From:	"A.Appleyard" <A DOT APPLEYARD AT fs2 DOT mt DOT umist DOT ac DOT uk>
To:	DJGPP AT SUN DOT SOE DOT CLARKSON DOT EDU
Date:	Sat, 7 Oct 1995 12:28:51 BST
Subject:	Lots of small files in DJGPP waste space & should be chained up!

  The great size of the full DJGPP puts a big strain on people's hard disk
stores. This is NOT helped by the current practice in many parts of it of
having stuff on a mass of little files. This has come to a head with me,
because, having got used to hard disk clusters being 2048 bytes each, my
department has just got 9 new Pentiums whose hard disk clusters are 32K bytes
each!! Each little file must start at the start of a cluster. Thus although
(e.g.) the 40 files in DJGPP\LIBSRC\C\IO\ are only 474 21250 18701 7290 245
1647 425 652 1879 3401 3267 1773 1849 458 1061 1269 1682 1892 1326 1434 500
581 574 30 7121 1535 512 492 562 1708 920 940 1250 1313 1645 781 714 1888 1570
1644 bytes long, each ties up a 32K-byte cluster for itself, total usage 40
clusters = 1.25 megabytes. But when I chained them all into one file (call it
say DJGPP\LIBSRC\C\IO.CC) (in that file I deleted many big duplicate copies of
licences, but altered nothing else), that single file only occupied 81296
bytes = 3 clusters, well over a megabyte saving!
  (It would also save bulk in MAKE files, not having to name each of those
files separately in them.)
  Repeat this by many of DJGPP's many directories of small or smallish *.C &
*.CC & *.S files. The space saving for Pentium users would be incredible.

- Raw text -