valgrind now reports only 4 memory leaks in test_eressea. That's pretty great, but they are all pointing to one function: cb_insert. I guess it's a bug in the critbit submodule, but would have to push on it a bit and reproduce it in a tighter test.
HEAP SUMMARY:
in use at exit: 5,663 bytes in 142 blocks
total heap usage: 297,153 allocs, 297,011 frees, 21,491,138 bytes allocated
32 bytes in 1 blocks are definitely lost in loss record 84 of 142
at 0x4C28BED: malloc (vg_replace_malloc.c:263)
by 0x518563: make_internal_node (critbit.c:69)
by 0x51883E: cb_insert (critbit.c:135)
by 0x4DFE7A: set_param (config.c:592)
by 0x4E0EB4: config_set (config.c:1050)
by 0x43921B: test_rules (otherfaction.test.c:29)
by 0x515C16: CuTestRun (CuTest.c:143)
by 0x51624F: CuSuiteRun (CuTest.c:289)
by 0x41B56C: RunTests (test_eressea.c:45)
by 0x41C909: RunAllTests (test_eressea.c:144)
by 0x41CA3D: main (test_eressea.c:169)
32 bytes in 1 blocks are definitely lost in loss record 85 of 142
at 0x4C28BED: malloc (vg_replace_malloc.c:263)
by 0x518563: make_internal_node (critbit.c:69)
by 0x51883E: cb_insert (critbit.c:135)
by 0x4DFE7A: set_param (config.c:592)
by 0x4E0EB4: config_set (config.c:1050)
by 0x42ABC1: test_peasant_luck_effect (laws.test.c:743)
by 0x515C16: CuTestRun (CuTest.c:143)
by 0x51624F: CuSuiteRun (CuTest.c:289)
by 0x41B56C: RunTests (test_eressea.c:45)
by 0x41C909: RunAllTests (test_eressea.c:144)
by 0x41CA3D: main (test_eressea.c:169)
32 bytes in 1 blocks are definitely lost in loss record 86 of 142
at 0x4C28BED: malloc (vg_replace_malloc.c:263)
by 0x518563: make_internal_node (critbit.c:69)
by 0x51883E: cb_insert (critbit.c:135)
by 0x4DFE7A: set_param (config.c:592)
by 0x442C5E: test_get_set_param (config.test.c:128)
by 0x515C16: CuTestRun (CuTest.c:143)
by 0x51624F: CuSuiteRun (CuTest.c:289)
by 0x41B56C: RunTests (test_eressea.c:45)
by 0x41C909: RunAllTests (test_eressea.c:144)
by 0x41CA3D: main (test_eressea.c:169)
32 bytes in 1 blocks are definitely lost in loss record 87 of 142
at 0x4C28BED: malloc (vg_replace_malloc.c:263)
by 0x518563: make_internal_node (critbit.c:69)
by 0x51883E: cb_insert (critbit.c:135)
by 0x4DFE7A: set_param (config.c:592)
by 0x442CE4: test_get_set_param (config.test.c:131)
by 0x515C16: CuTestRun (CuTest.c:143)
by 0x51624F: CuSuiteRun (CuTest.c:289)
by 0x41B56C: RunTests (test_eressea.c:45)
by 0x41C909: RunAllTests (test_eressea.c:144)
by 0x41CA3D: main (test_eressea.c:169)
LEAK SUMMARY:
definitely lost: 128 bytes in 4 blocks
indirectly lost: 0 bytes in 0 blocks
possibly lost: 0 bytes in 0 blocks
still reachable: 5,535 bytes in 138 blocks
suppressed: 0 bytes in 0 blocks
Reachable blocks (those to which a pointer was found) are not shown.
To see them, rerun with: --leak-check=full --show-reachable=yes
For counts of detected and suppressed errors, rerun with: -v
ERROR SUMMARY: 4 errors from 4 contexts (suppressed: 6 from 6)
valgrind now reports only 4 memory leaks in test_eressea. That's pretty great, but they are all pointing to one function: cb_insert. I guess it's a bug in the critbit submodule, but would have to push on it a bit and reproduce it in a tighter test.