> I don't think it's the same thing, as it's not even
> starting up; ...
Hmm, AFAIR, build 101 top wouldn't start when I tried
to run it with the libumem memory allocator.
With the standard libc malloc it started, but leaked memory.
But it did try to return an invalid pointer to the malloc
heap. I think that could explain the crash you're observing.
> ... here's the truss output tail:
>
> 11180: open("601/lpsinfo", O_RDONLY) = 26
> 11180: fstat(26, 0xFFFFFD7FFFDFEF60) = 0
> 11180: fstat(26, 0xFFFFFD7FFFDFF2F0) = 0
> 11180: pread(26, "13\0\0\0\0\0\0\090\0\0\0".., 2752, 0) = 2752
> 11180: Incurred fault #6, FLTBOUNDS %pc = 0x0040D480
> 11180: siginfo: SIGSEGV SEGV_MAPERR addr=0x00000011
> 11180: Received signal #11, SIGSEGV [default]
> 11180: siginfo: SIGSEGV SEGV_MAPERR addr=0x00000011
A stack backtrace from top's core dump would be better (pstack core)
Can you try the top binary that I attached to bug 5482,
comment 11,
http://defect.opensolaris.org/bz/show_bug.cgi?id=5482#c11
Does it core dump at startup, too?
--
This message posted from opensolaris.org
_______________________________________________
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
