Date: Wed, 18 Apr 2001 21:37:05 +0300 From: Valentin Nechayev <netch@netch.kiev.ua> To: "David O'Brien" <obrien@FreeBSD.ORG> Cc: Jordan Hubbard <jkh@osd.bsdi.com>, freebsd-stable@FreeBSD.ORG, phk@FreeBSD.ORG Subject: Re: Fw: fresh 4.3-RC: "microuptime went backwards", console lockup Message-ID: <20010418213703.I7367@netch.kiev.ua> In-Reply-To: <20010418111525.B22207@dragon.nuxi.com>; from obrien@FreeBSD.ORG on Wed, Apr 18, 2001 at 11:15:25AM -0700 References: <20010418091425L.jkh@osd.bsdi.com> <20010418111525.B22207@dragon.nuxi.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Wed, Apr 18, 2001 at 11:15:25, obrien wrote about "Re: Fw: fresh 4.3-RC: "microuptime went backwards", console lockup":
Sorry that I did not check that there were bogus commit in -current also.
The situation becomes more and more interesting ;|
But for release engineering, the only thing I tried to say is that without
reverting, 4.3-release will be broken on some hardware, similar to
described one of my workstation.
What is the most right place to discuss problems with time subsystem?
> On Wed, Apr 18, 2001 at 09:14:25AM -0700, Jordan Hubbard wrote:
> > Erm, I'd really like your opinion on this. The guy is feeling
> > naturally somewhat slighted.
>
> Very sorry about missing this.
>
> > MFC:
> > Convert all users of fldoff() to offsetof(). fldoff() is bad
> > because it only takes a struct tag which makes it impossible to
> > use unions, typedefs etc.
> ...
> > === cut ===
> > --- src/sys/i386/isa/clock.c 2000/08/03 00:09:29 1.149.2.1
> > +++ src/sys/i386/isa/clock.c 2001/03/05 13:09:01 1.149.2.2
> > [...skipped by netch...]
> > @@ -191,7 +191,7 @@ SYSCTL_OPAQUE(_debug, OID_AUTO, tsc_time
> > static struct timecounter i8254_timecounter = {
> > i8254_get_timecount, /* get_timecount */
> > 0, /* no poll_pps */
> > - ~0u, /* counter_mask */
> > + 0x1ffff, /* counter_mask */
> > 0, /* frequency */
> > "i8254" /* name */
> > };
> > === end cut ===
> >
> > and contains no more, hence commit description is _totally bogus_.
>
> The commit message is the same as the clock.c rev 1.162 change which was
> part of PHK's commit I was MFC'ing (I was using a script to pipe commit
> emails thru to produce the diffs).
>
> I do not know if this change in the original commit by PHK was an
> accident, or somewhere else in the system something is different from
> RELENG_4 such that it needs a different value in the struct.
>
> Maybe PHK can comment on that part of the change.
>
> I certainly do not mind backing out clock.c rev 1.149.2.2 if you want.
>
> --
> -- David (obrien@FreeBSD.org)
/netch
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010418213703.I7367>