Patchwork [04/13] eepro100: Fix network hang when rx buffers run out

login
register
mail settings
Submitter Stefan Hajnoczi
Date Sept. 14, 2012, 8:46 a.m.
Message ID <1347612420-5704-5-git-send-email-stefanha@gmail.com>
Download mbox | patch
Permalink /patch/183846/
State New
Headers show

Comments

Stefan Hajnoczi - Sept. 14, 2012, 8:46 a.m.
From: Bo Yang <boyang@suse.com>

This is reported by QA. When installing os with pxe, after the initial
kernel and initrd are loaded, the procedure tries to copy files from install
server to local harddisk, the network becomes stall because of running out of
receive descriptor.

[Whitespace fixes and removed qemu_notify_event() because Paolo's
earlier net patches have moved it into qemu_flush_queued_packets().

Additional info:

I can reproduce the network hang with a tap device doing a iPXE HTTP
boot as follows:

  $ qemu -enable-kvm -m 1024 \
    -netdev tap,id=netdev0,script=no,downscript=no \
    -device i82559er,netdev=netdev0,romfile=80861209.rom \
    -drive if=virtio,cache=none,file=test.img
  iPXE> ifopen net0
  iPXE> config # set static network configuration
  iPXE> kernel http://mirror.bytemark.co.uk/fedora/linux/releases/17/Fedora/x86_64/os/images/pxeboot/vmlinuz

I needed a vanilla iPXE ROM to get to the iPXE prompt.  I think the boot
prompt has been disabled in the ROMs that ship with QEMU to reduce boot
time.

During the vmlinuz HTTP download there is a network hang.  hw/eepro100.c
has reached the end of the rx descriptor list.  When the iPXE driver
replenishes the rx descriptor list we don't kick the QEMU net subsystem
and event loop, thereby leaving the tap netdev without its file
descriptor in select(2).

Stefan Hajnoczi <stefanha@gmail.com>]

Signed-off-by: Bo Yang <boyang@suse.com>
Signed-off-by: Stefan Hajnoczi <stefanha@gmail.com>
---
 hw/eepro100.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

Patch

diff --git a/hw/eepro100.c b/hw/eepro100.c
index 50d117e..5b23116 100644
--- a/hw/eepro100.c
+++ b/hw/eepro100.c
@@ -1036,6 +1036,7 @@  static void eepro100_ru_command(EEPRO100State * s, uint8_t val)
         }
         set_ru_state(s, ru_ready);
         s->ru_offset = e100_read_reg4(s, SCBPointer);
+        qemu_flush_queued_packets(&s->nic->nc);
         TRACE(OTHER, logout("val=0x%02x (rx start)\n", val));
         break;
     case RX_RESUME:
@@ -1770,7 +1771,8 @@  static ssize_t nic_receive(NetClientState *nc, const uint8_t * buf, size_t size)
     if (rfd_command & COMMAND_EL) {
         /* EL bit is set, so this was the last frame. */
         logout("receive: Running out of frames\n");
-        set_ru_state(s, ru_suspended);
+        set_ru_state(s, ru_no_resources);
+        eepro100_rnr_interrupt(s);
     }
     if (rfd_command & COMMAND_S) {
         /* S bit is set. */