From: "Steve Grubb" <ddata@gate.net>
To: <linux-kernel@vger.kernel.org>
Subject: Re: [Patch] performance enhancement for simple_strtoul
Date: Wed, 20 Dec 2000 13:42:56 -0500 [thread overview]
Message-ID: <001401c06ab4$ac8615e0$7d1a24cf@master> (raw)
In-Reply-To: <000e01c06a8e$6945db60$bc1a24cf@master> <20001220100446.A1249@inetnebr.com>
Hello,
I continued experimenting with the Test Case and found a further speed
improvement & I am re-submiting the patch. It is the same as the first one
with the two local variables changed to register storage types.
On a K6-2, I now see:
Base 10 - 28% speedup
Base 16 - 24% speedup
Base 8 - 30% speedup
On a P3 system, I now see:
Base 10 - 25% speedup
Base 16 - 17% speedup
Base 8 - 20% speedup
It seems gcc creates much better code with the variables set to register
types. Please apply the following patch. It should apply to any recent 2.2.x
without problems. In 2.4 the function starts 2 lines later.
Cheers,
Steve Grubb
----------------------------------
--- lib/vsprintf.orig Fri Dec 1 08:58:02 2000
+++ lib/vsprintf.c Wed Dec 20 13:14:13 2000
@@ -14,10 +14,13 @@
#include <linux/string.h>
#include <linux/ctype.h>
+/*
+* This function converts base 8, 10, or 16 only - Steve Grubb
+*/
unsigned long simple_strtoul(const char *cp,char **endp,unsigned int base)
{
- unsigned long result = 0,value;
-
+ register unsigned char c;
+ register unsigned long result = 0;
if (!base) {
base = 10;
if (*cp == '0') {
@@ -29,11 +32,36 @@
}
}
}
- while (isxdigit(*cp) && (value = isdigit(*cp) ? *cp-'0' : (islower(*cp)
- ? toupper(*cp) : *cp)-'A'+10) < base) {
- result = result*base + value;
- cp++;
- }
+ c = *cp;
+ switch (base) {
+ case 10:
+ while (isdigit(c)) {
+ result = (result*10) + (c & 0x0f);
+ c = *(++cp);
+ }
+ break;
+ case 16:
+ while (isxdigit(c)) {
+ result = result<<4;
+ if (c&0x40)
+ result += (c & 0x07) + 9;
+ else
+ result += c & 0x0f;
+ c = *(++cp);
+ }
+ break;
+ case 8:
+ while (isdigit(c)) {
+ if ((c&0x37) == c)
+ result = (result<<3) + (c & 0x07);
+ else
+ break;
+ c = *(++cp);
+ }
+ break;
+ default: /* Is anything else used by the kernel? */
+ break;
+ }
if (endp)
*endp = (char *)cp;
return result;
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/
next prev parent reply other threads:[~2000-12-20 19:13 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2000-12-20 14:09 [Patch] performance enhancement for simple_strtoul Steve Grubb
2000-12-20 16:04 ` Jeff Epler
2000-12-20 16:35 ` Steve Grubb
2000-12-20 18:42 ` Steve Grubb [this message]
2000-12-21 0:09 ` Jamie Lokier
2000-12-21 20:06 ` Pavel Machek
2000-12-28 5:15 ` Jamie Lokier
2000-12-21 9:16 ` Matthias Andree
2000-12-21 10:28 ` Bernd Schmidt
2000-12-21 16:07 ` Alan Cox
2000-12-21 20:05 ` Pavel Machek
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='001401c06ab4$ac8615e0$7d1a24cf@master' \
--to=ddata@gate.net \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox