qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
* [Qemu-devel] [PATCH v5 0/3] net: fix link status
@ 2012-09-28  2:05 Amos Kong
  2012-09-28  2:06 ` [Qemu-devel] [PATCH v5 1/3] rtl8139: implement 8139cp " Amos Kong
                   ` (3 more replies)
  0 siblings, 4 replies; 5+ messages in thread
From: Amos Kong @ 2012-09-28  2:05 UTC (permalink / raw)
  To: qemu-devel; +Cc: stefanha, jasowang, pbonzini, aliguori, mst

First patch implemeted link status setting of rtl8139, the rest patches
updated nc.link_down in post_load() of migration to keep it coincident
with real link status.

v5: fix bug in bit operating

Amos Kong (2):
  e1000: update nc.link_down in e1000_post_load()
  update nc.link_down in virtio_net_load()

Jason Wang (1):
  rtl8139: implement 8139cp link status

 hw/e1000.c      |   12 ++++++++++++
 hw/rtl8139.c    |   24 ++++++++++++++++++++++--
 hw/virtio-net.c |    5 +++++
 3 files changed, 39 insertions(+), 2 deletions(-)

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Qemu-devel] [PATCH v5 1/3] rtl8139: implement 8139cp link status
  2012-09-28  2:05 [Qemu-devel] [PATCH v5 0/3] net: fix link status Amos Kong
@ 2012-09-28  2:06 ` Amos Kong
  2012-09-28  2:06 ` [Qemu-devel] [PATCH v5 2/3] e1000: update nc.link_down in e1000_post_load() Amos Kong
                   ` (2 subsequent siblings)
  3 siblings, 0 replies; 5+ messages in thread
From: Amos Kong @ 2012-09-28  2:06 UTC (permalink / raw)
  To: qemu-devel; +Cc: aliguori, mst, stefanha, jasowang, pbonzini, Amos Kong

From: Jason Wang <jasowang@redhat.com>

Add a link status chang callback and change the link status bit in BMSR
& MSR accordingly. Tested in Linux/Windows guests.

The link status bit of MediaStatus is infered from BasicModeStatus,
they are inverse.

nc.link_down could not be migrated, this patch updates link_down in
rtl8139_post_load() to keep it coincident with real link status.

Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Amos Kong <akong@redhat.com>

---
v2: don't add MediaState in RTL8139State to avoid trouble
v3: adding an enum MediaStatusBits
v4: correct nc.link_down in the end of migration
v5: fix bug in bit operating
---
 hw/rtl8139.c |   24 ++++++++++++++++++++++--
 1 files changed, 22 insertions(+), 2 deletions(-)

diff --git a/hw/rtl8139.c b/hw/rtl8139.c
index 844f1b8..ae68f55 100644
--- a/hw/rtl8139.c
+++ b/hw/rtl8139.c
@@ -167,7 +167,7 @@ enum IntrStatusBits {
     PCIErr = 0x8000,
     PCSTimeout = 0x4000,
     RxFIFOOver = 0x40,
-    RxUnderrun = 0x20,
+    RxUnderrun = 0x20, /* Packet Underrun / Link Change */
     RxOverflow = 0x10,
     TxErr = 0x08,
     TxOK = 0x04,
@@ -3007,7 +3007,8 @@ static uint32_t rtl8139_io_readb(void *opaque, uint8_t addr)
             break;
 
         case MediaStatus:
-            ret = 0xd0;
+            /* The LinkDown bit of MediaStatus is inverse with link status */
+            ret = 0xd0 | (~s->BasicModeStatus & 0x04);
             DPRINTF("MediaStatus read 0x%x\n", ret);
             break;
 
@@ -3262,6 +3263,10 @@ static int rtl8139_post_load(void *opaque, int version_id)
         s->cplus_enabled = s->CpCmd != 0;
     }
 
+    /* nc.link_down can't be migrated, so infer link_down according
+     * to link status bit in BasicModeStatus */
+    s->nic->nc.link_down = (s->BasicModeStatus & 0x04) == 0;
+
     return 0;
 }
 
@@ -3453,12 +3458,27 @@ static void pci_rtl8139_uninit(PCIDevice *dev)
     qemu_del_net_client(&s->nic->nc);
 }
 
+static void rtl8139_set_link_status(NetClientState *nc)
+{
+    RTL8139State *s = DO_UPCAST(NICState, nc, nc)->opaque;
+
+    if (nc->link_down) {
+        s->BasicModeStatus &= ~0x04;
+    } else {
+        s->BasicModeStatus |= 0x04;
+    }
+
+    s->IntrStatus |= RxUnderrun;
+    rtl8139_update_irq(s);
+}
+
 static NetClientInfo net_rtl8139_info = {
     .type = NET_CLIENT_OPTIONS_KIND_NIC,
     .size = sizeof(NICState),
     .can_receive = rtl8139_can_receive,
     .receive = rtl8139_receive,
     .cleanup = rtl8139_cleanup,
+    .link_status_changed = rtl8139_set_link_status,
 };
 
 static int pci_rtl8139_init(PCIDevice *dev)
-- 
1.7.1

^ permalink raw reply related	[flat|nested] 5+ messages in thread

* [Qemu-devel] [PATCH v5 2/3] e1000: update nc.link_down in e1000_post_load()
  2012-09-28  2:05 [Qemu-devel] [PATCH v5 0/3] net: fix link status Amos Kong
  2012-09-28  2:06 ` [Qemu-devel] [PATCH v5 1/3] rtl8139: implement 8139cp " Amos Kong
@ 2012-09-28  2:06 ` Amos Kong
  2012-09-28  2:06 ` [Qemu-devel] [PATCH v5 3/3] update nc.link_down in virtio_net_load() Amos Kong
  2012-10-04 17:02 ` [Qemu-devel] [PATCH v5 0/3] net: fix link status Stefan Hajnoczi
  3 siblings, 0 replies; 5+ messages in thread
From: Amos Kong @ 2012-09-28  2:06 UTC (permalink / raw)
  To: qemu-devel; +Cc: aliguori, mst, stefanha, jasowang, pbonzini, Amos Kong

This patch introduced e1000_post_load(), it will be called in the end of
migration. nc.link_down could not be migrated, this patch updates
link_down in e1000_post_load() to keep it coincident with real link
status.

Signed-off-by: Amos Kong <akong@redhat.com>
---
 hw/e1000.c |   12 ++++++++++++
 1 files changed, 12 insertions(+), 0 deletions(-)

diff --git a/hw/e1000.c b/hw/e1000.c
index ec3a7c4..63fee10 100644
--- a/hw/e1000.c
+++ b/hw/e1000.c
@@ -1079,11 +1079,23 @@ static bool is_version_1(void *opaque, int version_id)
     return version_id == 1;
 }
 
+static int e1000_post_load(void *opaque, int version_id)
+{
+    E1000State *s = opaque;
+
+    /* nc.link_down can't be migrated, so infer link_down according
+     * to link status bit in mac_reg[STATUS] */
+    s->nic->nc.link_down = (s->mac_reg[STATUS] & E1000_STATUS_LU) == 0;
+
+    return 0;
+}
+
 static const VMStateDescription vmstate_e1000 = {
     .name = "e1000",
     .version_id = 2,
     .minimum_version_id = 1,
     .minimum_version_id_old = 1,
+    .post_load = e1000_post_load,
     .fields      = (VMStateField []) {
         VMSTATE_PCI_DEVICE(dev, E1000State),
         VMSTATE_UNUSED_TEST(is_version_1, 4), /* was instance id */
-- 
1.7.1

^ permalink raw reply related	[flat|nested] 5+ messages in thread

* [Qemu-devel] [PATCH v5 3/3] update nc.link_down in virtio_net_load()
  2012-09-28  2:05 [Qemu-devel] [PATCH v5 0/3] net: fix link status Amos Kong
  2012-09-28  2:06 ` [Qemu-devel] [PATCH v5 1/3] rtl8139: implement 8139cp " Amos Kong
  2012-09-28  2:06 ` [Qemu-devel] [PATCH v5 2/3] e1000: update nc.link_down in e1000_post_load() Amos Kong
@ 2012-09-28  2:06 ` Amos Kong
  2012-10-04 17:02 ` [Qemu-devel] [PATCH v5 0/3] net: fix link status Stefan Hajnoczi
  3 siblings, 0 replies; 5+ messages in thread
From: Amos Kong @ 2012-09-28  2:06 UTC (permalink / raw)
  To: qemu-devel; +Cc: aliguori, mst, stefanha, jasowang, pbonzini, Amos Kong

nc.link_down could not be migrated, this patch updates link_down in
virtio_post_load() to keep it coincident with real link status.

Signed-off-by: Amos Kong <akong@redhat.com>
---
 hw/virtio-net.c |    5 +++++
 1 files changed, 5 insertions(+), 0 deletions(-)

diff --git a/hw/virtio-net.c b/hw/virtio-net.c
index 6490743..f3d4acf 100644
--- a/hw/virtio-net.c
+++ b/hw/virtio-net.c
@@ -973,6 +973,11 @@ static int virtio_net_load(QEMUFile *f, void *opaque, int version_id)
         }
     }
     n->mac_table.first_multi = i;
+
+    /* nc.link_down can't be migrated, so infer link_down according
+     * to link status bit in n->status */
+    n->nic->nc.link_down = (n->status & VIRTIO_NET_S_LINK_UP) == 0;
+
     return 0;
 }
 
-- 
1.7.1

^ permalink raw reply related	[flat|nested] 5+ messages in thread

* Re: [Qemu-devel] [PATCH v5 0/3] net: fix link status
  2012-09-28  2:05 [Qemu-devel] [PATCH v5 0/3] net: fix link status Amos Kong
                   ` (2 preceding siblings ...)
  2012-09-28  2:06 ` [Qemu-devel] [PATCH v5 3/3] update nc.link_down in virtio_net_load() Amos Kong
@ 2012-10-04 17:02 ` Stefan Hajnoczi
  3 siblings, 0 replies; 5+ messages in thread
From: Stefan Hajnoczi @ 2012-10-04 17:02 UTC (permalink / raw)
  To: Amos Kong; +Cc: pbonzini, jasowang, aliguori, qemu-devel, mst

On Fri, Sep 28, 2012 at 4:05 AM, Amos Kong <akong@redhat.com> wrote:
> First patch implemeted link status setting of rtl8139, the rest patches
> updated nc.link_down in post_load() of migration to keep it coincident
> with real link status.
>
> v5: fix bug in bit operating
>
> Amos Kong (2):
>   e1000: update nc.link_down in e1000_post_load()
>   update nc.link_down in virtio_net_load()
>
> Jason Wang (1):
>   rtl8139: implement 8139cp link status
>
>  hw/e1000.c      |   12 ++++++++++++
>  hw/rtl8139.c    |   24 ++++++++++++++++++++++--
>  hw/virtio-net.c |    5 +++++
>  3 files changed, 39 insertions(+), 2 deletions(-)

Thanks, merged!

http://github.com/stefanha/qemu/commits/net

Stefan

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2012-10-04 17:02 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-09-28  2:05 [Qemu-devel] [PATCH v5 0/3] net: fix link status Amos Kong
2012-09-28  2:06 ` [Qemu-devel] [PATCH v5 1/3] rtl8139: implement 8139cp " Amos Kong
2012-09-28  2:06 ` [Qemu-devel] [PATCH v5 2/3] e1000: update nc.link_down in e1000_post_load() Amos Kong
2012-09-28  2:06 ` [Qemu-devel] [PATCH v5 3/3] update nc.link_down in virtio_net_load() Amos Kong
2012-10-04 17:02 ` [Qemu-devel] [PATCH v5 0/3] net: fix link status Stefan Hajnoczi

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).