[5/5] Set the RU state to ru_no_resources instead of asserting when we used up the last receive buffer. This should not usually happen with good drivers, but it can happen with the OS X drivers at least.

Submitted by Reimar Döffinger on Aug. 11, 2009, 9:15 p.m.

Details

Message ID 20090811211536.GF10500@1und1.de
State Superseded
Headers show

Commit Message

Reimar Döffinger Aug. 11, 2009, 9:15 p.m. 
Currently if the hardware runs out of receive buffers qemu
crashes with an assert.
Simply setting the RU state to "no resources" seems to work (though it
is hard to provoke the situation and thus hard to test well), at least
it can't be any worse than crashing due to an assert (IMO)...

Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>
---
 hw/eepro100.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

Patch hide | download patch | download mbox 

diff --git a/hw/eepro100.c b/hw/eepro100.c
index d2c18cc..8b343c1 100644
--- a/hw/eepro100.c
+++ b/hw/eepro100.c
@@ -1558,7 +1558,7 @@  static ssize_t nic_receive(VLANClientState *vc, const uint8_t * buf, size_t size
     s->ru_offset = le32_to_cpu(rx.link);
     if (rfd_command & 0x8000) {
         /* EL bit is set, so this was the last frame. */
-        assert(0);
+        set_ru_state(s, ru_no_resources);
     }
     if (rfd_command & 0x4000) {
         /* S bit is set. */