- Timestamp:
- Jun 28, 2011 4:02:10 PM (13 years ago)
- File:
-
- 1 edited
Legend:
- Unmodified
- Added
- Removed
-
trunk/src/VBox/Devices/USB/DevOHCI.cpp
r37636 r37668 1701 1701 { 1702 1702 int i = ohci_in_done_queue_find(pOhci, GCPhysTD); 1703 #if 0 1704 /* This condition has been observed with the USB tablet emulation or with 1705 * a real USB mouse and an SMP XP guest. I am also not sure if this is 1706 * really a problem for us. The assertion checks that the guest doesn't 1707 * re-submit a TD which is still in the done queue. It seems to me that 1708 * this should only be a problem if we either keep track of TDs in the done 1709 * queue somewhere else as well (in which case we should also free those 1710 * references in time, and I can't see any code doing that) or if we 1711 * manipulate TDs in the done queue in some way that might fail if they are 1712 * re-submitted (can't see anything like that either). 1713 */ 1703 1714 AssertMsg(i < 0, ("TD %#010x (i=%d)\n", GCPhysTD, i)); 1715 #endif 1704 1716 return i < 0; 1705 1717 }
Note:
See TracChangeset
for help on using the changeset viewer.