[GRASS5] Random G_malloc/malloc SIGSEGV in s.qcount.

I'd appreciate any help figuring out why malloc() is generating a
Segment Violation in s.qcount. The problem occurs randomly in the
G_malloc call of count.c. I'm guessing that it's related to the random
number generation or something in findquads.c (which is called first).
I want to know if these segment violations occur on systems other than
pc-linux-gnu.

I have definetly tracked it to one particular instance of a G_malloc
call like 'counts = (int *) G_malloc(nquads * sizeof(int));'. I'm
baffled.

--
#! /bin/sh
# ppp-address: What's my Internet Address for ppp0 ?
/sbin/ifconfig ppp0 2> /dev/null | grep 'inet addr:' | sed \
's=.*inet addr\:\([0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}\).*=\1='

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'

Problem found and fixed.

BTW, GNU Libc has a nifty mcheck function that can be enabled by
exporting the environment variable MALLOC_CHECK_ with a value equal to
0 <= NO CHECKING
1 <= REPORT ERROR
2 <= ABORT ON ERRORS

No recompilation required.

--
#! /bin/sh
# ppp-address: What's my Internet Address for ppp0 ?
/sbin/ifconfig ppp0 2> /dev/null | grep 'inet addr:' | sed \
's=.*inet addr\:\([0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}\).*=\1='

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'

On Sat, Oct 28, 2000 at 05:39:07PM -0700, Eric G . Miller wrote:

On Sun, Oct 29, 2000 at 01:16:55AM +0100, Markus Neteler wrote:
> Hi Eric,
>
> what do you think: can we rename s.rand to s.random to
> get it conform to v.random/r.random?

Sounds good.

Hi all,

if there are no objections, I would rename s.rand to s.random.

Markus

PS: Eric, could you please check in the missing s.buffer code?

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'

On Sun, Oct 29, 2000 at 11:05:25AM +0000, Markus Neteler wrote:

PS: Eric, could you please check in the missing s.buffer code?

I don't have the original (only an html man page). I have a partly
completed replacement, but I got a little stuck on the case where
buffered areas overlap. I didn't find any library code that would make
handling this easy, so I started working on my own merge code. Anyway,
I got sidetracked working on other things (guess I'll look at v.merge
and see how it does the job).

--
#! /bin/sh
# ppp-address: What's my Internet Address for ppp0 ?
/sbin/ifconfig ppp0 2> /dev/null | grep 'inet addr:' | sed \
's=.*inet addr\:\([0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}\).*=\1='

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'