linux-next: triage for April 19, 2012

The "undefined reference to `vga_default_device'" stuff appears to
still be unresolved, as it was in the Apr 19 fails still. :-(
[ https://lkml.org/lkml/2012/4/18/157 ]

The samples/seccomp/bpf-direct.c stuff had a fix posted, but not in time for Apr19 tree. Hopefully it is gone tomorrow... [ http://marc.info/?l=linux-next&m=133479674429442&w=2 ]

The parisc got borked by some kvm header shuffle it seems? Now complaining about "file 'asm-generic/kvm_para.h' is not exported" [ http://kisskb.ellerman.id.au/kisskb/buildresult/6137786/ ]

The gfs2 stuff still has issues, but Apr 19th should be the last day, as a fix was posted to follow up my bisect. [ http://marc.info/?l=linux-next&m=133479890929994&w=2 ]

The frv defconfig got its 1st ever linux-next passing build!

The recent cris breakage was fixed with the "-traditional" removal.

The xtensa lost its BUG issue, but still has a final link issue.

---------------------

New breakage since last report:
parisc:defconfig

Randconfig fails that may or may not be new (by their very nature):
i386: drivers/isdn/hardware/avm/b1.o copy_from_user() buffer
size is not provably correct
(http://kisskb.ellerman.id.au/kisskb/buildresult/6137763/)

Builds that are fixed since last report:
frv:defconfig
cris:several

----------------------

Note1: The randconfig fails are only listed once and not tracked in the below listing.

Note2: Build fails that are perpetual and/or unlikely to ever get fixed are not tracked here.

Note3: Issues that Stephen finds and reverts/avoids before kicking off the all-arch/all-config builds are also not tracked here.

Note4: I'm also not tracking the kallsyms extra pass on ARM on a per board basis anymore. It has nothing to do with the board. Why? See: http://marc.info/?l=linux-next&m=133267456809502

----------------------

Latest results at:
http://kisskb.ellerman.id.au/kisskb/branch/9/

Failures as of April 19th, 2012:
===============================

AM33:asb2303_defconfig,asb2364_defconfig

	when:	Mar 15
	why:	cc1: error: unrecognized command line option '-mmem-funcs'
	status:	Possible fix from TonyB sent for review.
	fix:	http://marc.info/?l=linux-kernel&m=133185020313818&w=2
	ref:	http://kisskb.ellerman.id.au/kisskb/buildresult/5878400/

----------------

ARM:mackerel_defconfig,ag5evm_defconfig

	when:	Apr 19 (possibly earlier?)
	why:	pfc-sh7372.c: undefined reference to `mmc_detect_change'
	why:	pfc-sh73a0.c: undefined reference to `mmc_detect_change'
	status:	
	fix:	
	ref:	http://article.gmane.org/gmane.linux.ports.sh.devel/14412
	ref:	http://article.gmane.org/gmane.linux.ports.sh.devel/14413
	ref:	http://kisskb.ellerman.id.au/kisskb/buildresult/6131930/
	ref:	http://kisskb.ellerman.id.au/kisskb/buildresult/6131908/

ARM:bcmring_defconfig
when: Apr 19 (possibly earlier?)
why: core.c: macro "AMBA_APB_DEVICE" requires 6 args, but only 5 given
status: fixed, pending feed into linux next.
fix:
ref: http://www.spinics.net/lists/arm-kernel/msg169724.html

ARM:ezx_defconfig
when: Apr 19 (possibly earlier?)
why: ezx-pcap.c: implicit declaration of function 'irq_to_gpio'
status:
fix:
ref: http://kisskb.ellerman.id.au/kisskb/buildresult/6131924/

ARM:mini2440_defconfig
when: Apr 19 (possibly earlier?)
why: arm-unknown-linux-gnueabi-ld: no machine record defined
status:
fix:
ref: http://kisskb.ellerman.id.au/kisskb/buildresult/6131932/

ARM:spear6xx_defconfig
when: Apr 19 (possibly earlier?)
why: asm/timex.h: fatal error: mach/timex.h: No such file
status:
fix:
ref:

ARM:netx_defconfig,pnx4008_defconfig

	when:	Mar 14
	why:	net/core/pktgen.c: can't find a register in class 'GENERAL_REGS'
	status:	
	fix:	
	ref:	http://kisskb.ellerman.id.au/kisskb/buildresult/5869486/
	ref:	https://lkml.org/lkml/2012/3/13/670

ARM:ixp4xx_defconfig

	when:	Oct 26
	why:	gpio.h: implicit declaration of function '__gpio_get_value'
	status:	fixed (Imre Kaloz) - but possibly needs rework?
	fix:	https://lkml.org/lkml/2012/2/8/288
	ref:	https://lkml.org/lkml/2012/2/8/287

ARM:raumfeld_defconfig

	when:	Nov 23
	why:	eeti_ts.c: implicit declaration of function 'irq_to_gpio'
	status:	perhaps broken since 2010?  See ref below.
	fix:	
	ref:	http://www.gossamer-threads.com/lists/linux/kernel/1291618?do=post_view_threaded#1291618

----------------

AVR32:allnoconfig

	when:	May 26 2010, 20:49
	why:	at32_map_usart: undefined reference to `atmel_default_console_device'
	status:
	fix:
	ref:

---------------

bfin: 50% of builds

	when:	Mar 31
	why:	"Error: pcrel too far BFD_RELOC_BFIN_10" from kernel/time/timekeeping.c
	status:	unknown, binutils issue?
	fix:
	ref:	http://kisskb.ellerman.id.au/kisskb/buildresult/6087324/

---------------

cris:allyesconfig,allmodconfig

	when:	Apr 3
	why:	too many to list here
	ref:	http://kisskb.ellerman.id.au/kisskb/buildresult/6035250/
	ref:	http://kisskb.ellerman.id.au/kisskb/buildresult/6035249/

----------------

i386:allmodconfig

	when:	Apr 17
	why:	samples/seccomp/bpf-direct.c:47:2: error: #error Unsupported platform
	status:	reported
	ref:	http://marc.info/?l=linux-kernel&m=133471624530782&w=2

----------------

h8300:defconfig

	when:	Mar 15
	why:	time.c: implicit declaration of function 'get_irq_regs'
	why:	also seems gcc suffers an ICE
	status:	ICE bug lodged with gcc folks (TonyB)
	ref:	http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52598
	ref:	http://kisskb.ellerman.id.au/kisskb/buildresult/5884668/

----------------

m32r:m32700ut.smp_defconfig

	when:	Mar 15
	why:	No rule to make target `arch/m32r/boot/compressed/vmlinux.bin
	ref:	http://kisskb.ellerman.id.au/kisskb/buildresult/5884670/

----------------

mips:ip27_defconfig

	when:	Apr 18
	why: iwlwifi/iwl-core.h: static decl of 'iwl_alloc_traffic_mem' follows non-static
	ref:	http://kisskb.ellerman.id.au/kisskb/buildresult/6131791/

mips:allmodconfig

	when:	Apr 17
	why:	sound/oss/soundcard.c: 'MAX_DMA_CHANNELS' undeclared
	status: one of many fixes mips needs...
	ref:	http://kisskb.ellerman.id.au/kisskb/buildresult/6091660/

----------------

parisc:defconfig

	when:	Apr 20
	why:	file 'asm-generic/kvm_para.h' is not exported
	ref:	http://kisskb.ellerman.id.au/kisskb/buildresult/6137786/

parisc:allmodconfig

	when:	?
	why:	include/linux/log2.h: implicit declaration of function 'fls'
	status:	reported by PaulG, WIP by David Howells 
	ref:	https://lkml.org/lkml/2012/2/29/409
	ref:	https://lkml.org/lkml/2012/3/6/228

parisc64:a500_defconfig

	when:	?
	why:	include/linux/log2.h: implicit declaration of function 'fls'
	status:	reported by PaulG, WIP by David Howells 
	ref:	https://lkml.org/lkml/2012/2/29/409
	ref:	https://lkml.org/lkml/2012/3/6/228

------------------

sparc/m68k:all

	when:	Apr 18
	why:	fs/gfs2/lops.c:379:42: error: request for member 'virtual' in something not a structure or union
	status:	reported/bisected by PaulG
	ref:	http://marc.info/?l=linux-next&m=133479890929994&w=2

------------------

xtensa:defconfig

	when:	Apr 20
	why:	kallsyms.c:(.text+0x1eb44): undefined reference to `ack_bad_irq'
	ref:	http://kisskb.ellerman.id.au/kisskb/buildresult/6138090/

Not only parisc.

This breakage has now entered mainline:

parisc deconfig http://kisskb.ellerman.id.au/kisskb/buildresult/6365677/
m68k allmodconfig: http://kisskb.ellerman.id.au/kisskb/buildresult/6365681/

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

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

See also this thread:

http://www.spinics.net/lists/kvm/msg72785.html

Paul.

Hi Paul,

I saw it. But it doesn't contain a fix for this, does it?

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

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

Does the following patch help?

I was unable to find a cross compiler, so I can't test it myself.

From: Avi Kivity <[email protected]>
Date: Mon, 28 May 2012 17:35:22 +0300
Subject: [PATCH] KVM: Export asm-generic/kvm_para.h

Prevents build failures on non-KVM archs.

---
 include/asm-generic/Kbuild |    1 +
 1 file changed, 1 insertion(+)

index 53f91b1..2c85a0f 100644
@@ -8,6 +8,7 @@ header-y += int-ll64.h
 header-y += ioctl.h
 header-y += ioctls.h
 header-y += ipcbuf.h
+header-y += kvm_para.h
 header-y += mman-common.h
 header-y += mman.h
 header-y += msgbuf.h

error compiling committee.c: too many arguments to function

Hi Avi,

Thanks, that fixes it!

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

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

Hi Kbuild people,

I just noticed include/asm-generic/Kbuild.asm already had

ifneq ($(wildcard $(srctree)/arch/$(SRCARCH)/include/asm/kvm_para.h \
                  $(srctree)/include/asm-$(SRCARCH)/kvm_para.h),)

header-y += kvm_para.h
endif

but this doesn't seem to work.

Kbuild people: which one is correct?

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

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

Can you say what arch/logs had this fail in it?

If its only m68k then hopefully Geert merges his patch to fix it.

Dave.

I never saw this happen on m68k.

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

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