From: Jim Meyering <meyering@redhat.com>
Date: Tue, 18 Sep 2012 06:26:17 +0000 (+0200)
Subject: factor: 25% speed-up, on output
X-Git-Tag: v8.20~15
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=06b7ed8051d07677572e074d3d45007f1282ed56;p=thirdparty%2Fcoreutils.git

factor: 25% speed-up, on output

* src/factor.c (print_factors_single): Use fputs and umaxtostr
rather than printf with "%ju".  This reduced the time required
to compute and print the factors of the first 10^7 integers from
over 8 seconds to 5.75s.  Run this command:
seq $((10**7)) | env time factor > /dev/null
---

diff --git a/src/factor.c b/src/factor.c
index 84d9721c4b..1ca5812276 100644
--- a/src/factor.c
+++ b/src/factor.c
@@ -2272,7 +2272,11 @@ print_factors_single (uintmax_t t1, uintmax_t t0)
 
   for (unsigned int j = 0; j < factors.nfactors; j++)
     for (unsigned int k = 0; k < factors.e[j]; k++)
-      printf (" %ju", factors.p[j]);
+      {
+        char buf[INT_BUFSIZE_BOUND (uintmax_t)];
+        putchar (' ');
+        fputs (umaxtostr (factors.p[j], buf), stdout);
+      }
 
   if (factors.plarge[1])
     {