ray@maxwell.physics.purdue.edu (Ray Moody) (03/13/89)
Index: sys/sys/kern_clock.c
Description:
Some variable names are messed up in sys/sys/kern_clock.c
Repeat-By:
Inspection, mainly.
We noticed this bug in our system when our clocks started loosing about
half an hour every day. (No, I don't happen to know the magic numbers
to give adjtime() to cause this.)
Fix:
*** /tmp/,RCSt1007315 Mon Mar 13 10:52:35 1989
--- /tmp/,RCSt2007315 Mon Mar 13 10:52:36 1989
***************
*** 78,84
register struct proc *p;
register int s;
int needsoft = 0;
! extern int tickdelta, tickadj;
#ifdef is68k
lbolt++; /* ISI: SYSV */
--- 78,84 -----
register struct proc *p;
register int s;
int needsoft = 0;
! extern int tickdelta, timedelta;
#ifdef is68k
lbolt++; /* ISI: SYSV */
***************
*** 202,208
* so we don't keep the relatively high clock interrupt
* priority any longer than necessary.
*/
! if (tickdelta == 0)
BUMPTIME(&time, tick)
else {
register delta;
--- 202,208 -----
* so we don't keep the relatively high clock interrupt
* priority any longer than necessary.
*/
! if (timedelta == 0)
BUMPTIME(&time, tick)
else {
register delta;
***************
*** 207,215
else {
register delta;
! if (tickdelta < 0) {
! delta = tick - tickadj;
! tickdelta += tickadj;
} else {
delta = tick + tickadj;
tickdelta -= tickadj;
--- 207,215 -----
else {
register delta;
! if (timedelta < 0) {
! delta = tick - tickdelta;
! timedelta += tickdelta;
} else {
delta = tick + tickdelta;
timedelta -= tickdelta;
***************
*** 211,218
delta = tick - tickadj;
tickdelta += tickadj;
} else {
! delta = tick + tickadj;
! tickdelta -= tickadj;
}
BUMPTIME(&time, delta);
/* KLUDGE SUN NFS code ????
--- 211,218 -----
delta = tick - tickdelta;
timedelta += tickdelta;
} else {
! delta = tick + tickdelta;
! timedelta -= tickdelta;
}
BUMPTIME(&time, delta);
/* KLUDGE SUN NFS code ????chris@mimsy.UUCP (Chris Torek) (03/15/89)
In article <2040@pur-phy> ray@maxwell.physics.purdue.edu (Ray Moody) writes: > #ifdef is68k This line is not found in 4BSD. (Not 4.0, nor 4.1, nor 4.2, nor 4.3, nor 4.3-tahoe.) An `index' line indicating the operating system version and, if the system is from a vendor that repackages 4BSD, the vendor, is helpful in avoiding confusion. Incidentally, 4.3BSD-tahoe, at least, does not have the bug. -- In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 7163) Domain: chris@mimsy.umd.edu Path: uunet!mimsy!chris
ray@maxwell.physics.purdue.edu (Ray Moody) (03/16/89)
In article <2040@pur-phy> ray@maxwell.physics.purdue.edu (Ray Moody) writes: > #ifdef is68k In article <16396@mimsy.UUCP>, chris@mimsy (Chris Torek) writes: >This line is not found in 4BSD. (Not 4.0, nor 4.1, nor 4.2, nor 4.3, nor >4.3-tahoe.) > >An `index' line indicating the operating system version and, if the >system is from a vendor that repackages 4BSD, the vendor, is helpful >in avoiding confusion. This bug was in Integrated Solutions Incorporated's version of 4.3BSD. It has been my experience that almost every bug we have is persent in one or more versions of Sun's UNIX, and many of our bugs are also in generic 4.3BSD. Ray