Messages in this thread | | | Date | Sun, 14 Apr 1996 19:08:23 +0200 (SAT) | From | Craig Schlenter <> | Subject | 1.3.88 stress testing (was: .. and oh, btw ..) |
| |
On Sat, 13 Apr 1996, Linus Torvalds wrote: [munch] > In mm/memory.c, function verify_area(), there is a test that looks like > if (expand_stack(vma, start)) > goto good_area; > but that test is actually reversed, and should be > if (expand_stack(vma, start) == 0) > goto good_area; [munch]
I have just tried running Bernd Schmidt's make -j script on a clean 1.3.88 and 1.3.88 with the above patch. The clean 1.3.88 fails with the odd SIG11 and SIG6 but the patched version is perfect. Excellent work!!
(I've tagged a copy of Bernd's script on below for those linux-kernel people who missed it the first time around and want to stress their machines a little. On the subject of SIG11: the memtest program on sunsite does an excellent job of sniffing out bad RAM and even found a buggy motherboard for me which kept gettting SIG11's)
--Craig
PS. Don't expect to get any useful work done while this script runs. Also I have 16M RAM and used about 20M swap space during this test. Had a load average of 64 for a while too! -------------------------CUT HERE-------------------------- #! /bin/sh
cp /usr/src/linux/kernel/sysctl.c sysctl-0.c cat >Makefile <<\EOF OBJS=\ EOF NUM=0 while [ $NUM != 100 ]; do echo "sysctl-$NUM.i \\" >>Makefile NEWNUM=`expr $NUM + 1` ln sysctl-$NUM.c sysctl-$NEWNUM.c NUM=$NEWNUM done cat >>Makefile <<\EOF
.SUFFIXES: .i .c all: $(OBJS)
clean: rm sysctl-*.[ci] rm Makefile
%.i: %.c gcc -E $< -o $@ EOF make -j
| |