Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 16 May 2004 14:41:00 -0700 (PDT)
From:      Peter Wemm <peter@FreeBSD.org>
To:        Perforce Change Reviews <perforce@freebsd.org>
Subject:   PERFORCE change 52884 for review
Message-ID:  <200405162141.i4GLf0fh021224@repoman.freebsd.org>

next in thread | raw e-mail | index | archive | help
http://perforce.freebsd.org/chv.cgi?CH=52884

Change 52884 by peter@peter_overcee on 2004/05/16 14:40:41

	add item for fixing 32/64 bit confustion in sysctl(2) for 32 bit
	binaries.

Affected files ...

.. //depot/projects/hammer/TODO.txt#15 edit

Differences ...

==== //depot/projects/hammer/TODO.txt#15 (text+ko) ====

@@ -1,4 +1,4 @@
-$P4: //depot/projects/hammer/TODO.txt#14 $
+$P4: //depot/projects/hammer/TODO.txt#15 $
 
 reread the logs etc and check out the "XXX worry about this later" stuff
 
@@ -36,6 +36,12 @@
 (XXX: kern.ps_strings and kern.usrstack are prime examples of the 4/8 byte
  long problem).
 
+Fix sysctl(2) int/long mess, especially for kern.ps_strings and
+kern.usrstack.  The current hack is Just Wrong.  What is needed is a
+syscall wrapper for it that downconverts the "long" values from 64 bit to
+32 bit for the 32 bit binaries.  This is going to need a table of important
+"long" sysctls to translate.
+
 ======= DONE =======
 ppp(8) (compiler bug, hack in http://people.freebsd.org/~peter/hammer.diff,
   but gcc-3.3 would be much better)



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200405162141.i4GLf0fh021224>