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

Groups > comp.lang.c > #126314

Re: Count string in string with loop increase 2, why is this not faster?

Subject	Re: Count string in string with loop increase 2, why is this not faster?
Newsgroups	comp.lang.c
References	(8 earlier) <Kz1eC.1462$nY2.602@fx28.iad> <iR2eC.760204$pP1.434568@fx21.am4> <M_3eC.5140$vP3.694@fx15.iad> <Vx4eC.534983$YE3.185238@fx06.am4> <j15eC.18818$s_2.335@fx42.iad>
From	bartc <bc@freeuk.com>
Message-ID	<8C7eC.446657$1t7.45854@fx03.am4> (permalink)
Organization	virginmedia.com
Date	2018-02-06 01:34 +0000

Show all headers | View raw

On 05/02/2018 22:38, Scott Lurndal wrote:
> bartc <bc@freeuk.com> writes:

>>>> How did you fare with PCMPEQW?
>>>
>>>
>>> http://www.alfredklomp.com/programming/sse-strings/
>>
>>
>> Those are for 8-bit, zero terminated strings.
> 
> That was an example.  Please feel free to do your own
> work on adapting the example.

Me? It's the OP who needs the work and he's more familiar with VBA. 
(I've been using x86 asm for decades and still have a hard time with SSE 
instructions.)

I thought maybe that you'd done tests comparing optimised C against a 
routine using these instructions, and had an idea what benefits there 
might have been before other people wasted /their/ time.

But I've since had another go at the C, and came up with this variation. 
Where occurrences of the search string are sparse, this code, both 
compiled with gcc-O3, can be three times as fast as the original routine:

int test1(wchar_t *strSearch, wchar_t *strFind, int nSearch, int nFind)
{
     int n,c,nbytes;
     int countstring = 0;

     if (nSearch == 0 || nFind == 0 || nFind > nSearch) {
         return 0;
     }
     n=nSearch - nFind+1;
     nbytes = nFind * 2;
     c = *strFind;

     while (n > 0) {
         if (*strSearch == c && memcmp(strSearch, strFind, nbytes)==0) {
             ++countstring;
             strSearch += nFind;
             n -= nFind;
         } else {
             ++strSearch;
             --n;
         }
     }

     return countstring;
}

The key is searching first for a matching first pair of characters 
before bothering to call memcmp. (Which then compares nFind characters, 
including the first pair again, but that was easier to do than to avoid).

It might be worth making nFind==1 a special case, as that can be greatly 
simplified if it is likely to be common.

(I've added a 'nFind>nSearch' check, but that is not really necessary, 
as it will bail out in that case when it gets to the loop. Take it out 
if that condition is unlikely. However, the way gcc-O3 works is so 
unpredictable, that when I did try, it was immediately 20% slower!)

-- 
bartc

Back to comp.lang.c | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-03 00:00 -0800
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-03 11:29 +0000
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-03 03:37 -0800
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-03 12:29 +0000
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-03 04:37 -0800
Re: Count string in string with loop increase 2, why is this not faster? Ben Bacarisse <ben.usenet@bsb.me.uk> - 2018-02-03 13:41 +0000
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-03 06:38 -0800
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-03 15:13 +0000
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-03 15:19 +0000
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-03 07:41 -0800
Re: Count string in string with loop increase 2, why is this not faster? Richard Damon <Richard@Damon-Family.org> - 2018-02-03 11:38 -0500
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-03 13:50 +0000
Re: Count string in string with loop increase 2, why is this not faster? Ben Bacarisse <ben.usenet@bsb.me.uk> - 2018-02-03 11:34 +0000
Re: Count string in string with loop increase 2, why is this not faster? Richard Damon <Richard@Damon-Family.org> - 2018-02-03 11:32 -0500
Re: Count string in string with loop increase 2, why is this not faster? Ben Bacarisse <ben.usenet@bsb.me.uk> - 2018-02-03 16:41 +0000
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-03 09:02 -0800
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-03 09:06 -0800
Re: Count string in string with loop increase 2, why is this not faster? Richard Damon <Richard@Damon-Family.org> - 2018-02-03 12:39 -0500
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-03 11:56 -0800
Re: Count string in string with loop increase 2, why is this not faster? Richard Damon <Richard@Damon-Family.org> - 2018-02-03 19:54 -0500
Re: Count string in string with loop increase 2, why is this not faster? Keith Thompson <kst-u@mib.org> - 2018-02-03 18:00 -0800
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-04 01:53 -0800
Re: Count string in string with loop increase 2, why is this not faster? Richard Damon <Richard@Damon-Family.org> - 2018-02-04 13:23 -0500
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-04 11:11 -0800
Re: Count string in string with loop increase 2, why is this not faster? Richard Damon <Richard@Damon-Family.org> - 2018-02-04 14:48 -0500
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-04 12:17 -0800
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-04 21:53 +0000
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-04 14:35 -0800
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-04 23:32 +0000
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-05 06:26 -0800
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-05 16:59 +0000
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-05 09:44 -0800
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-05 18:28 +0000
Re: Count string in string with loop increase 2, why is this not faster? scott@slp53.sl.home (Scott Lurndal) - 2018-02-05 18:41 +0000
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-05 20:08 +0000
Re: Count string in string with loop increase 2, why is this not faster? scott@slp53.sl.home (Scott Lurndal) - 2018-02-05 21:27 +0000
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-05 22:04 +0000
Re: Count string in string with loop increase 2, why is this not faster? scott@slp53.sl.home (Scott Lurndal) - 2018-02-05 22:38 +0000
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-06 01:34 +0000
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-06 06:39 -0800
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-06 18:07 +0000
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-06 12:34 -0800
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-06 21:12 +0000
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-06 13:42 -0800
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-06 16:43 -0800
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-07 01:10 +0000
Re: Count string in string with loop increase 2, why is this not faster? Keith Thompson <kst-u@mib.org> - 2018-02-06 18:47 -0800
Re: Count string in string with loop increase 2, why is this not faster? David Brown <david.brown@hesbynett.no> - 2018-02-07 15:13 +0100
Re: Count string in string with loop increase 2, why is this not faster? Ben Bacarisse <ben.usenet@bsb.me.uk> - 2018-02-07 14:48 +0000
Re: Count string in string with loop increase 2, why is this not faster? Ben Bacarisse <ben.usenet@bsb.me.uk> - 2018-02-07 17:13 +0000
Re: Count string in string with loop increase 2, why is this not faster? Ben Bacarisse <ben.usenet@bsb.me.uk> - 2018-02-07 17:27 +0000
Re: Count string in string with loop increase 2, why is this not faster? David Brown <david.brown@hesbynett.no> - 2018-02-07 17:01 +0100
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-07 02:00 -0800
Re: Count string in string with loop increase 2, why is this not faster? Ben Bacarisse <ben.usenet@bsb.me.uk> - 2018-02-07 12:02 +0000
Re: Count string in string with loop increase 2, why is this not faster? Ben Bacarisse <ben.usenet@bsb.me.uk> - 2018-02-07 14:46 +0000
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-07 13:56 +0000
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-07 11:17 +0000
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-06 23:59 -0800
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-07 05:10 -0800
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-07 11:01 +0000
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-07 14:24 -0800
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-07 14:44 -0800
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-07 22:59 +0000
Re: Count string in string with loop increase 2, why is this not faster? Keith Thompson <kst-u@mib.org> - 2018-02-07 15:20 -0800
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-07 23:31 +0000
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-07 15:32 -0800
Re: Count string in string with loop increase 2, why is this not faster? Keith Thompson <kst-u@mib.org> - 2018-02-07 16:05 -0800
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-05 15:29 -0800
Re: Count string in string with loop increase 2, why is this not faster? Richard Damon <Richard@Damon-Family.org> - 2018-02-04 17:41 -0500
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-04 23:11 +0000
Re: Count string in string with loop increase 2, why is this not faster? Richard Damon <Richard@Damon-Family.org> - 2018-02-04 23:45 -0500
Re: Count string in string with loop increase 2, why is this not faster? Keith Thompson <kst-u@mib.org> - 2018-02-04 15:11 -0800
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-04 23:37 +0000
Re: Count string in string with loop increase 2, why is this not faster? Keith Thompson <kst-u@mib.org> - 2018-02-04 18:08 -0800
Re: Count string in string with loop increase 2, why is this not faster? Richard Damon <Richard@Damon-Family.org> - 2018-02-04 23:12 -0500
Re: Count string in string with loop increase 2, why is this not faster? David Brown <david.brown@hesbynett.no> - 2018-02-05 13:08 +0100
Re: Count string in string with loop increase 2, why is this not faster? Richard Damon <Richard@Damon-Family.org> - 2018-02-05 08:24 -0500
Re: Count string in string with loop increase 2, why is this not faster? David Brown <david.brown@hesbynett.no> - 2018-02-05 14:39 +0100
Re: Count string in string with loop increase 2, why is this not faster? Robert Wessel <robertwessel2@yahoo.com> - 2018-02-05 10:37 -0600
Re: Count string in string with loop increase 2, why is this not faster? supercat@casperkitty.com - 2018-02-05 13:52 -0800
Re: Count string in string with loop increase 2, why is this not faster? Ben Bacarisse <ben.usenet@bsb.me.uk> - 2018-02-05 22:29 +0000
Re: Count string in string with loop increase 2, why is this not faster? Keith Thompson <kst-u@mib.org> - 2018-02-05 09:01 -0800
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-05 18:44 +0000
Re: Count string in string with loop increase 2, why is this not faster? David Brown <david.brown@hesbynett.no> - 2018-02-06 10:42 +0100
Re: Count string in string with loop increase 2, why is this not faster? Ben Bacarisse <ben.usenet@bsb.me.uk> - 2018-02-06 11:17 +0000
Re: Count string in string with loop increase 2, why is this not faster? Richard Damon <Richard@Damon-Family.org> - 2018-02-06 08:40 -0500
Re: Count string in string with loop increase 2, why is this not faster? bartc <bc@freeuk.com> - 2018-02-06 14:08 +0000
Re: Count string in string with loop increase 2, why is this not faster? Ben Bacarisse <ben.usenet@bsb.me.uk> - 2018-02-06 14:42 +0000
Re: Count string in string with loop increase 2, why is this not faster? Keith Thompson <kst-u@mib.org> - 2018-02-06 12:35 -0800
Re: Count string in string with loop increase 2, why is this not faster? David Brown <david.brown@hesbynett.no> - 2018-02-06 23:04 +0100
Re: Count string in string with loop increase 2, why is this not faster? Richard Damon <Richard@Damon-Family.org> - 2018-02-04 17:35 -0500
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-04 14:38 -0800
Re: Count string in string with loop increase 2, why is this not faster? Richard Damon <Richard@Damon-Family.org> - 2018-02-04 17:57 -0500
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-04 15:22 -0800
Re: Count string in string with loop increase 2, why is this not faster? Bart Smissaert <bart.smissaert@gmail.com> - 2018-02-04 12:19 -0800
Re: Count string in string with loop increase 2, why is this not faster? Richard Damon <Richard@Damon-Family.org> - 2018-02-04 16:42 -0500
Re: Count string in string with loop increase 2, why is this not faster? Keith Thompson <kst-u@mib.org> - 2018-02-03 13:38 -0800
Re: Count string in string with loop increase 2, why is this not faster? Jorgen Grahn <grahn+nntp@snipabacken.se> - 2018-02-03 22:03 +0000

csiph-web