core dump issue

[azuryl]

valgrind --tool=memcheck --leak-check=full --show-reachable=yes --trace-children=yes ./darknet classifier valid squeezenet/imagenet1k.data squeezenet/squeezenet.cfg squeezenet/squeezeNetCaffe.weights ==983== Memcheck, a memory error detector ==983== Copyright (C) 2002-2015, and GNU GPL'd, by Julian Seward et al. ==983== Using Valgrind-3.11.0 and LibVEX; rerun with -h for copyright info ==983== Command: ./darknet classifier valid squeezenet/imagenet1k.data squeezenet/squeezenet.cfg squeezenet/squeezeNetCaffe.weights ==983== batch: Using default '1' layer filters size input output 0 make l.delta91718832 conv 64 3 x 3 / 2 227 x 227 x 3 -> 113 x 113 x 64 0.044 BFLOPs 1 max 3 x 3 / 2 113 x 113 x 64 -> 57 x 57 x 64 2 make l.delta91732912 conv 16 1 x 1 / 1 57 x 57 x 64 -> 57 x 57 x 16 0.007 BFLOPs 3 make l.delta91746096 conv 64 1 x 1 / 1 57 x 57 x 16 -> 57 x 57 x 64 0.007 BFLOPs 4 route 2 5 make l.delta91857872 conv 64 3 x 3 / 1 57 x 57 x 16 -> 57 x 57 x 64 0.060 BFLOPs 6 route 3 5 7 make l.delta91883248 conv 16 1 x 1 / 1 57 x 57 x 128 -> 57 x 57 x 16 0.013 BFLOPs 8 make l.delta91896432 conv 64 1 x 1 / 1 57 x 57 x 16 -> 57 x 57 x 64 0.007 BFLOPs 9 route 7 10 make l.delta92008208 conv 64 3 x 3 / 1 57 x 57 x 16 -> 57 x 57 x 64 0.060 BFLOPs 11 route 8 10 12 max 3 x 3 / 2 57 x 57 x 128 -> 29 x 29 x 128 13 make l.delta92058480 conv 32 1 x 1 / 1 29 x 29 x 128 -> 29 x 29 x 32 0.007 BFLOPs 14 make l.delta92109040 conv 128 1 x 1 / 1 29 x 29 x 32 -> 29 x 29 x 128 0.007 BFLOPs 15 route 13 16 make l.delta92553104 conv 128 3 x 3 / 1 29 x 29 x 32 -> 29 x 29 x 128 0.062 BFLOPs 17 route 14 16 18 make l.delta92652336 conv 32 1 x 1 / 1 29 x 29 x 256 -> 29 x 29 x 32 0.014 BFLOPs 19 make l.delta92702896 conv 128 1 x 1 / 1 29 x 29 x 32 -> 29 x 29 x 128 0.007 BFLOPs 20 route 18 21 make l.delta93146960 conv 128 3 x 3 / 1 29 x 29 x 32 -> 29 x 29 x 128 0.062 BFLOPs 22 route 19 21 23 max 3 x 3 / 2 29 x 29 x 256 -> 15 x 15 x 256 24 make l.delta93295664 conv 48 1 x 1 / 1 15 x 15 x 256 -> 15 x 15 x 48 0.006 BFLOPs 25 make l.delta93408176 conv 192 1 x 1 / 1 15 x 15 x 48 -> 15 x 15 x 192 0.004 BFLOPs 26 route 24 27 make l.delta94405712 conv 192 3 x 3 / 1 15 x 15 x 48 -> 15 x 15 x 192 0.037 BFLOPs 28 route 25 27 29 make l.delta94627952 conv 48 1 x 1 / 1 15 x 15 x 384 -> 15 x 15 x 48 0.008 BFLOPs 30 make l.delta94740464 conv 192 1 x 1 / 1 15 x 15 x 48 -> 15 x 15 x 192 0.004 BFLOPs 31 route 29 32 make l.delta95925184 conv 192 3 x 3 / 1 15 x 15 x 48 -> 15 x 15 x 192 0.037 BFLOPs 33 route 30 32 34 make l.delta95504544 conv 64 1 x 1 / 1 15 x 15 x 384 -> 15 x 15 x 64 0.011 BFLOPs 35 make l.delta95572448 conv 256 1 x 1 / 1 15 x 15 x 64 -> 15 x 15 x 256 0.007 BFLOPs 36 route 34 37 make l.delta95575040 conv 256 3 x 3 / 1 15 x 15 x 64 -> 15 x 15 x 256 0.066 BFLOPs 38 route 35 37 39 make l.delta95576096 conv 64 1 x 1 / 1 15 x 15 x 512 -> 15 x 15 x 64 0.015 BFLOPs 40 make l.delta95578400 conv 256 1 x 1 / 1 15 x 15 x 64 -> 15 x 15 x 256 0.007 BFLOPs 41 route 39 42 make l.delta95580992 conv 256 3 x 3 / 1 15 x 15 x 64 -> 15 x 15 x 256 0.066 BFLOPs 43 route 40 42 44 make l.delta95589536 conv 1000 1 x 1 / 1 15 x 15 x 512 -> 15 x 15 x1000 0.230 BFLOPs 45 avg 15 x 15 x1000 -> 1000 46 softmax 1000 47 cost 1000 Loading weights from squeezenet/squeezeNetCaffe.weights...Done!

!!!l.delta:91641312## ==983== Invalid write of size 4 ==983== at 0x4514CB: fill_cpu (blas.c:193) ==983== by 0x45D1E3: forward_network (network.c:203) ==983== by 0x45E50A: network_predict (network.c:505) ==983== by 0x413F05: validate_classifier_single (classifier.c:403) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== Address 0x57784b0 is 0 bytes after a block of size 0 alloc'd ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x427628: make_convolutional_layer (convolutional_layer.c:219) ==983== by 0x4615AA: parse_convolutional (parser.c:202) ==983== by 0x4648FB: parse_network_cfg (parser.c:777) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== Invalid write of size 4 ==983== at 0x4514CB: fill_cpu (blas.c:193) ==983== by 0x428149: forward_convolutional_layer (convolutional_layer.c:449) ==983== by 0x45D231: forward_network (network.c:205) ==983== by 0x45E50A: network_predict (network.c:505) ==983== by 0x413F05: validate_classifier_single (classifier.c:403) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== Address 0x5778470 is 0 bytes after a block of size 0 alloc'd ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x427603: make_convolutional_layer (convolutional_layer.c:218) ==983== by 0x4615AA: parse_convolutional (parser.c:202) ==983== by 0x4648FB: parse_network_cfg (parser.c:777) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== Invalid read of size 4 ==983== at 0x47CA8A: gemm_cpu (gemm.c:155) ==983== by 0x47C5BC: gemm (gemm.c:71) ==983== by 0x42838F: forward_convolutional_layer (convolutional_layer.c:473) ==983== by 0x45D231: forward_network (network.c:205) ==983== by 0x45E50A: network_predict (network.c:505) ==983== by 0x413F05: validate_classifier_single (classifier.c:403) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== Address 0x5778470 is 0 bytes after a block of size 0 alloc'd ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x427603: make_convolutional_layer (convolutional_layer.c:218) ==983== by 0x4615AA: parse_convolutional (parser.c:202) ==983== by 0x4648FB: parse_network_cfg (parser.c:777) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== Invalid write of size 4 ==983== at 0x47CA93: gemm_cpu (gemm.c:155) ==983== by 0x47C5BC: gemm (gemm.c:71) ==983== by 0x42838F: forward_convolutional_layer (convolutional_layer.c:473) ==983== by 0x45D231: forward_network (network.c:205) ==983== by 0x45E50A: network_predict (network.c:505) ==983== by 0x413F05: validate_classifier_single (classifier.c:403) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== Address 0x5778470 is 0 bytes after a block of size 0 alloc'd ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x427603: make_convolutional_layer (convolutional_layer.c:218) ==983== by 0x4615AA: parse_convolutional (parser.c:202) ==983== by 0x4648FB: parse_network_cfg (parser.c:777) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== Invalid read of size 4 ==983== at 0x47C66C: gemm_nn (gemm.c:85) ==983== by 0x47CAF4: gemm_cpu (gemm.c:159) ==983== by 0x47C5BC: gemm (gemm.c:71) ==983== by 0x42838F: forward_convolutional_layer (convolutional_layer.c:473) ==983== by 0x45D231: forward_network (network.c:205) ==983== by 0x45E50A: network_predict (network.c:505) ==983== by 0x413F05: validate_classifier_single (classifier.c:403) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== Address 0x5778470 is 0 bytes after a block of size 0 alloc'd ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x427603: make_convolutional_layer (convolutional_layer.c:218) ==983== by 0x4615AA: parse_convolutional (parser.c:202) ==983== by 0x4648FB: parse_network_cfg (parser.c:777) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== Invalid write of size 4 ==983== at 0x47C69E: gemm_nn (gemm.c:85) ==983== by 0x47CAF4: gemm_cpu (gemm.c:159) ==983== by 0x47C5BC: gemm (gemm.c:71) ==983== by 0x42838F: forward_convolutional_layer (convolutional_layer.c:473) ==983== by 0x45D231: forward_network (network.c:205) ==983== by 0x45E50A: network_predict (network.c:505) ==983== by 0x413F05: validate_classifier_single (classifier.c:403) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== Address 0x5778470 is 0 bytes after a block of size 0 alloc'd ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x427603: make_convolutional_layer (convolutional_layer.c:218) ==983== by 0x4615AA: parse_convolutional (parser.c:202) ==983== by 0x4648FB: parse_network_cfg (parser.c:777) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== Invalid read of size 4 ==983== at 0x427F30: add_bias (convolutional_layer.c:417) ==983== by 0x42843F: forward_convolutional_layer (convolutional_layer.c:480) ==983== by 0x45D231: forward_network (network.c:205) ==983== by 0x45E50A: network_predict (network.c:505) ==983== by 0x413F05: validate_classifier_single (classifier.c:403) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== Address 0x5778470 is 0 bytes after a block of size 0 alloc'd ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x427603: make_convolutional_layer (convolutional_layer.c:218) ==983== by 0x4615AA: parse_convolutional (parser.c:202) ==983== by 0x4648FB: parse_network_cfg (parser.c:777) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== Invalid write of size 4 ==983== at 0x427F50: add_bias (convolutional_layer.c:417) ==983== by 0x42843F: forward_convolutional_layer (convolutional_layer.c:480) ==983== by 0x45D231: forward_network (network.c:205) ==983== by 0x45E50A: network_predict (network.c:505) ==983== by 0x413F05: validate_classifier_single (classifier.c:403) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== Address 0x5778470 is 0 bytes after a block of size 0 alloc'd ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x427603: make_convolutional_layer (convolutional_layer.c:218) ==983== by 0x4615AA: parse_convolutional (parser.c:202) ==983== by 0x4648FB: parse_network_cfg (parser.c:777) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== Invalid read of size 4 ==983== at 0x44FF2C: activate_array (activations.c:99) ==983== by 0x42845C: forward_convolutional_layer (convolutional_layer.c:483) ==983== by 0x45D231: forward_network (network.c:205) ==983== by 0x45E50A: network_predict (network.c:505) ==983== by 0x413F05: validate_classifier_single (classifier.c:403) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== Address 0x5778470 is 0 bytes after a block of size 0 alloc'd ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x427603: make_convolutional_layer (convolutional_layer.c:218) ==983== by 0x4615AA: parse_convolutional (parser.c:202) ==983== by 0x4648FB: parse_network_cfg (parser.c:777) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== Invalid write of size 4 ==983== at 0x44FF44: activate_array (activations.c:99) ==983== by 0x42845C: forward_convolutional_layer (convolutional_layer.c:483) ==983== by 0x45D231: forward_network (network.c:205) ==983== by 0x45E50A: network_predict (network.c:505) ==983== by 0x413F05: validate_classifier_single (classifier.c:403) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== Address 0x5778470 is 0 bytes after a block of size 0 alloc'd ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x427603: make_convolutional_layer (convolutional_layer.c:218) ==983== by 0x4615AA: parse_convolutional (parser.c:202) ==983== by 0x4648FB: parse_network_cfg (parser.c:777) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== +\ufffd\ufffd?\ufffd\u03a3? !!!l.delta:91642496## ==983== Invalid read of size 4 ==983== at 0x47ED5F: forward_maxpool_layer (maxpool_layer.c:103) ==983== by 0x45D231: forward_network (network.c:205) ==983== by 0x45E50A: network_predict (network.c:505) ==983== by 0x413F05: validate_classifier_single (classifier.c:403) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== Address 0x5778470 is 0 bytes after a block of size 0 alloc'd ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x427603: make_convolutional_layer (convolutional_layer.c:218) ==983== by 0x4615AA: parse_convolutional (parser.c:202) ==983== by 0x4648FB: parse_network_cfg (parser.c:777) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== Invalid write of size 4 ==983== at 0x47EDE3: forward_maxpool_layer (maxpool_layer.c:108) ==983== by 0x45D231: forward_network (network.c:205) ==983== by 0x45E50A: network_predict (network.c:505) ==983== by 0x413F05: validate_classifier_single (classifier.c:403) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== Address 0x5778970 is 0 bytes after a block of size 0 alloc'd ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x47E9AC: make_maxpool_layer (maxpool_layer.c:39) ==983== by 0x463098: parse_maxpool (parser.c:496) ==983== by 0x465125: parse_network_cfg (parser.c:818) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== Invalid write of size 4 ==983== at 0x47EDFE: forward_maxpool_layer (maxpool_layer.c:109) ==983== by 0x45D231: forward_network (network.c:205) ==983== by 0x45E50A: network_predict (network.c:505) ==983== by 0x413F05: validate_classifier_single (classifier.c:403) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== Address 0x5778930 is 0 bytes after a block of size 0 alloc'd ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x47E990: make_maxpool_layer (maxpool_layer.c:38) ==983== by 0x463098: parse_maxpool (parser.c:496) ==983== by 0x465125: parse_network_cfg (parser.c:818) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ,l.delta !!!l.delta:91643680## ==983== Invalid read of size 4 ==983== at 0x47C68F: gemm_nn (gemm.c:85) ==983== by 0x47CAF4: gemm_cpu (gemm.c:159) ==983== by 0x47C5BC: gemm (gemm.c:71) ==983== by 0x42838F: forward_convolutional_layer (convolutional_layer.c:473) ==983== by 0x45D231: forward_network (network.c:205) ==983== by 0x45E50A: network_predict (network.c:505) ==983== by 0x413F05: validate_classifier_single (classifier.c:403) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== Address 0x5778970 is 0 bytes after a block of size 0 alloc'd ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x47E9AC: make_maxpool_layer (maxpool_layer.c:39) ==983== by 0x463098: parse_maxpool (parser.c:496) ==983== by 0x465125: parse_network_cfg (parser.c:818) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ,l.delta !!!l.delta:91644864## ,l.delta !!!l.delta:91646048## ==983== Invalid read of size 4 ==983== at 0x45175D: copy_cpu (blas.c:229) ==983== by 0x469734: forward_route_layer (route_layer.c:83) ==983== by 0x45D231: forward_network (network.c:205) ==983== by 0x45E50A: network_predict (network.c:505) ==983== by 0x413F05: validate_classifier_single (classifier.c:403) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== Address 0x0 is not stack'd, malloc'd or (recently) free'd ==983== ==983== ==983== Process terminating with default action of signal 11 (SIGSEGV) ==983== Access not within mapped region at address 0x0 ==983== at 0x45175D: copy_cpu (blas.c:229) ==983== by 0x469734: forward_route_layer (route_layer.c:83) ==983== by 0x45D231: forward_network (network.c:205) ==983== by 0x45E50A: network_predict (network.c:505) ==983== by 0x413F05: validate_classifier_single (classifier.c:403) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== If you believe this happened as a result of a stack ==983== overflow in your program's main thread (unlikely but ==983== possible), you can try to increase the size of the ==983== main thread stack using the --main-stacksize= flag. ==983== The main thread stack size used in this run was 8388608. ,l.delta==983== ==983== HEAP SUMMARY: ==983== in use at exit: 56,969,076 bytes in 72,116 blocks ==983== total heap usage: 144,812 allocs, 72,696 frees, 62,783,419 bytes allocated ==983== ==983== 0 bytes in 1 blocks are still reachable in loss record 1 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x46CBA1: make_avgpool_layer (avgpool_layer.c:20) ==983== by 0x463175: parse_avgpool (parser.c:509) ==983== by 0x4651F7: parse_network_cfg (parser.c:822) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 0 bytes in 1 blocks are still reachable in loss record 2 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x46CBBD: make_avgpool_layer (avgpool_layer.c:21) ==983== by 0x463175: parse_avgpool (parser.c:509) ==983== by 0x4651F7: parse_network_cfg (parser.c:822) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 0 bytes in 1 blocks are still reachable in loss record 3 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x47EFD9: make_softmax_layer (softmax_layer.c:21) ==983== by 0x461C25: parse_softmax (parser.c:273) ==983== by 0x464FD8: parse_network_cfg (parser.c:811) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 0 bytes in 1 blocks are still reachable in loss record 4 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x47EFFC: make_softmax_layer (softmax_layer.c:22) ==983== by 0x461C25: parse_softmax (parser.c:273) ==983== by 0x464FD8: parse_network_cfg (parser.c:811) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 0 bytes in 1 blocks are still reachable in loss record 5 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x47F01F: make_softmax_layer (softmax_layer.c:23) ==983== by 0x461C25: parse_softmax (parser.c:273) ==983== by 0x464FD8: parse_network_cfg (parser.c:811) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 0 bytes in 1 blocks are still reachable in loss record 6 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x4606EB: make_cost_layer (cost_layer.c:52) ==983== by 0x462B0C: parse_cost (parser.c:434) ==983== by 0x464DCB: parse_network_cfg (parser.c:801) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 0 bytes in 1 blocks are still reachable in loss record 7 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x46070E: make_cost_layer (cost_layer.c:53) ==983== by 0x462B0C: parse_cost (parser.c:434) ==983== by 0x464DCB: parse_network_cfg (parser.c:801) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 0 bytes in 1 blocks are still reachable in loss record 8 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x4657AD: parse_network_cfg (parser.c:869) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 0 bytes in 1 blocks are still reachable in loss record 9 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x4657E7: parse_network_cfg (parser.c:870) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 0 bytes in 3 blocks are still reachable in loss record 10 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x47E990: make_maxpool_layer (maxpool_layer.c:38) ==983== by 0x463098: parse_maxpool (parser.c:496) ==983== by 0x465125: parse_network_cfg (parser.c:818) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 0 bytes in 3 blocks are still reachable in loss record 11 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x47E9AC: make_maxpool_layer (maxpool_layer.c:39) ==983== by 0x463098: parse_maxpool (parser.c:496) ==983== by 0x465125: parse_network_cfg (parser.c:818) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 0 bytes in 3 blocks are still reachable in loss record 12 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x47E9C8: make_maxpool_layer (maxpool_layer.c:40) ==983== by 0x463098: parse_maxpool (parser.c:496) ==983== by 0x465125: parse_network_cfg (parser.c:818) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 0 bytes in 16 blocks are still reachable in loss record 13 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x4692F7: make_route_layer (route_layer.c:25) ==983== by 0x463B81: parse_route (parser.c:623) ==983== by 0x465267: parse_network_cfg (parser.c:824) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 0 bytes in 16 blocks are still reachable in loss record 14 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x46931A: make_route_layer (route_layer.c:26) ==983== by 0x463B81: parse_route (parser.c:623) ==983== by 0x465267: parse_network_cfg (parser.c:824) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 0 bytes in 26 blocks are still reachable in loss record 15 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x427603: make_convolutional_layer (convolutional_layer.c:218) ==983== by 0x4615AA: parse_convolutional (parser.c:202) ==983== by 0x4648FB: parse_network_cfg (parser.c:777) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 0 bytes in 26 blocks are still reachable in loss record 16 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x427628: make_convolutional_layer (convolutional_layer.c:219) ==983== by 0x4615AA: parse_convolutional (parser.c:202) ==983== by 0x4648FB: parse_network_cfg (parser.c:777) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 4 bytes in 1 blocks are still reachable in loss record 17 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x424DEA: read_intlist (utils.c:52) ==983== by 0x4160D2: run_classifier (classifier.c:1098) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 4 bytes in 1 blocks are still reachable in loss record 18 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x45D0E2: make_network (network.c:183) ==983== by 0x4646E5: parse_network_cfg (parser.c:747) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 4 bytes in 1 blocks are still reachable in loss record 19 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x45D0FC: make_network (network.c:184) ==983== by 0x4646E5: parse_network_cfg (parser.c:747) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 4 bytes in 1 blocks are still reachable in loss record 20 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x47F035: make_softmax_layer (softmax_layer.c:24) ==983== by 0x461C25: parse_softmax (parser.c:273) ==983== by 0x464FD8: parse_network_cfg (parser.c:811) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 4 bytes in 1 blocks are still reachable in loss record 21 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x460724: make_cost_layer (cost_layer.c:54) ==983== by 0x462B0C: parse_cost (parser.c:434) ==983== by 0x464DCB: parse_network_cfg (parser.c:801) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 8 bytes in 1 blocks are still reachable in loss record 22 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x45D0C8: make_network (network.c:182) ==983== by 0x4646E5: parse_network_cfg (parser.c:747) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 20 bytes in 1 blocks are still reachable in loss record 23 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x413DEE: validate_classifier_single (classifier.c:386) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 24 bytes in 1 blocks are still reachable in loss record 24 of 42 ==983== at 0x4C2DB8F: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x428EE1: make_list (list.c:7) ==983== by 0x4674DD: read_data_cfg (option_list.c:13) ==983== by 0x413CA5: validate_classifier_single (classifier.c:368) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 96 bytes in 16 blocks are still reachable in loss record 25 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x463A4D: parse_route (parser.c:612) ==983== by 0x465267: parse_network_cfg (parser.c:824) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 96 bytes in 16 blocks are still reachable in loss record 26 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x463A69: parse_route (parser.c:613) ==983== by 0x465267: parse_network_cfg (parser.c:824) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 168 bytes in 7 blocks are still reachable in loss record 27 of 42 ==983== at 0x4C2DB8F: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x467744: option_insert (option_list.c:72) ==983== by 0x46771F: read_option (option_list.c:66) ==983== by 0x46752C: read_data_cfg (option_list.c:24) ==983== by 0x413CA5: validate_classifier_single (classifier.c:368) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 168 bytes in 7 blocks are still reachable in loss record 28 of 42 ==983== at 0x4C2DB8F: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x428FAC: list_insert (list.c:42) ==983== by 0x46777D: option_insert (option_list.c:76) ==983== by 0x46771F: read_option (option_list.c:66) ==983== by 0x46752C: read_data_cfg (option_list.c:24) ==983== by 0x413CA5: validate_classifier_single (classifier.c:368) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 272 bytes in 1 blocks are still reachable in loss record 29 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x45D08C: make_network (network.c:179) ==983== by 0x4646E5: parse_network_cfg (parser.c:747) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 3,584 bytes in 7 blocks are still reachable in loss record 30 of 42 ==983== at 0x4C2DB8F: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x425C26: fgetl (utils.c:339) ==983== by 0x46756E: read_data_cfg (option_list.c:14) ==983== by 0x413CA5: validate_classifier_single (classifier.c:368) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 15,776 bytes in 26 blocks are still reachable in loss record 31 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x427440: make_convolutional_layer (convolutional_layer.c:198) ==983== by 0x4615AA: parse_convolutional (parser.c:202) ==983== by 0x4648FB: parse_network_cfg (parser.c:777) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 15,776 bytes in 26 blocks are still reachable in loss record 32 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x42745C: make_convolutional_layer (convolutional_layer.c:199) ==983== by 0x4615AA: parse_convolutional (parser.c:202) ==983== by 0x4648FB: parse_network_cfg (parser.c:777) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 56,832 bytes in 1 blocks are still reachable in loss record 33 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x45D0AE: make_network (network.c:181) ==983== by 0x4646E5: parse_network_cfg (parser.c:747) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 174,736 bytes in 1 blocks are still reachable in loss record 34 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x4290EF: list_to_array (list.c:84) ==983== by 0x455ABF: get_labels (data.c:660) ==983== by 0x413D85: validate_classifier_single (classifier.c:377) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 400,000 bytes in 1 blocks are still reachable in loss record 35 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x4290EF: list_to_array (list.c:84) ==983== by 0x413DA5: validate_classifier_single (classifier.c:380) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 618,348 bytes in 1 blocks are still reachable in loss record 36 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x44B1E3: make_image (image.c:759) ==983== by 0x44E2E7: resize_image (image.c:1353) ==983== by 0x44B555: center_crop_image (image.c:800) ==983== by 0x413EEB: validate_classifier_single (classifier.c:398) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 1,871,424 bytes in 1 blocks are still reachable in loss record 37 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x465816: parse_network_cfg (parser.c:885) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 2,250,000 bytes in 1 blocks are still reachable in loss record 38 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x44B1E3: make_image (image.c:759) ==983== by 0x44EA7B: load_image_stb (image.c:1456) ==983== by 0x44EBCD: load_image (image.c:1475) ==983== by 0x44ECB2: load_image_color (image.c:1488) ==983== by 0x413EB6: validate_classifier_single (classifier.c:397) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 4,926,208 bytes in 26 blocks are still reachable in loss record 39 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x427424: make_convolutional_layer (convolutional_layer.c:196) ==983== by 0x4615AA: parse_convolutional (parser.c:202) ==983== by 0x4648FB: parse_network_cfg (parser.c:777) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 9,852,416 bytes in 26 blocks are still reachable in loss record 40 of 42 ==983== at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x4273F5: make_convolutional_layer (convolutional_layer.c:195) ==983== by 0x4615AA: parse_convolutional (parser.c:202) ==983== by 0x4648FB: parse_network_cfg (parser.c:777) ==983== by 0x45CAA8: load_network (network.c:55) ==983== by 0x413C6A: validate_classifier_single (classifier.c:364) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 11,183,104 bytes in 21,842 blocks are still reachable in loss record 41 of 42 ==983== at 0x4C2DB8F: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x425C26: fgetl (utils.c:339) ==983== by 0x45216C: get_paths (data.c:18) ==983== by 0x455AAF: get_labels (data.c:659) ==983== by 0x413D85: validate_classifier_single (classifier.c:377) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== 25,600,000 bytes in 50,000 blocks are still reachable in loss record 42 of 42 ==983== at 0x4C2DB8F: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==983== by 0x425C26: fgetl (utils.c:339) ==983== by 0x45216C: get_paths (data.c:18) ==983== by 0x413D95: validate_classifier_single (classifier.c:378) ==983== by 0x416461: run_classifier (classifier.c:1120) ==983== by 0x424512: main (darknet.c:453) ==983== ==983== LEAK SUMMARY: ==983== definitely lost: 0 bytes in 0 blocks ==983== indirectly lost: 0 bytes in 0 blocks ==983== possibly lost: 0 bytes in 0 blocks ==983== still reachable: 56,969,076 bytes in 72,116 blocks ==983== suppressed: 0 bytes in 0 blocks ==983== ==983== For counts of detected and suppressed errors, rerun with: -v ==983== ERROR SUMMARY: 360420 errors from 15 contexts (suppressed: 0 from 0) Segmentation fault (core dumped)

[azuryl]

l.delta address which alloced in make convolutional_layer was changed in network forward layer

[azuryl]

i fixed if batch=0 in make_convolutional_layer the program will core dump

make_convolutional_layer (batch=0, h=227, w=227, c=3, n=64, groups=1, size=3, stride=2, padding=0, activation=RELU, batch_normalize=0, binary=0, xnor=0, adam=0) at ./src/convolutional_layer.c:179

[azuryl]

but in fact I set batch=1 in cfg

[net] batch=1 subdivisions=2 height=227 width=227 channels=3 momentum=0.9 decay=0.0005

learning_rate=0.04 policy=poly power=1 max_batches=200000

[crop]

crop_width=227

crop_height=227

flip=0

angle=0

saturation=1.5

exposure=1.5

SqueezeNet: conv1

[convolutional] batch_normalize=0 filters=64 size=3 stride=2 activation=relu

SqueezeNet: pool1

[maxpool] size=3 stride=2

SqueezeNet: fire2/squeeze1x1

[convolutional] batch_normalize=0 filters=16 size=1 stride=1 activation=relu

SqueezeNet: fire2/expand1x1

[convolutional] batch_normalize=0 filters=64 size=1 stride=1 activation=relu

SqueezeNet: fire2/route

[route] layers=-2

SqueezeNet: fire2/expa

[azuryl]

so if batch=1 subdivisions=2 it must core dump

[dijkstraj]

I'm also getting a segmentation fault at blas.c:193 when running "rnn train" with the default rnn.train.cfg file. Is there any way of fixing this by changing the configuration?