linux-kernel - Re: [PATCH 2/2] mm,memory_hotplug: Fix shrink_{zone,node}

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]

Message-ID: <20190717073853.GA22253@linux>
Date:   Wed, 17 Jul 2019 09:38:58 +0200
From:   Oscar Salvador <osalvador@...e.de>
To:     Dan Williams <dan.j.williams@...el.com>
Cc:     "Aneesh Kumar K.V" <aneesh.kumar@...ux.ibm.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        David Hildenbrand <david@...hat.com>,
        Pavel Tatashin <pasha.tatashin@...een.com>,
        Michal Hocko <mhocko@...e.com>, Linux MM <linux-mm@...ck.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 2/2] mm,memory_hotplug: Fix shrink_{zone,node}_span

On Tue, Jul 16, 2019 at 07:28:54PM -0700, Dan Williams wrote:
> This makes it more clear that the problem is with the "start_pfn ==
> pfn" check relative to subsections, but it does not clarify why it
> needs to clear pfn_valid() before calling shrink_zone_span().
> Sections were not invalidated prior to shrink_zone_span() in the
> pre-subsection implementation and it seems all we need is to keep the
> same semantic. I.e. skip the range that is currently being removed:

Yes, as I said in my reply to Aneesh, that is the other way I thought
when fixing it.
The reason I went this way is because it seemed more reasonable and
natural to me that pfn_valid() would just return the next active
sub-section.

I just though that we could leverage the fact that we can deactivate
a sub-section before scanning for the next one.

On a second thought, the changes do not outweight the case, being the first
fix enough and less intrusive, so I will send a v2 with that instead.

-- 
Oscar Salvador
SUSE L3