[svn-r13012] Description:

    Refactor object header code to separate process of creating an object
header message from the process of writing to an existing one.

    Start renaming operations that deal with object header messages to have
"H5O_msg_" prefix...

Tested on:
    Mac OS X/32 10.4.8 (amazon)
    FreeBSD/32 4.11 (sleipnir)
    Linux/32 2.4 (heping)
    AIX/32 5.? (copper)

1 files changed, 5 insertions, 2 deletions

diff --git a/src/H5SM.c b/src/H5SM.c
index 7ab2f59..38b8c75 100755
--- a/src/H5SM.c
+++ b/src/H5SM.c
@@ -519,6 +519,9 @@ H5SM_try_share(H5F_t *f, hid_t dxpl_id, unsigned type_id, void *mesg)
 
     /* Type-specific checks */
     /* JAMES: should this go here? Should there be a "can share" callback? */
+    /* QAK: Yes, a "can share" callback would be very good here, this chunk of
+     *          code is really violating the encapsulation of the datatype class
+     */
     if(type_id == H5O_DTYPE_ID)
     {
         /* Don't share immutable datatypes */
@@ -573,12 +576,12 @@ H5SM_try_share(H5F_t *f, hid_t dxpl_id, unsigned type_id, void *mesg)
 done:
     /* Release the master SOHM table */
     if (table && H5AC_unprotect(f, dxpl_id, H5AC_SOHM_TABLE, f->shared->sohm_addr, table, cache_flags) < 0)
-	HGOTO_ERROR(H5E_CACHE, H5E_CANTRELEASE, FAIL, "unable to close SOHM master table")
+	HDONE_ERROR(H5E_CACHE, H5E_CANTRELEASE, FAIL, "unable to close SOHM master table")
 
     FUNC_LEAVE_NOAPI(ret_value)
 }
-
 
+
 /*-------------------------------------------------------------------------
  * Function:    H5SM_write_mesg
  *