From: "Philippe Mathieu-Daudé" <philmd@linaro.org>
To: qemu-devel@nongnu.org, Thomas Huth <thuth@redhat.com>
Cc: "Greg Kurz" <groug@kaod.org>,
"Stefan Hajnoczi" <stefanha@redhat.com>,
"Kevin Wolf" <kwolf@redhat.com>,
qemu-block@nongnu.org, "Paolo Bonzini" <pbonzini@redhat.com>,
"Michael S. Tsirkin" <mst@redhat.com>,
"Alex Bennée" <alex.bennee@linaro.org>,
"Jason Wang" <jasowang@redhat.com>,
"Marc-André Lureau" <marcandre.lureau@redhat.com>,
"Christian Schoenebeck" <qemu_oss@crudebyte.com>,
"Hanna Reitz" <hreitz@redhat.com>,
"Philippe Mathieu-Daudé" <philmd@linaro.org>
Subject: [RFC PATCH-for-8.0 00/10] hw/virtio: Build most objects as target independent units
Date: Tue, 13 Dec 2022 00:05:07 +0100 [thread overview]
Message-ID: <20221212230517.28872-1-philmd@linaro.org> (raw)
Currently the inlined virtio_access_is_big_endian() function
"hw/virtio/virtio-access.h" which is used by all I/O accesses
force any virtio device to be built as target-dependent object.
This series isolates the few VirtIO target specific bits, trying
to not impact the performance (a function is un-inlined once
not in the hot path).
At the end only 6 files remain in the target specific source set,
all other files are built once.
On a Linux/x86_64 host when building QEMU configured with
--disable-user, before this series a "make clean all" builds
7713 objects, after it only build 6831... I don't think my maths
are correct, because that would save building a few hundreds objects.
RFC because only build-tested.
Regards,
Phil.
Philippe Mathieu-Daudé (10):
hw/virtio: Add missing "hw/core/cpu.h" include
hw/virtio: Rename virtio_ss[] -> specific_virtio_ss[]
hw/virtio: Constify qmp_virtio_feature_map_t[]
hw/virtio: Extract config read/write accessors to virtio-config.c
hw/virtio: Extract QMP related code virtio-qmp.c
hw/virtio: Cache access_is_big_endian value in VirtIODevice state
hw/virtio: Directly access cached VirtIODevice::access_is_big_endian
hw/virtio: Un-inline virtio_access_is_big_endian()
hw/virtio: Extract vhost_user_ram_slots_max() to vhost-user-target.c
hw/virtio: Make most of virtio devices target-independent
hw/9pfs/meson.build | 2 +-
hw/block/dataplane/meson.build | 2 +-
hw/block/meson.build | 4 +-
hw/char/meson.build | 2 +-
hw/net/meson.build | 2 +-
hw/virtio/meson.build | 45 +-
hw/virtio/vhost-user-target.c | 29 ++
hw/virtio/vhost-user.c | 26 +-
hw/virtio/virtio-config.c | 224 +++++++++
hw/virtio/virtio-qmp.c | 631 +++++++++++++++++++++++
hw/virtio/virtio-qmp.h | 20 +
hw/virtio/virtio.c | 799 +-----------------------------
include/hw/virtio/vhost-user.h | 7 +
include/hw/virtio/virtio-access.h | 63 +--
include/hw/virtio/virtio.h | 1 +
15 files changed, 974 insertions(+), 883 deletions(-)
create mode 100644 hw/virtio/vhost-user-target.c
create mode 100644 hw/virtio/virtio-config.c
create mode 100644 hw/virtio/virtio-qmp.c
create mode 100644 hw/virtio/virtio-qmp.h
--
2.38.1
next reply other threads:[~2022-12-12 23:06 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-12-12 23:05 Philippe Mathieu-Daudé [this message]
2022-12-12 23:05 ` [PATCH-for-8.0 01/10] hw/virtio: Add missing "hw/core/cpu.h" include Philippe Mathieu-Daudé
2022-12-12 23:05 ` [PATCH-for-8.0 02/10] hw/virtio: Rename virtio_ss[] -> specific_virtio_ss[] Philippe Mathieu-Daudé
2022-12-12 23:05 ` [PATCH-for-8.0 03/10] hw/virtio: Constify qmp_virtio_feature_map_t[] Philippe Mathieu-Daudé
2022-12-13 0:02 ` Richard Henderson
2022-12-13 7:35 ` Philippe Mathieu-Daudé
2022-12-12 23:05 ` [PATCH-for-8.0 04/10] hw/virtio: Extract config read/write accessors to virtio-config.c Philippe Mathieu-Daudé
2022-12-12 23:05 ` [PATCH-for-8.0 05/10] hw/virtio: Extract QMP related code virtio-qmp.c Philippe Mathieu-Daudé
2022-12-12 23:05 ` [RFC PATCH-for-8.0 06/10] hw/virtio: Cache access_is_big_endian value in VirtIODevice state Philippe Mathieu-Daudé
2022-12-13 0:14 ` Richard Henderson
2022-12-13 7:30 ` Philippe Mathieu-Daudé
2022-12-13 8:03 ` Thomas Huth
2022-12-13 8:32 ` Philippe Mathieu-Daudé
2022-12-13 8:47 ` Thomas Huth
2022-12-13 8:22 ` Philippe Mathieu-Daudé
2022-12-13 15:41 ` Richard Henderson
2022-12-12 23:05 ` [RFC PATCH-for-8.0 07/10] hw/virtio: Directly access cached VirtIODevice::access_is_big_endian Philippe Mathieu-Daudé
2022-12-13 10:50 ` Greg Kurz
2022-12-12 23:05 ` [PATCH-for-8.0 08/10] hw/virtio: Un-inline virtio_access_is_big_endian() Philippe Mathieu-Daudé
2022-12-12 23:05 ` [RFC PATCH-for-8.0 09/10] hw/virtio: Extract vhost_user_ram_slots_max() to vhost-user-target.c Philippe Mathieu-Daudé
2025-04-10 12:14 ` Philippe Mathieu-Daudé
2025-04-10 14:36 ` Pierrick Bouvier
2025-04-10 17:21 ` Philippe Mathieu-Daudé
2025-04-10 17:29 ` Pierrick Bouvier
2025-04-11 10:15 ` Philippe Mathieu-Daudé
2022-12-12 23:05 ` [RFC PATCH-for-8.0 10/10] hw/virtio: Make most of virtio devices target-independent Philippe Mathieu-Daudé
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=20221212230517.28872-1-philmd@linaro.org \
--to=philmd@linaro.org \
--cc=alex.bennee@linaro.org \
--cc=groug@kaod.org \
--cc=hreitz@redhat.com \
--cc=jasowang@redhat.com \
--cc=kwolf@redhat.com \
--cc=marcandre.lureau@redhat.com \
--cc=mst@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-block@nongnu.org \
--cc=qemu-devel@nongnu.org \
--cc=qemu_oss@crudebyte.com \
--cc=stefanha@redhat.com \
--cc=thuth@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 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).