On Mar 30, 2009, at 1:24 PM, Iain Bason wrote:
> But don't we need the whole area to be zero filled?
It will be zero-filled on demand using the lseek/touch method.
Ok.
However, the OS may not reserve space for the skipped pages or disk
blocks. Thus one could still get out of memory or file system full
errors at arbitrary points. Presumably one could also get segfaults
from an mmap'ed segment whose pages couldn't be allocated when the
demand came.
Should we touch every page at init, then? (I think someone else --
Eugene? -- mentioned that already during this thread) FWIW, it would
be better to die during init then at some random point during
execution later...
--
Jeff Squyres
Cisco Systems
