This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: gcc on OSR5 status. was: bootstraps broken during last 24 hours
On Mon, Jan 08, 2001 at 06:58:37PM -0600, Robert Lipe wrote:
> > | > And I also just did a suuccessful bootstrap for i686-pc-udk using
> > |
> > | Excellent. This is with or without the MAP_ANONYMOUS hack?
> >
> > Was that something introduced in the tree over the weekend?
> > It's nothing that I tickled from my keyboard. :-)
>
> This was the mmap thing in ggc-page where it tries to use MAP_ANON
> becuase it sees a #define and then fails becuase it doesn't actually
> work. You were the one that reported it many months ago.
Being as how I am supposed to fix the mmap checks so they don't think
cygwin's mmap works to the extent GCC needs (it either crashes, or
leaks memory, depending on which version you have), I'd appreciate
more info on this problem, so I can fix it at the same time.
Specifically:
- we do mmap(NULL, size, READ/WRITE, MAP_ANON, -1, 0) if we think we
have MAP_ANON. -1 is the file descriptor argument. What does -udk
do when handed this?
- Is MAP_ANONYMOUS also defined, and does *that* work in this context?
- Is there a different, functional way to use MAP_ANON(YMOUS) on -udk?
- Can we do instead
mmap(NULL, size, READ/WRITE, MAP_PRIVATE, /dev/zero, 0) ?
zw