[Bug 43340] New: valgrind shows several possible leaks in several tests ( e.g., dlls/amstream/tests/amstream.c) (AMD+llvm mesa)

https://bugs.winehq.org/show_bug.cgi?id=43340

Bug ID: 43340
           Summary: valgrind shows several possible leaks in several tests
                    (e.g., dlls/amstream/tests/amstream.c) (AMD+llvm mesa)
           Product: Wine
           Version: 2.12
          Hardware: x86
                OS: Linux
            Status: NEW
          Keywords: download, source, testcase, valgrind
          Severity: normal
          Priority: P2
         Component: advapi32
          Assignee: [e-mail]
          Reporter: [e-mail]
      Distribution: ---

Glx info:
OpenGL renderer string: Gallium 0.4 on AMD OLAND (DRM 2.49.0 /
4.9.33-vanilla-workstation-3, LLVM 3.9.1)
OpenGL core profile version string: 4.5 (Core Profile) Mesa 17.0.6
OpenGL core profile shading language version string: 4.50
OpenGL version string: 3.0 Mesa 17.0.6

Henri and I briefly discussed on IRC, not likely an issue:
<wizardedit> strfllw, looking at valgrind + mesa with debugging info fixed
(march disabled); I see a lot of possible leaks, could you take a quick look
please? e.g., d3d10core/device: https://pastebin.com/URY5kwxu
<strfllw> wizardedit: I'd have to take a much closer look to say for sure, but
purely by "gut feeling" I'd say probably not
<strfllw> my guess would be valgrind not understanding ralloc in the first one,
and not understanding Mesa's shader cache for the others

Filing a bug for reference / place holder in case someone does want to
investigate.

../../../tools/runtest -q -P wine -T ../../.. -M amstream.dll -p
amstream_test.exe.so amstream && touch amstream.ok
==25729== 252 bytes in 1 blocks are possibly lost in loss record 2,505 of 3,085
==25729==    at 0x402C267: malloc (vg_replace_malloc.c:299)
==25729==    by 0x6928B70: ralloc_size (ralloc.c:126)
==25729==    by 0x6928C8E: rzalloc_size (ralloc.c:158)
==25729==    by 0x6928151: _mesa_hash_table_rehash (hash_table.c:269)
==25729==    by 0x69280AC: hash_table_insert (hash_table.c:301)
==25729==    by 0x6D52812: si_shader_cache_insert_shader
(si_state_shaders.c:198)
==25729==    by 0x6D52812: si_init_shader_selector_async
(si_state_shaders.c:1353)
==25729==    by 0x6A780B6: util_queue_thread_func (u_queue.c:154)
==25729==    by 0x6A77C04: impl_thrd_routine (threads_posix.h:87)
==25729==    by 0x4245249: start_thread (pthread_create.c:333)
==25729==    by 0x4340D6D: clone (clone.S:114)
==25729==

==25729== 520 bytes in 6 blocks are possibly lost in loss record 2,715 of 3,085
==25729==    at 0x402C267: malloc (vg_replace_malloc.c:299)
==25729==    by 0x6D4FA9A: read_chunk (si_state_shaders.c:99)
==25729==    by 0x6D524CD: si_load_shader_binary (si_state_shaders.c:165)
==25729==    by 0x6D524CD: si_shader_cache_load_shader (si_state_shaders.c:216)
==25729==    by 0x6D524CD: si_init_shader_selector_async
(si_state_shaders.c:1336)
==25729==    by 0x6A780B6: util_queue_thread_func (u_queue.c:154)
==25729==    by 0x6A77C04: impl_thrd_routine (threads_posix.h:87)
==25729==    by 0x4245249: start_thread (pthread_create.c:333)
==25729==    by 0x4340D6D: clone (clone.S:114)
==25729==

==25729== 2,304 bytes in 6 blocks are possibly lost in loss record 2,935 of
3,085
==25729==    at 0x402DDE5: calloc (vg_replace_malloc.c:711)
==25729==    by 0x6D51FA6: si_init_shader_selector_async
(si_state_shaders.c:1319)
==25729==    by 0x6A780B6: util_queue_thread_func (u_queue.c:154)
==25729==    by 0x6A77C04: impl_thrd_routine (threads_posix.h:87)
==25729==    by 0x4245249: start_thread (pthread_create.c:333)
==25729==    by 0x4340D6D: clone (clone.S:114)
==25729==

==25729== 7,260 bytes in 6 blocks are possibly lost in loss record 3,014 of
3,085
==25729==    at 0x402C267: malloc (vg_replace_malloc.c:299)
==25729==    by 0x6D4FA9A: read_chunk (si_state_shaders.c:99)
==25729==    by 0x6D52514: si_load_shader_binary (si_state_shaders.c:171)
==25729==    by 0x6D52514: si_shader_cache_load_shader (si_state_shaders.c:216)
==25729==    by 0x6D52514: si_init_shader_selector_async
(si_state_shaders.c:1336)
==25729==    by 0x6A780B6: util_queue_thread_func (u_queue.c:154)
==25729==    by 0x6A77C04: impl_thrd_routine (threads_posix.h:87)
==25729==    by 0x4245249: start_thread (pthread_create.c:333)
==25729==    by 0x4340D6D: clone (clone.S:114)
==25729==

==25729== 9,668 bytes in 15 blocks are possibly lost in loss record 3,028 of
3,085
==25729==    at 0x402C267: malloc (vg_replace_malloc.c:299)
==25729==    by 0x6D52001: si_get_tgsi_binary (si_state_shaders.c:50)
==25729==    by 0x6D52001: si_init_shader_selector_async
(si_state_shaders.c:1330)
==25729==    by 0x6A780B6: util_queue_thread_func (u_queue.c:154)
==25729==    by 0x6A77C04: impl_thrd_routine (threads_posix.h:87)
==25729==    by 0x4245249: start_thread (pthread_create.c:333)
==25729==    by 0x4340D6D: clone (clone.S:114)
==25729==

==25729== 21,668 bytes in 15 blocks are possibly lost in loss record 3,061 of
3,085
==25729==    at 0x402DDE5: calloc (vg_replace_malloc.c:711)
==25729==    by 0x6D52699: si_get_shader_binary (si_state_shaders.c:125)
==25729==    by 0x6D52699: si_shader_cache_insert_shader
(si_state_shaders.c:194)
==25729==    by 0x6D52699: si_init_shader_selector_async
(si_state_shaders.c:1353)
==25729==    by 0x6A780B6: util_queue_thread_func (u_queue.c:154)
==25729==    by 0x6A77C04: impl_thrd_routine (threads_posix.h:87)
==25729==    by 0x4245249: start_thread (pthread_create.c:333)
==25729==    by 0x4340D6D: clone (clone.S:114)
==25729==

shows up in (at least):
amstream/amstream.c
d2d1/d2d1.c
d3d11/d3d11.c
d3d10core/device.c
etc.

[Bug 43340] valgrind shows several possible leaks in several tests (e.g., dlls/amstream/tests/amstream.c) (AMD+llvm mesa)

https://bugs.winehq.org/show_bug.cgi?id=43340

--- Comment #1 from Austin English  ---
Created attachment 58682
  --> https://bugs.winehq.org/attachment.cgi?id=58682
glxinfo

[Bug 43340] valgrind shows several possible leaks in several tests (e.g., dlls/amstream/tests/amstream.c) (AMD+llvm mesa)

https://bugs.winehq.org/show_bug.cgi?id=43340

joaopa  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |[e-mail]

--- Comment #2 from joaopa  ---
Does the bug still occur with wine-4.9?