>The problem with your analysis is that you can't guarantee some other client
>has not written to the file while you are in afs_GetDCache. There are windows
>where a callback break would not be recognized. This is filling in one of those
>windows. If you want fewer reads, you'll have to use NFS, but then you'll have
I understand that, but I think you missed one of my points; this is
writing _a completely new file_. There is obviously logic in the
cache manager to handle this case (see the comments from the code snippet
in my previous email); it just seems to be broken. Is performing no-op
reads on a newly created file really worth doing?
--Ken
