Date: Sun, 18 Oct 2009 09:24:42 -0600 (MDT) From: "M. Warner Losh" <imp@bsdimp.com> To: alancyang@gmail.com Cc: freebsd-mips@FreeBSD.org Subject: Re: Cavium port Message-ID: <20091018.092442.329271515.imp@bsdimp.com> In-Reply-To: <290865fd0910161902y626f8375m5d46881f4883c1ec@mail.gmail.com> References: <290865fd0910161801x35bbdcd2hd77626d4dfe2465a@mail.gmail.com> <290865fd0910161856l3d988d0bqf4c600bd76b37037@mail.gmail.com> <290865fd0910161902y626f8375m5d46881f4883c1ec@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
In message: <290865fd0910161902y626f8375m5d46881f4883c1ec@mail.gmail.co=
m>
alan yang <alancyang@gmail.com> writes:
: but ran into another failure, wonder there is a quick work around:
: ---
: -c -O -pipe -std=3Dc99 -g -Wall -Wredundant-decls -Wnested-externs
: -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -Winline
: -Wcast-qual -Wundef -Wno-pointer-sign -fformat-extensions -nostdinc
: -I. -I/usr/home/alan/genova/mips/sys
: -I/usr/home/alan/genova/mips/sys/contrib/altq -D_KERNEL
: -DHAVE_KERNEL_OPTION_HEADERS -include opt_global.h -fno-common
: -finline-limit=3D8000 --param inline-unit-growth=3D100 --param
: large-function-growth=3D1000 -EB -fno-pic -mno-abicalls -G0 -EB
: -march=3Dmips64 -mabi=3D64 -msoft-float -ffreestanding
: /usr/home/alan/genova/mips/sys/mips/mips/trap.c
: /usr/home/alan/genova/mips/sys/mips/mips/trap.c:1063: error:
: conflicting types for 'MipsEmulateBranch'
: ./machine/md_var.h:51: error: previous declaration of
: 'MipsEmulateBranch' was here
: *** Error code 1
Hmmm... Looks like there were some changes that were stomped on by
the latest MFC. We'll have to investigate why that was... In the
mean time, I've restored them and fixed a few new issues and committed
the results. You can at least build OCTEON1 now. I've not tried to
run it, and I'd expect that it will die somewhere in the context
switch for the first task, if not sooner....
Warner
: On Fri, Oct 16, 2009 at 6:56 PM, alan yang <alancyang@gmail.com> wrot=
e:
: > seems to find the place: =A0kern.pre.mk file to comment out WERROR =
as a
: > work around and redo buildkernel.
: >
: > On Fri, Oct 16, 2009 at 6:01 PM, alan yang <alancyang@gmail.com> wr=
ote:
: >> thanks Warner! =A0got that.
: >>
: >> the buildkernel failed on some files due to warning of cast pointe=
r
: >> and -Werror flag. =A0wonder how to remove -Werror, should regen
: >> toolchain / Makefile... ?
: >>
: >> thanks again!!
: >>
: >> On Fri, Oct 16, 2009 at 3:24 PM, M. Warner Losh <imp@bsdimp.com> w=
rote:
: >>> Looks like you are trying to use the old-school way of building t=
he
: >>> kernel (which is what I do all the time). =A0Problem is that you =
need to
: >>> have the right compilers in your path to do this.
: >>>
: >>> Here's what I do:
: >>>
: >>> setenv TARGET mips
: >>> setenv TARGET_BIG_ENDIAN t
: >>> make kernel-toolchain
: >>> make buildkernel KERNCONF=3DOCTEON1
: >>>
: >>> and then to do an incremental build:
: >>>
: >>> make buildkernel KERNFAST=3DOCTEON1
: >>>
: >>> I should turn this into a readme file.
: >>>
: >>> Warner
: >>>
: >>>
: >>>
: >>> In message: <290865fd0910161428g45ab74a0x272c9b9674c9d5fd@mail.gm=
ail.com>
: >>> =A0 =A0 =A0 =A0 =A0 =A0alan yang <alancyang@gmail.com> writes:
: >>> : Hello,
: >>> :
: >>> : I checked out from svn repo the project/mips and try to build k=
ernel
: >>> : for OCTEON, but ran into the following errors. =A0Wonder any re=
adme
: >>> : available describes procedures, appreciate people could shed so=
me
: >>> : light.
: >>> :
: >>> : Thanks in advance.
: >>> : ---
: >>> :
: >>> : tsu# config OCTEON1
: >>> : Kernel build directory is ../compile/OCTEON1
: >>> : Don't forget to do ``make cleandepend && make depend''
: >>> : Fujitsu# cd ../compile/OCTEON1/
: >>> : Fujitsu# make cleandepend
: >>> : rm -f .depend machine
: >>> : cd ../../../modules;
: >>> : MAKEOBJDIRPREFIX=3D/usr/home/alan/genova/mips/sys/mips/compile/=
OCTEON1/modules
: >>> : KMODDIR=3D/boot/kernel MODULES_OVERRIDE=3D"""" DEBUG_FLAGS=3D"-=
g"
: >>> : MACHINE=3Dmips MACHINE_ARCH=3Dmips
: >>> : KERNBUILDDIR=3D"/usr/home/alan/genova/mips/sys/mips/compile/OCT=
EON1"
: >>> : SYSDIR=3D"/usr/home/alan/genova/mips/sys/mips/compile/OCTEON1/.=
./../.."
: >>> : make =A0cleandepend
: >>> : Fujitsu# make depend
: >>> : machine -> ../../../mips/include
: >>> : cc -c -O -pipe -std=3Dc99 -g -Wall -Wredundant-decls -Wnested-e=
xterns
: >>> : -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -Winli=
ne
: >>> : -Wcast-qual -Wundef -Wno-pointer-sign -fformat-extensions -nost=
dinc
: >>> : -I. -I../../.. -I../../../contrib/altq -I../../../contrib/ipfil=
ter
: >>> : -I../../../contrib/pf -I../../../dev/ath -I../../../dev/ath/ath=
_hal
: >>> : -I../../../contrib/ngatm -I../../../dev/twa
: >>> : -I../../../gnu/fs/xfs/FreeBSD -I../../../gnu/fs/xfs/FreeBSD/sup=
port
: >>> : -I../../../gnu/fs/xfs -I../../../contrib/opensolaris/compat
: >>> : -I../../../dev/cxgb -D_KERNEL -DHAVE_KERNEL_OPTION_HEADERS -inc=
lude
: >>> : opt_global.h -finline-limit=3D8000 --param inline-unit-growth=3D=
100
: >>> : --param large-function-growth=3D1000 -EB -fno-pic -mno-abicalls=
-G0 -EB
: >>> : -march=3Dmips64 -mabi=3D64 -msoft-float -ffreestanding
: >>> : ../../../mips/mips/genassym.c
: >>> : cc: unrecognized option '-EB'
: >>> : cc: unrecognized option '-G0'
: >>> : cc: unrecognized option '-EB'
: >>> : cc1: error: unrecognized command line option "-mno-abicalls"
: >>> : cc1: error: unrecognized command line option "-mabi=3D64"
: >>> : ../../../mips/mips/genassym.c:1: error: bad value (mips64) for =
-march=3D switch
: >>> : ../../../mips/mips/genassym.c:1: error: bad value (mips64) for =
-mtune=3D switch
: >>> : *** Error code 1
: >>> :
: >>> :
: >>> :
: >>> : On Wed, Oct 14, 2009 at 3:45 PM, alan yang <alancyang@gmail.com=
> wrote:
: >>> : > Thanks Warner for the link, i checkout the project mips from =
svn repo,
: >>> : > and wonder is there any readme like for procedures to build/r=
un on
: >>> : > CN8560 base eval board as the target.
: >>> : >
: >>> : > Thanks for shedding the light...
: >>> : >
: >>> : >
: >>> : > On Tue, Oct 13, 2009 at 9:04 PM, M. Warner Losh <imp@bsdimp.c=
om> wrote:
: >>> : >> In message: <290865fd0910132051l6a4a6280id401cab8ef5a1c0c@ma=
il.gmail.com>
: >>> : >> =A0 =A0 =A0 =A0 =A0 =A0alan yang <alancyang@gmail.com> write=
s:
: >>> : >> : Hi, Warner,
: >>> : >> :
: >>> : >> : Thanks for the info. =A0I wonder this port could be transf=
erred, to
: >>> : >> : allow me to participate and continue ..., i know this is q=
uite naive
: >>> : >> : and don't know will this have issue.
: >>> : >>
: >>> : >> The current port is in the svn repo under projects/mips.
: >>> : >>
: >>> : >> Warner
: >>> : >>
: >>> : >> : On Tue, Oct 13, 2009 at 3:45 PM, M. Warner Losh <imp@bsdim=
p.com> wrote:
: >>> : >> : > In message: <290865fd0910131105i6daa9a17y5f6fe2a4447d723=
3@mail.gmail.com>
: >>> : >> : > =A0 =A0 =A0 =A0 =A0 =A0alan yang <alancyang@gmail.com> w=
rites:
: >>> : >> : > : last email thread is about 1 year ago, and i wonder if=
things changed
: >>> : >> : > : that would allow your code to be available and / or pr=
ovide some
: >>> : >> : > : guidance on porting...
: >>> : >> : >
: >>> : >> : > Cavium released a port to me earlier this year. =A0It is=
substantially
: >>> : >> : > similar to the port I started from that I referred to in=
this post.
: >>> : >> : > I've been making good progress, but the efforts have bee=
n stalled by
: >>> : >> : > my job search...
: >>> : >> : >
: >>> : >> : > Warner
: >>> : >> : >
: >>> : >> : >
: >>> : >> : > : many thanks!!
: >>> : >> : > :
: >>> : >> : > : On Thu, Oct 9, 2008 at 12:52 PM, M. Warner Losh <imp@b=
sdimp.com> wrote:
: >>> : >> : > : > In message: <86068e730810091235q3f41b5a3w4dbc38fb231=
64483@mail.gmail.com>
: >>> : >> : > : > =A0 =A0 =A0 =A0 =A0 =A0"Jerry Toung" <jrytoung@gmail=
.com> writes:
: >>> : >> : > : > : On Wed, Oct 8, 2008 at 7:54 PM, M. Warner Losh <im=
p@bsdimp.com> wrote:
: >>> : >> : > : > :
: >>> : >> : > : > : >
: >>> : >> : > : > : >
: >>> : >> : > : > : >
: >>> : >> : > : > : > The issues aren't technical, but the short versi=
on is that I've not
: >>> : >> : > : > : > received permission from the copyright owner to =
release the code.
: >>> : >> : > : > : >
: >>> : >> : > : > : >
: >>> : >> : > : > : > Hi Warner,
: >>> : >> : > : > : are you able to run the build on FreeBSD or is it =
cross-built ?
: >>> : >> : > : >
: >>> : >> : > : > Both, although the version I'm running now is cross-=
built. =A0I'll be
: >>> : >> : > : > doing native builds soon enough...
: >>> : >> : > : >
: >>> : >> : > : > : what version of your FreeBSD is running on CN38xx =
?
: >>> : >> : > : >
: >>> : >> : > : > FreeBSD -current as of Tuesday. =A0Prior versions of=
-current have
: >>> : >> : > : > worked as well.
: >>> : >> : > : >
: >>> : >> : > : > Warner
: >>> : >> : > : > _______________________________________________
: >>> : >> : > : > freebsd-mips@freebsd.org mailing list
: >>> : >> : > : > http://lists.freebsd.org/mailman/listinfo/freebsd-mi=
ps
: >>> : >> : > : > To unsubscribe, send any mail to "freebsd-mips-unsub=
scribe@freebsd.org"
: >>> : >> : > : >
: >>> : >> : > :
: >>> : >> : > :
: >>> : >> : >
: >>> : >> :
: >>> : >> :
: >>> : >>
: >>> : >
: >>> : _______________________________________________
: >>> : freebsd-mips@freebsd.org mailing list
: >>> : http://lists.freebsd.org/mailman/listinfo/freebsd-mips
: >>> : To unsubscribe, send any mail to "freebsd-mips-unsubscribe@free=
bsd.org"
: >>> :
: >>> :
: >>>
: >>
: >
: _______________________________________________
: freebsd-mips@freebsd.org mailing list
: http://lists.freebsd.org/mailman/listinfo/freebsd-mips
: To unsubscribe, send any mail to "freebsd-mips-unsubscribe@freebsd.or=
g"
: =
: =
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20091018.092442.329271515.imp>