Commit Graph

279 Commits

Author SHA1 Message Date
William Skellenger
5da7e95fa9 block with five characters is now handled correctly 2024-06-07 13:40:15 -04:00
William Skellenger
930c187c75 added new ipv6 valid addr test 2024-05-28 18:42:59 -04:00
Erik Ekman
1cc1536e6a test/unit: fix inconsistent nullness check error
Inside a testcase, but still. Trying to address #160
2024-04-09 20:11:39 +02:00
Erik Ekman
20c9b117e1 test: Fix speling issues
Found by codespell
2023-10-14 17:51:01 +02:00
Simon Goldschmidt
9e2fb2806f unittests: fix compiling in release mode 2023-10-10 20:14:52 +02:00
Simon Goldschmidt
66ef250b8c Fix running IPv6 reassembly unit tests on 64-bit 2023-10-10 19:59:06 +02:00
Simon Goldschmidt
c8f42025ce ipv6 reassembly: fix detecting holes in reassembled packets
Added unit test for reassembly.

See bug #64767
2023-10-10 16:31:16 +02:00
Simon Goldschmidt
732d3411bf fuzz: count packets for debugging 2023-10-03 21:37:14 +02:00
Simon Goldschmidt
9509acd876 fuzz: fix warning casting u64_t to s32_t/u32_t 2023-10-03 21:11:57 +02:00
Simon Goldschmidt
14444c1c7f mqtt: support binary Will Message
see patch #10049
2023-09-29 22:08:03 +02:00
Simon Goldschmidt
9672405eb9 test: unit: tcp: fix tcp state tests to not break other tests
On platform where check cannot fork (win32), the tcp_state tests crashed other tests.

Signed-off-by: Simon Goldschmidt <goldsimon@gmx.de>
2023-06-29 21:32:20 +02:00
Jonathan Neuschäfer
c2f9244137 test/fuzz: Fix use of htons vs. ntohs
Although ntohs and htons perform equivalent operations (a byte swap on
little-endian platforms), their semantic roles (as implied by the names)
are different.

In fuzz_common.c, we get integers from simulated network packages, and
convert them for host CPU use, so ntohs is the right variant to use.
2023-05-11 22:10:32 +02:00
Jonathan Neuschäfer
5514d8eb63 test/fuzz: Build with -O2 by default
Fuzzers should run fast by default, since the goal is to try many
variations of input data.
2023-05-11 22:08:46 +02:00
Jonathan Neuschäfer
911da26f96 test/fuzz: Fix CC default value in makefile
Unfortunately, CC ?= afl-gcc doesn't work. This is because CC has a
default value (of "cc"), and the ?= operator only assigns a value if the
variable previously had none. "make" currently compiles with cc.

In this patch, I implemented the more elaborate way to achieve what was
probably intended: "make" will use afl-gcc now, and "make CC=foo" will
compile with "foo".
2023-05-11 22:08:40 +02:00
Simon Goldschmidt
90c9a482a6 ci: fix compiling test_tcp_state.c; add to msvc project 2023-05-11 20:26:21 +02:00
Chen Yu Dong
f50c87da6f ci: Add unit test suite to validate tcp state transitions 2023-05-10 22:02:43 +02:00
Simon Goldschmidt
64f75dbd89 pppos: fix compiling new pppos unit test
Signed-off-by: Simon Goldschmidt <goldsimon@gmx.de>
2023-05-10 20:58:23 +02:00
David Cermak
6b734bbb2c PPP: Add test exhibiting empty packet null-deref
Also updates lcp.c to compile with PPP_AUTH_SUPPORT=0 in clang
2023-05-10 20:42:27 +02:00
Jonathan Neuschäfer
95a63bc352 test/fuzz: Use HTTPS URLs and mention AFL++ 2023-04-07 22:27:30 +02:00
Jonathan Neuschäfer
e8b0324f81 test/fuzz: Don't crash with assertion failure when a packet is too big
A fuzzer may generate big packets, and it will interpret assertion
failures as bugs. Instead of asserting that the packet size is
reasonable, simply skip the packet when it's not the case.
2023-04-07 22:25:50 +02:00
Florian La Roche
7c3aab2ea2 gcc warning for sign conversion
See patch #10143
2021-12-30 22:48:18 +01:00
Wouter van Gulik
5ab46f82f9 Fix failing fuzzer if compiling with ubsan
See patch #10163

Signed-off-by: Simon Goldschmidt <goldsimon@gmx.de>
2021-12-30 22:04:23 +01:00
Simon Goldschmidt
bb5ee3783a fix compiling/running tests on win32
- new files
- different warning settings
- all tests run in a single process (no fork)
2021-05-21 22:00:49 +02:00
Erik Ekman
988ce1e80e netif: Add test for netif_find 2021-02-27 12:28:14 +01:00
Erik Ekman
74231d6873 netif: Add test for netif_is_flag_set() 2021-02-08 19:29:33 +01:00
Erik Ekman
c825eb3604 tcp: Check tcp_output return values in tests
Found with clang-analyzer
2021-01-19 20:19:02 +01:00
Erik Ekman
21fba21054 test/unit: Update convenience makefile
Allow cleaning - useful when switching compiler between GCC/clang.
2021-01-19 20:19:02 +01:00
Erik Ekman
7b77378ed8 test/unit: Fix compilation of ip4 test
../../../../src/../test/unit/ip4/test_ip4.c: In function ‘test_ip4_icmp_replylen_short’:
../../../../src/../test/unit/ip4/test_ip4.c:291:35: error: comparison between signed and unsigned integer expressions [-Werror=sign-compare]
   fail_unless(linkoutput_byte_ctr == icmp_len + sizeof(unknown_proto));

Plus minor cleanup in second icmp reply test
2021-01-13 21:31:19 +01:00
Erik Ekman
d0c214feff test/unit: Add convenience Makefile
To allow running them easily on *nix without changing directory.
2021-01-13 20:57:23 +01:00
Erik Ekman
35550eeab3 icmp: Fix copied length in response packets
Fixes bug #59364, reported by Yi Guai
2021-01-13 20:31:55 +01:00
Erik Ekman
d6a37b1ad6 fuzz: Improve Makefile
Similar to recent change to ports/unix/check:
- Split depfile into three different files
- Fix make dep
- Skip depfile generation for 'make clean'
2020-12-04 10:51:11 +01:00
Erik Ekman
264a5a3e97 Rename IP and Ethernet equality checkers from _cmp to _eq
Functions ending in cmp are expected to return 0 on equality but these
return non-zero.

eth_addr_cmp -> eth_addr_eq

ip_addr_cmp -> ip_addr_eq
ip4_addr_cmp -> ip4_addr_eq
ip6_addr_cmp -> ip6_addr_eq

ip_addr_netcmp -> ip_addr_net_eq
ip4_addr_netcmp -> ip4_addr_net_eq
ip6_addr_netcmp -> ip6_addr_net_eq

ip_addr_cmp_zoneless -> ip_addr_zoneless_eq
ip6_addr_cmp_zoneless -> ip6_addr_zoneless_eq

ip6_addr_cmp_zone -> ip6_addr_zone_eq
ip6_addr_netcmp_zoneless -> ip6_addr_net_zoneless_eq
ip6_addr_nethostcmp -> ip6_addr_nethost_eq
ip6_addr_cmp_packed -> ip6_addr_packed_eq
ip6_addr_cmp_solicitednode -> ip6_addr_solicitednode_eq

All call sites have been changed, and fallback macros have been added to not
break external users.
2020-07-07 18:51:45 +02:00
Erik Ekman
047f3b3306 dns: Verify setting/getting servers in unit test 2020-07-04 18:57:41 +02:00
Erik Ekman
5b7524e10e unit: Enable DNS
Avoid allocating in dns_init by enabling random source port.
2020-07-04 14:24:40 +02:00
Erik Ekman
2bf8fcd7c2 pbuf: Add pbuf_copy_partial_pbuf library function
Like pbuf_copy, but can copy part of a pbuf to an offset in another.
pbuf_copy now uses this function internally.

Replace pbuf_take_at loop in icmp6 with pbuf_copy_partial_pbuf().
2020-07-03 18:55:02 +02:00
Erik Ekman
942ca825d0 ip4: Fix missing END_TEST in ip4 test case 2020-07-01 21:09:49 +02:00
Erik Ekman
ec5f3a1563 ip6: Add ipv6 fragmentation test
Seems to work the same as ipv4, and that I was just confused by all the
loops when reading the code.
2020-07-01 20:17:29 +02:00
Erik Ekman
0bda621d3d ip4: Verify ipv4 fragmentation sends all chunks right away
Ahead of investigation how ipv6 code behaves.
2020-07-01 20:16:24 +02:00
Erik Ekman
2faa2d7009 ip6: Rename pbuf in unit test
../../../../src/../test/unit/ip6/test_ip6.c: In function ‘test_ip6_dest_unreachable_chained_pbuf’:
../../../../src/../test/unit/ip6/test_ip6.c:314:16: error: ‘main’ is usually a function [-Werror=main]
   struct pbuf *main = pbuf_alloc(PBUF_RAW, sizeof(udp_hdr), PBUF_ROM);
                ^~~~
cc1: all warnings being treated as errors
2020-07-01 10:46:33 +02:00
Erik Ekman
e2ae25d158 ip6: Add test for fragmenting special packet
Reproduces bug #57374 (Found by Hiromasa Ito).
2020-06-30 01:27:32 +02:00
Erik Ekman
90b3e45c31 pbuf: Add pbuf_copy test operating on chained pbufs
With the payload lengths not matching between source and destination.

Also remove redundant allocation check in other test, it is now done
in the test setup code.
2020-06-29 19:45:18 +02:00
Erik Ekman
bc1da0bddb pbuf: Remove redundant tests that pbufs are freed
Already checked by lwip_check_ensure_no_alloc() in teardown function.
2020-06-24 01:00:12 +02:00
Erik Ekman
dd913250a9 icmp6: add test case for returning copy of chained pbufs
Currently fails due to bug #58553
2020-06-24 01:00:12 +02:00
Erik Ekman
b58e0061e0 unit: Improve message when the mem heap has usage
Show "mem heap still has 216 bytes allocated"
instead of "Assertion 'lwip_stats.mem.used == 0' failed"
2020-06-24 01:00:02 +02:00
Erik Ekman
ff67ce6e07 unit: Improve message when a memp pool has entries
Show "memp pool 'PBUF_REF/ROM' still has 2 entries allocated"
instead of "Assertion 'lwip_stats.memp[i]->used == 0' failed"
2020-06-22 00:51:09 +02:00
Erik Ekman
ea111c511e unit: Support check v0.13.0 and later
Function name no longer needs to be fed separately when adding tests.

Also fix collision of non-static net_test variables in dhcp and netif
tests.
2020-06-19 15:39:15 +02:00
Erik Ekman
9b1056ef0e fuzz: Fix compile error in simulated glibc rand
When using LWIP_RAND_FOR_FUZZ_SIMULATE_GLIBC:

fuzz_common.c: In function ‘lwip_fuzz_rand’:
fuzz_common.c:683:11: error: comparison of integer expressions of different signedness: ‘int’ and ‘long unsigned int’ [-Werror=sign-compare]
  683 |   if (idx >= sizeof(rand_nrs)/sizeof((rand_nrs)[0])) {
      |           ^~
cc1: all warnings being treated as errors
2020-06-19 13:26:58 +02:00
Simon Goldschmidt
e30d50710d fuzz: allow overriding LWIP_RAND for fuzz tests
Fuzz tests need reproducible code, so we need an "unsafe" version of
LWIP_RAND() in this case...

Also, to reproduce fuzz tests cases from Linux on Windows,
LWIP_RAND_FOR_FUZZ_SIMULATE_GLIBC provides the first 20 random numbers that
glibc would have...
2020-03-27 22:42:06 +01:00
Simon Goldschmidt
003f5cc12d fuzz: whitespace fixes 2020-03-09 21:50:50 +01:00
Simon Goldschmidt
b795784832 fuzz: make CC overridable from 'afl-gcc'
e.g. for debugging crashes
2020-03-09 21:40:30 +01:00