Date: Sat, 16 Mar 1996 06:06:26 -0800 From: David Greenman <davidg@Root.COM> To: "Jordan K. Hubbard" <jkh@time.cdrom.com> Cc: David Dawes <dawes@rf900.physics.usyd.edu.au>, sos@FreeBSD.org, dyson@FreeBSD.org, current@FreeBSD.org, mmead@Glock.COM Subject: Re: Try this vm_mmap.c -- please Message-ID: <199603161406.GAA07890@Root.COM> In-Reply-To: Your message of "Sat, 16 Mar 1996 04:36:29 PST." <244.826979789@time.cdrom.com>
next in thread | previous in thread | raw e-mail | index | archive | help
>> Does accelx by any chance use MAP_PRIVATE?
>
>Bingo!
>
>Between David's comment here and John's own comment in the latest
>vm_mmap.c he sent out which says:
>
> * On device mmaps we default to map shared -- makes the most
> * sense in this case. Does MAP_PRIVATE even make sense on
> * devices mmaps???
>
>I made the following change:
>
> if ((type == OBJT_DEVICE)) {
> if (flags & (MAP_PRIVATE|MAP_SHARED) == 0)
> flags |= MAP_SHARED;
> else if (flags & MAP_PRIVATE)
> flags &= ~MAP_PRIVATE;
> }
Could be written as:
/*
* Force device mappings to be shared.
*/
if (type == OBJT_DEVICE) {
flags &= ~MAP_PRIVATE;
flags |= MAP_SHARED;
}
>Which turns MAP_PRIVATE *off* for device objects. My Xaccel 1.2 now
>works perfectly! Yay!
Yes, as John and I have discussed, it really doesn't make any sense to
allow COW (MAP_PRIVATE) mappings of devices. I think forcing them shared
is reasonable.
-DG
David Greenman
Core-team/Principal Architect, The FreeBSD Project
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199603161406.GAA07890>