From: Peter Xu <peterx@redhat.com>
To: linux-kernel@vger.kernel.org, linux-mm@kvack.org, kvm@vger.kernel.org
Cc: Andrew Morton <akpm@linux-foundation.org>,
Alex Williamson <alex.williamson@redhat.com>,
Zi Yan <ziy@nvidia.com>, Jason Gunthorpe <jgg@nvidia.com>,
Alex Mastro <amastro@fb.com>,
David Hildenbrand <david@redhat.com>,
Nico Pache <npache@redhat.com>,
peterx@redhat.com
Subject: [PATCH 4/5] vfio: Introduce vfio_device_ops.get_unmapped_area hook
Date: Fri, 13 Jun 2025 09:41:10 -0400 [thread overview]
Message-ID: <20250613134111.469884-5-peterx@redhat.com> (raw)
In-Reply-To: <20250613134111.469884-1-peterx@redhat.com>
Add a hook to vfio_device_ops to allow sub-modules provide virtual
addresses for an mmap() request.
Note that the fallback will be mm_get_unmapped_area(), which should
maintain the old behavior of generic VA allocation (__get_unmapped_area).
It's a bit unfortunate that is needed, as the current get_unmapped_area()
file ops cannot support a retval which fallbacks to the default. So that
is needed both here and whenever sub-module will opt-in with its own.
Signed-off-by: Peter Xu <peterx@redhat.com>
---
drivers/vfio/vfio_main.c | 18 ++++++++++++++++++
include/linux/vfio.h | 7 +++++++
2 files changed, 25 insertions(+)
diff --git a/drivers/vfio/vfio_main.c b/drivers/vfio/vfio_main.c
index 1fd261efc582..19db8e58d223 100644
--- a/drivers/vfio/vfio_main.c
+++ b/drivers/vfio/vfio_main.c
@@ -1354,6 +1354,23 @@ static int vfio_device_fops_mmap(struct file *filep, struct vm_area_struct *vma)
return device->ops->mmap(device, vma);
}
+static unsigned long vfio_device_get_unmapped_area(struct file *file,
+ unsigned long addr,
+ unsigned long len,
+ unsigned long pgoff,
+ unsigned long flags)
+{
+ struct vfio_device_file *df = file->private_data;
+ struct vfio_device *device = df->device;
+
+ if (!device->ops->get_unmapped_area)
+ return mm_get_unmapped_area(current->mm, file, addr,
+ len, pgoff, flags);
+
+ return device->ops->get_unmapped_area(device, file, addr, len,
+ pgoff, flags);
+}
+
const struct file_operations vfio_device_fops = {
.owner = THIS_MODULE,
.open = vfio_device_fops_cdev_open,
@@ -1363,6 +1380,7 @@ const struct file_operations vfio_device_fops = {
.unlocked_ioctl = vfio_device_fops_unl_ioctl,
.compat_ioctl = compat_ptr_ioctl,
.mmap = vfio_device_fops_mmap,
+ .get_unmapped_area = vfio_device_get_unmapped_area,
};
static struct vfio_device *vfio_device_from_file(struct file *file)
diff --git a/include/linux/vfio.h b/include/linux/vfio.h
index 707b00772ce1..48fe71c61ed2 100644
--- a/include/linux/vfio.h
+++ b/include/linux/vfio.h
@@ -108,6 +108,7 @@ struct vfio_device {
* @dma_unmap: Called when userspace unmaps IOVA from the container
* this device is attached to.
* @device_feature: Optional, fill in the VFIO_DEVICE_FEATURE ioctl
+ * @get_unmapped_area: Optional, provide virtual address hint for mmap()
*/
struct vfio_device_ops {
char *name;
@@ -135,6 +136,12 @@ struct vfio_device_ops {
void (*dma_unmap)(struct vfio_device *vdev, u64 iova, u64 length);
int (*device_feature)(struct vfio_device *device, u32 flags,
void __user *arg, size_t argsz);
+ unsigned long (*get_unmapped_area)(struct vfio_device *device,
+ struct file *file,
+ unsigned long addr,
+ unsigned long len,
+ unsigned long pgoff,
+ unsigned long flags);
};
#if IS_ENABLED(CONFIG_IOMMUFD)
--
2.49.0
next prev parent reply other threads:[~2025-06-13 13:41 UTC|newest]
Thread overview: 78+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-06-13 13:41 [PATCH 0/5] mm/vfio: huge pfnmaps with !MAP_FIXED mappings Peter Xu
2025-06-13 13:41 ` [PATCH 1/5] mm: Deduplicate mm_get_unmapped_area() Peter Xu
2025-06-13 14:12 ` Jason Gunthorpe
2025-06-13 14:55 ` Oscar Salvador
2025-06-13 14:58 ` Zi Yan
2025-06-13 15:57 ` Lorenzo Stoakes
2025-06-13 17:00 ` Pedro Falcato
2025-06-13 18:00 ` David Hildenbrand
2025-06-16 8:01 ` David Laight
2025-06-17 21:13 ` Peter Xu
2025-06-13 13:41 ` [PATCH 2/5] mm/hugetlb: Remove prepare_hugepage_range() Peter Xu
2025-06-13 14:12 ` Jason Gunthorpe
2025-06-13 14:59 ` Oscar Salvador
2025-06-13 15:13 ` Zi Yan
2025-06-13 16:24 ` Peter Xu
2025-06-13 18:01 ` David Hildenbrand
2025-06-14 4:11 ` Liam R. Howlett
2025-06-17 21:07 ` Peter Xu
2025-06-13 13:41 ` [PATCH 3/5] mm: Rename __thp_get_unmapped_area to mm_get_unmapped_area_aligned Peter Xu
2025-06-13 14:17 ` Jason Gunthorpe
2025-06-13 15:13 ` Peter Xu
2025-06-13 16:00 ` Jason Gunthorpe
2025-06-13 18:31 ` Peter Xu
2025-06-13 15:19 ` Zi Yan
2025-06-13 18:33 ` Peter Xu
2025-06-13 15:36 ` Lorenzo Stoakes
2025-06-13 18:45 ` Peter Xu
2025-06-13 19:18 ` Lorenzo Stoakes
2025-06-13 20:34 ` Peter Xu
2025-06-14 5:58 ` Lorenzo Stoakes
2025-06-14 5:23 ` Liam R. Howlett
2025-06-16 12:14 ` Jason Gunthorpe
2025-06-16 12:20 ` Lorenzo Stoakes
2025-06-16 12:26 ` Jason Gunthorpe
2025-06-13 13:41 ` Peter Xu [this message]
2025-06-13 14:18 ` [PATCH 4/5] vfio: Introduce vfio_device_ops.get_unmapped_area hook Jason Gunthorpe
2025-06-13 18:03 ` David Hildenbrand
2025-06-14 14:46 ` kernel test robot
2025-06-17 15:39 ` Peter Xu
2025-06-17 15:41 ` Jason Gunthorpe
2025-06-17 16:47 ` Peter Xu
2025-06-17 19:39 ` Peter Xu
2025-06-17 19:46 ` Jason Gunthorpe
2025-06-17 20:01 ` Peter Xu
2025-06-17 23:00 ` Jason Gunthorpe
2025-06-17 23:26 ` Peter Xu
2025-06-13 13:41 ` [PATCH 5/5] vfio-pci: Best-effort huge pfnmaps with !MAP_FIXED mappings Peter Xu
2025-06-13 14:29 ` Jason Gunthorpe
2025-06-13 15:26 ` Peter Xu
2025-06-13 16:09 ` Jason Gunthorpe
2025-06-13 19:15 ` Peter Xu
2025-06-13 23:16 ` Jason Gunthorpe
2025-06-16 22:06 ` Peter Xu
2025-06-16 23:00 ` Jason Gunthorpe
2025-06-17 20:56 ` Peter Xu
2025-06-17 23:18 ` Jason Gunthorpe
2025-06-17 23:36 ` Peter Xu
2025-06-18 16:56 ` Peter Xu
2025-06-18 17:46 ` Jason Gunthorpe
2025-06-18 19:15 ` Peter Xu
2025-06-19 13:58 ` Jason Gunthorpe
2025-06-19 14:55 ` Peter Xu
2025-06-19 18:40 ` Jason Gunthorpe
2025-06-24 20:37 ` Peter Xu
2025-06-24 20:51 ` Peter Xu
2025-06-24 23:40 ` Jason Gunthorpe
2025-06-25 0:48 ` Peter Xu
2025-06-25 13:07 ` Jason Gunthorpe
2025-06-25 17:12 ` Peter Xu
2025-06-25 18:41 ` Jason Gunthorpe
2025-06-25 19:26 ` Peter Xu
2025-06-30 14:05 ` Jason Gunthorpe
2025-07-02 20:58 ` Peter Xu
2025-07-02 23:32 ` Jason Gunthorpe
2025-06-13 17:44 ` Alex Mastro
2025-06-13 18:53 ` Peter Xu
2025-06-13 18:09 ` David Hildenbrand
2025-06-13 19:21 ` Peter Xu
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=20250613134111.469884-5-peterx@redhat.com \
--to=peterx@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=alex.williamson@redhat.com \
--cc=amastro@fb.com \
--cc=david@redhat.com \
--cc=jgg@nvidia.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=npache@redhat.com \
--cc=ziy@nvidia.com \
/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;
as well as URLs for NNTP newsgroup(s).