597d10b33f
Add correct parsers
2019-07-29 16:29:12 +02:00
0c68b1b82c
++notes
2019-07-29 10:43:54 +02:00
bbe78a44c1
Shift first
2019-07-29 09:40:20 +02:00
62e0279b25
try to emit on ANY other port
2019-07-29 09:37:01 +02:00
eed0f3c56e
Let's see if 16 == the first port of the other card
2019-07-29 02:00:13 +02:00
53ce47a54e
Make it more stupid - forget about tables
2019-07-28 20:08:28 +02:00
c3331bcea7
Only use v4_networks() and set default port to 4
2019-07-28 16:56:30 +02:00
58d7d91358
Re-organising
2019-07-28 16:27:53 +02:00
db5bae89ad
++ various logs
2019-07-28 13:40:49 +02:00
cbf2a05ffd
++notes
2019-07-28 12:19:59 +02:00
1ae0294a39
Re-add table entries in commands.txt
2019-07-27 21:35:47 +02:00
1c661f6990
Try to workaround compiler bug with markers
2019-07-25 15:06:56 +02:00
809466cf61
re-enable other code that might not trigger compiler bug
2019-07-25 14:52:57 +02:00
7bba816f2c
v4 if next to v6 if, not to nat64
2019-07-25 14:40:20 +02:00
5478e9fad7
Begin to write defines for netpfga instead of actions
2019-07-24 23:45:40 +02:00
a3dd7ee106
++log
2019-07-24 23:33:09 +02:00
8c94c1ac09
[netpfga] Fix default actions / cannot have arguments
2019-07-24 22:52:23 +02:00
dc7d78252e
Merge full IPv6 handling code from BMV2 side
2019-07-24 22:48:21 +02:00
6b176fdc06
table size up, down, up...
...
Exact matches don't like 63
actions_nat64_generic.p4(173): error: table size too small for match_type(EM): 63 < 64
size = 63;
^^
actions_nat64_generic.p4(173): error: could not not map table size size
size = 63;
^^^^
2019-07-24 22:42:05 +02:00
a055f55c8d
++notes
2019-07-24 22:39:59 +02:00
4afbf053af
[netpfga] change table key to exact instead of LPM
...
Known bug of netpfga not correctly supporting LPM. Current bug
might be triggered by it:
2019-07-24 22:38:55 +02:00
9e6bb893ac
[netpfga] remove ARP support
...
Due to
minip4_solution.p4(35)
parser RealParser(
^^^^^^^^^^
error: table match_types are not the same
actions_arp.p4(35): error: could not map table key(s) KeyElement
hdr.arp.dst_ipv4_addr: lpm;
^^^^^^^^^^^^^^^^^^^^^
Makefile:34: recipe for target 'all' failed
make[1]: *** [all] Error 1
caused by
table v4_arp {
key = {
hdr.ethernet.dst_addr: exact;
hdr.arp.opcode: exact;
hdr.arp.dst_ipv4_addr: lpm;
}
2019-07-24 22:27:36 +02:00
306d52176f
Set table size to 2^n -1 for everything
...
Motivated by v6.1 error:
make[1]: Entering directory '/home/nico/projects/P4-NetFPGA/contrib-projects/sume-sdnet-switch/projects/minip4/src'
p4c-sdnet -o minip4.sdnet --sdnet_info .sdnet_switch_info.dat minip4_solution.p4
actions_egress.p4(52): warning: Table v6_networks is not used; removing
table v6_networks {
^^^^^^^^^^^
actions_egress.p4(69): warning: Table v4_networks is not used; removing
table v4_networks {
^^^^^^^^^^^
actions_nat64_generic.p4(174): warning: Table nat46 is not used; removing
table nat46 {
^^^^^
minip4_solution.p4(38): [--Wwarn=uninitialized_out_param] warning: out parameter meta may be uninitialized when RealParser terminates
out metadata meta,
^^^^
minip4_solution.p4(35)
parser RealParser(
^^^^^^^^^^
error: LPM table size should be 2^n - 1
actions_nat64_generic.p4(169): error: could not not map table size size
size = 64;
^^^^
error: table match_types are not the same
actions_arp.p4(35): error: could not map table key(s) KeyElement
hdr.arp.dst_ipv4_addr: lpm;
^^^^^^^^^^^^^^^^^^^^^
error: LPM table size should be 2^n - 1
actions_arp.p4(55): error: could not not map table size size
size = 64;
^^^^
Makefile:34: recipe for target 'all' failed
make[1]: *** [all] Error 1
make[1]: Leaving directory '/home/nico/projects/P4-NetFPGA/contrib-projects/sume-sdnet-switch/projects/minip4/src'
Makefile:31: recipe for target 'frontend' failed
make: *** [frontend] Error 2
nico@nsg-System:~/master-thesis/netpfga/log$
2019-07-24 13:20:35 +02:00
ee0d500355
[netpfga] phase in first applied tables/nat64 code
2019-07-24 13:08:02 +02:00
3559637b56
--typo
2019-07-24 11:52:42 +02:00
94e06be245
[netpfga] include UNUSED nat64 actions and tables
...
Will probably be removed, but need to check whether it still compiles.
2019-07-24 11:45:37 +02:00
4eabf8cc38
state name change
2019-07-24 09:53:13 +02:00
181870d14d
reset the code base to working version again
...
Signed-off-by: Nico Schottelius <nico@nico-notebook.schottelius.org>
2019-07-24 09:36:24 +02:00
93c98323f4
add ifdef for controller actions
2019-07-23 15:50:31 +02:00
6e162ca63b
Rewrite the netpfga p4 program
2019-07-23 12:21:49 +02:00
1059e8d0e0
Begin from the beginning: reset to port1 only
2019-07-23 10:20:18 +02:00
5064efda2d
++notes
2019-07-21 22:30:49 +02:00
d97864ffa7
+notes -comment
2019-07-21 16:39:15 +02:00
71404d9a58
++notes
2019-07-21 14:52:43 +02:00
55a363055b
++debug of netfpga compile
2019-07-21 14:37:31 +02:00
2462356e68
log update
2019-07-21 14:08:47 +02:00
9458507ee9
Start pktcnt at 1 instead of 0
2019-07-21 12:15:04 +02:00
6b478c87f1
Try fix shifting errors in first block
2019-07-21 10:50:26 +02:00
266ab1d036
Try fixing shift errors (precendence!)
2019-07-21 10:48:35 +02:00
dda1aa9c3c
Split ipv4 header into 16 bit words
2019-07-21 10:45:43 +02:00
018e4cc9ff
In theory: implement NAT64 from v6 to v4 without externs
2019-07-21 10:23:49 +02:00
020601f463
[checksum] add v6->v4 translations delta based, too
2019-07-17 18:06:49 +02:00
55cc9db975
Fix meaning of "send_to_port1"
2019-07-17 17:41:54 +02:00
831d69fe9c
Remove function, because functions don't compile
2019-07-17 17:37:37 +02:00
87d2db4b5f
[netpfga] Add default action for matching, remove commands.txt
2019-07-17 17:24:34 +02:00
512fca1349
update checksum code
2019-07-17 15:46:54 +02:00
bc90421426
Get rid of negative wrap around
2019-07-16 13:05:13 +02:00
811e4c2b11
Separate checksum addition/subtraction into 2 steps
2019-07-16 12:30:29 +02:00
ec48d89c67
get the v6sum
2019-07-16 12:14:46 +02:00
ecab110d7c
Back to 16 bit arithmetic
2019-07-16 12:06:26 +02:00
1a39c35e3b
Check how wrapping is done in P4
2019-07-16 11:53:00 +02:00
baa50b91c6
Use own filtering code
2019-07-15 16:53:57 +02:00
f0900c3821
update tcp checksumming
2019-07-15 16:48:24 +02:00
26c27cefa8
Change order of complement & filtering
2019-07-15 16:20:51 +02:00
a9cc93281e
Change testdata offset by 1 and find checksum error
2019-07-13 22:07:53 +02:00
c243778bc3
++log ++change gen_testdata.py
2019-07-13 19:10:05 +02:00
86885760b2
update gen_testdata
...
- add more packets
- fix counter for one packet
2019-07-13 17:28:03 +02:00
75c185794e
++notes
2019-07-11 10:58:06 +02:00
e9997964cd
Split of checksumming of bmv2
2019-07-11 10:50:00 +02:00
378e16f48c
++stuff
2019-07-11 08:47:34 +02:00
04d0fbc85a
send udp from scapy
2019-07-10 14:37:58 +02:00
476b04c97c
++doc ++egress port
2019-07-10 08:55:07 +02:00
ff383e3b08
add test json file for bmv2
2019-07-10 08:44:56 +02:00
a69c4060e2
begin implementing diff based checksum in p4
2019-07-06 20:57:11 +02:00
ce212ed9e1
Checksums: ipv4 diff not needed; minip4: enable icmp, arp
2019-07-01 11:03:10 +02:00
92e1407729
checksumming part finish
2019-07-01 09:36:14 +02:00
b757a3c2b0
Use UDP for testing checksums
2019-06-29 17:13:28 +02:00
80aeaf164a
port controller sending code partially to netpfga
2019-06-24 14:24:51 +02:00
0a225479c7
Generate different test data
2019-06-24 13:45:06 +02:00
47a7c966bf
Try matching different compiler output
2019-06-24 13:20:20 +02:00
8b8f70e6a0
Begin to introduce commented out code, use metadata
2019-06-24 13:05:42 +02:00
408dd6b782
Delta checksum approach
2019-06-23 14:19:06 +02:00
2b5d5ea62d
++netpfga updates
...
Signed-off-by: Nico Schottelius <nico@nico-notebook.schottelius.org>
2019-06-17 19:39:07 +02:00
04d174c01f
++ meeting notes
2019-06-17 12:35:12 +02:00
03317a6c11
++notes
2019-06-17 09:41:17 +02:00
42af5cd2d8
++ notes
2019-06-15 22:49:54 +02:00
47d49638a8
Try implementing bcast option in gen_testdata
2019-06-12 19:46:22 +02:00
dec8a02a26
+doc +log +allports
2019-06-12 19:06:18 +02:00
9cebd4024a
++ notes
2019-06-11 17:54:22 +02:00
36ccd6e021
++ notes
2019-06-10 11:20:11 +02:00
2130f14d02
add switch to skip long compile step
2019-06-06 22:30:31 +02:00
ead0276527
add bin/ with test script for sending 1 packet
2019-06-04 19:41:57 +02:00
5e10401f96
++doc ++sudo for drivers
2019-06-04 00:21:28 +02:00
63ec17b9a4
begin to integrate headers of real code into netpfga
2019-06-03 22:33:37 +02:00
e0226c4c90
try 1 == nf0
2019-06-03 22:01:58 +02:00
442ea1a7da
++doc, use nf3
2019-06-03 21:56:45 +02:00
6c895c583a
netpfga: always emit on port1
2019-06-03 21:27:07 +02:00
61b807bd2e
++doc ++setup egress
2019-05-26 11:12:49 +02:00
fa254011e6
++notes
2019-05-26 10:58:35 +02:00
880c228a76
++ notes
2019-05-25 14:18:06 +02:00
4059701f63
++netfpga logs
2019-05-20 11:12:48 +02:00
533ae3e080
++ log netpfga
2019-05-18 15:33:21 +02:00
5592d45e7b
++notes
2019-05-10 23:28:24 +02:00
7be9c01b48
stop processing v6 when matching for nat64 session
2019-05-06 12:50:45 +02:00
0525f2b4b5
[dummy commit]
2019-05-06 12:19:26 +02:00
b0b9cd7461
[controller] insert 6->4 session
2019-05-06 12:16:22 +02:00
b972accc73
[refactor] nat64 now - not only static mapping anymore
2019-05-05 13:48:05 +02:00
02fc065c1d
Begin adding session tables, debug infos
2019-05-02 13:54:31 +02:00
879abe94c2
update doc
2019-04-21 12:24:15 +02:00
079699c687
++update
...
Signed-off-by: Nico Schottelius <nico@nico-notebook.schottelius.org>
2019-04-11 10:57:30 +02:00
Nico Schottelius