casaroli opened a new pull request, #21908:
URL: https://github.com/apache/nuttx/pull/21908

   depends-on: apache/nuttx-apps/pull/4063
   
   ## Summary
   
   This PR enables the packet socket test `cmocka_net_pkt` (nuttx-apps 
testing/nettest/pkt) in rv-virt:citest, so that CI builds it and NTFC runs it. 
No CI configuration builds testing/nettest with `NET_PKT` today.
   
   The defconfig enables `NET_LOOPBACK`, `NET_PKT` with write buffers (two 
preallocated connections) and `TESTING_NET_TEST`. The other nettest programs 
that these options enable (`local_dgram`, `others`) stay off, because no NTFC 
case runs them. `CONFIG_IOB_THROTTLE=0` keeps the old value, because packet 
write buffers change its default to 8. The test task gets 4096 bytes of stack: 
the existing `test_pkt_lo_send` overflows the default 2048 bytes on this board.
   
   This PR builds on #21906: its branch contains that commit, so merge it after 
#21906. The new case `test_pkt_lo_close` (apache/nuttx-apps#4063) fails without 
#21906.
   
   ## Impact
   
   rv-virt:citest only: a larger image (packet sockets, loopback device, one 
more cmocka program). No code change.
   
   ## Testing
   
   rv-virt:citest (flat rv32, QEMU TCG), with the nuttx-apps PR above, built as 
CI builds it (`configure.sh rv-virt:citest`, make):
   
   | Tree | test_pkt_lo_send | test_pkt_lo_close |
   |---|---|---|
   | this PR (with #21906) | OK | OK |
   | this PR with #21906 reverted | OK | FAILED: 4 IOBs lost |
   
   `test_pkt_lo_frag` is SKIPPED (no NET_IPFRAG). The NTFC command `cmocka -s 
cmocka_net_pkt -t test_pkt_lo_close` gives the same results. The defconfig is 
the output of `make savedefconfig`. checkpatch: `tools/checkpatch.sh -c -u -m 
-g` passes.
   
   NTFC runs a cmocka program only through a case in apache/nuttx-ntfc-testing, 
and CI uses its tag release-0.0.2. A case for `cmocka_net_pkt` is proposed 
there separately.
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to