Jeff King <peff@xxxxxxxx> writes: > I think diff_filespec_clear() would not be quite right. It is freeing > only the allocated _data_, but leaving the other portions intact. You are (as usual) more right than I am ;-) Yes, the free_data() is designed to retain enough information about the filespec so that the data can be re-read by the next person who needs to access it after free_data() is called; i.e. it was a measure to reduce the memory pressure by trading it off with I/O cost. It is wrong to name it "clear", which is to "clear the slate, make it into pristine state" whose side effect is to release held resources. -- To unsubscribe from this list: send the line "unsubscribe git" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html