This is the mail archive of the
cygwin@sources.redhat.com
mailing list for the Cygwin project.
Re: [PATCH] Problems with Cygwin mmap/munmap (mmap.cc)
- To: "cygwin at cygwin dot com" <cygwin at cygwin dot com>
- Subject: Re: [PATCH] Problems with Cygwin mmap/munmap (mmap.cc)
- From: Christopher Faylor <cgf at redhat dot com>
- Date: Tue, 28 Nov 2000 22:52:59 -0500
- Cc: zackw at stanford dot edu
- References: <200011281912.eASJCmf13405@plmlir5.mail.eds.com> <20001128114558.R2285@wolery.stanford.edu>
- Reply-To: cygwin at cygwin dot com
On Tue, Nov 28, 2000 at 11:45:58AM -0800, Zack Weinberg wrote:
>I might point out that we only want these things for mmap of anonymous
>memory. If Cygwin used VirtualAlloc instead of CreateFileMapping with
>an invalid handle to implement anonymous mmap, then both behaviors
>would be implementable easily. (I don't know if this would interfere
>with cygwin's brk() implementation, though.)
I think that I raised this specter before but, on reflection, this
implementation detail should matter. We'd have the same issue with
VirtualAlloc or CreateFileMapping.
I think that you (Zack) noted that using VirtualAlloc actually might
save us some precious upper memory in the Windows 9x "OS"'s.
The problem, I think, is that VirtualAlloc space is not inherited. We
could kludge around this, like we do with fork, by copying the regions
but that could slow down things a lot.
The Microsoft docs sometimes seem to imply that you can mark a
VirtualAlloc'ed region as shareable but I've never been able to get that
to work.
cgf
--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe@sourceware.cygnus.com