More doc updates.

 -Erik

1 files changed, 37 insertions, 22 deletions

diff --git a/TODO b/TODO
index 3aba136..39fa37e 100644
--- a/TODO
+++ b/TODO
@@ -16,32 +16,58 @@ around to it some time. If you have any good ideas, please let me know.
 
 -----------
 
-Bugs that need fixing before the 0.44 release goes out the door:
+Bugs that need fixing:
 
  - "more" doesn't accept " " to scroll by one page when BB_FEATURE_USE_TERMIOS
     is not on.
- - doc / website updates to tell what the real mailing list is for busybox.
  - fix 'cp' (or test suite script) so that it passes the test suite.
-
-These could get done before the 0.44 release if we have time, but we won't
-hold up the 0.44 release just because of these:
-
  - ps dirent race bug (need to stat the file before attempting chdir)
  - Make 'ln -s /tmp/file .' work the way GNU ln does (i.e. makes a link to 
     /tmp/file in the current directory, rather then trying and failing to create
     a symlink named "." in the current working directory).
  - Prune sfdisk
- - Graft fdisk
+ - Graft fdisk instead
 
-These will be deferred til after 0.44 is released. We will rework these to use
-libc regex functions instead (as per the mailing list discussion):
 
+We will rework these to use libc regex functions instead (as per the mailing
+list discussion):
  - 'grep foo$ file' doesn't work
  - 'grep *foo file' segfaults
 
 
 -----------
 
+Linux 2.4.x kernels
+
+BusyBox 0.45 currently will not work with the Linux 2.4.x kernels.  
+I know of the following problems:
+
+1) The sysinfo syscall has changed what it does (binary incompatable), breaking
+    init and free.
+2) BusyBox NFS support is broken with 2.4.x (needs to be adjusted for NFSv3 and
+    kernel header changes).
+3) mount,umount,and df are all broken by the "none" entries for fake filesystems
+    such as the shared mem one.  Al Viro claims these will be disappearing soon...
+
+I made a kernel patch that reverts the sysinfo changes
+    http://kernelnotes.org/lnxlists/linux-kernel/lk_0006_01/msg00619.html
+
+and I have been fighting with Alan Cox to get these changes fixed in a binary
+compatable way, but Alan has so far not been very receptive.  I am planning on
+appealing to Linus (when he gets back from vacation) and then going with
+whatever he decides...
+
+So my thought is, 2.4.x just isn't ready for BusyBox to target it, and even if
+it was, BusyBox isn't ready yet either.  Seems to me like this will not be
+ready for a while, and we should just not worry about it yet.
+
+As long as I have BB_FEATURE_NFSMOUNT turned off, everything compiles cleanly
+for me with linux2.4.0test1-ac22-riel (i.e. I don't see the freeramdisk.c
+problem you reported).  I use Debian potato (gcc 2.95.2, GNU libc 2.1.3).
+Of course, as noted above, compiling != working.
+
+-----------
+
 * Make insmod actually work
 * dnsdomainname
 * traceroute/netstat
@@ -53,7 +79,8 @@ libc regex functions instead (as per the mailing list discussion):
 * tftp
 * ftp
 * group/commonize strings, remove dups (for i18n, l10n)
-* consider making a unified option parser (if it can be done modular, small, etc.) 
+* consider making a unified option parser (if it can be done 
+    modular, small, etc.) 
 
 -----------------------
 
@@ -77,18 +104,6 @@ Most wanted list:
 Policy violation.  getgroups uses libc nss, which is unlikely
 to be present in an embedded system.
 
-To be replaced with a busybox local glob routine:
-
-    [andersen@slag busybox]$ grep -l glob *.[ch]
-    gunzip.c
-    gzip.c
-    sh.c
-    tar.c
-    telnet.c
-
-Can check_wildcard_match() from utility.c do this job?
-
-
 -----------------------
 
 Compile with debugging on, run 'nm --size-sort ./busybox'