[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20161110162633.GA5301@aepfle.de>
Date: Thu, 10 Nov 2016 17:26:33 +0100
From: Olaf Hering <olaf@...fle.de>
To: Boris Ostrovsky <boris.ostrovsky@...cle.com>
Cc: konrad.wilk@...cle.com, david.vrabel@...rix.com,
linux-kernel@...r.kernel.org, stable@...r.kernel.org,
xen-devel@...ts.xen.org
Subject: Re: [Xen-devel] [PATCH RESEND] xen/gntdev: Grant maps should not be
subject to NUMA balancing
On Tue, Nov 10, Boris Ostrovsky wrote:
> Doing so will cause the grant to be unmapped and then, during
> fault handling, the fault to be mistakenly treated as NUMA hint
> fault.
>
> In addition, even if those maps could partcipate in NUMA
> balancing, it wouldn't provide any benefit since we are unable
> to determine physical page's node (even if/when VNUMA is
> implemented).
>
> Marking grant maps' VMAs as VM_IO will exclude them from being
> part of NUMA balancing.
This breaks qdisk+aio because now such pages are rejected 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
domU.cfg:
builder=hvm
disk=['vdev=xvda, direct-io-safe, backendtype=qdisk, target=img.raw']
> @@ -802,7 +802,7 @@ static int gntdev_mmap(struct file *flip, struct vm_area_struct *vma)
> - vma->vm_flags |= VM_DONTEXPAND | VM_DONTDUMP;
> + vma->vm_flags |= VM_DONTEXPAND | VM_DONTDUMP | VM_IO;
Olaf
Download attachment "signature.asc" of type "application/pgp-signature" (164 bytes)
Powered by blists - more mailing lists