On Fri, Mar 11, 2011 at 3:19 PM, Brian C. Lane <bcl@fedoraproject.org> wrote:
 HACKING                     |    4 ++--
 tools/livecd-iso-to-disk.sh |   26 +++++++++++++++++++-------
 2 files changed, 21 insertions(+), 9 deletions(-)

New commits:
commit 46f2867369802e38db1d8c9fa2c87dbc33c48371
Author: Lubomir Rintel <lkundrak@v3.sk>
Date:   Fri Mar 11 12:19:05 2011 -0800

   Mailing list address changed

diff --git a/HACKING b/HACKING
index e912112..00fbe3c 100644
--- a/HACKING
+++ b/HACKING
@@ -17,9 +17,9 @@ Change history etc. can be obtained from the web interface
 PATCHES
 =======

-Send patches to the fedora-livecd-list@redhat.com mailing list
+Send patches to the livecd@lists.fedoraproject.org mailing list

- http://www.redhat.com/mailman/listinfo/fedora-livecd-list
+ https://lists.fedoraproject.org/mailman/listinfo/livecd

 If you have a number of patches and want to have them pulled from a
 public git repository, please post a pointer to it as well as sending


commit f3f30ae700ed02be26cd24841d40d9283d62b348
Author: Brian C. Lane <bcl@redhat.com>
Date:   Fri Mar 11 12:16:44 2011 -0800

   Fall back to to msdos format if no extlinux

   And fall back to ext3 if an older version of extlinux is being
   used (thanks to Lubomir Rintel for this suggestion).

   Add --msdos to force --format to use msdos instead of ext4

diff --git a/tools/livecd-iso-to-disk.sh b/tools/livecd-iso-to-disk.sh
index 8cdf86e..3489d51 100755
--- a/tools/livecd-iso-to-disk.sh
+++ b/tools/livecd-iso-to-disk.sh
@@ -26,8 +26,8 @@ shortusage() {
    echo "
    SYNTAX

-    livecd-iso-to-disk [--help] [--noverify] [--format] [--reset-mbr] [--efi]
-                       [--skipcopy] [--force] [--xo] [--xo-no-home]
+    livecd-iso-to-disk [--help] [--noverify] [--format] [--msdos] [--reset-mbr]
+                       [--efi] [--skipcopy] [--force] [--xo] [--xo-no-home]
                       [--timeout <time>] [--totaltimeout <time>]
                       [--extra-kernel-args <args>] [--multi] [--livedir <dir>]
                       [--compress] [--skipcompress] [--swap-size-mb <size>]
@@ -121,6 +121,9 @@ usage() {
        Formats the target device and creates an MS-DOS partition table (or GPT
        partition table, if the --efi option is passed).

+    --msdos
+        Forces format to use msdos instead of ext4.
+
    --reset-mbr
        Sets the Master Boot Record (MBR) of the target storage device to the
        mbr.bin file from the installation system's syslinux directory.  This
@@ -472,7 +475,13 @@ createEXTFSLayout() {
    getpartition ${device#/dev/}
    TGTDEV=${device}${partnum}
    umount $TGTDEV &> /dev/null
-    /sbin/mkfs.ext4 -L LIVE $TGTDEV
+
+    # Check extlinux version
+    if extlinux -v 2>&1 | grep -q 'extlinux 3'; then
+        /sbin/mkfs.ext3 -L LIVE $TGTDEV
+    else
+        /sbin/mkfs.ext4 -L LIVE $TGTDEV
+    fi
    TGTLABEL="UUID=$(/sbin/blkid -s UUID -o value $TGTDEV)"
 }

@@ -655,6 +664,9 @@ while [ $# -gt 2 ]; do
        --format)
            format=1
            ;;
+        --msdos)
+            usemsdos=1
+            ;;
        --reset-mbr|--resetmbr)
            resetmbr=1
            ;;
@@ -749,7 +761,6 @@ if [ ! -b "$SRC" -a ! -f "$SRC" ]; then
    exit 1
 fi

-# FIXME: If --format is given, we shouldn't care and just use /dev/foo1
 if [ -z "$TGTDEV" ]; then
    echo "Missing target device"
    shortusage
@@ -773,15 +784,16 @@ if [ -z "$noverify" ]; then
    fi
 fi

-#checkFilesystem $TGTDEV
 # do some basic sanity checks.
 checkMounted $TGTDEV
+
+# Format the device
 if [ -n "$format" -a -z "$skipcopy" ]; then
    checkLVM $TGTDEV
-    # checks for a valid filesystem
+
    if [ -n "$efi" ]; then
        createGPTLayout $TGTDEV
-    elif [ "$TGTFS" == "vfat" -o "$TGTFS" == "msdos" ]; then
+    elif [ -n "$usemsdos" -o ! -x /sbin/extlinux ]; then
        createMSDOSLayout $TGTDEV
    else
        createEXTFSLayout $TGTDEV

Am I reading this patch correctly to mean that the previous target filesystem structure will be ignored in the format choice?

If so, I object.

Users are often unaware of the the implications of such a change, and a change may have effects we don't understand.  See for example,
http://wiki.laptop.org/go/How_to_Damage_a_FLASH_Storage_Device

        --Fred