h index ea17cca..9653fdb 100644 --- a/include/linux/bitmap.h +++ b/include/linux/bitmap.h @@ -295,7 +295,7 @@ static inline int bitmap_full(const unsigned long *src, unsigned int nbits) return find_first_zero_bit(src, nbits) == nbits; } -static inline int bitmap_weight(const unsigned long *src, unsigned int nbits) +static __always_inline int bitmap_weight(const unsigned long *src, unsigned int nbits) { if (small_const_nbits(nbits)) return hweight_long(*src & BITMAP_LAST_WORD_MASK(nbits)); diff --git a/include/linux/bitops.h b/include/linux/bitops.h index 297f5bd..e635533 100644 --- a/include/linux/bitops.h +++ b/include/linux/bitops.h @@ -57,7 +57,7 @@ extern unsigned long __sw_hweight64(__u64 w); (bit) < (size); \ (bit) = find_next_zero_bit((addr), (size), (bit) + 1)) -static __inline__ int get_bitmask_order(unsigned int count) +static inline int get_bitmask_order(unsigned int count) { int order; @@ -65,7 +65,7 @@ static __inline__ int get_bitmask_order(unsigned int count) return order; /* We could be slightly more clever with -1 here... */ } -static __inline__ int get_count_order(unsigned int count) +static inline int get_count_order(unsigned int count) { int order; @@ -75,7 +75,7 @@ static __inline__ int get_count_order(unsigned int count) return order; } -static inline unsigned long hweight_long(unsigned long w) +static __always_inline unsigned long hweight_long(unsigned long w) { return sizeof(w) == 4 ? hweight32(w) : hweight64(w); } -- 1.8.1.4