hw/virtio/vhost-user.c | 14 ----- hw/virtio/vhost-vdpa.c | 1 - hw/virtio/vhost.c | 85 ++++++++++++++++++++----------- include/hw/virtio/vhost-backend.h | 4 -- 4 files changed, 56 insertions(+), 48 deletions(-)
Following up on my previous work to make virtio-mem consume multiple
memslots dynamically [1] that requires precise accounting between used vs.
reserved memslots, I realized that vhost makes this extra hard by
filtering out some memory region sections (so they don't consume a
memslot) in the vhost-user case, which messes up the whole memslot
accounting.
This series fixes what I found to be broken and prepares for more work on
[1]. Further, it cleanes up the merge checks that I consider unnecessary.
[1] https://lkml.kernel.org/r/20211027124531.57561-8-david@redhat.com
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Cc: Dr. David Alan Gilbert <dgilbert@redhat.com>
David Hildenbrand (2):
vhost: Defer filtering memory sections until building the vhost memory
structure
vhost: Remove vhost_backend_can_merge() callback
hw/virtio/vhost-user.c | 14 -----
hw/virtio/vhost-vdpa.c | 1 -
hw/virtio/vhost.c | 85 ++++++++++++++++++++-----------
include/hw/virtio/vhost-backend.h | 4 --
4 files changed, 56 insertions(+), 48 deletions(-)
--
2.39.1
On Thu, Feb 16, 2023 at 12:47:50PM +0100, David Hildenbrand wrote: > Following up on my previous work to make virtio-mem consume multiple > memslots dynamically [1] that requires precise accounting between used vs. > reserved memslots, I realized that vhost makes this extra hard by > filtering out some memory region sections (so they don't consume a > memslot) in the vhost-user case, which messes up the whole memslot > accounting. > > This series fixes what I found to be broken and prepares for more work on > [1]. Further, it cleanes up the merge checks that I consider unnecessary. > > [1] https://lkml.kernel.org/r/20211027124531.57561-8-david@redhat.com > > Cc: "Michael S. Tsirkin" <mst@redhat.com> > Cc: Stefan Hajnoczi <stefanha@redhat.com> > Cc: Dr. David Alan Gilbert <dgilbert@redhat.com> Igor worked on memslots a lot previously and he asked for a bit of time to review this, so I'll wait a bit before applying. > David Hildenbrand (2): > vhost: Defer filtering memory sections until building the vhost memory > structure > vhost: Remove vhost_backend_can_merge() callback > > hw/virtio/vhost-user.c | 14 ----- > hw/virtio/vhost-vdpa.c | 1 - > hw/virtio/vhost.c | 85 ++++++++++++++++++++----------- > include/hw/virtio/vhost-backend.h | 4 -- > 4 files changed, 56 insertions(+), 48 deletions(-) > > -- > 2.39.1
On Fri, 17 Feb 2023 09:20:27 -0500 "Michael S. Tsirkin" <mst@redhat.com> wrote: > On Thu, Feb 16, 2023 at 12:47:50PM +0100, David Hildenbrand wrote: > > Following up on my previous work to make virtio-mem consume multiple > > memslots dynamically [1] that requires precise accounting between used vs. > > reserved memslots, I realized that vhost makes this extra hard by > > filtering out some memory region sections (so they don't consume a > > memslot) in the vhost-user case, which messes up the whole memslot > > accounting. > > > > This series fixes what I found to be broken and prepares for more work on > > [1]. Further, it cleanes up the merge checks that I consider unnecessary. > > > > [1] https://lkml.kernel.org/r/20211027124531.57561-8-david@redhat.com > > > > Cc: "Michael S. Tsirkin" <mst@redhat.com> > > Cc: Stefan Hajnoczi <stefanha@redhat.com> > > Cc: Dr. David Alan Gilbert <dgilbert@redhat.com> > > > Igor worked on memslots a lot previously and he asked for > a bit of time to review this, so I'll wait a bit before > applying. I've reviewed it as much as I could. (That said, vhost mem map code was mostly rewritten by dgilbert, since the last time I've touched it, so his review would be more valuable in this case than mine) > > > David Hildenbrand (2): > > vhost: Defer filtering memory sections until building the vhost memory > > structure > > vhost: Remove vhost_backend_can_merge() callback > > > > hw/virtio/vhost-user.c | 14 ----- > > hw/virtio/vhost-vdpa.c | 1 - > > hw/virtio/vhost.c | 85 ++++++++++++++++++++----------- > > include/hw/virtio/vhost-backend.h | 4 -- > > 4 files changed, 56 insertions(+), 48 deletions(-) > > > > -- > > 2.39.1 > >
On 07.03.23 12:14, Igor Mammedov wrote: > On Fri, 17 Feb 2023 09:20:27 -0500 > "Michael S. Tsirkin" <mst@redhat.com> wrote: > >> On Thu, Feb 16, 2023 at 12:47:50PM +0100, David Hildenbrand wrote: >>> Following up on my previous work to make virtio-mem consume multiple >>> memslots dynamically [1] that requires precise accounting between used vs. >>> reserved memslots, I realized that vhost makes this extra hard by >>> filtering out some memory region sections (so they don't consume a >>> memslot) in the vhost-user case, which messes up the whole memslot >>> accounting. >>> >>> This series fixes what I found to be broken and prepares for more work on >>> [1]. Further, it cleanes up the merge checks that I consider unnecessary. >>> >>> [1] https://lkml.kernel.org/r/20211027124531.57561-8-david@redhat.com >>> >>> Cc: "Michael S. Tsirkin" <mst@redhat.com> >>> Cc: Stefan Hajnoczi <stefanha@redhat.com> >>> Cc: Dr. David Alan Gilbert <dgilbert@redhat.com> >> >> >> Igor worked on memslots a lot previously and he asked for >> a bit of time to review this, so I'll wait a bit before >> applying. > > I've reviewed it as much as I could. > (That said, vhost mem map code was mostly rewritten by dgilbert, > since the last time I've touched it, so his review would be > more valuable in this case than mine) Thanks for the review! I'll resend (extending the patch description) and will move patch #1 last, so we can decide if we want to leave that broken in corner cases. -- Thanks, David / dhildenb
On 17.02.23 15:20, Michael S. Tsirkin wrote: > On Thu, Feb 16, 2023 at 12:47:50PM +0100, David Hildenbrand wrote: >> Following up on my previous work to make virtio-mem consume multiple >> memslots dynamically [1] that requires precise accounting between used vs. >> reserved memslots, I realized that vhost makes this extra hard by >> filtering out some memory region sections (so they don't consume a >> memslot) in the vhost-user case, which messes up the whole memslot >> accounting. >> >> This series fixes what I found to be broken and prepares for more work on >> [1]. Further, it cleanes up the merge checks that I consider unnecessary. >> >> [1] https://lkml.kernel.org/r/20211027124531.57561-8-david@redhat.com >> >> Cc: "Michael S. Tsirkin" <mst@redhat.com> >> Cc: Stefan Hajnoczi <stefanha@redhat.com> >> Cc: Dr. David Alan Gilbert <dgilbert@redhat.com> > > > Igor worked on memslots a lot previously and he asked for > a bit of time to review this, so I'll wait a bit before > applying. Sure, no need to rush. I'm still working on the other bits of the virtio-mem approach. -- Thanks, David / dhildenb
On Thu, Feb 16, 2023 at 12:47:50PM +0100, David Hildenbrand wrote: > Following up on my previous work to make virtio-mem consume multiple > memslots dynamically [1] that requires precise accounting between used vs. > reserved memslots, I realized that vhost makes this extra hard by > filtering out some memory region sections (so they don't consume a > memslot) in the vhost-user case, which messes up the whole memslot > accounting. > > This series fixes what I found to be broken and prepares for more work on > [1]. Further, it cleanes up the merge checks that I consider unnecessary. > > [1] https://lkml.kernel.org/r/20211027124531.57561-8-david@redhat.com > > Cc: "Michael S. Tsirkin" <mst@redhat.com> > Cc: Stefan Hajnoczi <stefanha@redhat.com> > Cc: Dr. David Alan Gilbert <dgilbert@redhat.com> > > David Hildenbrand (2): > vhost: Defer filtering memory sections until building the vhost memory > structure > vhost: Remove vhost_backend_can_merge() callback > > hw/virtio/vhost-user.c | 14 ----- > hw/virtio/vhost-vdpa.c | 1 - > hw/virtio/vhost.c | 85 ++++++++++++++++++++----------- > include/hw/virtio/vhost-backend.h | 4 -- > 4 files changed, 56 insertions(+), 48 deletions(-) > > -- > 2.39.1 > I'm not familiar enough with the memslot code to review this properly, but overall it looks okay: Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
© 2016 - 2024 Red Hat, Inc.