[MLton] interrupted system call

[Henry Cejtin]

re too expensive, my interpretation of Matthews thought was that we alway run
with all signals blocked (more on that below),  but  when  the  current  code
checks  to  see  if  a signal has occured, instead we unblock the signals and
then immediately reblock them and then check the flag.  This way system calls
would  be  un-interruptable,  but  that  must  be true any way because the ML
handler can't be run then, and we get the right behavior for free.

The problem is that unblocking and then reblocking  signals  would  mean  two
system calls (both sigblock()) and that is probably too expensive.

With regards to `all signals blocked', note that C libraries must be handling
interrupts.  I guess that the right thing would be to only block signals that
are handled by ML code.

I'm  still  a bit nervous about this.  Note, signals are really quite a botch
in Unix.