Date: Mon, 6 Nov 2000 01:57:59 -0500 From: "John W. De Boskey" <jwd@bsdwins.com> To: Marcel Moolenaar <marcel@cup.hp.com> Cc: "David O'Brien" <obrien@FreeBSD.org>, "Jordan K. Hubbard" <jkh@FreeBSD.org>, cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src Makefile Message-ID: <20001106015759.A16228@bsdwins.com> In-Reply-To: <3A06250E.E752CF4B@cup.hp.com>; from marcel@cup.hp.com on Sun, Nov 05, 2000 at 07:27:10PM -0800 References: <200011052244.OAA60716@freefall.freebsd.org> <20001105161514.A7914@dragon.nuxi.com> <20001105193652.A14868@bsdwins.com> <3A06250E.E752CF4B@cup.hp.com>
next in thread | previous in thread | raw e-mail | index | archive | help
How about something like the following:
STARTSECS!= date +%s
all:
@echo ">>> ${OBJFORMAT} make world started on `LC_TIME=C date`"
sleep 2
@echo ">>> ${OBJFORMAT} make world completed on `LC_TIME=C date`"
@echo ">>> ${OBJFORMAT} make world duration `TZ=GMT date -r \`expr \\\`date +%s\\\` - ${STA RTSECS}\` +%H:%M:%S`"
That last line is a bit dicey, but it works...
>>> elf make world started on Mon Nov 6 01:54:39 EST 2000
sleep 2
>>> elf make world completed on Mon Nov 6 01:54:41 EST 2000
>>> elf make world duration 00:00:02
I'll turn this into a patch if you folks think it's reasonable. I can
either send it to Jordan, or commit it.
-John
----- Marcel Moolenaar's Original Message -----
> "John W. De Boskey" wrote:
> >
> > ----- David O'Brien's Original Message -----
> > > On Sun, Nov 05, 2000 at 02:44:49PM -0800, Jordan K. Hubbard wrote:
> > > > jkh 2000/11/05 14:44:49 PST
> > > >
> > > > Modified files: (Branch: RELENG_4)
> > > > . Makefile
> > > > Log:
> > > > removed duplicated "start" message for world target.
> > >
> > > Adding the duplicated "start" message was intentional -- so it would be
> > > easy to see how long the `make world' took.
> > >
> > > Personally I prefer the duplicated message and would like to see it
> > > return. With it I then don't have to either save the output to a file to
> > > see how long it took, or the times I do save the build output, I don't
> > > have to go looking thru the output file.
> > >
> > > --
> > > -- David (obrien@FreeBSD.org)
> > > GNU is Not Unix / Linux Is Not UniX
> >
> > I have to agree with David. I have code which depends on these
> > values at the end of the world/release logs.
> >
> > May I be so bold to ask where this was discussed beforehand?
> > I'd like to see other folks' pros and cons.
>
> >From Jordan's point of view it makes sense to remove the duplicate at
> the end. Having the start, milestones and finish in the proper order
> without duplicates, makes the grep more pleasing.
>
> >From David's point of view, it make sense to know when it started if you
> would only see the last lines of the output. Only in this case the time
> elapsed is the information that's actaully wanted.
>
> If we combine these views, we have a "start" mesage at the beginning and
> use that and the "end" message to synthesize the elapsed time, which we
> emit after the "end" message. This should work, no matter what viewpoint
> is taken, right?
>
> --
> Marcel Moolenaar
> mail: marcel@cup.hp.com / marcel@FreeBSD.org
> tel: (408) 447-4222
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20001106015759.A16228>