Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1487098 > unrolled thread

linux-next: build failure after merge of the akpm-current tree

Back to article view | Back to linux.kernel

Contents

  linux-next: build failure after merge of the akpm-current tree Stephen Rothwell <sfr@canb.auug.org.au> - 2016-09-20 08:50 +0200
    Re: linux-next: build failure after merge of the akpm-current tree Stephen Rothwell <sfr@canb.auug.org.au> - 2016-09-20 09:40 +0200
      Re: linux-next: build failure after merge of the akpm-current tree Stephen Rothwell <sfr@canb.auug.org.au> - 2016-09-21 01:20 +0200
        Re: linux-next: build failure after merge of the akpm-current tree Michael Ellerman <mpe@ellerman.id.au> - 2016-09-21 02:30 +0200
          Re: linux-next: build failure after merge of the akpm-current tree Thiago Jung Bauermann <bauerman@linux.vnet.ibm.com> - 2016-09-21 03:20 +0200
          Re: linux-next: build failure after merge of the akpm-current tree Stephen Rothwell <sfr@canb.auug.org.au> - 2016-09-21 04:10 +0200
        Re: linux-next: build failure after merge of the akpm-current tree Stephen Rothwell <sfr@canb.auug.org.au> - 2016-09-21 08:10 +0200
      Re: linux-next: build failure after merge of the akpm-current tree Thiago Jung Bauermann <bauerman@linux.vnet.ibm.com> - 2016-09-21 04:30 +0200

#1487098 — linux-next: build failure after merge of the akpm-current tree

Hi Andrew,

After merging the akpm-current tree, today's linux-next build (powerpc
ppc64_defconfig) failed like this:

powerpc-linux-gcc: error: unrecognized command line option '-m'
scripts/Makefile.build:290: recipe for target 'arch/powerpc/purgatory/printf.o' failed

and a few more like that.

Caused by commit

  8635b1186486 ("powerpc: add purgatory for kexec_file_load implementation")

I have disabled KEXEC_FILE on powerpc again for now.

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Mon, 5 Sep 2016 19:12:42 +1000
Subject: [PATCH] disable KEXEC_FILE on powerpc for now

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 arch/powerpc/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 41300c3a1bfe..86ea07d7ead2 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -467,6 +467,7 @@ config KEXEC_FILE
 	depends on PPC64
 	depends on CRYPTO=y
 	depends on CRYPTO_SHA256=y
+	depends on BROKEN
 	help
 	  This is a new version of the kexec system call. This call is
 	  file based and takes in file descriptors as system call arguments
-- 
2.8.1

-- 
Cheers,
Stephen Rothwell

[toc] | [next] | [standalone]

#1487119

Hi all,

On Tue, 20 Sep 2016 16:44:14 +1000 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>
> After merging the akpm-current tree, today's linux-next build (powerpc
> ppc64_defconfig) failed like this:
> 
> powerpc-linux-gcc: error: unrecognized command line option '-m'
> scripts/Makefile.build:290: recipe for target 'arch/powerpc/purgatory/printf.o' failed
> 
> and a few more like that.

OK, this is (most likely) caused by the CONFIG_WORD_SIZE -> BITS change
interacting with the KEXEC_FILE changes. Tomorrow I will do a better
merge conflict resolution patch that fixes all those.

-- 
Cheers,
Stephen Rothwell

[toc] | [prev] | [next] | [standalone]

#1487717

Hi Michael,

On Tue, 20 Sep 2016 22:00:32 +1000 Michael Ellerman <mpe@ellerman.id.au> wrote:
>
> Ah yep looks like that's the problem, patch below should fix it?

Yeah, I am just going to (logically) run "sed 's/CONFIG_WORD_SIZE/BITS/'"
over the tree during the merge of the apm-current tree today.

> I think I'd actually prefer it if purgatory didn't redefine the CFLAGS
> from scratch, so I'll see if Thiago can do that and send a new version.

That could be better, but there are still some additions of
CONFIG_WORD_SIZE elsewhere :-(

-- 
Cheers,
Stephen Rothwell

[toc] | [prev] | [next] | [standalone]

#1487737

Stephen Rothwell <sfr@canb.auug.org.au> writes:

> On Tue, 20 Sep 2016 22:00:32 +1000 Michael Ellerman <mpe@ellerman.id.au> wrote:
>>
>> Ah yep looks like that's the problem, patch below should fix it?
>
> Yeah, I am just going to (logically) run "sed 's/CONFIG_WORD_SIZE/BITS/'"
> over the tree during the merge of the apm-current tree today.
>
>> I think I'd actually prefer it if purgatory didn't redefine the CFLAGS
>> from scratch, so I'll see if Thiago can do that and send a new version.
>
> That could be better, but there are still some additions of
> CONFIG_WORD_SIZE elsewhere :-(

I don't see any others in yesterday's next?

cheers

[toc] | [prev] | [next] | [standalone]

#1487748

Am Mittwoch, 21 September 2016, 10:27:46 schrieb Michael Ellerman:
> Stephen Rothwell <sfr@canb.auug.org.au> writes:
> > On Tue, 20 Sep 2016 22:00:32 +1000 Michael Ellerman <mpe@ellerman.id.au> 
wrote:
> >> Ah yep looks like that's the problem, patch below should fix it?
> > 
> > Yeah, I am just going to (logically) run "sed
> > 's/CONFIG_WORD_SIZE/BITS/'"
> > over the tree during the merge of the apm-current tree today.
> > 
> >> I think I'd actually prefer it if purgatory didn't redefine the CFLAGS
> >> from scratch, so I'll see if Thiago can do that and send a new version.
> > 
> > That could be better, but there are still some additions of
> > CONFIG_WORD_SIZE elsewhere :-(
> 
> I don't see any others in yesterday's next?

This kbuild failure is one case:

https://lists.ozlabs.org/pipermail/linuxppc-dev/2016-September/148898.html

elf_util_64.o is only built if CONFIG_WORD_SIZE=64. This is affects the 
bisectabilty of many patches in the kexec_file_load series. Should I post a 
new version rebased on powerpc/next?

-- 
[]'s
Thiago Jung Bauermann
IBM Linux Technology Center

[toc] | [prev] | [next] | [standalone]

#1487763

Hi Michael,

On Wed, 21 Sep 2016 10:27:46 +1000 Michael Ellerman <mpe@ellerman.id.au> wrote:
>
> Stephen Rothwell <sfr@canb.auug.org.au> writes:
> 
> > On Tue, 20 Sep 2016 22:00:32 +1000 Michael Ellerman <mpe@ellerman.id.au> wrote:  
> >>
> >> Ah yep looks like that's the problem, patch below should fix it?  
> >
> > Yeah, I am just going to (logically) run "sed 's/CONFIG_WORD_SIZE/BITS/'"
> > over the tree during the merge of the apm-current tree today.
> >  
> >> I think I'd actually prefer it if purgatory didn't redefine the CFLAGS
> >> from scratch, so I'll see if Thiago can do that and send a new version.  
> >
> > That could be better, but there are still some additions of
> > CONFIG_WORD_SIZE elsewhere :-(  
> 
> I don't see any others in yesterday's next?

I fixed the other one I found as it caused a build failure as well but
was more obvious (and I had had time to think about what was happening).

-- 
Cheers,
Stephen Rothwell

[toc] | [prev] | [next] | [standalone]

#1487829

Hi all,

On Wed, 21 Sep 2016 09:16:22 +1000 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>
> On Tue, 20 Sep 2016 22:00:32 +1000 Michael Ellerman <mpe@ellerman.id.au> wrote:
> >
> > Ah yep looks like that's the problem, patch below should fix it?  
> 
> Yeah, I am just going to (logically) run "sed 's/CONFIG_WORD_SIZE/BITS/'"
> over the tree during the merge of the apm-current tree today.
> 
> > I think I'd actually prefer it if purgatory didn't redefine the CFLAGS
> > from scratch, so I'll see if Thiago can do that and send a new version.  
> 
> That could be better, but there are still some additions of
> CONFIG_WORD_SIZE elsewhere :-(

So here is the relevant parts of the merge resolution for merging the
akpm-current tree today.

diff --cc arch/powerpc/kernel/Makefile
index 1925341dbb9c,d12a84003283..e59ed6af0523
--- a/arch/powerpc/kernel/Makefile
+++ b/arch/powerpc/kernel/Makefile
@@@ -104,11 -104,12 +104,12 @@@ obj-$(CONFIG_STACKTRACE)	+= stacktrace.
  obj-$(CONFIG_SWIOTLB)		+= dma-swiotlb.o
  
  pci64-$(CONFIG_PPC64)		+= pci_dn.o pci-hotplug.o isa-bridge.o
 -obj-$(CONFIG_PCI)		+= pci_$(CONFIG_WORD_SIZE).o $(pci64-y) \
 +obj-$(CONFIG_PCI)		+= pci_$(BITS).o $(pci64-y) \
  				   pci-common.o pci_of_scan.o
  obj-$(CONFIG_PCI_MSI)		+= msi.o
- obj-$(CONFIG_KEXEC)		+= machine_kexec.o crash.o \
+ obj-$(CONFIG_KEXEC_CORE)	+= machine_kexec.o crash.o \
 -				   machine_kexec_$(CONFIG_WORD_SIZE).o
 -obj-$(CONFIG_KEXEC_FILE)	+= kexec_elf_$(CONFIG_WORD_SIZE).o
 +				   machine_kexec_$(BITS).o
++obj-$(CONFIG_KEXEC_FILE)	+= kexec_elf_$(BITS).o
  obj-$(CONFIG_AUDIT)		+= audit.o
  obj64-$(CONFIG_AUDIT)		+= compat_audit.o
  
@@@ -123,6 -124,12 +124,12 @@@ ifneq ($(CONFIG_PPC_INDIRECT_PIO),y
  obj-y				+= iomap.o
  endif
  
+ ifneq ($(CONFIG_MODULES)$(CONFIG_KEXEC_FILE),)
 -ifeq ($(CONFIG_WORD_SIZE),64)
++ifeq ($(BITS),64)
+ obj-y				+= elf_util.o elf_util_64.o
+ endif
+ endif
+ 
  obj64-$(CONFIG_PPC_TRANSACTIONAL_MEM)	+= tm.o
  
  obj-$(CONFIG_PPC64)		+= $(obj64-y)
diff --cc arch/powerpc/purgatory/Makefile
index 000000000000,fea0308c7f01..bc3b9dc6eead
mode 000000,100644..100644
--- a/arch/powerpc/purgatory/Makefile
+++ b/arch/powerpc/purgatory/Makefile
@@@ -1,0 -1,48 +1,48 @@@
+ OBJECT_FILES_NON_STANDARD := y
+ 
+ purgatory-y := purgatory.o printf.o string.o v2wrap.o hvCall.o \
+ 		purgatory-ppc64.o console-ppc64.o crashdump_backup.o \
+ 		crtsavres.o sha256.o
+ 
+ targets += $(purgatory-y)
+ PURGATORY_OBJS = $(addprefix $(obj)/,$(purgatory-y))
+ 
+ LDFLAGS_purgatory.ro := -e purgatory_start -r --no-undefined -nostartfiles \
+ 			-nostdlib -nodefaultlibs
+ targets += purgatory.ro
+ 
+ # Default KBUILD_CFLAGS can have -pg option set when FTRACE is enabled. That
+ # in turn leaves some undefined symbols like __fentry__ in purgatory and not
+ # sure how to relocate those. Like kexec-tools, use custom flags.
+ 
+ KBUILD_CFLAGS := -Wall -Wstrict-prototypes -fno-strict-aliasing \
+ 		-fno-zero-initialized-in-bss -fno-builtin -ffreestanding \
+ 		-fno-PIC -fno-PIE -fno-stack-protector  -fno-exceptions \
 -		-msoft-float -MD -Os -m$(CONFIG_WORD_SIZE)
 -KBUILD_AFLAGS := -fno-exceptions -msoft-float -m$(CONFIG_WORD_SIZE) \
++		-msoft-float -MD -Os -m$(BITS)
++KBUILD_AFLAGS := -fno-exceptions -msoft-float -m$(BITS) \
+ 		-D__ASSEMBLY__
+ 
+ ifeq ($(CONFIG_CPU_LITTLE_ENDIAN),y)
+ KBUILD_CFLAGS += $(call cc-option,-mabi=elfv2,$(call cc-option,-mcall-aixdesc)) \
+ 		-mlittle-endian
+ KBUILD_AFLAGS += $(call cc-option,-mabi=elfv2) -mlittle-endian
+ else
+ KBUILD_CFLAGS += $(call cc-option,-mcall-aixdesc) -mbig-endian
+ KBUILD_AFLAGS += -mbig-endian
+ endif
+ 
+ $(obj)/purgatory.ro: $(PURGATORY_OBJS) FORCE
+ 		$(call if_changed,ld)
+ 
+ targets += kexec-purgatory.c
+ 
+ CMD_BIN2C = $(objtree)/scripts/basic/bin2c
+ quiet_cmd_bin2c = BIN2C   $@
+       cmd_bin2c = $(CMD_BIN2C) kexec_purgatory < $< > $@
+ 
+ $(obj)/kexec-purgatory.c: $(obj)/purgatory.ro FORCE
+ 	$(call if_changed,bin2c)
+ 	@:
+ 
+ 
+ obj-$(CONFIG_KEXEC_FILE)	+= kexec-purgatory.o

-- 
Cheers,
Stephen Rothwell

[toc] | [prev] | [next] | [standalone]

#1487767

Am Dienstag, 20 September 2016, 22:00:32 schrieb Michael Ellerman:
> Stephen Rothwell <sfr@canb.auug.org.au> writes:
> > On Tue, 20 Sep 2016 16:44:14 +1000 Stephen Rothwell 
<sfr@canb.auug.org.au> wrote:
> >> After merging the akpm-current tree, today's linux-next build (powerpc
> >> ppc64_defconfig) failed like this:
> >> 
> >> powerpc-linux-gcc: error: unrecognized command line option '-m'
> >> scripts/Makefile.build:290: recipe for target
> >> 'arch/powerpc/purgatory/printf.o' failed
> >> 
> >> and a few more like that.
> > 
> > OK, this is (most likely) caused by the CONFIG_WORD_SIZE -> BITS change
> > interacting with the KEXEC_FILE changes. Tomorrow I will do a better
> > merge conflict resolution patch that fixes all those.
> 
> Ah yep looks like that's the problem, patch below should fix it?
> 
> I think I'd actually prefer it if purgatory didn't redefine the CFLAGS
> from scratch, so I'll see if Thiago can do that and send a new version.

I could move the purgatory inside arch/powerpc/boot and use its BOOTCFLAGS. 
The needs of the purgatory and the boot wrapper are very similar. Would that 
be better?
-- 
[]'s
Thiago Jung Bauermann
IBM Linux Technology Center

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web