Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 03 Mar 2008 18:49:58 +0100
From:      Remko Lodder <remko@FreeBSD.org>
To:        jau@iki.fi
Cc:        freebsd-bugs@FreeBSD.org, kris@FreeBSD.org
Subject:   Re: kern/121292: FreeBSD-7.0 kernel build fails on FreeBSD-6.3
Message-ID:  <47CC3A46.4080906@FreeBSD.org>
In-Reply-To: <200803031700.m23H0W7w064934@jau.iki.fi>
References:  <200803031700.m23H0W7w064934@jau.iki.fi>

next in thread | previous in thread | raw e-mail | index | archive | help
Jukka A. Ukkonen wrote:
> Quoting remko@FreeBSD.org:
>> Synopsis: FreeBSD-7.0 kernel build fails on FreeBSD-6.3
>>
>> State-Changed-From-To: open->feedback
>> State-Changed-By: remko
>> State-Changed-When: Sun Mar 2 20:18:05 UTC 2008
>> State-Changed-Why: 
>> Kris asked for feedback
> 
> 	Yes he did, and his assumption is wrong. - I pasted all the
> 	error info that was shown by the build run.
> 	The very odd thing is that there were no further details
> 	between the output lines...
> 
> 	objcopy --strip-debug if_zyd.ko
> 	1 error

Kris is not wrong here, you are confused ;)

> 
> 	The bleeping thing just goes south without so much as a word.
> 	I'd say this is kind of disturbing, isn't it?

No, it shows the error, though not where -you- would expect it. You 
prove it to yourself later in the email.

> 
> 
> 	I have tried with "nodevice zyd" in the config file but in
> 	vain.

The error didn't say if_zyd broke, all it reported was error 1 (from 
your output).

> 
> 	I also tried without "-j5" in the make options and without
> 	"-march=athlon-xp" in the cc options while having "-j5" in
> 	the make options, all in vain.

No you are wrong here, see below.

> 
> 	The error shown below is from a build attempt without "-j5" in
> 	the make options and without "-march=athlon-xp" in cc options.
> 
> 	If I try building without "-j5" I get this far but eventually
> 	still fail.
> 
> cc -c -O2 -pipe -fno-strict-aliasing  -std=c99  -Wall -Wredundant-decls -Wnested-externs -Wstrict-prototypes  -Wmissing-prototypes -Wpointer-arith -Winline -Wcast-qual  -Wundef -Wno-pointer-sign -fformat-extensions -nostdinc  -I. -I/usr/src/sys -I/usr/src/sys/contrib/altq -D_KERNEL -DHAVE_KERNEL_OPTION_HEADERS -include opt_global.h -fno-common -finline-limit=8000 --param inline-unit-growth=100 --param large-function-growth=1000  -mno-align-long-strings -mpreferred-stack-boundary=2  -mno-mmx -mno-3dnow -mno-sse -mno-sse2 -mno-sse3 -ffreestanding -Werror  vers.c
> linking kernel
> if_ural.o(.text+0xe43): In function `ural_attach':
> : undefined reference to `ieee80211_amrr_init'
> if_ural.o(.text+0x2c09): In function `ural_amrr_update':
> : undefined reference to `ieee80211_amrr_choose'
> if_ural.o(.text+0x32ac): In function `ural_task':
> : undefined reference to `ieee80211_amrr_node_init'
> *** Error code 1


So if_ural has problems, if you comment that out it'll probably build. 
It should't do that btw, but it did, probably because you didn't attach 
all wireless modules in your configuration or simply forgot something.
You are not running a GENERIC so I'll assume you stripped out things
and added things as you saw fit. That's fine, but that could cause 
problems.  :)

> 
> Stop in /usr/obj/usr/src/sys/Mimir.
> *** Error code 1
> 
> Stop in /usr/src.
> *** Error code 1
> 
> Stop in /usr/src.
> 
> 	So, it seems the original (ominously silent) error might be
> 	due to parallel make failing somehow.

well yes, thread 1 is building if_ural for example, and thread 2 is 
doing something else, which it can do much quicker, the logs are kept 
being spit out to the console you  have active. As you should expect...

> 
> 	OTOH it does not exactly make one smile happily if giving up
> 	parallel make only gets one to another equally disturbing
> 	error, this time a bit more verbose one, though.

You should smile, the world is becoming green again (spring is coming) 
so all the reasons to be happy :-), but if you consider it some thought 
it makes sense ;-)

> 
> 	Trying the 7.0 kernel build on 6.3-STABLE is probably the only
> 	way to repeat the failures which probably do not show themselves
> 	on vanilla 7.0 at all.
> 
> 	I hope this helps.

Can you please retry with make -j4 buildkernel KERNCONF=GENERIC ?

Thanks,
remko

-- 
/"\   Best regards,                      | remko@FreeBSD.org
\ /   Remko Lodder                       | remko@EFnet
  X    http://www.evilcoder.org/          |
/ \   ASCII Ribbon Campaign              | Against HTML Mail and News



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?47CC3A46.4080906>