Граф коммитов

7 Коммитов

Автор SHA1 Сообщение Дата
Peter Zijlstra df65bba1dc lib/bsearch: Provide __always_inline variant
For code that needs the ultimate performance (it can inline the @cmp
function too) or simply needs to avoid calling external functions for
whatever reason, provide an __always_inline variant of bsearch().

[ tglx: Renamed to __inline_bsearch() as suggested by Andy ]

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Alexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: Andy Lutomirski <luto@kernel.org>
Link: https://lkml.kernel.org/r/20200505135313.624443814@linutronix.de
2020-06-11 15:14:53 +02:00
Andy Shevchenko e8877ec5db lib/bsearch: Use generic type for comparator function
Comparator function type, cmp_func_t, is defined in the types.h,
use it in bsearch() and, thus, add more sense to the corresponding
comment in the code.

Link: http://lkml.kernel.org/r/20191007135656.37734-2-andriy.shevchenko@linux.intel.com

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2019-11-14 13:15:11 -05:00
Thomas Gleixner a10e763b87 treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 372
Based on 1 normalized pattern(s):

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license as published by
  the free software foundation version 2

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-only

has been chosen to replace the boilerplate/reference in 135 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Allison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190531081036.435762997@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2019-06-05 17:37:10 +02:00
Andrea Righi 02106f883c kprobes: Prohibit probing on bsearch()
Since kprobe breakpoing handler is using bsearch(), probing on this
routine can cause recursive breakpoint problem.

int3
 ->do_int3()
   ->ftrace_int3_handler()
     ->ftrace_location()
       ->ftrace_location_range()
         ->bsearch() -> int3

Prohibit probing on bsearch().

Signed-off-by: Andrea Righi <righi.andrea@gmail.com>
Acked-by: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/154998813406.31052.8791425358974650922.stgit@devbox
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2019-02-13 08:16:41 +01:00
Sergey Senozhatsky 166a0f780a lib/bsearch.c: micro-optimize pivot position calculation
There is a slightly faster way (in terms of the number of instructions
being used) to calculate the position of a middle element, preserving
integer overflow safeness.

./scripts/bloat-o-meter lib/bsearch.o.old lib/bsearch.o.new
add/remove: 0/0 grow/shrink: 0/1 up/down: 0/-24 (-24)
function                                     old     new   delta
bsearch                                      122      98     -24

TEST

INT array of size 100001, elements [0..100000]. gcc 7.1, Os, x86_64.

a) bsearch() of existing key "100001 - 2":

BASE
====

$ perf stat ./a.out

 Performance counter stats for './a.out':

        619.445196      task-clock:u (msec)       #    0.999 CPUs utilized
                 0      context-switches:u        #    0.000 K/sec
                 0      cpu-migrations:u          #    0.000 K/sec
               133      page-faults:u             #    0.215 K/sec
     1,949,517,279      cycles:u                  #    3.147 GHz                      (83.06%)
       181,017,938      stalled-cycles-frontend:u #    9.29% frontend cycles idle     (83.05%)
        82,959,265      stalled-cycles-backend:u  #    4.26% backend cycles idle      (67.02%)
     4,355,706,383      instructions:u            #    2.23  insn per cycle
                                                  #    0.04  stalled cycles per insn  (83.54%)
     1,051,539,242      branches:u                # 1697.550 M/sec                    (83.54%)
        15,263,381      branch-misses:u           #    1.45% of all branches          (83.43%)

       0.620082548 seconds time elapsed

PATCHED
=======

$ perf stat ./a.out

 Performance counter stats for './a.out':

        475.097316      task-clock:u (msec)       #    0.999 CPUs utilized
                 0      context-switches:u        #    0.000 K/sec
                 0      cpu-migrations:u          #    0.000 K/sec
               135      page-faults:u             #    0.284 K/sec
     1,487,467,717      cycles:u                  #    3.131 GHz                      (82.95%)
       186,537,162      stalled-cycles-frontend:u #   12.54% frontend cycles idle     (82.93%)
        28,797,869      stalled-cycles-backend:u  #    1.94% backend cycles idle      (67.10%)
     3,807,564,203      instructions:u            #    2.56  insn per cycle
                                                  #    0.05  stalled cycles per insn  (83.57%)
     1,049,344,291      branches:u                # 2208.693 M/sec                    (83.60%)
             5,485      branch-misses:u           #    0.00% of all branches          (83.58%)

       0.475760235 seconds time elapsed

b) bsearch() of un-existing key "100001 + 2":

BASE
====

$ perf stat ./a.out

 Performance counter stats for './a.out':

        499.244480      task-clock:u (msec)       #    0.999 CPUs utilized
                 0      context-switches:u        #    0.000 K/sec
                 0      cpu-migrations:u          #    0.000 K/sec
               132      page-faults:u             #    0.264 K/sec
     1,571,194,855      cycles:u                  #    3.147 GHz                      (83.18%)
        13,450,980      stalled-cycles-frontend:u #    0.86% frontend cycles idle     (83.18%)
        21,256,072      stalled-cycles-backend:u  #    1.35% backend cycles idle      (66.78%)
     4,171,197,909      instructions:u            #    2.65  insn per cycle
                                                  #    0.01  stalled cycles per insn  (83.68%)
     1,009,175,281      branches:u                # 2021.405 M/sec                    (83.79%)
             3,122      branch-misses:u           #    0.00% of all branches          (83.37%)

       0.499871144 seconds time elapsed

PATCHED
=======

$ perf stat ./a.out

 Performance counter stats for './a.out':

        399.023499      task-clock:u (msec)       #    0.998 CPUs utilized
                 0      context-switches:u        #    0.000 K/sec
                 0      cpu-migrations:u          #    0.000 K/sec
               134      page-faults:u             #    0.336 K/sec
     1,245,793,991      cycles:u                  #    3.122 GHz                      (83.39%)
        11,529,273      stalled-cycles-frontend:u #    0.93% frontend cycles idle     (83.46%)
        12,116,311      stalled-cycles-backend:u  #    0.97% backend cycles idle      (66.92%)
     3,679,710,005      instructions:u            #    2.95  insn per cycle
                                                  #    0.00  stalled cycles per insn  (83.47%)
     1,009,792,625      branches:u                # 2530.660 M/sec                    (83.46%)
             2,590      branch-misses:u           #    0.00% of all branches          (83.12%)

       0.399733539 seconds time elapsed

Link: http://lkml.kernel.org/r/20170607150457.5905-1-sergey.senozhatsky@gmail.com
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-07-10 16:32:35 -07:00
Paul Gortmaker 8bc3bcc93a lib: reduce the use of module.h wherever possible
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 <paul.gortmaker@windriver.com>
2012-03-07 15:04:04 -05:00
Tim Abbott 1a94dc35bc lib: Add generic binary search function to the kernel.
There a large number hand-coded binary searches in the kernel (run
"git grep search | grep binary" to find many of them).  Since in my
experience, hand-coding binary searches can be error-prone, it seems
worth cleaning this up by providing a generic binary search function.

This generic binary search implementation comes from Ksplice.  It has
the same basic API as the C library bsearch() function.  Ksplice uses
it in half a dozen places with 4 different comparison functions, and I
think our code is substantially cleaner because of this.

Signed-off-by: Tim Abbott <tabbott@ksplice.com>
Extra-bikeshedding-by: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
Extra-bikeshedding-by: André Goddard Rosa <andre.goddard@gmail.com>
Extra-bikeshedding-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Alessio Igor Bogani <abogani@kernel.org>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2011-05-19 16:55:27 +09:30