[svn-r22600] Description:

Bring r22251:22599 from trunk to revise_chunks branch.

Tested on:
    Mac OSX/64 10.7.4 (amazon) w/gcc 4.7.1, C++ & FORTRAN
    (Not h5committested yet, as this branch doesn't require it)

tools/testfiles/tpbitsMaxExceeded.ddl

@@ -2,6 +2,7 @@ usage: h5dump [OPTIONS] files
   OPTIONS
      -h, --help           Print a usage message and exit
      -n, --contents       Print a list of the file contents and exit
+                          Optional value 1 also prints attributes.
      -B, --superblock     Print the content of the super block
      -H, --header         Print the header only; no data is displayed
      -A, --onlyattr       Print the header and value of attributes
@@ -61,10 +62,20 @@ usage: h5dump [OPTIONS] files
         number of dimensions in the dataspace being queried
 
   D - is the file driver to use in opening the file. Acceptable values
-        are "sec2", "family", "split", "multi", "direct", and "stream". Without
-        the file driver flag, the file will be opened with each driver in
-        turn and in the order specified above until one driver succeeds
-        in opening the file.
+      are "sec2", "family", "split", "multi", "direct", and "stream". Without
+      the file driver flag, the file will be opened with each driver in
+      turn and in the order specified above until one driver succeeds
+      in opening the file.
+      These are the letters that are appended to the file name(without .h5) when opening
+      names for the split(m,r) and multi(s,b,r,g,l,o) drivers. They are:
+         m: All meta data when using the split driver.
+         s: The userblock, superblock, and driver info block
+         b: B-tree nodes
+         r: Dataset raw data
+         g: Global heap
+         l: local heap (object names)
+         o: object headers
+
   F - is a filename.
   P - is the full path from the root group to the object.
   N - is an integer greater than 1.
@@ -101,8 +112,12 @@ usage: h5dump [OPTIONS] files
 
       h5dump -d /dset -M 0,1,4,3 quux.h5
 
-  5) Dataset foo in files multi1.h5 multi2.h5 multi3.h5
+  5) Dataset foo in files file1.h5 file2.h5 file3.h5
 
-      h5dump -d /foo multi1.h5 multi2.h5 multi3.h5
+      h5dump -d /foo file1.h5 file2.h5 file3.h5
+
+  6) Dataset foo in split files splitfile-m.h5 splitfile-r.h5
+
+      h5dump -d /foo -f split splitfile
 
 h5dump error: Too many masks requested (max. 8). Mask list(0,1,0,1,1,1,2,1,3,1,4,1,5,1,6,1,7,1)