robin@medstar.UUCP (Robin Cutshaw) (07/10/85)
Here is a quick way to panic the IBM Xenix kernel (both original and updated).
This works for both the small and medium models. (use cc shr.c -o shr -lx).
---------------CUT HERE-----------------------
#include <stdio.h>
#include <sd.h>
#include <errno.h>
main()
{
char *share, pnbuf[128];
extern int errno;
extern char *getcwd();
if ( (share = sdget("/tmp/shx", SD_CREAT|SD_WRITE,12L,0777)) == (char *)-1) {
fprintf(stderr,"Share = %d %d\n",errno,share);
exit(share);
}
sdenter(share, SD_WRITE);
(void )getcwd(pnbuf,100); /* THIS IS THE KILLER LINE */
sdleave(share);
sdfree(share);
}
--
----
Robin Cutshaw
uucp: ...!{akgua,gatech}!medstar!robinforceten@dalcs.UUCP (ForceTen Enterprises) (07/16/85)
][b In article 181@medstar, Robin Cutshaw writes: > Here is a quick way to panic the IBM Xenix kernel (both original and updated). > This works for both the small and medium models. (use cc shr.c -o shr -lx). > ... Code leading up to > sdenter(share, SD_WRITE); > (void )getcwd(pnbuf,100); /* THIS IS THE KILLER LINE */ > sdleave(share); > > sdfree(share); Although the crash is tragic, I would guess that removing the cast to (void) would cause the code to work correctly. Microsoft documents this behaviour in some obscure place (does the same on both XT and AT versions of the compiler) in the Xenix manuals. A similar piece of code crashes my XT, and causes the AT to segment fault. On the AT, a simpler means of instantly halting my machine without so much as a panic (power cycle to reboot) $ cat > /dev/monochrome I use an AT w 1 Meg memory, Paradise Systems Multidisplay Card. Has anyone encountered this on a system with a standard IBM colour adaptor? ||!][b --------------------- Neil S Erskine decvax!dartvax!dalcs!force10!erskine ForceTen Enterprises Inc. 3845 Dutch Village Rd. Halifax, N.S. (902) 453-0040 -- Neil S Erskine decvax!dartvax!dalcs!force10!erskine ForceTen Enterprises Inc. 3845 Dutch Village Rd. Halifax, N.S. (902) 453-0040
caf@omen.UUCP (Chuck Forsberg WA7KGX) (07/19/85)
In article <1535@dalcs.UUCP> forceten@dalcs.UUCP (ForceTen Enterprises) writes: > > On the AT, a simpler means of instantly halting my machine >without so much as a panic (power cycle to reboot) > > $ cat > /dev/monochrome > > I use an AT w 1 Meg memory, Paradise Systems Multidisplay Card. >Has anyone encountered this on a system with a standard IBM colour adaptor? My machine has an IBM monochrome board and a Paradise Systems Multidisplay Card configured as color only (take too much current to run in my PC!). Memory is 1152k and two serial + two parallel ports. Default monitor is monochrome. cat > /dev/monochrome elicits a "can't create" message, even if superuser. cat > /dev/color works as expected, accepting keyboard input until ^D, at which time it all scrolls out on the color monitor. In fact, I often redirect output to the color monitor, sort of a hardware version of windows but with only one virtual terminal. There are some strangenesses with the Paradise board if I try to run Xenix with the displays in any other configuration that the one described. -- Chuck Forsberg WA7KGX ...!tektronix!reed!omen!caf CIS:70715,131 Omen Technology Inc 17505-V NW Sauvie Island Road Portland OR 97231 Voice: 503-621-3406 Modem: 503-621-3746 (Hit CR's for speed detect) Home of Professional-YAM, the most powerful COMM program for the IBM PC
robin@medstar.UUCP (Robin Cutshaw) (07/19/85)
> Although the crash is tragic, I would guess that removing the > cast to (void) would cause the code to work correctly. Microsoft > documents this behaviour in some obscure place (does the same on both > XT and AT versions of the compiler) in the Xenix manuals. A similar > piece of code crashes my XT, and causes the AT to segment fault. (void) makes no difference... -robin -- ---- Robin Cutshaw uucp: ...!{akgua,gatech}!medstar!robin