From: Al Viro <viro@zeniv.linux.org.uk>
To: Xie Yongji <xieyongji@bytedance.com>
Cc: axboe@kernel.dk, corbet@lwn.net, linux-kernel@vger.kernel.org,
kvm@vger.kernel.org, mst@redhat.com, netdev@vger.kernel.org,
jasowang@redhat.com, rdunlap@infradead.org,
iommu@lists.linux-foundation.org, willy@infradead.org,
virtualization@lists.linux-foundation.org, hch@infradead.org,
christian.brauner@canonical.com, bcrl@kvack.org,
mika.penttila@nextfour.com, stefanha@redhat.com,
linux-fsdevel@vger.kernel.org, dan.carpenter@oracle.com,
sgarzare@redhat.com
Subject: Re: [PATCH v7 11/12] vduse: Introduce VDUSE - vDPA Device in Userspace
Date: Thu, 20 May 2021 06:28:37 +0000 [thread overview]
Message-ID: <YKYBle/F8aOgHO9p@zeniv-ca.linux.org.uk> (raw)
In-Reply-To: <20210517095513.850-12-xieyongji@bytedance.com>
On Mon, May 17, 2021 at 05:55:12PM +0800, Xie Yongji wrote:
> + case VDUSE_IOTLB_GET_FD: {
> + struct vduse_iotlb_entry entry;
> + struct vhost_iotlb_map *map;
> + struct vdpa_map_file *map_file;
> + struct vduse_iova_domain *domain = dev->domain;
> + struct file *f = NULL;
> +
> + ret = -EFAULT;
> + if (copy_from_user(&entry, argp, sizeof(entry)))
> + break;
return -EFAULT;
surely?
> +
> + ret = -EINVAL;
> + if (entry.start > entry.last)
> + break;
... and similar here, etc.
> + spin_lock(&domain->iotlb_lock);
> + map = vhost_iotlb_itree_first(domain->iotlb,
> + entry.start, entry.last);
> + if (map) {
> + map_file = (struct vdpa_map_file *)map->opaque;
> + f = get_file(map_file->file);
> + entry.offset = map_file->offset;
> + entry.start = map->start;
> + entry.last = map->last;
> + entry.perm = map->perm;
> + }
> + spin_unlock(&domain->iotlb_lock);
> + ret = -EINVAL;
> + if (!f)
> + break;
> +
> + ret = -EFAULT;
> + if (copy_to_user(argp, &entry, sizeof(entry))) {
> + fput(f);
> + break;
> + }
> + ret = receive_fd(f, perm_to_file_flags(entry.perm));
> + fput(f);
> + break;
IDGI. The main difference between receive_fd() and plain old
get_unused_fd_flags() + fd_install() is __receive_sock() call.
Which does nothing whatsoever in case of non-sockets. Can you
get a socket here?
IOW, why bother with that crap at all, nevermind exporting it?
_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu
WARNING: multiple messages have this Message-ID (diff)
From: Al Viro <viro@zeniv.linux.org.uk>
To: Xie Yongji <xieyongji@bytedance.com>
Cc: axboe@kernel.dk, corbet@lwn.net, linux-kernel@vger.kernel.org,
kvm@vger.kernel.org, mst@redhat.com, netdev@vger.kernel.org,
rdunlap@infradead.org, iommu@lists.linux-foundation.org,
willy@infradead.org, virtualization@lists.linux-foundation.org,
hch@infradead.org, christian.brauner@canonical.com,
bcrl@kvack.org, mika.penttila@nextfour.com, stefanha@redhat.com,
linux-fsdevel@vger.kernel.org, joro@8bytes.org,
dan.carpenter@oracle.com
Subject: Re: [PATCH v7 11/12] vduse: Introduce VDUSE - vDPA Device in Userspace
Date: Thu, 20 May 2021 06:28:37 +0000 [thread overview]
Message-ID: <YKYBle/F8aOgHO9p@zeniv-ca.linux.org.uk> (raw)
In-Reply-To: <20210517095513.850-12-xieyongji@bytedance.com>
On Mon, May 17, 2021 at 05:55:12PM +0800, Xie Yongji wrote:
> + case VDUSE_IOTLB_GET_FD: {
> + struct vduse_iotlb_entry entry;
> + struct vhost_iotlb_map *map;
> + struct vdpa_map_file *map_file;
> + struct vduse_iova_domain *domain = dev->domain;
> + struct file *f = NULL;
> +
> + ret = -EFAULT;
> + if (copy_from_user(&entry, argp, sizeof(entry)))
> + break;
return -EFAULT;
surely?
> +
> + ret = -EINVAL;
> + if (entry.start > entry.last)
> + break;
... and similar here, etc.
> + spin_lock(&domain->iotlb_lock);
> + map = vhost_iotlb_itree_first(domain->iotlb,
> + entry.start, entry.last);
> + if (map) {
> + map_file = (struct vdpa_map_file *)map->opaque;
> + f = get_file(map_file->file);
> + entry.offset = map_file->offset;
> + entry.start = map->start;
> + entry.last = map->last;
> + entry.perm = map->perm;
> + }
> + spin_unlock(&domain->iotlb_lock);
> + ret = -EINVAL;
> + if (!f)
> + break;
> +
> + ret = -EFAULT;
> + if (copy_to_user(argp, &entry, sizeof(entry))) {
> + fput(f);
> + break;
> + }
> + ret = receive_fd(f, perm_to_file_flags(entry.perm));
> + fput(f);
> + break;
IDGI. The main difference between receive_fd() and plain old
get_unused_fd_flags() + fd_install() is __receive_sock() call.
Which does nothing whatsoever in case of non-sockets. Can you
get a socket here?
IOW, why bother with that crap at all, nevermind exporting it?
_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization
WARNING: multiple messages have this Message-ID (diff)
From: Al Viro <viro@zeniv.linux.org.uk>
To: Xie Yongji <xieyongji@bytedance.com>
Cc: mst@redhat.com, jasowang@redhat.com, stefanha@redhat.com,
sgarzare@redhat.com, parav@nvidia.com, hch@infradead.org,
christian.brauner@canonical.com, rdunlap@infradead.org,
willy@infradead.org, axboe@kernel.dk, bcrl@kvack.org,
corbet@lwn.net, mika.penttila@nextfour.com,
dan.carpenter@oracle.com, joro@8bytes.org,
virtualization@lists.linux-foundation.org,
netdev@vger.kernel.org, kvm@vger.kernel.org,
linux-fsdevel@vger.kernel.org, iommu@lists.linux-foundation.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH v7 11/12] vduse: Introduce VDUSE - vDPA Device in Userspace
Date: Thu, 20 May 2021 06:28:37 +0000 [thread overview]
Message-ID: <YKYBle/F8aOgHO9p@zeniv-ca.linux.org.uk> (raw)
In-Reply-To: <20210517095513.850-12-xieyongji@bytedance.com>
On Mon, May 17, 2021 at 05:55:12PM +0800, Xie Yongji wrote:
> + case VDUSE_IOTLB_GET_FD: {
> + struct vduse_iotlb_entry entry;
> + struct vhost_iotlb_map *map;
> + struct vdpa_map_file *map_file;
> + struct vduse_iova_domain *domain = dev->domain;
> + struct file *f = NULL;
> +
> + ret = -EFAULT;
> + if (copy_from_user(&entry, argp, sizeof(entry)))
> + break;
return -EFAULT;
surely?
> +
> + ret = -EINVAL;
> + if (entry.start > entry.last)
> + break;
... and similar here, etc.
> + spin_lock(&domain->iotlb_lock);
> + map = vhost_iotlb_itree_first(domain->iotlb,
> + entry.start, entry.last);
> + if (map) {
> + map_file = (struct vdpa_map_file *)map->opaque;
> + f = get_file(map_file->file);
> + entry.offset = map_file->offset;
> + entry.start = map->start;
> + entry.last = map->last;
> + entry.perm = map->perm;
> + }
> + spin_unlock(&domain->iotlb_lock);
> + ret = -EINVAL;
> + if (!f)
> + break;
> +
> + ret = -EFAULT;
> + if (copy_to_user(argp, &entry, sizeof(entry))) {
> + fput(f);
> + break;
> + }
> + ret = receive_fd(f, perm_to_file_flags(entry.perm));
> + fput(f);
> + break;
IDGI. The main difference between receive_fd() and plain old
get_unused_fd_flags() + fd_install() is __receive_sock() call.
Which does nothing whatsoever in case of non-sockets. Can you
get a socket here?
IOW, why bother with that crap at all, nevermind exporting it?
next prev parent reply other threads:[~2021-05-20 6:29 UTC|newest]
Thread overview: 123+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-05-17 9:55 [PATCH v7 00/12] Introduce VDUSE - vDPA Device in Userspace Xie Yongji
2021-05-17 9:55 ` Xie Yongji
2021-05-17 9:55 ` [PATCH v7 01/12] iova: Export alloc_iova_fast() Xie Yongji
2021-05-17 9:55 ` Xie Yongji
2021-05-26 2:36 ` Jason Wang
2021-05-26 2:36 ` Jason Wang
2021-05-26 2:36 ` Jason Wang
2021-05-26 2:43 ` Yongji Xie
2021-05-26 2:43 ` Yongji Xie
2021-05-17 9:55 ` [PATCH v7 02/12] file: Export receive_fd() to modules Xie Yongji
2021-05-17 9:55 ` Xie Yongji
2021-05-20 6:18 ` Al Viro
2021-05-20 6:18 ` Al Viro
2021-05-20 6:18 ` Al Viro
2021-05-20 6:32 ` Yongji Xie
2021-05-20 6:32 ` Yongji Xie
2021-05-17 9:55 ` [PATCH v7 03/12] eventfd: Increase the recursion depth of eventfd_signal() Xie Yongji
2021-05-17 9:55 ` Xie Yongji
2021-05-17 9:55 ` [PATCH v7 04/12] virtio-blk: Add validation for block size in config space Xie Yongji
2021-05-17 9:55 ` Xie Yongji
2021-05-19 13:39 ` Yongji Xie
2021-05-19 13:39 ` Yongji Xie
2021-05-19 14:42 ` Dan Carpenter
2021-05-19 14:42 ` Dan Carpenter
2021-05-19 14:42 ` Dan Carpenter
2021-05-20 5:25 ` Yongji Xie
2021-05-20 5:25 ` Yongji Xie
2021-05-20 5:43 ` Michael S. Tsirkin
2021-05-20 5:43 ` Michael S. Tsirkin
2021-05-20 5:43 ` Michael S. Tsirkin
2021-05-20 7:08 ` Yongji Xie
2021-05-20 7:08 ` Yongji Xie
2021-05-17 9:55 ` [PATCH v7 05/12] virtio_scsi: Add validation for residual bytes from response Xie Yongji
2021-05-17 9:55 ` Xie Yongji
2021-05-26 2:41 ` Jason Wang
2021-05-26 2:41 ` Jason Wang
2021-05-26 2:41 ` Jason Wang
2021-05-17 9:55 ` [PATCH v7 06/12] vhost-iotlb: Add an opaque pointer for vhost IOTLB Xie Yongji
2021-05-17 9:55 ` Xie Yongji
2021-05-17 9:55 ` [PATCH v7 07/12] vdpa: Add an opaque pointer for vdpa_config_ops.dma_map() Xie Yongji
2021-05-17 9:55 ` Xie Yongji
2021-05-17 9:55 ` [PATCH v7 08/12] vdpa: factor out vhost_vdpa_pa_map() and vhost_vdpa_pa_unmap() Xie Yongji
2021-05-17 9:55 ` Xie Yongji
2021-05-17 9:55 ` [PATCH v7 09/12] vdpa: Support transferring virtual addressing during DMA mapping Xie Yongji
2021-05-17 9:55 ` Xie Yongji
2021-05-17 9:55 ` [PATCH v7 10/12] vduse: Implement an MMU-based IOMMU driver Xie Yongji
2021-05-17 9:55 ` Xie Yongji
2021-05-17 9:55 ` [PATCH v7 11/12] vduse: Introduce VDUSE - vDPA Device in Userspace Xie Yongji
2021-05-17 9:55 ` Xie Yongji
2021-05-20 6:28 ` Al Viro [this message]
2021-05-20 6:28 ` Al Viro
2021-05-20 6:28 ` Al Viro
2021-05-20 7:03 ` Yongji Xie
2021-05-20 7:03 ` Yongji Xie
2021-05-27 4:12 ` Jason Wang
2021-05-27 4:12 ` Jason Wang
2021-05-27 4:12 ` Jason Wang
2021-05-27 4:57 ` Yongji Xie
2021-05-27 4:57 ` Yongji Xie
2021-05-27 5:00 ` Jason Wang
2021-05-27 5:00 ` Jason Wang
2021-05-27 5:00 ` Jason Wang
2021-05-27 5:08 ` Yongji Xie
2021-05-27 5:08 ` Yongji Xie
2021-05-27 5:40 ` Jason Wang
2021-05-27 5:40 ` Jason Wang
2021-05-27 5:40 ` Jason Wang
2021-05-27 7:34 ` Yongji Xie
2021-05-27 7:34 ` Yongji Xie
2021-05-27 8:41 ` Jason Wang
2021-05-27 8:41 ` Jason Wang
2021-05-27 8:41 ` Jason Wang
2021-05-27 8:43 ` Jason Wang
2021-05-27 8:43 ` Jason Wang
2021-05-27 8:43 ` Jason Wang
2021-05-27 10:14 ` Yongji Xie
2021-05-27 10:14 ` Yongji Xie
2021-05-28 1:33 ` Jason Wang
2021-05-28 1:33 ` Jason Wang
2021-05-28 1:33 ` Jason Wang
2021-05-28 3:54 ` Yongji Xie
2021-05-28 3:54 ` Yongji Xie
2021-05-28 6:38 ` Jason Wang
2021-05-28 6:38 ` Jason Wang
2021-05-28 6:38 ` Jason Wang
2021-05-27 13:17 ` Yongji Xie
2021-05-27 13:17 ` Yongji Xie
2021-05-28 2:31 ` Jason Wang
2021-05-28 2:31 ` Jason Wang
2021-05-28 2:31 ` Jason Wang
2021-05-31 4:27 ` Yongji Xie
2021-05-31 4:27 ` Yongji Xie
2021-05-31 4:38 ` Jason Wang
2021-05-31 4:38 ` Jason Wang
2021-05-31 4:38 ` Jason Wang
2021-05-31 6:24 ` Yongji Xie
2021-05-31 6:24 ` Yongji Xie
2021-05-31 4:56 ` Greg KH
2021-05-31 4:56 ` Greg KH
2021-05-31 4:56 ` Greg KH
2021-05-31 6:19 ` Yongji Xie
2021-05-31 6:19 ` Yongji Xie
2021-05-31 6:32 ` Greg KH
2021-05-31 6:32 ` Greg KH
2021-05-31 6:32 ` Greg KH
2021-05-31 7:13 ` Yongji Xie
2021-05-31 7:13 ` Yongji Xie
2021-05-17 9:55 ` [PATCH v7 12/12] Documentation: Add documentation for VDUSE Xie Yongji
2021-05-17 9:55 ` Xie Yongji
2021-05-20 6:06 ` [PATCH v7 00/12] Introduce VDUSE - vDPA Device in Userspace Michael S. Tsirkin
2021-05-20 6:06 ` Michael S. Tsirkin
2021-05-20 6:06 ` Michael S. Tsirkin
2021-05-20 9:06 ` Yongji Xie
2021-05-20 9:06 ` Yongji Xie
2021-05-25 6:40 ` Jason Wang
2021-05-25 6:40 ` Jason Wang
2021-05-25 6:40 ` Jason Wang
2021-05-25 6:48 ` Michael S. Tsirkin
2021-05-25 6:48 ` Michael S. Tsirkin
2021-05-25 6:48 ` Michael S. Tsirkin
2021-05-25 7:11 ` Jason Wang
2021-05-25 7:11 ` Jason Wang
2021-05-25 7:11 ` Jason Wang
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=YKYBle/F8aOgHO9p@zeniv-ca.linux.org.uk \
--to=viro@zeniv.linux.org.uk \
--cc=axboe@kernel.dk \
--cc=bcrl@kvack.org \
--cc=christian.brauner@canonical.com \
--cc=corbet@lwn.net \
--cc=dan.carpenter@oracle.com \
--cc=hch@infradead.org \
--cc=iommu@lists.linux-foundation.org \
--cc=jasowang@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mika.penttila@nextfour.com \
--cc=mst@redhat.com \
--cc=netdev@vger.kernel.org \
--cc=rdunlap@infradead.org \
--cc=sgarzare@redhat.com \
--cc=stefanha@redhat.com \
--cc=virtualization@lists.linux-foundation.org \
--cc=willy@infradead.org \
--cc=xieyongji@bytedance.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 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.