Hello vpp-dev, I am trying to pin down an elusive memory leak and from previous messages its understood that memory-trace is the best tool for the job. From the docs it uses two commands:
$ vppctl memory-trace on main-heap $ vppctl show memory-trace on main-heap Thread 0 vpp_main base 0x7fffb6422000, size 1g, locked, unmap-on-destroy, name 'main heap' page stats: page-size 4K, total 262144, mapped 30343, not-mapped 231801 numa 0: 30343 pages, 118.53m bytes total: 1023.99M, used: 115.49M, free: 908.50M, trimmable: 908.48M free chunks 451 free fastbin blks 0 max total allocated 1023.99M Bytes Count Sample Traceback 31457440 1 0x7fffbb31ad00 clib_mem_alloc_aligned_at_offset + 0x80 clib_mem_alloc_aligned + 0x26 alloc_aligned_8_8 + 0xe1 clib_bihash_instantiate_8_8 + 0x76 clib_bihash_init2_8_8 + 0x2ec clib_bihash_init_8_8 + 0x6a l2fib_table_init + 0x54 set_int_l2_mode + 0x89 int_l3 + 0xb4 vlib_cli_dispatch_sub_commands + 0xeee vlib_cli_dispatch_sub_commands + 0xc62 vlib_cli_dispatch_sub_commands + 0xc62 266768 5222 0x7fffbd79f978 clib_mem_alloc_aligned_at_offset + 0x80 vec_resize_allocate_memory + 0xa8 _vec_resize_inline + 0x240 In the output what does Bytes, Count, Sample and Traceback denote? To check if there is a memory leak, do I have to leave the memory-trace on and get another output after say 24 hours? How do I compare the two outputs? Sorry for too many questions, but this aspect has been poorly documented and if documented in this thread, will help others in same situation in future. best, X.
-=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#21603): https://lists.fd.io/g/vpp-dev/message/21603 Mute This Topic: https://lists.fd.io/mt/92144131/21656 Group Owner: vpp-dev+ow...@lists.fd.io Unsubscribe: https://lists.fd.io/g/vpp-dev/leave/1480452/21656/631435203/xyzzy [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
