[comp.unix.ultrix] KM_ALLOC: corrupt buckets

cal@felix.UUCP (Cal J. Myers) (04/10/90)

We use to get these all the time some time, 3 or 4 a day.  At first
DEC support sent us several lat patches.  But the one routine that
ended up fixing it was init_sysent.o.  Output of what:

init_sysent.o
	init_sysent.c	9.1BLD	(ULTRIX)	11/4/88

Since we installed this we haven't had any problems.  Just for the curious
we have an 8820 with Ultrix 3.1.
/cal

grr@cbmvax.commodore.com (George Robbins) (04/11/90)

In article <142859@felix.UUCP> cal@felix.UUCP (Cal J. Myers) writes:
> We use to get these all the time some time, 3 or 4 a day.  At first
> DEC support sent us several lat patches.  But the one routine that
> ended up fixing it was init_sysent.o.  Output of what:
> 
> init_sysent.o
> 	init_sysent.c	9.1BLD	(ULTRIX)	11/4/88
> 
> Since we installed this we haven't had any problems.  Just for the curious
> we have an 8820 with Ultrix 3.1.

From previous references to this problem, this particular patch probably
only applies to multi-processor systems.  I beleive it marks one syscall
as non-reentrant (or whatever the multi-processor buzzword is).

-- 
George Robbins - now working for,     uucp:   {uunet|pyramid|rutgers}!cbmvax!grr
but no way officially representing:   domain: grr@cbmvax.commodore.com
Commodore, Engineering Department     phone:  215-431-9349 (only by moonlite)