linux-kernel - Re: [PATCH 0/6] Overhaul memalloc

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Sat, 27 Jun 2020 09:09:09 -0400 (EDT)
From:   Mikulas Patocka <mpatocka@...hat.com>
To:     Dave Chinner <david@...morbit.com>
cc:     "Matthew Wilcox (Oracle)" <willy@...radead.org>,
        linux-kernel@...r.kernel.org, linux-mm@...ck.org,
        linux-xfs@...r.kernel.org, dm-devel@...hat.com,
        Jens Axboe <axboe@...nel.dk>, NeilBrown <neilb@...e.de>
Subject: Re: [PATCH 0/6] Overhaul memalloc_no*



On Sat, 27 Jun 2020, Dave Chinner wrote:

> On Fri, Jun 26, 2020 at 11:02:19AM -0400, Mikulas Patocka wrote:
> > Hi
> > 
> > I suggest to join memalloc_noio and memalloc_nofs into just one flag that 
> > prevents both filesystem recursion and i/o recursion.
> > 
> > Note that any I/O can recurse into a filesystem via the loop device, thus 
> > it doesn't make much sense to have a context where PF_MEMALLOC_NOFS is set 
> > and PF_MEMALLOC_NOIO is not set.
> 
> Correct me if I'm wrong, but I think that will prevent swapping from
> GFP_NOFS memory reclaim contexts.

Yes.

> IOWs, this will substantially
> change the behaviour of the memory reclaim system under sustained
> GFP_NOFS memory pressure. Sustained GFP_NOFS memory pressure is
> quite common, so I really don't think we want to telling memory
> reclaim "you can't do IO at all" when all we are trying to do is
> prevent recursion back into the same filesystem.

So, we can define __GFP_ONLY_SWAP_IO and __GFP_IO.

> Given that the loop device IO path already operates under
> memalloc_noio context, (i.e. the recursion restriction is applied in
> only the context that needs is) I see no reason for making that a
> global reclaim limitation....

I think this is a problem.

Suppose that a filesystem does GFP_NOFS allocation, the allocation 
triggers an IO and waits for it to finish, the loop device driver 
redirects the IO to the same filesystem that did the GFP_NOFS allocation.

I saw this deadlock in the past in the dm-bufio subsystem - see the commit 
9d28eb12447ee08bb5d1e8bb3195cf20e1ecd1c0 that fixed it.

Other subsystems that do IO in GFP_NOFS context may deadlock just like 
bufio.

Mikulas