HEADS UP ON -DEVELOPMENT

Matthew Dillon dillon at apollo.backplane.com
Fri Aug 5 09:50:48 PDT 2005


:
:Matthew Dillon wrote:
:>     Joerg and I have discussed the struct stat and struct dirent changes.
:>     So far the stat work has been committed and the dirent work is still
:>     in progress.  This is work to increase ino_t from 32 to 64 bits and
:>     nlink_t from 16 to 32 bits.
:
:I forgot to discuss this with Joerg and you, but why not increase nlink_t 
:all the way up to 64-bits?  I am aware it will take 4 bytes more of store 
:space but atleast we will not need to mess with these things for the near 
:future.
:
:It might be an idea to do bump other fields as well.  I am currently 
:preparing a patch set for bumping up daddr_t to 64-bits.  Personally if it 
:was possible I would push the latter all the way to 128-bits but not quite 
:possible at this moment in time.
:
:What about time related fields, are they currently 64-bits wide?
:
:Regards,
:
:				Hiten Pandya

    I considered bumping nlink_t to 64 bits but I don't think its necessary
    (famous last words).  It's just the hard link count, after all.  While
    it is clear that a 65535 link limit for a file is not sufficient, I
    can't think of a scenario where you would need more then 2^32 hard
    links for any given file, even using something like rsync's link
    destination option.

    daddr_t does not need to be 128 bits.  64 bits is just fine.

					-Matt
					Matthew Dillon 
					<dillon at xxxxxxxxxxxxx>





More information about the Users mailing list