From: Vivek Goyal <vgoyal@redhat.com>
To: Chirantan Ekbote <chirantan@chromium.org>
Cc: mst@redhat.com,
"Dr. David Alan Gilbert \(git\)" <dgilbert@redhat.com>,
qemu-devel@nongnu.org, virtio-fs-list <virtio-fs@redhat.com>,
Stefan Hajnoczi <stefanha@redhat.com>,
marcandre.lureau@redhat.com
Subject: Re: [Virtio-fs] [PATCH 07/24] DAX: virtio-fs: Add vhost-user slave commands for mapping
Date: Mon, 15 Feb 2021 09:24:15 -0500 [thread overview]
Message-ID: <20210215142415.GA3331@redhat.com> (raw)
In-Reply-To: <CAJFHJrqioXu7K8Z=dFgMQDWoKoNK3pddcgf9znCS72BR4WHSYA@mail.gmail.com>
On Mon, Feb 15, 2021 at 07:35:53PM +0900, Chirantan Ekbote wrote:
> On Wed, Feb 10, 2021 at 4:04 AM Dr. David Alan Gilbert (git)
> <dgilbert@redhat.com> wrote:
> >
> > From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
> > +
> > +typedef struct {
> > + /* Offsets within the file being mapped */
> > + uint64_t fd_offset[VHOST_USER_FS_SLAVE_ENTRIES];
> > + /* Offsets within the cache */
> > + uint64_t c_offset[VHOST_USER_FS_SLAVE_ENTRIES];
> > + /* Lengths of sections */
> > + uint64_t len[VHOST_USER_FS_SLAVE_ENTRIES];
> > + /* Flags, from VHOST_USER_FS_FLAG_* */
> > + uint64_t flags[VHOST_USER_FS_SLAVE_ENTRIES];
> > +} VhostUserFSSlaveMsg;
> > +
>
> Is it too late to change this? This struct allocates space for up to
> 8 entries but most of the time the server will only try to set up one
> mapping at a time so only 32 out of the 256 bytes in the message are
> actually being used. We're just wasting time memcpy'ing bytes that
> will never be used. Is there some reason this can't be dynamically
> sized? Something like:
>
> typedef struct {
> /* Number of mapping requests */
> uint16_t num_requests;
> /* `num_requests` mapping requests */
> MappingRequest requests[];
> } VhostUserFSSlaveMsg;
>
> typedef struct {
> /* Offset within the file being mapped */
> uint64_t fd_offset;
> /* Offset within the cache */
> uint64_t c_offset;
> /* Length of section */
> uint64_t len;
> /* Flags, from VHOST_USER_FS_FLAG_* */
> uint64_t flags;
> } MappingRequest;
>
> The current pre-allocated structure both wastes space when there are
> fewer than 8 requests and requires extra messages to be sent when
> there are more than 8 requests. I realize that in the grand scheme of
> things copying 224 extra bytes is basically not noticeable but it just
> irks me that we could fix this really easily before it gets propagated
> to too many other places.
Sounds like a reasonable idea. We probably will have to dynamically
allocate memory for removemapping, hopefully that does not have a
performance impact.
Vivek
WARNING: multiple messages have this Message-ID (diff)
From: Vivek Goyal <vgoyal@redhat.com>
To: Chirantan Ekbote <chirantan@chromium.org>
Cc: mst@redhat.com, qemu-devel@nongnu.org,
virtio-fs-list <virtio-fs@redhat.com>,
marcandre.lureau@redhat.com
Subject: Re: [Virtio-fs] [PATCH 07/24] DAX: virtio-fs: Add vhost-user slave commands for mapping
Date: Mon, 15 Feb 2021 09:24:15 -0500 [thread overview]
Message-ID: <20210215142415.GA3331@redhat.com> (raw)
In-Reply-To: <CAJFHJrqioXu7K8Z=dFgMQDWoKoNK3pddcgf9znCS72BR4WHSYA@mail.gmail.com>
On Mon, Feb 15, 2021 at 07:35:53PM +0900, Chirantan Ekbote wrote:
> On Wed, Feb 10, 2021 at 4:04 AM Dr. David Alan Gilbert (git)
> <dgilbert@redhat.com> wrote:
> >
> > From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
> > +
> > +typedef struct {
> > + /* Offsets within the file being mapped */
> > + uint64_t fd_offset[VHOST_USER_FS_SLAVE_ENTRIES];
> > + /* Offsets within the cache */
> > + uint64_t c_offset[VHOST_USER_FS_SLAVE_ENTRIES];
> > + /* Lengths of sections */
> > + uint64_t len[VHOST_USER_FS_SLAVE_ENTRIES];
> > + /* Flags, from VHOST_USER_FS_FLAG_* */
> > + uint64_t flags[VHOST_USER_FS_SLAVE_ENTRIES];
> > +} VhostUserFSSlaveMsg;
> > +
>
> Is it too late to change this? This struct allocates space for up to
> 8 entries but most of the time the server will only try to set up one
> mapping at a time so only 32 out of the 256 bytes in the message are
> actually being used. We're just wasting time memcpy'ing bytes that
> will never be used. Is there some reason this can't be dynamically
> sized? Something like:
>
> typedef struct {
> /* Number of mapping requests */
> uint16_t num_requests;
> /* `num_requests` mapping requests */
> MappingRequest requests[];
> } VhostUserFSSlaveMsg;
>
> typedef struct {
> /* Offset within the file being mapped */
> uint64_t fd_offset;
> /* Offset within the cache */
> uint64_t c_offset;
> /* Length of section */
> uint64_t len;
> /* Flags, from VHOST_USER_FS_FLAG_* */
> uint64_t flags;
> } MappingRequest;
>
> The current pre-allocated structure both wastes space when there are
> fewer than 8 requests and requires extra messages to be sent when
> there are more than 8 requests. I realize that in the grand scheme of
> things copying 224 extra bytes is basically not noticeable but it just
> irks me that we could fix this really easily before it gets propagated
> to too many other places.
Sounds like a reasonable idea. We probably will have to dynamically
allocate memory for removemapping, hopefully that does not have a
performance impact.
Vivek
next prev parent reply other threads:[~2021-02-15 14:25 UTC|newest]
Thread overview: 138+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-09 19:02 [PATCH 00/24] virtiofs dax patches Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [PATCH 01/24] DAX: vhost-user: Rework slave return values Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-11 9:59 ` Stefan Hajnoczi
2021-02-11 9:59 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-11 15:27 ` Vivek Goyal
2021-02-11 15:27 ` [Virtio-fs] " Vivek Goyal
2021-02-18 12:18 ` Dr. David Alan Gilbert
2021-02-18 12:18 ` [Virtio-fs] " Dr. David Alan Gilbert
2021-02-09 19:02 ` [PATCH 02/24] DAX: libvhost-user: Route slave message payload Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-11 10:05 ` Stefan Hajnoczi
2021-02-11 10:05 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-09 19:02 ` [PATCH 03/24] DAX: libvhost-user: Allow popping a queue element with bad pointers Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-11 10:12 ` Stefan Hajnoczi
2021-02-11 10:12 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-09 19:02 ` [PATCH 04/24] DAX subprojects/libvhost-user: Add virtio-fs slave types Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-11 10:16 ` Stefan Hajnoczi
2021-02-11 10:16 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-09 19:02 ` [PATCH 05/24] DAX: virtio: Add shared memory capability Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-11 10:17 ` Stefan Hajnoczi
2021-02-11 10:17 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-09 19:02 ` [PATCH 06/24] DAX: virtio-fs: Add cache BAR Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-11 10:25 ` Stefan Hajnoczi
2021-02-11 10:25 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-18 17:33 ` Dr. David Alan Gilbert
2021-02-18 17:33 ` [Virtio-fs] " Dr. David Alan Gilbert
2021-02-09 19:02 ` [PATCH 07/24] DAX: virtio-fs: Add vhost-user slave commands for mapping Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-11 10:32 ` Stefan Hajnoczi
2021-02-11 10:32 ` [Virtio-fs] " Stefan Hajnoczi
2021-03-08 17:04 ` Dr. David Alan Gilbert
2021-03-08 17:04 ` [Virtio-fs] " Dr. David Alan Gilbert
2021-02-15 10:35 ` Chirantan Ekbote
2021-02-15 10:35 ` Chirantan Ekbote
2021-02-15 13:25 ` Dr. David Alan Gilbert
2021-02-15 13:25 ` Dr. David Alan Gilbert
2021-02-15 14:24 ` Vivek Goyal [this message]
2021-02-15 14:24 ` Vivek Goyal
2021-03-11 12:15 ` Dr. David Alan Gilbert
2021-03-11 12:15 ` Dr. David Alan Gilbert
2021-03-11 13:50 ` Vivek Goyal
2021-03-11 13:50 ` Vivek Goyal
2021-03-11 18:52 ` Dr. David Alan Gilbert
2021-03-11 18:52 ` Dr. David Alan Gilbert
2021-02-09 19:02 ` [PATCH 08/24] DAX: virtio-fs: Fill in " Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-11 10:57 ` Stefan Hajnoczi
2021-02-11 10:57 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-18 10:59 ` Dr. David Alan Gilbert
2021-02-18 10:59 ` [Virtio-fs] " Dr. David Alan Gilbert
2021-02-09 19:02 ` [PATCH 09/24] DAX: virtiofsd Add cache accessor functions Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-11 12:31 ` Stefan Hajnoczi
2021-02-11 12:31 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-09 19:02 ` [PATCH 10/24] DAX: virtiofsd: Add setup/remove mappings fuse commands Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-11 12:37 ` Stefan Hajnoczi
2021-02-11 12:37 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-11 16:39 ` Dr. David Alan Gilbert
2021-02-11 16:39 ` [Virtio-fs] " Dr. David Alan Gilbert
2021-02-11 18:30 ` Vivek Goyal
2021-02-11 18:30 ` [Virtio-fs] " Vivek Goyal
2021-02-11 19:50 ` Dr. David Alan Gilbert
2021-02-11 19:50 ` [Virtio-fs] " Dr. David Alan Gilbert
2021-02-11 20:15 ` Vivek Goyal
2021-02-11 20:15 ` [Virtio-fs] " Vivek Goyal
2021-02-09 19:02 ` [PATCH 11/24] DAX: virtiofsd: Add setup/remove mapping handlers to passthrough_ll Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-11 12:37 ` Stefan Hajnoczi
2021-02-11 12:37 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-09 19:02 ` [PATCH 12/24] DAX: virtiofsd: Wire up passthrough_ll's lo_setupmapping Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-11 12:41 ` Stefan Hajnoczi
2021-02-11 12:41 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-11 16:05 ` Vivek Goyal
2021-02-11 16:05 ` [Virtio-fs] " Vivek Goyal
2021-02-09 19:02 ` [PATCH 13/24] DAX: virtiofsd: Make lo_removemapping() work Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-11 12:41 ` Stefan Hajnoczi
2021-02-11 12:41 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-09 19:02 ` [PATCH 14/24] DAX: virtiofsd: route se down to destroy method Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-11 12:42 ` Stefan Hajnoczi
2021-02-11 12:42 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-09 19:02 ` [PATCH 15/24] DAX: virtiofsd: Perform an unmap on destroy Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-11 12:42 ` Stefan Hajnoczi
2021-02-11 12:42 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-09 19:02 ` [PATCH 16/24] DAX/unmap: virtiofsd: Add VHOST_USER_SLAVE_FS_IO Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-11 14:17 ` Stefan Hajnoczi
2021-02-11 14:17 ` [Virtio-fs] " Stefan Hajnoczi
2021-03-16 19:59 ` Dr. David Alan Gilbert
2021-03-16 19:59 ` [Virtio-fs] " Dr. David Alan Gilbert
2021-03-31 10:12 ` Stefan Hajnoczi
2021-03-31 10:12 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-09 19:02 ` [PATCH 17/24] DAX/unmap virtiofsd: Add wrappers for VHOST_USER_SLAVE_FS_IO Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-11 14:18 ` Stefan Hajnoczi
2021-02-11 14:18 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-09 19:02 ` [PATCH 18/24] DAX/unmap virtiofsd: Parse unmappable elements Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-11 14:29 ` Stefan Hajnoczi
2021-02-11 14:29 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-25 10:19 ` Dr. David Alan Gilbert
2021-02-25 10:19 ` [Virtio-fs] " Dr. David Alan Gilbert
2021-03-31 10:14 ` Stefan Hajnoczi
2021-03-31 10:14 ` [Virtio-fs] " Stefan Hajnoczi
2021-03-17 10:33 ` Dr. David Alan Gilbert
2021-03-17 10:33 ` [Virtio-fs] " Dr. David Alan Gilbert
2021-02-09 19:02 ` [PATCH 19/24] DAX/unmap virtiofsd: Route unmappable reads Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [PATCH 20/24] DAX/unmap virtiofsd: route unmappable write to slave command Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [PATCH 21/24] DAX:virtiofsd: implement FUSE_INIT map_alignment field Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [PATCH 22/24] vhost-user-fs: Extend VhostUserFSSlaveMsg to pass additional info Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [PATCH 23/24] vhost-user-fs: Implement drop CAP_FSETID functionality Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
2021-02-11 14:35 ` Stefan Hajnoczi
2021-02-11 14:35 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-11 14:40 ` Vivek Goyal
2021-02-11 14:40 ` [Virtio-fs] " Vivek Goyal
2021-02-15 15:57 ` Stefan Hajnoczi
2021-02-15 15:57 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-16 15:57 ` Vivek Goyal
2021-02-16 15:57 ` [Virtio-fs] " Vivek Goyal
2021-02-22 16:53 ` Stefan Hajnoczi
2021-02-22 16:53 ` [Virtio-fs] " Stefan Hajnoczi
2021-02-09 19:02 ` [PATCH 24/24] virtiofsd: Ask qemu to drop CAP_FSETID if client asked for it Dr. David Alan Gilbert (git)
2021-02-09 19:02 ` [Virtio-fs] " Dr. David Alan Gilbert (git)
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=20210215142415.GA3331@redhat.com \
--to=vgoyal@redhat.com \
--cc=chirantan@chromium.org \
--cc=dgilbert@redhat.com \
--cc=marcandre.lureau@redhat.com \
--cc=mst@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=stefanha@redhat.com \
--cc=virtio-fs@redhat.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.