aboutsummaryrefslogtreecommitdiffstats
path: root/firmware/microblaze/lib/net_common.h
Commit message (Collapse)AuthorAgeFilesLines
* zpu: renamed the directory for the usrp2 fw to zpu to reflect the cpu typeJosh Blum2010-12-221-49/+0
|
* usrp2: got fw working on usrp2+nseries (crosses fingers)Josh Blum2010-12-221-0/+2
| | | | | | The abort issue seemed to be cause by the listeners table not being initialized. Also gave the pic handler the initialization works to be consistent but not needed as it was static initialized. And finally, doubly set the packet router handshake ctrl before entering the claim calls (seemed to fix the lockup, perhaps a bad init state?)
* packet_router: dont register mac, also reorganized some tidbitsJosh Blum2010-12-111-0/+7
|
* packet_router: implemented code to program the addresses into the routerJosh Blum2010-12-111-3/+1
|
* usrp2: pretty much gutted the buffer pool stuff with other misc changes, not ↵Josh Blum2010-12-111-10/+0
| | | | in working shape
* usrp2: fw sends gratuitous arp on link-upJosh Blum2010-10-221-0/+2
|
* Added recovery app to use raw socket to burn known ip addr.Josh Blum2010-03-161-11/+2
| | | | Fixed up fw so the device can reply after new addrs are set.
* Moved timeouts into the udp transports.Josh Blum2010-03-041-2/+0
| | | | | | Simplified the fast path checking in the fw, but it turns out this was not the issue. Fixed some bad bit operations with the 16sc words (dont forget sign extension). Added some more documentation to the headers....
* The net common is too slow in usrp2 firmware to figure out if its vrt data.Josh Blum2010-03-021-0/+1
| | | | | Added a custom function to tell if a packet is vrt data, seems to be feeding fast enough at this rate... Fixed some buffer size calculation logic.
* net_common working in this state with control udp packetsJosh Blum2010-02-101-0/+60