From: Collin Funk <collin.funk1@gmail.com>
Date: Sat, 21 Feb 2026 08:06:01 +0000 (-0800)
Subject: shuf: avoid locking standard output when using --input-range
X-Git-Url: http://git.ipfire.org/gitweb.cgi?a=commitdiff_plain;h=89735ea830b293d50a9323770d6089bac7324aab;p=thirdparty%2Fcoreutils.git

shuf: avoid locking standard output when using --input-range

Here is the throughput before this patch:

    # write_permuted_numbers
    $ ./src/shuf-prev -i 0-100000000 | pv -r > /dev/null
    [ 153MiB/s]

    # write_random_numbers
    $ timeout 10 ./src/shuf-prev -i 0-100000 -r | pv -r > /dev/null
    [78.6MiB/s]

Here is the throughput after this patch:

    # write_permuted_numbers
    $ timeout 10 ./src/shuf -i 0-100000000 | pv -r > /dev/null
    [ 308MiB/s]

    # write_random_numbers
    $ timeout 10 ./src/shuf -i 0-100000 -r | pv -r > /dev/null
    [ 196MiB/s]

* NEWS: Mention the performance improvement.
* src/shuf.c (write_permuted_numbers, write_random_numbers): Prefer
fputs and fputc which may be unlocked over printf which locks standard
output.
---

diff --git a/NEWS b/NEWS
index c509cbc1ea..cfa6b4abbe 100644
--- a/NEWS
+++ b/NEWS
@@ -20,6 +20,9 @@ GNU coreutils NEWS                                    -*- outline -*-
 
   'nl' now supports multi-byte --section-delimiter characters.
 
+  'shuf -i' now operates up to two times faster on systems with unlocked stdio
+  functions.
+
   'wc -l' now operates up to three times faster on hosts that support Neon
   instructions.
 
diff --git a/src/shuf.c b/src/shuf.c
index f87b85dd0f..948ee88f3d 100644
--- a/src/shuf.c
+++ b/src/shuf.c
@@ -325,7 +325,9 @@ write_permuted_numbers (size_t n_lines, size_t lo_input,
   for (size_t i = 0; i < n_lines; i++)
     {
       unsigned long int n = lo_input + permutation[i];
-      if (printf ("%lu%c", n, eolbyte) < 0)
+      char buf[INT_BUFSIZE_BOUND (uintmax_t)];
+      if (fputs (umaxtostr (n, buf), stdout) < 0
+          || fputc (eolbyte, stdout) < 0)
         return -1;
     }
 
@@ -343,7 +345,9 @@ write_random_numbers (struct randint_source *s, size_t count,
   for (size_t i = 0; i < count; i++)
     {
       unsigned long int j = lo_input + randint_choose (s, range);
-      if (printf ("%lu%c", j, eolbyte) < 0)
+      char buf[INT_BUFSIZE_BOUND (uintmax_t)];
+      if (fputs (umaxtostr (j, buf), stdout) < 0
+          || fputc (eolbyte, stdout) < 0)
         return -1;
     }