From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
To: Michal Hocko <mhocko@kernel.org>,
        Luiz Capitulino <lcapitulino@redhat.com>,
        Mike Kravetz <mike.kravetz@oracle.com>
Cc: linux-mm@kvack.org, LKML <linux-kernel@vger.kernel.org>
Subject: Re: gigantic hugepages vs. movable zones
In-Reply-To: <20170726105004.GI2981@dhcp22.suse.cz>
References: <20170726105004.GI2981@dhcp22.suse.cz>
Date: Thu, 27 Jul 2017 07:52:08 +0530
MIME-Version: 1.0
Content-Type: text/plain
Message-Id: <87inie1uwf.fsf@linux.vnet.ibm.com>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 910
Lines: 22

Michal Hocko <mhocko@kernel.org> writes:

> Hi,
> I've just noticed that alloc_gigantic_page ignores movability of the
> gigantic page and it uses any existing zone. Considering that
> hugepage_migration_supported only supports 2MB and pgd level hugepages
> then 1GB pages are not migratable and as such allocating them from a
> movable zone will break the basic expectation of this zone. Standard
> hugetlb allocations try to avoid that by using htlb_alloc_mask and I
> believe we should do the same for gigantic pages as well.
>
> I suspect this behavior is not intentional. What do you think about the
> following untested patch?


I also noticed an unrelated issue with the usage of
start_isolate_page_range. On error we set the migrate type to
MIGRATE_MOVABLE. That may conflict with CMA pages ? Wondering whether
we should check for page's pageblock migrate type in
pfn_range_valid_gigantic() ?

-aneesh