[Qemu-devel] [PULL v2 00/12] Net patches

[Qemu-devel] [PULL v2 00/12] Net patches

[Jason Wang]

The following changes since commit 31e49ac192f782d594bbd04070fe79e800b7813f:

  Merge remote-tracking branch 'remotes/cohuck/tags/s390x-20151111' into staging (2015-11-11 18:23:08 +0000)

are available in the git repository at:

  https://github.com/jasowang/qemu.git tags/net-pull-request

for you to fetch changes up to 39bec4f38b028a2cff8c38f3455aef44d7b3b6c4:

  net: netmap: use error_setg() helpers in place of error_report() (2015-11-12 15:31:52 +0800)

----------------------------------------------------------------
Changes from V1:
- no changes in the patches, V1 just misses the list

----------------------------------------------------------------
Leonid Bloch (9):
      e1000: Cosmetic and alignment fixes
      e1000: Add support for migrating the entire MAC registers' array
      e1000: Introduced an array to control the access to the MAC registers
      e1000: Trivial implementation of various MAC registers
      e1000: Fixing the received/transmitted packets' counters
      e1000: Fixing the received/transmitted octets' counters
      e1000: Fixing the packet address filtering procedure
      e1000: Implementing various counters
      e1000: Introducing backward compatibility command line parameter

Stefan Weil (1):
      slirp: Fix type casts and format strings in debug code

Vincenzo Maffione (2):
      net: netmap: Fix compilation issue
      net: netmap: use error_setg() helpers in place of error_report()

 hw/net/e1000.c      | 476 ++++++++++++++++++++++++++++++++++++++++------------
 hw/net/e1000_regs.h |   8 +-
 include/hw/compat.h |   6 +-
 net/netmap.c        |  24 +--
 slirp/bootp.c       |  12 +-
 slirp/if.c          |   4 +-
 slirp/ip_icmp.c     |   4 +-
 slirp/ip_input.c    |  10 +-
 slirp/ip_output.c   |   4 +-
 slirp/mbuf.c        |   6 +-
 slirp/misc.c        |   6 +-
 slirp/sbuf.c        |   4 +-
 slirp/socket.c      |  18 +-
 slirp/tcp_input.c   |  14 +-
 slirp/tcp_output.c  |   2 +-
 slirp/tcp_subr.c    |  16 +-
 slirp/udp.c         |   6 +-
 17 files changed, 448 insertions(+), 172 deletions(-)

-- 
2.1.4

Re: [Qemu-devel] [PULL v2 00/12] Net patches

[Peter Maydell]

On 12 November 2015 at 08:32, Jason Wang <jasowang@redhat.com> wrote:
> The following changes since commit 31e49ac192f782d594bbd04070fe79e800b7813f:
>
>   Merge remote-tracking branch 'remotes/cohuck/tags/s390x-20151111' into staging (2015-11-11 18:23:08 +0000)
>
> are available in the git repository at:
>
>   https://github.com/jasowang/qemu.git tags/net-pull-request
>
> for you to fetch changes up to 39bec4f38b028a2cff8c38f3455aef44d7b3b6c4:
>
>   net: netmap: use error_setg() helpers in place of error_report() (2015-11-12 15:31:52 +0800)
>
> ----------------------------------------------------------------
> Changes from V1:
> - no changes in the patches, V1 just misses the list
>
> ----------------------------------------------------------------

Applied, thanks.

-- PMM

[Qemu-devel] [PULL V2 00/12] Net patches

[Jason Wang]

The following changes since commit 97556fe80e4f7252300b3498b3477fb4295153a3:

  Merge remote-tracking branch 'remotes/bonzini/tags/for-upstream' into staging (2016-03-08 04:53:37 +0000)

are available in the git repository at:

  https://github.com/jasowang/qemu.git tags/net-pull-request

for you to fetch changes up to 362786f14a753d8a5256ef97d7c10ed576d6572b:

  net: check packet payload length (2016-03-08 15:34:18 +0800)

----------------------------------------------------------------

- netfilter could be disabled and enabled through qom-set now
- fix netfilter crash when specifiying wrong parameters
- rocker switch now can allow user to specifiy world
- fix OOB access for ne2000

Changes from V1:
- drop mirror from the pull request

----------------------------------------------------------------
Jason Wang (1):
      net: filter: correctly remove filter from the list during finalization

Jiri Pirko (4):
      rocker: forbid to change world type
      rocker: return -ENOMEM in case of some world alloc fails
      rocker: add name field into WorldOps ale let world specify its name
      rocker: allow user to specify rocker world by property

Paolo Bonzini (1):
      net: simplify net_init_tap_one logic

Prasad J Pandit (2):
      net: ne2000: check ring buffer control registers
      net: check packet payload length

Thomas Huth (1):
      MAINTAINERS: Add entries for include/net/ files

Vincenzo Maffione (1):
      net: netmap: probe netmap interface for virtio-net header

zhanghailiang (2):
      filter: Add 'status' property for filter object
      filter-buffer: Add status_changed callback processing

 MAINTAINERS                   |  2 ++
 hw/net/ne2000.c               |  4 +++
 hw/net/rocker/rocker.c        | 38 ++++++++++++++++++++++++++--
 hw/net/rocker/rocker_fp.c     |  5 ++++
 hw/net/rocker/rocker_fp.h     |  1 +
 hw/net/rocker/rocker_of_dpa.c |  1 +
 hw/net/rocker/rocker_world.c  |  7 +----
 hw/net/rocker/rocker_world.h  |  1 +
 include/net/filter.h          |  4 +++
 net/checksum.c                | 10 ++++++--
 net/filter-buffer.c           | 34 ++++++++++++++++++++-----
 net/filter.c                  | 44 +++++++++++++++++++++++++++++++-
 net/netmap.c                  | 59 ++++++++++++++++++++++++++++---------------
 net/tap.c                     |  4 +--
 qemu-options.hx               |  4 ++-
 15 files changed, 177 insertions(+), 41 deletions(-)

[Qemu-devel] [PULL V2 01/12] net: ne2000: check ring buffer control registers

[Jason Wang]

From: Prasad J Pandit <pjp@fedoraproject.org>

Ne2000 NIC uses ring buffer of NE2000_MEM_SIZE(49152)
bytes to process network packets. Registers PSTART & PSTOP
define ring buffer size & location. Setting these registers
to invalid values could lead to infinite loop or OOB r/w
access issues. Add check to avoid it.

Reported-by: Yang Hongke <yanghongke@huawei.com>
Tested-by: Yang Hongke <yanghongke@huawei.com>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 hw/net/ne2000.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/hw/net/ne2000.c b/hw/net/ne2000.c
index e408083..f0feaf9 100644
--- a/hw/net/ne2000.c
+++ b/hw/net/ne2000.c
@@ -155,6 +155,10 @@ static int ne2000_buffer_full(NE2000State *s)
 {
     int avail, index, boundary;
 
+    if (s->stop <= s->start) {
+        return 1;
+    }
+
     index = s->curpag << 8;
     boundary = s->boundary << 8;
     if (index < boundary)
-- 
2.5.0

[Qemu-devel] [PULL V2 02/12] net: filter: correctly remove filter from the list during finalization

[Jason Wang]

Qemu may crash when we want to add two filters on the same netdev but
the initialization of second fails (e.g missing parameters):

./qemu-system-x86_64 -netdev user,id=un0 \
 -object filter-buffer,id=f0,netdev=un0,interval=10 \
 -object filter-buffer,id=f1,netdev=un0
Segmentation fault (core dumped)

This is because we don't check whether or not the filter was in the
list of netdev. This patch fixes this.

Cc: Yang Hongyang <hongyang.yang@easystack.cn>
Reviewed-by: Yang Hongyang <hongyang.yang@easystack.cn>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 net/filter.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/filter.c b/net/filter.c
index d2a514e..7cdbc6c 100644
--- a/net/filter.c
+++ b/net/filter.c
@@ -196,7 +196,8 @@ static void netfilter_finalize(Object *obj)
         nfc->cleanup(nf);
     }
 
-    if (nf->netdev && !QTAILQ_EMPTY(&nf->netdev->filters)) {
+    if (nf->netdev && !QTAILQ_EMPTY(&nf->netdev->filters) &&
+        nf->next.tqe_prev) {
         QTAILQ_REMOVE(&nf->netdev->filters, nf, next);
     }
     g_free(nf->netdev_id);
-- 
2.5.0

[Qemu-devel] [PULL V2 03/12] MAINTAINERS: Add entries for include/net/ files

[Jason Wang]

From: Thomas Huth <thuth@redhat.com>

The include/net/ files correspond to the files in the net/ directory,
thus there should be corresponding entries in the MAINTAINERS file.

Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 MAINTAINERS | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 9d0255c..c47709d 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1123,6 +1123,7 @@ Network device backends
 M: Jason Wang <jasowang@redhat.com>
 S: Maintained
 F: net/
+F: include/net/
 T: git git://github.com/jasowang/qemu.git net
 
 Netmap network backend
@@ -1222,6 +1223,7 @@ M: Jan Kiszka <jan.kiszka@siemens.com>
 S: Maintained
 F: slirp/
 F: net/slirp.c
+F: include/net/slirp.h
 T: git git://git.kiszka.org/qemu.git queues/slirp
 
 Tracing
-- 
2.5.0

[Qemu-devel] [PULL V2 04/12] net: simplify net_init_tap_one logic

[Jason Wang]

From: Paolo Bonzini <pbonzini@redhat.com>

net_init_tap_one receives in vhostfdname a fd name from vhostfd= or
vhostfds=, or NULL if there is no vhostfd=/vhostfds=.  It is simpler
to just check vhostfdname, than it is to check for vhostfd= or
vhostfds=.  This also calms down Coverity, which otherwise thinks
that monitor_fd_param could dereference a NULL vhostfdname.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 net/tap.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/tap.c b/net/tap.c
index cfb6831..cd7a7fc 100644
--- a/net/tap.c
+++ b/net/tap.c
@@ -662,7 +662,7 @@ static void net_init_tap_one(const NetdevTapOptions *tap, NetClientState *peer,
         options.backend_type = VHOST_BACKEND_TYPE_KERNEL;
         options.net_backend = &s->nc;
 
-        if (tap->has_vhostfd || tap->has_vhostfds) {
+        if (vhostfdname) {
             vhostfd = monitor_fd_param(cur_mon, vhostfdname, &err);
             if (vhostfd == -1) {
                 error_propagate(errp, err);
@@ -684,7 +684,7 @@ static void net_init_tap_one(const NetdevTapOptions *tap, NetClientState *peer,
                        "vhost-net requested but could not be initialized");
             return;
         }
-    } else if (tap->has_vhostfd || tap->has_vhostfds) {
+    } else if (vhostfdname) {
         error_setg(errp, "vhostfd= is not valid without vhost");
     }
 }
-- 
2.5.0

[Qemu-devel] [PULL V2 05/12] net: netmap: probe netmap interface for virtio-net header

[Jason Wang]

From: Vincenzo Maffione <v.maffione@gmail.com>

Previous implementation of has_ufo, has_vnet_hdr, has_vnet_hdr_len, etc.
did not really probe for virtio-net header support for the netmap
interface attached to the backend. These callbacks were correct for
VALE ports, but incorrect for hardware NICs, pipes, monitors, etc.

This patch fixes the implementation to work properly with all kinds
of netmap ports.

Signed-off-by: Vincenzo Maffione <v.maffione@gmail.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 net/netmap.c | 59 ++++++++++++++++++++++++++++++++++++++---------------------
 1 file changed, 38 insertions(+), 21 deletions(-)

diff --git a/net/netmap.c b/net/netmap.c
index 9710321..1b42728 100644
--- a/net/netmap.c
+++ b/net/netmap.c
@@ -323,20 +323,47 @@ static void netmap_cleanup(NetClientState *nc)
 }
 
 /* Offloading manipulation support callbacks. */
-static bool netmap_has_ufo(NetClientState *nc)
+static int netmap_fd_set_vnet_hdr_len(NetmapState *s, int len)
 {
-    return true;
+    struct nmreq req;
+
+    /* Issue a NETMAP_BDG_VNET_HDR command to change the virtio-net header
+     * length for the netmap adapter associated to 's->ifname'.
+     */
+    memset(&req, 0, sizeof(req));
+    pstrcpy(req.nr_name, sizeof(req.nr_name), s->ifname);
+    req.nr_version = NETMAP_API;
+    req.nr_cmd = NETMAP_BDG_VNET_HDR;
+    req.nr_arg1 = len;
+
+    return ioctl(s->nmd->fd, NIOCREGIF, &req);
 }
 
-static bool netmap_has_vnet_hdr(NetClientState *nc)
+static bool netmap_has_vnet_hdr_len(NetClientState *nc, int len)
 {
+    NetmapState *s = DO_UPCAST(NetmapState, nc, nc);
+    int prev_len = s->vnet_hdr_len;
+
+    /* Check that we can set the new length. */
+    if (netmap_fd_set_vnet_hdr_len(s, len)) {
+        return false;
+    }
+
+    /* Restore the previous length. */
+    if (netmap_fd_set_vnet_hdr_len(s, prev_len)) {
+        error_report("Failed to restore vnet-hdr length %d on %s: %s",
+                     prev_len, s->ifname, strerror(errno));
+        abort();
+    }
+
     return true;
 }
 
-static bool netmap_has_vnet_hdr_len(NetClientState *nc, int len)
+/* A netmap interface that supports virtio-net headers always
+ * supports UFO, so we use this callback also for the has_ufo hook. */
+static bool netmap_has_vnet_hdr(NetClientState *nc)
 {
-    return len == 0 || len == sizeof(struct virtio_net_hdr) ||
-                len == sizeof(struct virtio_net_hdr_mrg_rxbuf);
+    return netmap_has_vnet_hdr_len(nc, sizeof(struct virtio_net_hdr));
 }
 
 static void netmap_using_vnet_hdr(NetClientState *nc, bool enable)
@@ -347,20 +374,11 @@ static void netmap_set_vnet_hdr_len(NetClientState *nc, int len)
 {
     NetmapState *s = DO_UPCAST(NetmapState, nc, nc);
     int err;
-    struct nmreq req;
 
-    /* Issue a NETMAP_BDG_VNET_HDR command to change the virtio-net header
-     * length for the netmap adapter associated to 's->ifname'.
-     */
-    memset(&req, 0, sizeof(req));
-    pstrcpy(req.nr_name, sizeof(req.nr_name), s->ifname);
-    req.nr_version = NETMAP_API;
-    req.nr_cmd = NETMAP_BDG_VNET_HDR;
-    req.nr_arg1 = len;
-    err = ioctl(s->nmd->fd, NIOCREGIF, &req);
+    err = netmap_fd_set_vnet_hdr_len(s, len);
     if (err) {
-        error_report("Unable to execute NETMAP_BDG_VNET_HDR on %s: %s",
-                     s->ifname, strerror(errno));
+        error_report("Unable to set vnet-hdr length %d on %s: %s",
+                     len, s->ifname, strerror(errno));
     } else {
         /* Keep track of the current length. */
         s->vnet_hdr_len = len;
@@ -373,8 +391,7 @@ static void netmap_set_offload(NetClientState *nc, int csum, int tso4, int tso6,
     NetmapState *s = DO_UPCAST(NetmapState, nc, nc);
 
     /* Setting a virtio-net header length greater than zero automatically
-     * enables the offloadings.
-     */
+     * enables the offloadings. */
     if (!s->vnet_hdr_len) {
         netmap_set_vnet_hdr_len(nc, sizeof(struct virtio_net_hdr));
     }
@@ -388,7 +405,7 @@ static NetClientInfo net_netmap_info = {
     .receive_iov = netmap_receive_iov,
     .poll = netmap_poll,
     .cleanup = netmap_cleanup,
-    .has_ufo = netmap_has_ufo,
+    .has_ufo = netmap_has_vnet_hdr,
     .has_vnet_hdr = netmap_has_vnet_hdr,
     .has_vnet_hdr_len = netmap_has_vnet_hdr_len,
     .using_vnet_hdr = netmap_using_vnet_hdr,
-- 
2.5.0

[Qemu-devel] [PULL V2 06/12] rocker: forbid to change world type

[Jason Wang]

From: Jiri Pirko <jiri@mellanox.com>

Port to world assignment should be permitted only by qemu user. Driver
should not be able to do it, so forbid that possibility.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 hw/net/rocker/rocker.c    | 8 +++++++-
 hw/net/rocker/rocker_fp.c | 5 +++++
 hw/net/rocker/rocker_fp.h | 1 +
 3 files changed, 13 insertions(+), 1 deletion(-)

diff --git a/hw/net/rocker/rocker.c b/hw/net/rocker/rocker.c
index f3e994d..a1d921d 100644
--- a/hw/net/rocker/rocker.c
+++ b/hw/net/rocker/rocker.c
@@ -400,7 +400,13 @@ static int cmd_set_port_settings(Rocker *r,
 
     if (tlvs[ROCKER_TLV_CMD_PORT_SETTINGS_MODE]) {
         mode = rocker_tlv_get_u8(tlvs[ROCKER_TLV_CMD_PORT_SETTINGS_MODE]);
-        fp_port_set_world(fp_port, r->worlds[mode]);
+        if (mode >= ROCKER_WORLD_TYPE_MAX) {
+            return -ROCKER_EINVAL;
+        }
+        /* We don't support world change. */
+        if (!fp_port_check_world(fp_port, r->worlds[mode])) {
+            return -ROCKER_EINVAL;
+        }
     }
 
     if (tlvs[ROCKER_TLV_CMD_PORT_SETTINGS_LEARNING]) {
diff --git a/hw/net/rocker/rocker_fp.c b/hw/net/rocker/rocker_fp.c
index af37fef..0149899 100644
--- a/hw/net/rocker/rocker_fp.c
+++ b/hw/net/rocker/rocker_fp.c
@@ -186,6 +186,11 @@ void fp_port_set_world(FpPort *port, World *world)
     port->world = world;
 }
 
+bool fp_port_check_world(FpPort *port, World *world)
+{
+    return port->world == world;
+}
+
 bool fp_port_enabled(FpPort *port)
 {
     return port->enabled;
diff --git a/hw/net/rocker/rocker_fp.h b/hw/net/rocker/rocker_fp.h
index ab80fd8..04592bb 100644
--- a/hw/net/rocker/rocker_fp.h
+++ b/hw/net/rocker/rocker_fp.h
@@ -40,6 +40,7 @@ int fp_port_set_settings(FpPort *port, uint32_t speed,
 bool fp_port_from_pport(uint32_t pport, uint32_t *port);
 World *fp_port_get_world(FpPort *port);
 void fp_port_set_world(FpPort *port, World *world);
+bool fp_port_check_world(FpPort *port, World *world);
 bool fp_port_enabled(FpPort *port);
 void fp_port_enable(FpPort *port);
 void fp_port_disable(FpPort *port);
-- 
2.5.0

[Qemu-devel] [PULL V2 07/12] rocker: return -ENOMEM in case of some world alloc fails

[Jason Wang]

From: Jiri Pirko <jiri@mellanox.com>

Until now, 0 is returned in this error case. Fix it ro return -ENOMEM.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 hw/net/rocker/rocker.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/hw/net/rocker/rocker.c b/hw/net/rocker/rocker.c
index a1d921d..104c097 100644
--- a/hw/net/rocker/rocker.c
+++ b/hw/net/rocker/rocker.c
@@ -1301,6 +1301,7 @@ static int pci_rocker_init(PCIDevice *dev)
 
     for (i = 0; i < ROCKER_WORLD_TYPE_MAX; i++) {
         if (!r->worlds[i]) {
+            err = -ENOMEM;
             goto err_world_alloc;
         }
     }
-- 
2.5.0

[Qemu-devel] [PULL V2 08/12] rocker: add name field into WorldOps ale let world specify its name

[Jason Wang]

From: Jiri Pirko <jiri@mellanox.com>

Also use this in world_name getter function.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 hw/net/rocker/rocker_of_dpa.c | 1 +
 hw/net/rocker/rocker_world.c  | 7 +------
 hw/net/rocker/rocker_world.h  | 1 +
 3 files changed, 3 insertions(+), 6 deletions(-)

diff --git a/hw/net/rocker/rocker_of_dpa.c b/hw/net/rocker/rocker_of_dpa.c
index da3fc54..0a134eb 100644
--- a/hw/net/rocker/rocker_of_dpa.c
+++ b/hw/net/rocker/rocker_of_dpa.c
@@ -2614,6 +2614,7 @@ RockerOfDpaGroupList *qmp_query_rocker_of_dpa_groups(const char *name,
 }
 
 static WorldOps of_dpa_ops = {
+    .name = "ofdpa",
     .init = of_dpa_init,
     .uninit = of_dpa_uninit,
     .ig = of_dpa_ig,
diff --git a/hw/net/rocker/rocker_world.c b/hw/net/rocker/rocker_world.c
index 1ed0fcd..89777e9 100644
--- a/hw/net/rocker/rocker_world.c
+++ b/hw/net/rocker/rocker_world.c
@@ -98,10 +98,5 @@ enum rocker_world_type world_type(World *world)
 
 const char *world_name(World *world)
 {
-    switch (world->type) {
-    case ROCKER_WORLD_TYPE_OF_DPA:
-        return "OF_DPA";
-    default:
-        return "unknown";
-    }
+    return world->ops->name;
 }
diff --git a/hw/net/rocker/rocker_world.h b/hw/net/rocker/rocker_world.h
index 18d277b..58ade47 100644
--- a/hw/net/rocker/rocker_world.h
+++ b/hw/net/rocker/rocker_world.h
@@ -33,6 +33,7 @@ typedef int (world_cmd)(World *world, DescInfo *info,
                         RockerTlv *cmd_info_tlv);
 
 typedef struct world_ops {
+    const char *name;
     world_init *init;
     world_uninit *uninit;
     world_ig *ig;
-- 
2.5.0

[Qemu-devel] [PULL V2 09/12] rocker: allow user to specify rocker world by property

[Jason Wang]

From: Jiri Pirko <jiri@mellanox.com>

Add property to specify rocker world. All ports will be assigned to this
world.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 hw/net/rocker/rocker.c | 29 ++++++++++++++++++++++++++++-
 1 file changed, 28 insertions(+), 1 deletion(-)

diff --git a/hw/net/rocker/rocker.c b/hw/net/rocker/rocker.c
index 104c097..30f2ce4 100644
--- a/hw/net/rocker/rocker.c
+++ b/hw/net/rocker/rocker.c
@@ -43,6 +43,7 @@ struct rocker {
 
     /* switch configuration */
     char *name;                  /* switch name */
+    char *world_name;            /* world name */
     uint32_t fp_ports;           /* front-panel port count */
     NICPeers *fp_ports_peers;
     MACAddr fp_start_macaddr;    /* front-panel port 0 mac addr */
@@ -1286,6 +1287,18 @@ static void rocker_msix_uninit(Rocker *r)
     rocker_msix_vectors_unuse(r, ROCKER_MSIX_VEC_COUNT(r->fp_ports));
 }
 
+static World *rocker_world_type_by_name(Rocker *r, const char *name)
+{
+    int i;
+
+    for (i = 0; i < ROCKER_WORLD_TYPE_MAX; i++) {
+        if (strcmp(name, world_name(r->worlds[i])) == 0) {
+            return r->worlds[i];
+	}
+    }
+    return NULL;
+}
+
 static int pci_rocker_init(PCIDevice *dev)
 {
     Rocker *r = to_rocker(dev);
@@ -1297,7 +1310,6 @@ static int pci_rocker_init(PCIDevice *dev)
     /* allocate worlds */
 
     r->worlds[ROCKER_WORLD_TYPE_OF_DPA] = of_dpa_world_alloc(r);
-    r->world_dflt = r->worlds[ROCKER_WORLD_TYPE_OF_DPA];
 
     for (i = 0; i < ROCKER_WORLD_TYPE_MAX; i++) {
         if (!r->worlds[i]) {
@@ -1306,6 +1318,19 @@ static int pci_rocker_init(PCIDevice *dev)
         }
     }
 
+    if (!r->world_name) {
+        r->world_name = g_strdup(world_name(r->worlds[ROCKER_WORLD_TYPE_OF_DPA]));
+    }
+
+    r->world_dflt = rocker_world_type_by_name(r, r->world_name);
+    if (!r->world_dflt) {
+        fprintf(stderr,
+                "rocker: requested world \"%s\" does not exist\n",
+                r->world_name);
+        err = -EINVAL;
+        goto err_world_type_by_name;
+    }
+
     /* set up memory-mapped region at BAR0 */
 
     memory_region_init_io(&r->mmio, OBJECT(r), &rocker_mmio_ops, r,
@@ -1439,6 +1464,7 @@ err_duplicate:
 err_msix_init:
     object_unparent(OBJECT(&r->msix_bar));
     object_unparent(OBJECT(&r->mmio));
+err_world_type_by_name:
 err_world_alloc:
     for (i = 0; i < ROCKER_WORLD_TYPE_MAX; i++) {
         if (r->worlds[i]) {
@@ -1510,6 +1536,7 @@ static void rocker_reset(DeviceState *dev)
 
 static Property rocker_properties[] = {
     DEFINE_PROP_STRING("name", Rocker, name),
+    DEFINE_PROP_STRING("world", Rocker, world_name),
     DEFINE_PROP_MACADDR("fp_start_macaddr", Rocker,
                         fp_start_macaddr),
     DEFINE_PROP_UINT64("switch_id", Rocker,
-- 
2.5.0

[Qemu-devel] [PULL V2 10/12] filter: Add 'status' property for filter object

[Jason Wang]

From: zhanghailiang <zhang.zhanghailiang@huawei.com>

With this property, users can control if this filter is 'on'
or 'off'. The default behavior for filter is 'on'.

For some types of filters, they may need to react to status changing,
So here, we introduced status changing callback/notifier for filter class.

We will skip the disabled ('off') filter when delivering packets in net layer.

Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Cc: Jason Wang <jasowang@redhat.com>
Cc: Yang Hongyang <hongyang.yang@easystack.cn>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 include/net/filter.h |  4 ++++
 net/filter.c         | 41 +++++++++++++++++++++++++++++++++++++++++
 qemu-options.hx      |  4 +++-
 3 files changed, 48 insertions(+), 1 deletion(-)

diff --git a/include/net/filter.h b/include/net/filter.h
index 5639976..cfb1172 100644
--- a/include/net/filter.h
+++ b/include/net/filter.h
@@ -36,12 +36,15 @@ typedef ssize_t (FilterReceiveIOV)(NetFilterState *nc,
                                    int iovcnt,
                                    NetPacketSent *sent_cb);
 
+typedef void (FilterStatusChanged) (NetFilterState *nf, Error **errp);
+
 typedef struct NetFilterClass {
     ObjectClass parent_class;
 
     /* optional */
     FilterSetup *setup;
     FilterCleanup *cleanup;
+    FilterStatusChanged *status_changed;
     /* mandatory */
     FilterReceiveIOV *receive_iov;
 } NetFilterClass;
@@ -55,6 +58,7 @@ struct NetFilterState {
     char *netdev_id;
     NetClientState *netdev;
     NetFilterDirection direction;
+    bool on;
     QTAILQ_ENTRY(NetFilterState) next;
 };
 
diff --git a/net/filter.c b/net/filter.c
index 7cdbc6c..a08ef68 100644
--- a/net/filter.c
+++ b/net/filter.c
@@ -17,6 +17,11 @@
 #include "qom/object_interfaces.h"
 #include "qemu/iov.h"
 
+static inline bool qemu_can_skip_netfilter(NetFilterState *nf)
+{
+    return !nf->on;
+}
+
 ssize_t qemu_netfilter_receive(NetFilterState *nf,
                                NetFilterDirection direction,
                                NetClientState *sender,
@@ -25,6 +30,9 @@ ssize_t qemu_netfilter_receive(NetFilterState *nf,
                                int iovcnt,
                                NetPacketSent *sent_cb)
 {
+    if (qemu_can_skip_netfilter(nf)) {
+        return 0;
+    }
     if (nf->direction == direction ||
         nf->direction == NET_FILTER_DIRECTION_ALL) {
         return NETFILTER_GET_CLASS(OBJECT(nf))->receive_iov(
@@ -134,8 +142,38 @@ static void netfilter_set_direction(Object *obj, int direction, Error **errp)
     nf->direction = direction;
 }
 
+static char *netfilter_get_status(Object *obj, Error **errp)
+{
+    NetFilterState *nf = NETFILTER(obj);
+
+    return nf->on ? g_strdup("on") : g_strdup("off");
+}
+
+static void netfilter_set_status(Object *obj, const char *str, Error **errp)
+{
+    NetFilterState *nf = NETFILTER(obj);
+    NetFilterClass *nfc = NETFILTER_GET_CLASS(obj);
+
+    if (strcmp(str, "on") && strcmp(str, "off")) {
+        error_setg(errp, "Invalid value for netfilter status, "
+                         "should be 'on' or 'off'");
+        return;
+    }
+    if (nf->on == !strcmp(str, "on")) {
+        return;
+    }
+    nf->on = !nf->on;
+    if (nfc->status_changed) {
+        nfc->status_changed(nf, errp);
+    }
+}
+
 static void netfilter_init(Object *obj)
 {
+    NetFilterState *nf = NETFILTER(obj);
+
+    nf->on = true;
+
     object_property_add_str(obj, "netdev",
                             netfilter_get_netdev_id, netfilter_set_netdev_id,
                             NULL);
@@ -143,6 +181,9 @@ static void netfilter_init(Object *obj)
                              NetFilterDirection_lookup,
                              netfilter_get_direction, netfilter_set_direction,
                              NULL);
+    object_property_add_str(obj, "status",
+                            netfilter_get_status, netfilter_set_status,
+                            NULL);
 }
 
 static void netfilter_complete(UserCreatable *uc, Error **errp)
diff --git a/qemu-options.hx b/qemu-options.hx
index 2aa6577..2b3ed86 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -3788,11 +3788,13 @@ version by providing the @var{passwordid} parameter. This provides
 the ID of a previously created @code{secret} object containing the
 password for decryption.
 
-@item -object filter-buffer,id=@var{id},netdev=@var{netdevid},interval=@var{t}[,queue=@var{all|rx|tx}]
+@item -object filter-buffer,id=@var{id},netdev=@var{netdevid},interval=@var{t}[,queue=@var{all|rx|tx}][,status=@var{on|off}]
 
 Interval @var{t} can't be 0, this filter batches the packet delivery: all
 packets arriving in a given interval on netdev @var{netdevid} are delayed
 until the end of the interval. Interval is in microseconds.
+@option{status} is optional that indicate whether the netfilter is
+on (enabled) or off (disabled), the default status for netfilter will be 'on'.
 
 queue @var{all|rx|tx} is an option that can be applied to any netfilter.
 
-- 
2.5.0

[Qemu-devel] [PULL V2 11/12] filter-buffer: Add status_changed callback processing

[Jason Wang]

From: zhanghailiang <zhang.zhanghailiang@huawei.com>

While the status of filter-buffer changing from 'on' to 'off',
it need to release all the buffered packets, and delete the related
timer, while switch from 'off' to 'on', it need to resume the release
packets timer.

Here, we extract the process of setup timer into a new helper,
which will be used in the new status_changed callback.

Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Cc: Jason Wang <jasowang@redhat.com>
Cc: Yang Hongyang <hongyang.yang@easystack.cn>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 net/filter-buffer.c | 34 ++++++++++++++++++++++++++++------
 1 file changed, 28 insertions(+), 6 deletions(-)

diff --git a/net/filter-buffer.c b/net/filter-buffer.c
index 12ad2e3..972177b 100644
--- a/net/filter-buffer.c
+++ b/net/filter-buffer.c
@@ -100,6 +100,19 @@ static void filter_buffer_cleanup(NetFilterState *nf)
     }
 }
 
+static void filter_buffer_setup_timer(NetFilterState *nf)
+{
+    FilterBufferState *s = FILTER_BUFFER(nf);
+
+    if (s->interval) {
+        timer_init_us(&s->release_timer, QEMU_CLOCK_VIRTUAL,
+                      filter_buffer_release_timer, nf);
+        /* Timer armed to fire in s->interval microseconds. */
+        timer_mod(&s->release_timer,
+                  qemu_clock_get_us(QEMU_CLOCK_VIRTUAL) + s->interval);
+    }
+}
+
 static void filter_buffer_setup(NetFilterState *nf, Error **errp)
 {
     FilterBufferState *s = FILTER_BUFFER(nf);
@@ -115,12 +128,20 @@ static void filter_buffer_setup(NetFilterState *nf, Error **errp)
     }
 
     s->incoming_queue = qemu_new_net_queue(qemu_netfilter_pass_to_next, nf);
-    if (s->interval) {
-        timer_init_us(&s->release_timer, QEMU_CLOCK_VIRTUAL,
-                      filter_buffer_release_timer, nf);
-        /* Timer armed to fire in s->interval microseconds. */
-        timer_mod(&s->release_timer,
-                  qemu_clock_get_us(QEMU_CLOCK_VIRTUAL) + s->interval);
+    filter_buffer_setup_timer(nf);
+}
+
+static void filter_buffer_status_changed(NetFilterState *nf, Error **errp)
+{
+    FilterBufferState *s = FILTER_BUFFER(nf);
+
+    if (!nf->on) {
+        if (s->interval) {
+            timer_del(&s->release_timer);
+        }
+        filter_buffer_flush(nf);
+    } else {
+        filter_buffer_setup_timer(nf);
     }
 }
 
@@ -131,6 +152,7 @@ static void filter_buffer_class_init(ObjectClass *oc, void *data)
     nfc->setup = filter_buffer_setup;
     nfc->cleanup = filter_buffer_cleanup;
     nfc->receive_iov = filter_buffer_receive_iov;
+    nfc->status_changed = filter_buffer_status_changed;
 }
 
 static void filter_buffer_get_interval(Object *obj, Visitor *v,
-- 
2.5.0

[Qemu-devel] [PULL V2 12/12] net: check packet payload length

[Jason Wang]

From: Prasad J Pandit <pjp@fedoraproject.org>

While computing IP checksum, 'net_checksum_calculate' reads
payload length from the packet. It could exceed the given 'data'
buffer size. Add a check to avoid it.

Reported-by: Liu Ling <liuling-it@360.cn>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 net/checksum.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/net/checksum.c b/net/checksum.c
index b5016ab..d0fa424 100644
--- a/net/checksum.c
+++ b/net/checksum.c
@@ -60,6 +60,11 @@ void net_checksum_calculate(uint8_t *data, int length)
     int hlen, plen, proto, csum_offset;
     uint16_t csum;
 
+    /* Ensure data has complete L2 & L3 headers. */
+    if (length < 14 + 20) {
+        return;
+    }
+
     if ((data[14] & 0xf0) != 0x40)
 	return; /* not IPv4 */
     hlen  = (data[14] & 0x0f) * 4;
@@ -77,8 +82,9 @@ void net_checksum_calculate(uint8_t *data, int length)
 	return;
     }
 
-    if (plen < csum_offset+2)
-	return;
+    if (plen < csum_offset + 2 || 14 + hlen + plen > length) {
+        return;
+    }
 
     data[14+hlen+csum_offset]   = 0;
     data[14+hlen+csum_offset+1] = 0;
-- 
2.5.0

Re: [Qemu-devel] [PULL V2 00/12] Net patches

[Peter Maydell]

On 8 March 2016 at 14:52, Jason Wang <jasowang@redhat.com> wrote:
> The following changes since commit 97556fe80e4f7252300b3498b3477fb4295153a3:
>
>   Merge remote-tracking branch 'remotes/bonzini/tags/for-upstream' into staging (2016-03-08 04:53:37 +0000)
>
> are available in the git repository at:
>
>   https://github.com/jasowang/qemu.git tags/net-pull-request
>
> for you to fetch changes up to 362786f14a753d8a5256ef97d7c10ed576d6572b:
>
>   net: check packet payload length (2016-03-08 15:34:18 +0800)
>
> ----------------------------------------------------------------
>
> - netfilter could be disabled and enabled through qom-set now
> - fix netfilter crash when specifiying wrong parameters
> - rocker switch now can allow user to specifiy world
> - fix OOB access for ne2000
>
> Changes from V1:
> - drop mirror from the pull request
>
> ----------------------------------------------------------------

Applied, thanks.

-- PMM