lkml.org 
[lkml]   [2021]   [Sep]   [29]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [RFC PATCH 0/8] mm/madvise: support process_madvise(MADV_DONTNEED)
On Mon 27-09-21 12:12:46, Nadav Amit wrote:
>
> > On Sep 27, 2021, at 5:16 AM, Michal Hocko <mhocko@suse.com> wrote:
> >
> > On Mon 27-09-21 05:00:11, Nadav Amit wrote:
> > [...]
> >> The manager is notified on memory regions that it should monitor
> >> (through PTRACE/LD_PRELOAD/explicit-API). It then monitors these regions
> >> using the remote-userfaultfd that you saw on the second thread. When it wants
> >> to reclaim (anonymous) memory, it:
> >>
> >> 1. Uses UFFD-WP to protect that memory (and for this matter I got a vectored
> >> UFFD-WP to do so efficiently, a patch which I did not send yet).
> >> 2. Calls process_vm_readv() to read that memory of that process.
> >> 3. Write it back to “swap”.
> >> 4. Calls process_madvise(MADV_DONTNEED) to zap it.
> >
> > Why cannot you use MADV_PAGEOUT/MADV_COLD for this usecase?
>
> Providing hints to the kernel takes you so far to a certain extent.
> The kernel does not want to (for a good reason) to be completely
> configurable when it comes to reclaim and prefetch policies. Doing
> so from userspace allows you to be fully configurable.

I am sorry but I do not follow. Your scenario is describing a user
space driven reclaim. Something that MADV_{COLD,PAGEOUT} have been
designed for. What are you missing in the existing functionality?

> > MADV_DONTNEED on a remote process has been proposed in the past several
> > times and it has always been rejected because it is a free ticket to all
> > sorts of hard to debug problems as it is just a free ticket for a remote
> > memory corruption. An additional capability requirement might reduce the
> > risk to some degree but I still do not think this is a good idea.
>
> I would argue that there is nothing bad that remote MADV_DONTNEED can do
> that process_vm_writev() cannot do as well (putting aside ptrace).

I am not arguing this would be the first syscall to allow tricky and
hard to debug corruptions if used without care.

> process_vm_writev() is checking:
>
> mm = mm_access(task, PTRACE_MODE_ATTACH_REALCREDS)
>
> Wouldn't adding such a condition suffice?

This would be a minimum requirement. Another one is a sensible usecase
that is not covered by an existing functionality.

--
Michal Hocko
SUSE Labs

\
 
 \ /
  Last update: 2021-09-29 09:52    [W:0.472 / U:0.716 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site