From: Andrea Arcangeli <aarcange@redhat.com>
To: "Aneesh Kumar K.V" <aneesh.kumar@linux.ibm.com>
Cc: Alexey Kardashevskiy <aik@ozlabs.ru>,
linux-kernel@vger.kernel.org, Michal Hocko <mhocko@kernel.org>,
linux-mm@kvack.org, akpm@linux-foundation.org,
linuxppc-dev@lists.ozlabs.org,
David Gibson <david@gibson.dropbear.id.au>
Subject: Re: [PATCH V6 3/4] powerpc/mm/iommu: Allow migration of cma allocated pages during mm_iommu_get
Date: Tue, 8 Jan 2019 20:53:59 -0500 [thread overview]
Message-ID: <20190109015359.GE20586@redhat.com> (raw)
In-Reply-To: <20190108045110.28597-4-aneesh.kumar@linux.ibm.com>
Hello,
On Tue, Jan 08, 2019 at 10:21:09AM +0530, Aneesh Kumar K.V wrote:
> @@ -187,41 +149,25 @@ static long mm_iommu_do_alloc(struct mm_struct *mm, unsigned long ua,
> goto unlock_exit;
> }
>
> + ret = get_user_pages_cma_migrate(ua, entries, 1, mem->hpages);
In terms of gup APIs, I've been wondering if this shall become
get_user_pages_longerm(FOLL_CMA_MIGRATE). So basically moving this
CMA migrate logic inside get_user_pages_longerm.
It depends if powerpc will ever need to bail on dax and/or if other
non-powerpc vfio drivers which are already bailing on dax may also
later optionally need to avoid interfering with CMA.
Aside from the API detail above, this CMA page migration logic seems a
good solution for the problem.
Thanks,
Andrea
WARNING: multiple messages have this Message-ID (diff)
From: Andrea Arcangeli <aarcange@redhat.com>
To: "Aneesh Kumar K.V" <aneesh.kumar@linux.ibm.com>
Cc: akpm@linux-foundation.org, Michal Hocko <mhocko@kernel.org>,
Alexey Kardashevskiy <aik@ozlabs.ru>,
David Gibson <david@gibson.dropbear.id.au>,
mpe@ellerman.id.au, linux-mm@kvack.org,
linux-kernel@vger.kernel.org, linuxppc-dev@lists.ozlabs.org
Subject: Re: [PATCH V6 3/4] powerpc/mm/iommu: Allow migration of cma allocated pages during mm_iommu_get
Date: Tue, 8 Jan 2019 20:53:59 -0500 [thread overview]
Message-ID: <20190109015359.GE20586@redhat.com> (raw)
In-Reply-To: <20190108045110.28597-4-aneesh.kumar@linux.ibm.com>
Hello,
On Tue, Jan 08, 2019 at 10:21:09AM +0530, Aneesh Kumar K.V wrote:
> @@ -187,41 +149,25 @@ static long mm_iommu_do_alloc(struct mm_struct *mm, unsigned long ua,
> goto unlock_exit;
> }
>
> + ret = get_user_pages_cma_migrate(ua, entries, 1, mem->hpages);
In terms of gup APIs, I've been wondering if this shall become
get_user_pages_longerm(FOLL_CMA_MIGRATE). So basically moving this
CMA migrate logic inside get_user_pages_longerm.
It depends if powerpc will ever need to bail on dax and/or if other
non-powerpc vfio drivers which are already bailing on dax may also
later optionally need to avoid interfering with CMA.
Aside from the API detail above, this CMA page migration logic seems a
good solution for the problem.
Thanks,
Andrea
next prev parent reply other threads:[~2019-01-09 1:55 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-08 4:51 [PATCH V6 0/4] mm/kvm/vfio/ppc64: Migrate compound pages out of CMA region Aneesh Kumar K.V
2019-01-08 4:51 ` Aneesh Kumar K.V
2019-01-08 4:51 ` [PATCH V6 1/4] mm/cma: Add PF flag to force non cma alloc Aneesh Kumar K.V
2019-01-08 4:51 ` Aneesh Kumar K.V
2019-01-09 2:38 ` Andrea Arcangeli
2019-01-09 2:38 ` Andrea Arcangeli
2019-01-08 4:51 ` [PATCH V6 2/4] mm: Add get_user_pages_cma_migrate Aneesh Kumar K.V
2019-01-08 4:51 ` Aneesh Kumar K.V
2019-01-08 4:51 ` [PATCH V6 3/4] powerpc/mm/iommu: Allow migration of cma allocated pages during mm_iommu_get Aneesh Kumar K.V
2019-01-08 4:51 ` Aneesh Kumar K.V
2019-01-09 1:53 ` Andrea Arcangeli [this message]
2019-01-09 1:53 ` Andrea Arcangeli
2019-01-09 8:40 ` Aneesh Kumar K.V
2019-01-09 8:40 ` Aneesh Kumar K.V
2019-01-08 4:51 ` [PATCH V6 4/4] powerpc/mm/iommu: Allow large IOMMU page size only for hugetlb backing Aneesh Kumar K.V
2019-01-08 4:51 ` Aneesh Kumar K.V
2019-01-08 19:56 ` [PATCH V6 0/4] mm/kvm/vfio/ppc64: Migrate compound pages out of CMA region Andrew Morton
2019-01-08 19:56 ` Andrew Morton
2019-01-09 8:41 ` Aneesh Kumar K.V
2019-01-09 8:41 ` Aneesh Kumar K.V
2019-01-10 4:11 ` David Gibson
2019-01-10 4:11 ` David Gibson
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190109015359.GE20586@redhat.com \
--to=aarcange@redhat.com \
--cc=aik@ozlabs.ru \
--cc=akpm@linux-foundation.org \
--cc=aneesh.kumar@linux.ibm.com \
--cc=david@gibson.dropbear.id.au \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=mhocko@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.