[MonoDevelop] Memory leak

[Michael Hutchinson]

On Wed, Aug 13, 2008 at 3:50 PM, Casey Marshall
<casey.s.marshall at gmail.com> wrote:
> Mike Krüger wrote:
>> Hi
>>
>> Yes, we're aware of memory problems and working on fixing them.
>>
>> Regards
>> Mike
>>
>>> I think I've seen references to this before, but it looks like
>>> MonoDevelop has a somewhat slow, but constant, memory leak. Are you guys
>>> aware of this? Does anyone have an inkling on where it could be? The
>>> leak seems pretty slow -- the heap grows by about 100K every two seconds
>>> -- but it's constantly going up at that rate.
>>>
>>> I've tried running MonoDevelop both using heap-shot and valgrind, and
>>> neither one seems to be pointing to any obvious issues, at least not
>>> after a few hours.
>>>
>>> Thanks.
>
> WRT that, though, any clue why I'm getting this result running
> MonoDevelop under valgrind? 822 MiB is a pretty sizable percentage of
> the memory MonoDevelop accumulated overnight:
>
>> ==7947== 862,223,392 bytes in 1,737,838 blocks are still reachable in loss record 238 of 238
>> ==7947==    at 0x4C21F8F: memalign (vg_replace_malloc.c:460)
>> ==7947==    by 0x4C22028: posix_memalign (vg_replace_malloc.c:569)
>> ==7947==    by 0x507D299: (within /usr/lib/libglib-2.0.so.0.1600.4)
>> ==7947==    by 0x507E0F0: g_slice_alloc (in /usr/lib/libglib-2.0.so.0.1600.4)
>> ==7947==    by 0x506035D: g_list_prepend (in /usr/lib/libglib-2.0.so.0.1600.4)
>> ==7947==    by 0x432953: mono_arch_get_allocatable_int_vars (mini-amd64.c:910)
>> ==7947==    by 0x5579A9: mini_method_compile (mini.c:12490)
>> ==7947==    by 0x558CF8: mono_jit_compile_method (mini.c:12819)
>> ==7947==    by 0x42C5A2: mono_magic_trampoline (mini-trampolines.c:249)
>> ==7947==    by 0x415B164: ???
>> ==7947==    by 0x8FCA917: ???
>> ==7947==    by 0x85E7E9F: ???

Am not familiar with valgrind output. What does this mean? Is it
possible to see what's being leaked?

If that's the stack trace of the leak, it looks like a JIT leak to me
(not that I have any real idea). This may be more likely when Mono's
in debug mode (which "make run" uses). Can you repro this in release
mode and/or Mono from the 2.0?

-- 
Michael Hutchinson
http://mjhutchinson.com