On Tue, May 20, 2008 at 09:23:41AM +0100, Peter Denison wrote:
> On Mon, 19 May 2008, Steve Brown wrote:
>
> > Dirk Nehring wrote:
> >> [...]
> >> make[5]: Leaving directory 
> >> `/home/technik/dnehring/openwrt/build_dir/linux-brcm47xx/linux-2.6.25.1'
> >> make[5]: Entering directory 
> >> `/home/technik/dnehring/openwrt/build_dir/linux-brcm47xx/linux-2.6.25.1'
> >>   CHK     include/linux/version.h
> >>   CHK     include/linux/utsrelease.h
> >>   CALL    scripts/checksyscalls.sh
> >>   CHK     include/linux/compile.h
> >>   CC      drivers/ssb/driver_pcicore.o
> >> drivers/ssb/driver_pcicore.c: In function 'ssb_pcicore_fixup_pcibridge':
> >> drivers/ssb/driver_pcicore.c:314: error: implicit declaration of function 
> >> 'pcibios_enable_device'
> >> make[7]: *** [drivers/ssb/driver_pcicore.o] Error 1
> >> make[6]: *** [drivers/ssb] Error 2
> >> make[5]: *** [drivers] Error 2
> >> make[5]: Leaving directory 
> >> `/home/technik/dnehring/openwrt/build_dir/linux-brcm47xx/linux-2.6.25.1'
> >> make[4]: *** 
> >> [/home/technik/dnehring/openwrt/build_dir/linux-brcm47xx/linux-2.6.25.1/.image]
> >>  Error 2
> >> make[4]: Leaving directory 
> >> `/home/technik/dnehring/openwrt/target/linux/brcm47xx'
> >>
> >> And, iproute2 must be updated (works with iproute2-2.6.25.tar.bz2).
>
> Yes. I have a patch waiting to go in, but I'm waiting for confirmation (or
> time to test it) that it works under older kernels as well.
Me too, I just drop most of the included patches (I don't understand why
people are heavily patching stuff like new options for the htb queuing
class. This does not belong into the mainstream trunk, imho).

iproute2$ svn st
D      patches/003-iproute2-htb_overhead.patch
D      patches/004-darwin_fixes.patch
D      patches/000-debian_patches_3.patch
D      patches/005-flex-generated.patch
D      patches/006-iproute2-tc_esfq.patch
D      patches/002-iproute2-ipxfrm_no_sctp.patch
M      Makefile

Index: Makefile
===================================================================
--- Makefile    (revision 11218)
+++ Makefile    (working copy)
@@ -1,4 +1,4 @@
-#
+#
 # Copyright (C) 2006 OpenWrt.org
 #
 # This is free software, licensed under the GNU General Public License
 v2.
@@ -9,14 +9,14 @@
 include $(TOPDIR)/rules.mk

 PKG_NAME:=iproute2
-PKG_VERSION:=2.6.20-070313
+PKG_VERSION:=2.6.25
 PKG_RELEASE:=1

-PKG_SOURCE:=$(PKG_NAME)-$(PKG_VERSION).tar.gz
+PKG_SOURCE:=$(PKG_NAME)-$(PKG_VERSION).tar.bz2
 PKG_SOURCE_URL:=http://developer.osdl.org/dev/iproute2/download/
-PKG_MD5SUM:=7bc5883aadf740761fa2dd70b661e8cc
+PKG_MD5SUM:=5737bade2f5e03fad0e2c81da91e551e

-PKG_BUILD_DIR:=$(BUILD_DIR)/iproute-$(PKG_VERSION)
+PKG_BUILD_DIR:=$(BUILD_DIR)/iproute2-$(PKG_VERSION)

 include $(INCLUDE_DIR)/package.mk

@@ -49,7 +49,7 @@
        $(SED) "s,^LIBC_INCLUDE.*,LIBC_INCLUDE=$(STAGING_DIR)/include," \
                $(PKG_BUILD_DIR)/Makefile
        # For now disable compiling of the misc directory because it seems to 
fail
-       rm -rf $(PKG_BUILD_DIR)/misc
+       rm -rf $(PKG_BUILD_DIR)/misc
        $(SED) "s, misc,," $(PKG_BUILD_DIR)/Makefile
 endef


> >>
> > I added the following to config-2.6.25.
> >
> > CONFIG_PCI=y
> > CONFIG_SYS_HAS_EARLY_PRINTK=y
> > CONFIG_EARLY_PRINTK=y
> > # CONFIG_MEMSTICK is not set
> > # CONFIG_SGI_IP28 is not set
> > # CONFIG_THERMAL is not set
> >
> > The config_pci is to get the declaration for pcibios_enable_device to be
> > included. A better solution would be to add the missing ifdef around
> > that code.
>
> Not really. The platform has PCI, so that config setting should be set.
> Possibly there are some config dependencies that should go in so that you
> don't get a compilation failure.
>
> > I was pretty surprised at having to add the commented lines. If they
> > aren't there, you get questions from the 'make oldconfig' during the
> > kernel build.
>
> Sorry. If I understood exactly where to set things to get the right
> resultant .config, it would work more often.
>
> > The early printk stuff was to try to figure why it would boot on a
> > Buffalo whr-hp-54g, but not an Asus wl520gu. I still don't know why.
>
> CONFIG_BCM47XX sets CONFIG_SYS_HAS_EARLY_PRINTK, so you shouldn't need to
> set that. I can't work out where CONFIG_EARLY_PRINTK is set in my config,
> maybe by hand? That should probably go in.

Why not just taking the resulting .config from the kernel dir? It
includes all relevant stuff and it is up-to-date every time.

Dirk
_______________________________________________
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
http://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel

Reply via email to