Re: Build regressions/improvements in v4.7-rc1
On Wed, Jun 01, 2016 at 06:18:02PM -0400, Rich Felker wrote: > On Mon, May 30, 2016 at 09:34:07AM +0200, Geert Uytterhoeven wrote: > > > + /home/kisskb/slave/src/arch/sh/kernel/setup.c: error: implicit > > > declaration of function 'early_init_dt_scan' > > > [-Werror=implicit-function-declaration]: => 256:2 > > > > sh-randconfig > > Nice find. I think I'm incorrectly checking for CONFIG_OF in places > that should actually be checking for CONFIG_OF_FLATTREE. Does that > look right to you? Nothing breaks in actual configurations that make > sense now, only if you enable OF manually without selecting > SH_DEVICE_TREE as your board. > > If you can confirm that my assessment is correct I'll include a fix in > my (now-late) pull request. BTW it seems a bigger version of the same issue affects my J2 support patch series: CONFIG_CPU_J2 enables code that performs an early FDT scan to obtain cache information, so it needs to select OF and OF_FLATTREE to avoid build errors. Eventually none of this will matter once we drop legacy support and make it so SUPERH implies OF and OF_EARLY_FLATTREE. Rich
Re: Build regressions/improvements in v4.7-rc1
On Mon, May 30, 2016 at 09:34:07AM +0200, Geert Uytterhoeven wrote: > > + /home/kisskb/slave/src/arch/sh/kernel/setup.c: error: implicit > > declaration of function 'early_init_dt_scan' > > [-Werror=implicit-function-declaration]: => 256:2 > > sh-randconfig Nice find. I think I'm incorrectly checking for CONFIG_OF in places that should actually be checking for CONFIG_OF_FLATTREE. Does that look right to you? Nothing breaks in actual configurations that make sense now, only if you enable OF manually without selecting SH_DEVICE_TREE as your board. If you can confirm that my assessment is correct I'll include a fix in my (now-late) pull request. Rich
Re: Build regressions/improvements in v4.7-rc1
On Mon, May 30, 2016 at 9:28 AM, Geert Uytterhoeven
wrote:
> + /home/kisskb/slave/src/arch/powerpc/include/asm/nohash/64/pgalloc.h:
> error: implicit declaration of function 'pte_fragment_fre'
> [-Werror=implicit-function-declaration]/home/kisskb/slave/src/arch/powerpc/mm/pgtable.c:
> At top level:: => 175:2
> + /home/kisskb/slave/src/arch/powerpc/include/asm/nohash/64/pgalloc.h:
> error: implicit declaration of function 'pte_fragment_fre'
> [-Werror=implicit-function-declaration]: => 175:2
> + /home/kisskb/slave/src/arch/powerpc/kernel/pci_dn.c: error: 'struct
> pci_dn' has no member named 'edev': => 272:8
> + /home/kisskb/slave/src/arch/powerpc/kernel/pci_dn.c: error: dereferencing
> pointer to incomplete type: => 215:7
> + /home/kisskb/slave/src/arch/powerpc/kernel/pci_dn.c: error: implicit
> declaration of function 'eeh_dev_init'
> [-Werror=implicit-function-declaration]: => 212:3
> + /home/kisskb/slave/src/arch/powerpc/mm/init_64.c: error: 'PUD_TABLE_SIZE'
> undeclared (first use in this function): => 90:18
powerpc-randconfig
> + /home/kisskb/slave/src/arch/sh/kernel/setup.c: error: implicit
> declaration of function 'early_init_dt_scan'
> [-Werror=implicit-function-declaration]: => 256:2
sh-randconfig
> + error: fou6.c: undefined reference to `ip6_tnl_encap_add_ops': =>
> .init.text+0x4b0f), .init.text+0x4b24)
> + error: fou6.c: undefined reference to `ip6_tnl_encap_del_ops': =>
> .exit.text+0x438), .init.text+0x4b39), .exit.text+0x447)
i386-randconfig (fix from Arnd just applied by DaveM).
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
