MFC of ZFSv15

[Paul Mather]

On Sep 17, 2010, at 8:37 AM, Paul Mather wrote:

> On Sep 16, 2010, at 1:44 AM, Martin Matuska wrote:
> 
>> I have fixed the missing bits in r212688.
>> 
>> Thanks for the notice.
>> 
>> Dňa 15. 9. 2010 21:12, Xin LI  wrote / napísal(a):
>>> On 2010/09/15 11:30, Mike Tancsa wrote:
>>>> At 02:07 PM 9/15/2010, Pascal Stumpf wrote:
>>>>> First of all, a great thanks to mm@ and pjd@ for the excellent work on
>>>>> ZFS in FreeBSD. :) And especially for the MFC of v15 a few hours ago.
>>> 
>>> [...]
>>>> here too.  RELENG_8 AMD64.  The tinderboxes havent hit that branch yet
>>>> (<http://tinderbox.freebsd.org/>http://tinderbox.freebsd.org/), so it
>>>> will be a few hrs before they get to test RELENG_8
>>> [...]
>>>> -lsbuf  -lm -lnvpair -luutil -lutil
>>>> /usr/obj/usr/src/tmp/usr/lib/libzfs.so: undefined reference to `getmntent'
>>>> *** Error code 1
>>> 
>>> Sorry for that, it seems to be caused by a partial merge
>>> (cddl/compat/opensolaris/misc/mnttab.c).  mm@ is going to fix that ASAP.
>>> 
>>> Cheers,
> 
> I am getting a build failure on 8.1-STABLE:
> 
> =====
> [[...]]
> cc -c -O2 -pipe -fno-strict-aliasing -march=pentium4 -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 -fstack-protector -Werror  /usr/src/sys/kern/p1003_1b.c
> cc -c -O2 -pipe -fno-strict-aliasing -march=pentium4 -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 -fstack-protector -Werror  /usr/src/sys/kern/posix4_mib.c
> cc -c -O2 -pipe -fno-strict-aliasing -march=pentium4 -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 -fstack-protector -Werror  /usr/src/sys/kern/sched_ule.c
> cc1: warnings being treated as errors
> /usr/src/sys/kern/sched_ule.c: In function 'sched_switch':
> /usr/src/sys/kern/sched_ule.c:1807: warning: implicit declaration of function 'sched_pickcpu'
> /usr/src/sys/kern/sched_ule.c:1807: warning: nested extern declaration of 'sched_pickcpu'
> *** Error code 1
> 
> Stop in /usr/obj/usr/src/sys/BACKUP.
> *** Error code 1
> 
> Stop in /usr/src.
> *** Error code 1
> 
> Stop in /usr/src.
> =====
> 
> Unfortunately (for me, I guess), GENERIC will build successfully on this system.  It's only my custom kernel config file that is failing "make buildkernel."  The custom kernel config is GENERIC with various inapplicable drivers removed, plus some non-GENERIC things added in (such as ALTQ support options).  I am building via the standard "make buildworld" followed by "make buildkernel" method.  Can anyone spot anything obviously awry?  I've included my config file at the end.

Just to follow up myself, here: I added "options SMP" to my custom kernel config file and that allowed me successfully to finish "make buildkernel ..." without error.

So, is "options SMP" mandatory now on FreeBSD/i386 (even on uniprocessor systems), and, if so, shouldn't it be included in DEFAULTS?

Cheers,

Paul.