Re: Memory again/ NS crash

[emacs user]

Please see below another crash that seems NS-related...  happy to
provide additional details/ do tests, if anyone is interested in
further exploring this.

On Sat, Jan 21, 2012 at 6:37 AM, YAMAMOTO Mitsuharu wrote:

>> Program received signal EXC_BAD_ACCESS, Could not access memory.
>> Reason: KERN_PROTECTION_FAILURE at address: 0x00007fff5f3aefe8
>
> According to the output of the `vmmap' command, 0x00007fff5f3aefe8 is
> in the stack guard for the main thread.  Together with the long stack
> trace, this crash seems to be due to too deeply nested recursive
> calls.  It might be caused by too deeply nested data structures that
> lead to deep the GC traversal, or some NS-specific problem again.  It
> would be useful to post to the list whether it also happens with an
> unoptimized X11 build.

Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_PROTECTION_FAILURE at address: 0x00007fff5f3aefe8
0x00000001001d1e12 in mark_object (arg=4571286741) at alloc.c:5579
5579            mark_vectorlike (XVECTOR (obj));
(gdb) where
#0  0x00000001001d1e12 in mark_object (arg=4571286741) at alloc.c:5579
#1  0x00000001001d2354 in mark_object (arg=4468885238) at alloc.c:5702
#2  0x00000001001d1edc in mark_object (arg=4552798930) at alloc.c:5595
#3  0x00000001001d17bd in mark_vectorlike (ptr=0x107b8b050) at alloc.c:5391
#4  0x00000001001d1e17 in mark_object (arg=4424511573) at alloc.c:5579
#5  0x00000001001d17bd in mark_vectorlike (ptr=0x107b90390) at alloc.c:5391
#6  0x00000001001d1e17 in mark_object (arg=4424532885) at alloc.c:5579
#7  0x00000001001d2354 in mark_object (arg=4468885222) at alloc.c:5702
#8  0x00000001001d1edc in mark_object (arg=4552799074) at alloc.c:5595
#9  0x00000001001d17bd in mark_vectorlike (ptr=0x107bc6fe0) at alloc.c:5391
#10 0x00000001001d1e17 in mark_object (arg=4424757221) at alloc.c:5579
#11 0x00000001001d17bd in mark_vectorlike (ptr=0x107b48eb0) at alloc.c:5391
#12 0x00000001001d1e17 in mark_object (arg=4424240821) at alloc.c:5579
#13 0x00000001001d2354 in mark_object (arg=4468885206) at alloc.c:5702
#14 0x00000001001d1edc in mark_object (arg=4552770562) at alloc.c:5595
#15 0x00000001001d17bd in mark_vectorlike (ptr=0x107b2a360) at alloc.c:5391
#16 0x00000001001d1e17 in mark_object (arg=4424115045) at alloc.c:5579
#17 0x00000001001d17bd in mark_vectorlike (ptr=0x107b49170) at alloc.c:5391
#18 0x00000001001d1e17 in mark_object (arg=4424241525) at alloc.c:5579
#19 0x00000001001d2354 in mark_object (arg=4468885190) at alloc.c:5702
#20 0x00000001001d1edc in mark_object (arg=4552770706) at alloc.c:5595
#21 0x00000001001d17bd in mark_vectorlike (ptr=0x107bc8640) at alloc.c:5391
#22 0x00000001001d1e17 in mark_object (arg=4424762949) at alloc.c:5579
...
this seems to be going on forever, I killed it at some point and am
attaching only the beginning of the traceback above.