On 04/11/2018 08:37 AM, Jann Horn wrote: > On Wed, Apr 11, 2018 at 2:04 PM, <mho...@kernel.org> wrote: >> From: Michal Hocko <mho...@suse.com> >> >> 4.17+ kernels offer a new MAP_FIXED_NOREPLACE flag which allows the caller to >> atomicaly probe for a given address range. >> >> [wording heavily updated by John Hubbard <jhubb...@nvidia.com>] >> Signed-off-by: Michal Hocko <mho...@suse.com> >> --- >> Hi, >> Andrew's sent the MAP_FIXED_NOREPLACE to Linus for the upcoming merge >> window. So here we go with the man page update. >> >> man2/mmap.2 | 27 +++++++++++++++++++++++++++ >> 1 file changed, 27 insertions(+) >> >> diff --git a/man2/mmap.2 b/man2/mmap.2 >> index ea64eb8f0dcc..f702f3e4eba2 100644 >> --- a/man2/mmap.2 >> +++ b/man2/mmap.2 >> @@ -261,6 +261,27 @@ Examples include >> and the PAM libraries >> .UR http://www.linux-pam.org >> .UE . >> +Newer kernels >> +(Linux 4.17 and later) have a >> +.B MAP_FIXED_NOREPLACE >> +option that avoids the corruption problem; if available, MAP_FIXED_NOREPLACE >> +should be preferred over MAP_FIXED. > > This still looks wrong to me. There are legitimate uses for MAP_FIXED, > and for most users of MAP_FIXED that I'm aware of, MAP_FIXED_NOREPLACE > wouldn't work while MAP_FIXED works perfectly well. > > MAP_FIXED is for when you have already reserved the targeted memory > area using another VMA; MAP_FIXED_NOREPLACE is for when you haven't.
That's a nice summary, I hope it shows up in your upcoming patch. I recall that we went back and forth, trying to find a balance of explaining this feature, without providing overly-elaborate examples (which I tend toward). > Please don't make it sound as if MAP_FIXED is always wrong. > Agreed. thanks, -- John Hubbard NVIDIA