From: Juergen Gross <jgross@suse.com>
To: Hugh Dickins <hughd@google.com>,
Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: david.vrabel@citrix.com, xen-devel@lists.xenproject.org,
linux-kernel@vger.kernel.org, olaf@aepfle.de
Subject: Re: [PATCH v4] xen/gntdev: Use VM_MIXEDMAP instead of VM_IO to avoid NUMA balancing
Date: Tue, 22 Nov 2016 06:21:15 +0100 [thread overview]
Message-ID: <ffb6def6-4f89-fa23-2a52-f7b948f2dbb6@suse.com> (raw)
In-Reply-To: <alpine.LSU.2.11.1611211952080.1287@eggly.anvils>
On 22/11/16 04:59, Hugh Dickins wrote:
> On Mon, 21 Nov 2016, Boris Ostrovsky wrote:
>
>> Commit 9c17d96500f7 ("xen/gntdev: Grant maps should not be subject to
>> NUMA balancing") set VM_IO flag to prevent grant maps from being
>> subjected to NUMA balancing.
>>
>> It was discovered recently that this flag causes get_user_pages() to
>> always fail with -EFAULT.
>>
>> check_vma_flags
>> __get_user_pages
>> __get_user_pages_locked
>> __get_user_pages_unlocked
>> get_user_pages_fast
>> iov_iter_get_pages
>> dio_refill_pages
>> do_direct_IO
>> do_blockdev_direct_IO
>> do_blockdev_direct_IO
>> ext4_direct_IO_read
>> generic_file_read_iter
>> aio_run_iocb
>>
>> (which can happen if guest's vdisk has direct-io-safe option).
>>
>> To avoid this let's use VM_MIXEDMAP flag instead --- it prevents
>> NUMA balancing just as VM_IO does and has no effect on
>> check_vma_flags().
>>
>> Reported-by: Olaf Hering <olaf@aepfle.de>
>> Suggested-by: Hugh Dickins <hughd@google.com>
>> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
>
> Acked-by: Hugh Dickins <hughd@google.com>
>
> But nicer with the addition of a code comment saying
> /* VM_MIXEDMAP to forbid NUMA balancing but permit get_user_pages() */
I can fix that up when committing.
Juergen
>
>> ---
>> drivers/xen/gntdev.c | 2 +-
>> 1 files changed, 1 insertions(+), 1 deletions(-)
>>
>> diff --git a/drivers/xen/gntdev.c b/drivers/xen/gntdev.c
>> index bb95212..2ef2b61 100644
>> --- a/drivers/xen/gntdev.c
>> +++ b/drivers/xen/gntdev.c
>> @@ -1007,7 +1007,7 @@ static int gntdev_mmap(struct file *flip, struct vm_area_struct *vma)
>>
>> vma->vm_ops = &gntdev_vmops;
>>
>> - vma->vm_flags |= VM_DONTEXPAND | VM_DONTDUMP | VM_IO;
>> + vma->vm_flags |= VM_DONTEXPAND | VM_DONTDUMP | VM_MIXEDMAP;
>>
>> if (use_ptemod)
>> vma->vm_flags |= VM_DONTCOPY;
>> --
>> 1.7.1
>
next prev parent reply other threads:[~2016-11-22 5:21 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-11-21 14:56 Boris Ostrovsky
2016-11-22 3:59 ` Hugh Dickins
2016-11-22 5:21 ` Juergen Gross [this message]
2016-11-23 10:16 ` Olaf Hering
2016-11-28 13:16 ` Juergen Gross
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=ffb6def6-4f89-fa23-2a52-f7b948f2dbb6@suse.com \
--to=jgross@suse.com \
--cc=boris.ostrovsky@oracle.com \
--cc=david.vrabel@citrix.com \
--cc=hughd@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=olaf@aepfle.de \
--cc=xen-devel@lists.xenproject.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®