Re: [E-devel] valgrind show error when debug efl demo
On Fri, 6 Dec 2019 10:16:41 +1000 Imran Syed said: > 'still reachable' and 'in use at exit' are generally fine. The ones where > it says 'definitely lost' and 'indirectly lost' are the only ones you > should investigate. As this usually means you've lost a reference (pointer) > to the object. not necessarily. like some of those may be instances created only once ever but the pointer was stored in an allocated object. but they are not related to Jing's issue. I've taken his samples and can't spot a leak no matter how long I run it for and keep objects cycling. > On Thu, Dec 5, 2019 at 11:45 PM Carsten Haitzler > wrote: > > > On Wed, 27 Nov 2019 17:00:13 +0800 (CST) Jing said: > > > > > Hi all, > > > I use valgrind to check memory on efl demo, but valgrind show error as > > shown > > > below, attached are the efl demo codes, please help to advise where went > > > wrong ? > > > > Don't worry about the below. They are the kind of "one off" allocations > > that > > are allocated just once for the life of a process and just aren't being > > explicitly freed when the process exits. All memory for a process is > > cleaned up > > by the host kernel anyway so this isn't a problem really other than > > creating > > noise. > > > > > sudo valgrind --tool=memcheck --leak-check=full ./edje-anchors > > > ==37097== Memcheck, a memory error detector > > > ==37097== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et al. > > > ==37097== Using Valgrind-3.15.0 and LibVEX; rerun with -h for copyright > > info > > > ==37097== Command: ./edje-anchors > > > ==37097== > > > key:Escape > > > ==37097== > > > ==37097== HEAP SUMMARY: > > > ==37097== in use at exit: 246,790 bytes in 2,198 blocks > > > ==37097== total heap usage: 17,746 allocs, 15,548 frees, 6,874,546 > > bytes > > > allocated ==37097== > > > ==37097== 10 bytes in 1 blocks are definitely lost in loss record 10 of > > 238 > > > ==37097==at 0x4C2DE96: malloc (vg_replace_malloc.c:309) > > > ==37097==by 0x5D07489: strdup (strdup.c:42) > > > ==37097==by 0x72A8CBE: eina_vpath_resolve (eina_vpath.c:285) > > > ==37097==by 0x90128D9: _efreet_efreet_app_interface_set (efreet.c:49) > > > ==37097==by 0x9012F27: efreet_init (efreet.c:149) > > > ==37097==by 0x5803F36: edje_init (edje_main.c:78) > > > ==37097==by 0x400EED: main > > > (in /home/lg/GSC/efl-1.21.1/src/examples/edje/edje-anchors) ==37097== > > > ==37097== 32 bytes in 1 blocks are definitely lost in loss record 49 of > > 238 > > > ==37097==at 0x4C2FFAC: calloc (vg_replace_malloc.c:762) > > > ==37097==by 0x4FD5B28: generic_cache_new > > (evas_common_generic_cache.c:7) > > > ==37097==by 0x4FE9AE6: eng_engine_new (evas_engine.c:3880) > > > ==37097==by 0x4EAC544: evas_output_method_set (evas_main.c:1224) > > > ==37097==by 0x12C3F004: ecore_evas_software_x11_new_internal > > > (ecore_evas_x.c:4153) ==37097==by 0x554F606: > > ecore_evas_software_x11_new > > > (ecore_evas.c:3831) ==37097==by 0x554F7EA: > > > _ecore_evas_constructor_software_x11 (ecore_evas.c:780) ==37097==by > > > 0x55499EF: _ecore_evas_new_auto_discover (ecore_evas.c:1021) ==37097== > > by > > > 0x55499EF: ecore_evas_new (ecore_evas.c:1046) ==37097==by 0x400F12: > > main > > > (in /home/lg/GSC/efl-1.21.1/src/examples/edje/edje-anchors) ==37097== > > > ==37097== 82 bytes in 1 blocks are definitely lost in loss record 154 of > > 238 > > > ==37097==at 0x4C2DE96: malloc (vg_replace_malloc.c:309) > > > ==37097==by 0x727B586: eina_module_new (eina_module.c:272) > > > ==37097==by 0x4F530F6: evas_module_find_type.part.3 > > (evas_module.c:601) > > > ==37097==by 0x4EA9E52: evas_render_method_lookup (evas_main.c:644) > > > ==37097==by 0x12C3EE2D: ecore_evas_software_x11_new_internal > > > (ecore_evas_x.c:4086) ==37097==by 0x554F606: > > ecore_evas_software_x11_new > > > (ecore_evas.c:3831) ==37097==by 0x554F7EA: > > > _ecore_evas_constructor_software_x11 (ecore_evas.c:780) ==37097==by > > > 0x55499EF: _ecore_evas_new_auto_discover (ecore_evas.c:1021) ==37097== > > by > > > 0x55499EF: ecore_evas_new (ecore_evas.c:1046) ==37097==by 0x400F12: > > main > > > (in /home/lg/GSC/efl-1.21.1/src/examples/edje/edje-anchors) ==37097== > > > ==37097== 232 (192 direct, 40 indirect) bytes in 1 blocks are definitely > > lost > > > in loss record 190 of 238 ==37097==at 0x4C2FFAC: calloc > > > (vg_replace_malloc.c:762) ==37097==by 0x703431E: > > _efl_add_internal_start > > > (eo.c:896) ==37097==by 0x4E8F279: evas_device_add_full > > (evas_device.c:190) > > > ==37097==by 0x4EAC655: evas_output_method_set (evas_main.c:1251) > > > ==37097==by 0x12C3F004: ecore_evas_software_x11_new_internal > > > (ecore_evas_x.c:4153) ==37097==by 0x554F606: > > ecore_evas_software_x11_new > > > (ecore_evas.c:3831) ==37097==by 0x554F7EA: > > > _ecore_evas_constructor_software_x11 (ecore_evas.c:780) ==37097==by > > > 0x55499EF: _ecore_evas_new_auto_discover
Re: [E-devel] valgrind show error when debug efl demo
'still reachable' and 'in use at exit' are generally fine. The ones where it says 'definitely lost' and 'indirectly lost' are the only ones you should investigate. As this usually means you've lost a reference (pointer) to the object. On Thu, Dec 5, 2019 at 11:45 PM Carsten Haitzler wrote: > On Wed, 27 Nov 2019 17:00:13 +0800 (CST) Jing said: > > > Hi all, > > I use valgrind to check memory on efl demo, but valgrind show error as > shown > > below, attached are the efl demo codes, please help to advise where went > > wrong ? > > Don't worry about the below. They are the kind of "one off" allocations > that > are allocated just once for the life of a process and just aren't being > explicitly freed when the process exits. All memory for a process is > cleaned up > by the host kernel anyway so this isn't a problem really other than > creating > noise. > > > sudo valgrind --tool=memcheck --leak-check=full ./edje-anchors > > ==37097== Memcheck, a memory error detector > > ==37097== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et al. > > ==37097== Using Valgrind-3.15.0 and LibVEX; rerun with -h for copyright > info > > ==37097== Command: ./edje-anchors > > ==37097== > > key:Escape > > ==37097== > > ==37097== HEAP SUMMARY: > > ==37097== in use at exit: 246,790 bytes in 2,198 blocks > > ==37097== total heap usage: 17,746 allocs, 15,548 frees, 6,874,546 > bytes > > allocated ==37097== > > ==37097== 10 bytes in 1 blocks are definitely lost in loss record 10 of > 238 > > ==37097==at 0x4C2DE96: malloc (vg_replace_malloc.c:309) > > ==37097==by 0x5D07489: strdup (strdup.c:42) > > ==37097==by 0x72A8CBE: eina_vpath_resolve (eina_vpath.c:285) > > ==37097==by 0x90128D9: _efreet_efreet_app_interface_set (efreet.c:49) > > ==37097==by 0x9012F27: efreet_init (efreet.c:149) > > ==37097==by 0x5803F36: edje_init (edje_main.c:78) > > ==37097==by 0x400EED: main > > (in /home/lg/GSC/efl-1.21.1/src/examples/edje/edje-anchors) ==37097== > > ==37097== 32 bytes in 1 blocks are definitely lost in loss record 49 of > 238 > > ==37097==at 0x4C2FFAC: calloc (vg_replace_malloc.c:762) > > ==37097==by 0x4FD5B28: generic_cache_new > (evas_common_generic_cache.c:7) > > ==37097==by 0x4FE9AE6: eng_engine_new (evas_engine.c:3880) > > ==37097==by 0x4EAC544: evas_output_method_set (evas_main.c:1224) > > ==37097==by 0x12C3F004: ecore_evas_software_x11_new_internal > > (ecore_evas_x.c:4153) ==37097==by 0x554F606: > ecore_evas_software_x11_new > > (ecore_evas.c:3831) ==37097==by 0x554F7EA: > > _ecore_evas_constructor_software_x11 (ecore_evas.c:780) ==37097==by > > 0x55499EF: _ecore_evas_new_auto_discover (ecore_evas.c:1021) ==37097== > by > > 0x55499EF: ecore_evas_new (ecore_evas.c:1046) ==37097==by 0x400F12: > main > > (in /home/lg/GSC/efl-1.21.1/src/examples/edje/edje-anchors) ==37097== > > ==37097== 82 bytes in 1 blocks are definitely lost in loss record 154 of > 238 > > ==37097==at 0x4C2DE96: malloc (vg_replace_malloc.c:309) > > ==37097==by 0x727B586: eina_module_new (eina_module.c:272) > > ==37097==by 0x4F530F6: evas_module_find_type.part.3 > (evas_module.c:601) > > ==37097==by 0x4EA9E52: evas_render_method_lookup (evas_main.c:644) > > ==37097==by 0x12C3EE2D: ecore_evas_software_x11_new_internal > > (ecore_evas_x.c:4086) ==37097==by 0x554F606: > ecore_evas_software_x11_new > > (ecore_evas.c:3831) ==37097==by 0x554F7EA: > > _ecore_evas_constructor_software_x11 (ecore_evas.c:780) ==37097==by > > 0x55499EF: _ecore_evas_new_auto_discover (ecore_evas.c:1021) ==37097== > by > > 0x55499EF: ecore_evas_new (ecore_evas.c:1046) ==37097==by 0x400F12: > main > > (in /home/lg/GSC/efl-1.21.1/src/examples/edje/edje-anchors) ==37097== > > ==37097== 232 (192 direct, 40 indirect) bytes in 1 blocks are definitely > lost > > in loss record 190 of 238 ==37097==at 0x4C2FFAC: calloc > > (vg_replace_malloc.c:762) ==37097==by 0x703431E: > _efl_add_internal_start > > (eo.c:896) ==37097==by 0x4E8F279: evas_device_add_full > (evas_device.c:190) > > ==37097==by 0x4EAC655: evas_output_method_set (evas_main.c:1251) > > ==37097==by 0x12C3F004: ecore_evas_software_x11_new_internal > > (ecore_evas_x.c:4153) ==37097==by 0x554F606: > ecore_evas_software_x11_new > > (ecore_evas.c:3831) ==37097==by 0x554F7EA: > > _ecore_evas_constructor_software_x11 (ecore_evas.c:780) ==37097==by > > 0x55499EF: _ecore_evas_new_auto_discover (ecore_evas.c:1021) ==37097== > by > > 0x55499EF: ecore_evas_new (ecore_evas.c:1046) ==37097==by 0x400F12: > main > > (in /home/lg/GSC/efl-1.21.1/src/examples/edje/edje-anchors) ==37097== > > ==37097== 232 (192 direct, 40 indirect) bytes in 1 blocks are definitely > lost > > in loss record 191 of 238 ==37097==at 0x4C2FFAC: calloc > > (vg_replace_malloc.c:762) ==37097==by 0x703431E: > _efl_add_internal_start > > (eo.c:896) ==37097==by 0x4E8F279: evas_device_add_full > (evas_device.c:190) > > ==37
Re: [E-devel] valgrind show error when debug efl demo
On Wed, 27 Nov 2019 17:00:13 +0800 (CST) Jing said: > Hi all, > I use valgrind to check memory on efl demo, but valgrind show error as shown > below, attached are the efl demo codes, please help to advise where went > wrong ? Don't worry about the below. They are the kind of "one off" allocations that are allocated just once for the life of a process and just aren't being explicitly freed when the process exits. All memory for a process is cleaned up by the host kernel anyway so this isn't a problem really other than creating noise. > sudo valgrind --tool=memcheck --leak-check=full ./edje-anchors > ==37097== Memcheck, a memory error detector > ==37097== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et al. > ==37097== Using Valgrind-3.15.0 and LibVEX; rerun with -h for copyright info > ==37097== Command: ./edje-anchors > ==37097== > key:Escape > ==37097== > ==37097== HEAP SUMMARY: > ==37097== in use at exit: 246,790 bytes in 2,198 blocks > ==37097== total heap usage: 17,746 allocs, 15,548 frees, 6,874,546 bytes > allocated ==37097== > ==37097== 10 bytes in 1 blocks are definitely lost in loss record 10 of 238 > ==37097==at 0x4C2DE96: malloc (vg_replace_malloc.c:309) > ==37097==by 0x5D07489: strdup (strdup.c:42) > ==37097==by 0x72A8CBE: eina_vpath_resolve (eina_vpath.c:285) > ==37097==by 0x90128D9: _efreet_efreet_app_interface_set (efreet.c:49) > ==37097==by 0x9012F27: efreet_init (efreet.c:149) > ==37097==by 0x5803F36: edje_init (edje_main.c:78) > ==37097==by 0x400EED: main > (in /home/lg/GSC/efl-1.21.1/src/examples/edje/edje-anchors) ==37097== > ==37097== 32 bytes in 1 blocks are definitely lost in loss record 49 of 238 > ==37097==at 0x4C2FFAC: calloc (vg_replace_malloc.c:762) > ==37097==by 0x4FD5B28: generic_cache_new (evas_common_generic_cache.c:7) > ==37097==by 0x4FE9AE6: eng_engine_new (evas_engine.c:3880) > ==37097==by 0x4EAC544: evas_output_method_set (evas_main.c:1224) > ==37097==by 0x12C3F004: ecore_evas_software_x11_new_internal > (ecore_evas_x.c:4153) ==37097==by 0x554F606: ecore_evas_software_x11_new > (ecore_evas.c:3831) ==37097==by 0x554F7EA: > _ecore_evas_constructor_software_x11 (ecore_evas.c:780) ==37097==by > 0x55499EF: _ecore_evas_new_auto_discover (ecore_evas.c:1021) ==37097==by > 0x55499EF: ecore_evas_new (ecore_evas.c:1046) ==37097==by 0x400F12: main > (in /home/lg/GSC/efl-1.21.1/src/examples/edje/edje-anchors) ==37097== > ==37097== 82 bytes in 1 blocks are definitely lost in loss record 154 of 238 > ==37097==at 0x4C2DE96: malloc (vg_replace_malloc.c:309) > ==37097==by 0x727B586: eina_module_new (eina_module.c:272) > ==37097==by 0x4F530F6: evas_module_find_type.part.3 (evas_module.c:601) > ==37097==by 0x4EA9E52: evas_render_method_lookup (evas_main.c:644) > ==37097==by 0x12C3EE2D: ecore_evas_software_x11_new_internal > (ecore_evas_x.c:4086) ==37097==by 0x554F606: ecore_evas_software_x11_new > (ecore_evas.c:3831) ==37097==by 0x554F7EA: > _ecore_evas_constructor_software_x11 (ecore_evas.c:780) ==37097==by > 0x55499EF: _ecore_evas_new_auto_discover (ecore_evas.c:1021) ==37097==by > 0x55499EF: ecore_evas_new (ecore_evas.c:1046) ==37097==by 0x400F12: main > (in /home/lg/GSC/efl-1.21.1/src/examples/edje/edje-anchors) ==37097== > ==37097== 232 (192 direct, 40 indirect) bytes in 1 blocks are definitely lost > in loss record 190 of 238 ==37097==at 0x4C2FFAC: calloc > (vg_replace_malloc.c:762) ==37097==by 0x703431E: _efl_add_internal_start > (eo.c:896) ==37097==by 0x4E8F279: evas_device_add_full (evas_device.c:190) > ==37097==by 0x4EAC655: evas_output_method_set (evas_main.c:1251) > ==37097==by 0x12C3F004: ecore_evas_software_x11_new_internal > (ecore_evas_x.c:4153) ==37097==by 0x554F606: ecore_evas_software_x11_new > (ecore_evas.c:3831) ==37097==by 0x554F7EA: > _ecore_evas_constructor_software_x11 (ecore_evas.c:780) ==37097==by > 0x55499EF: _ecore_evas_new_auto_discover (ecore_evas.c:1021) ==37097==by > 0x55499EF: ecore_evas_new (ecore_evas.c:1046) ==37097==by 0x400F12: main > (in /home/lg/GSC/efl-1.21.1/src/examples/edje/edje-anchors) ==37097== > ==37097== 232 (192 direct, 40 indirect) bytes in 1 blocks are definitely lost > in loss record 191 of 238 ==37097==at 0x4C2FFAC: calloc > (vg_replace_malloc.c:762) ==37097==by 0x703431E: _efl_add_internal_start > (eo.c:896) ==37097==by 0x4E8F279: evas_device_add_full (evas_device.c:190) > ==37097==by 0x4EAC6A0: evas_output_method_set (evas_main.c:1255) > ==37097==by 0x12C3F004: ecore_evas_software_x11_new_internal > (ecore_evas_x.c:4153) ==37097==by 0x554F606: ecore_evas_software_x11_new > (ecore_evas.c:3831) ==37097==by 0x554F7EA: > _ecore_evas_constructor_software_x11 (ecore_evas.c:780) ==37097==by > 0x55499EF: _ecore_evas_new_auto_discover (ecore_evas.c:1021) ==37097==by > 0x55499EF: ecore_evas_new (ecore_evas.c:1046) ==37097==by 0x400F
Re: [E-devel] valgrind show error when debug efl demo
I havent looked at e stuff in ages. Your code does not invoke ecore_evas_free() to free the allocated memory from the ecore_evas_new() call you have. Not sure about the efreet stuff, I thought that was deprecated. Regards. On Wed, Nov 27, 2019 at 7:00 PM Jing wrote: > Hi all, > I use valgrind to check memory on efl demo, but valgrind show error as > shown below, attached are the efl demo codes, please help to advise where > went wrong ? > > > sudo valgrind --tool=memcheck --leak-check=full ./edje-anchors > ==37097== Memcheck, a memory error detector > ==37097== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et al. > ==37097== Using Valgrind-3.15.0 and LibVEX; rerun with -h for copyright > info > ==37097== Command: ./edje-anchors > ==37097== > key:Escape > ==37097== > ==37097== HEAP SUMMARY: > ==37097== in use at exit: 246,790 bytes in 2,198 blocks > ==37097== total heap usage: 17,746 allocs, 15,548 frees, 6,874,546 bytes > allocated > ==37097== > ==37097== 10 bytes in 1 blocks are definitely lost in loss record 10 of 238 > ==37097==at 0x4C2DE96: malloc (vg_replace_malloc.c:309) > ==37097==by 0x5D07489: strdup (strdup.c:42) > ==37097==by 0x72A8CBE: eina_vpath_resolve (eina_vpath.c:285) > ==37097==by 0x90128D9: _efreet_efreet_app_interface_set (efreet.c:49) > ==37097==by 0x9012F27: efreet_init (efreet.c:149) > ==37097==by 0x5803F36: edje_init (edje_main.c:78) > ==37097==by 0x400EED: main (in > /home/lg/GSC/efl-1.21.1/src/examples/edje/edje-anchors) > ==37097== > ==37097== 32 bytes in 1 blocks are definitely lost in loss record 49 of 238 > ==37097==at 0x4C2FFAC: calloc (vg_replace_malloc.c:762) > ==37097==by 0x4FD5B28: generic_cache_new > (evas_common_generic_cache.c:7) > ==37097==by 0x4FE9AE6: eng_engine_new (evas_engine.c:3880) > ==37097==by 0x4EAC544: evas_output_method_set (evas_main.c:1224) > ==37097==by 0x12C3F004: ecore_evas_software_x11_new_internal > (ecore_evas_x.c:4153) > ==37097==by 0x554F606: ecore_evas_software_x11_new (ecore_evas.c:3831) > ==37097==by 0x554F7EA: _ecore_evas_constructor_software_x11 > (ecore_evas.c:780) > ==37097==by 0x55499EF: _ecore_evas_new_auto_discover > (ecore_evas.c:1021) > ==37097==by 0x55499EF: ecore_evas_new (ecore_evas.c:1046) > ==37097==by 0x400F12: main (in > /home/lg/GSC/efl-1.21.1/src/examples/edje/edje-anchors) > ==37097== > ==37097== 82 bytes in 1 blocks are definitely lost in loss record 154 of > 238 > ==37097==at 0x4C2DE96: malloc (vg_replace_malloc.c:309) > ==37097==by 0x727B586: eina_module_new (eina_module.c:272) > ==37097==by 0x4F530F6: evas_module_find_type.part.3 (evas_module.c:601) > ==37097==by 0x4EA9E52: evas_render_method_lookup (evas_main.c:644) > ==37097==by 0x12C3EE2D: ecore_evas_software_x11_new_internal > (ecore_evas_x.c:4086) > ==37097==by 0x554F606: ecore_evas_software_x11_new (ecore_evas.c:3831) > ==37097==by 0x554F7EA: _ecore_evas_constructor_software_x11 > (ecore_evas.c:780) > ==37097==by 0x55499EF: _ecore_evas_new_auto_discover > (ecore_evas.c:1021) > ==37097==by 0x55499EF: ecore_evas_new (ecore_evas.c:1046) > ==37097==by 0x400F12: main (in > /home/lg/GSC/efl-1.21.1/src/examples/edje/edje-anchors) > ==37097== > ==37097== 232 (192 direct, 40 indirect) bytes in 1 blocks are definitely > lost in loss record 190 of 238 > ==37097==at 0x4C2FFAC: calloc (vg_replace_malloc.c:762) > ==37097==by 0x703431E: _efl_add_internal_start (eo.c:896) > ==37097==by 0x4E8F279: evas_device_add_full (evas_device.c:190) > ==37097==by 0x4EAC655: evas_output_method_set (evas_main.c:1251) > ==37097==by 0x12C3F004: ecore_evas_software_x11_new_internal > (ecore_evas_x.c:4153) > ==37097==by 0x554F606: ecore_evas_software_x11_new (ecore_evas.c:3831) > ==37097==by 0x554F7EA: _ecore_evas_constructor_software_x11 > (ecore_evas.c:780) > ==37097==by 0x55499EF: _ecore_evas_new_auto_discover > (ecore_evas.c:1021) > ==37097==by 0x55499EF: ecore_evas_new (ecore_evas.c:1046) > ==37097==by 0x400F12: main (in > /home/lg/GSC/efl-1.21.1/src/examples/edje/edje-anchors) > ==37097== > ==37097== 232 (192 direct, 40 indirect) bytes in 1 blocks are definitely > lost in loss record 191 of 238 > ==37097==at 0x4C2FFAC: calloc (vg_replace_malloc.c:762) > ==37097==by 0x703431E: _efl_add_internal_start (eo.c:896) > ==37097==by 0x4E8F279: evas_device_add_full (evas_device.c:190) > ==37097==by 0x4EAC6A0: evas_output_method_set (evas_main.c:1255) > ==37097==by 0x12C3F004: ecore_evas_software_x11_new_internal > (ecore_evas_x.c:4153) > ==37097==by 0x554F606: ecore_evas_software_x11_new (ecore_evas.c:3831) > ==37097==by 0x554F7EA: _ecore_evas_constructor_software_x11 > (ecore_evas.c:780) > ==37097==by 0x55499EF: _ecore_evas_new_auto_discover > (ecore_evas.c:1021) > ==37097==by 0x55499EF: ecore_evas_new (ecore_evas.c:1046) > ==37097==by 0x400F12: main (in > /home/lg/GSC/efl-1.21.1/src/examples/edje
