henry@utzoo.UUCP (Henry Spencer) (11/28/84)
Dump(1) has been improved in some small ways (use of getopt(3), internal cleanup), and has been changed to check for write errors (!) on the tape. -- Henry Spencer @ U of Toronto Zoology {allegra,ihnp4,linus,decvax}!utzoo!henry
henry@utzoo.UUCP (Henry Spencer) (11/28/84)
Dump(1) has been further modified to not insist on having its tape mounted until after pass II; this will let us overlap rewind time with the "think time" in pass I and II. As a side effect of this, for complicated and confused reasons, dump(1) actually temporarily is *not* using getopt(3), contrary to the previous posting. This will be fixed, probably tomorrow. -- Henry Spencer @ U of Toronto Zoology {allegra,ihnp4,linus,decvax}!utzoo!henry
henry@utzoo.UUCP (Henry Spencer) (12/02/84)
We're now running (actually, this happened very late yesterday afternoon) a version of dump(1) that forks off an auxiliary process to do its tape writes. This was done in hopes of greater speed, through concurrent disk and tape operations. It may have speeded things up some; the effect was not spectacular. More to come. -- Henry Spencer @ U of Toronto Zoology {allegra,ihnp4,linus,decvax}!utzoo!henry