linux-next: build failure after merge of the final tree (akpm tree related)

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 718 bytes --]

Hi Andrew,

After merging the final tree, today's linux-next build (powerpc
allyesconfig) failed like this:

In file included from include/linux/netdevice.h:54:0,
                 from arch/powerpc/net/bpf_jit_comp.c:14:
include/net/netprio_cgroup.h:34:12: error: 'net_prio_subsys_id' redeclared as different kind of symbol
include/linux/cgroup_subsys.h:71:1: note: previous definition of 'net_prio_subsys_id' was here

And many more similar.

Caused (or exposed) by commit 9222aa56c0ce
("include/net/netprio_cgroup.h: various fixes") from the akpm tree.

I have reverted that commit for today.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[David Miller]

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Wed, 30 Nov 2011 15:52:26 +1100

> And many more similar.
> 
> Caused (or exposed) by commit 9222aa56c0ce
> ("include/net/netprio_cgroup.h: various fixes") from the akpm tree.
> 
> I have reverted that commit for today.

Andrew, please submit networking bug fixes to the networking maintainers
in order to avoid problems like this in the future.

Unlike other subsystems, I guarentee to handle it within 24 hours, often
much faster.

Thanks.

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[Andrew Morton]

On Wed, 30 Nov 2011 00:40:10 -0500 (EST) David Miller <davem@davemloft.net> wrote:

> From: Stephen Rothwell <sfr@canb.auug.org.au>
> Date: Wed, 30 Nov 2011 15:52:26 +1100
> 
> > And many more similar.
> > 
> > Caused (or exposed) by commit 9222aa56c0ce
> > ("include/net/netprio_cgroup.h: various fixes") from the akpm tree.
> > 
> > I have reverted that commit for today.
> 
> Andrew, please submit networking bug fixes to the networking maintainers
> in order to avoid problems like this in the future.
> 
> Unlike other subsystems, I guarentee to handle it within 24 hours, often
> much faster.
> 

This is my attempt to address the issues I mentioned last week.  It is
still under development and doesn't work yet.  I thought it did.

I'm now trying to get my brain around what that code is doing with
Kconfig symbols and net_prio_subsys_id.  I'm suspecting it's all to
make cgroup-subsys-within-a-module appear to work.

afaict net_prio_subsys_id is an enum if CONFIG_NETPRIO_CGROUP=y and is
an `extern int' when CONFIG_NETPRIO_CGROUP=m.  It's unclear to me why
the extern int version or net_prio_subsys_id exists at all, really.

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[Neil Horman]

On Tue, Nov 29, 2011 at 10:07:28PM -0800, Andrew Morton wrote:
> On Wed, 30 Nov 2011 00:40:10 -0500 (EST) David Miller <davem@davemloft.net> wrote:
> 
> > From: Stephen Rothwell <sfr@canb.auug.org.au>
> > Date: Wed, 30 Nov 2011 15:52:26 +1100
> > 
> > > And many more similar.
> > > 
> > > Caused (or exposed) by commit 9222aa56c0ce
> > > ("include/net/netprio_cgroup.h: various fixes") from the akpm tree.
> > > 
> > > I have reverted that commit for today.
> > 
> > Andrew, please submit networking bug fixes to the networking maintainers
> > in order to avoid problems like this in the future.
> > 
> > Unlike other subsystems, I guarentee to handle it within 24 hours, often
> > much faster.
> > 
> 
> This is my attempt to address the issues I mentioned last week.  It is
> still under development and doesn't work yet.  I thought it did.
> 
> I'm now trying to get my brain around what that code is doing with
> Kconfig symbols and net_prio_subsys_id.  I'm suspecting it's all to
> make cgroup-subsys-within-a-module appear to work.
> 
> afaict net_prio_subsys_id is an enum if CONFIG_NETPRIO_CGROUP=y and is
> an `extern int' when CONFIG_NETPRIO_CGROUP=m.  It's unclear to me why
> the extern int version or net_prio_subsys_id exists at all, really.
> 
> 
I implemented this code the same way that the net_cls cgroup does.  I'm not at
all sure whats going on in your tree, as its building as both a module and
monolitically in net-next.  I'll get it sorted today though.
Neil

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 696 bytes --]

Hi all,

After merging the final tree, today's linux-next build (sparc32 defconfig)
failed like this:

mm/page_alloc.c: In function 'free_area_init_node':
mm/page_alloc.c:4801:2: error: implicit declaration of function 'get_pfn_range_for_nid' [-Werror=implicit-function-declaration]

Caused by commit 55ce7b776831 ("mm: remove duplicated call to
get_pfn_range_for_nid()").  This build does not have
CONFIG_HAVE_MEMBLOCK_NODE_MAP defined, so that get_pfn_range_for_nid() is
not defined.  It used to be called from within the protection of
CONFIG_HAVE_MEMBLOCK_NODE_MAP.

I have reverted that commit for today.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 3285 bytes --]

Hi all,

After merging the final tree, today's linux-next build (sparc32 defconfig)
failed like this:

In file included from arch/sparc/include/asm/pgtable.h:6:0,
                 from include/linux/mm.h:50,
                 from include/linux/memcontrol.h:27,
                 from include/linux/swap.h:8,
                 from include/linux/suspend.h:4,
                 from init/do_mounts.c:16:
arch/sparc/include/asm/pgtable_32.h:357:40: error: unknown type name 'swp_entry_t'
arch/sparc/include/asm/pgtable_32.h:362:42: error: unknown type name 'swp_entry_t'
arch/sparc/include/asm/pgtable_32.h:367:1: error: unknown type name 'swp_entry_t'
arch/sparc/include/asm/pgtable_32.h: In function '__swp_entry':
arch/sparc/include/asm/pgtable_32.h:369:10: error: 'swp_entry_t' undeclared (first use in this function)
arch/sparc/include/asm/pgtable_32.h:369:10: note: each undeclared identifier is reported only once for each function it appears in
arch/sparc/include/asm/pgtable_32.h:369:23: error: expected ';' before '{' token
In file included from include/linux/memcontrol.h:27:0,
                 from include/linux/swap.h:8,
                 from arch/sparc/include/asm/pgtable_32.h:17,
                 from arch/sparc/include/asm/pgtable.h:6,
                 from arch/sparc/kernel/traps_32.c:23:
include/linux/mm.h: In function 'is_vmalloc_addr':
include/linux/mm.h:322:17: error: 'VMALLOC_START' undeclared (first use in this function)
include/linux/mm.h:322:17: note: each undeclared identifier is reported only once for each function it appears in
include/linux/mm.h:322:41: error: 'VMALLOC_END' undeclared (first use in this function)
include/linux/mm.h: In function 'maybe_mkwrite':
include/linux/mm.h:527:3: error: implicit declaration of function 'pte_mkwrite' [-Werror=implicit-function-declaration]
In file included from arch/sparc/include/asm/pgtable.h:6:0,
                 from arch/sparc/kernel/traps_32.c:23:
arch/sparc/include/asm/pgtable_32.h: At top level:
arch/sparc/include/asm/pgtable_32.h:251:21: error: conflicting types for 'pte_mkwrite'
include/linux/mm.h:527:9: note: previous implicit declaration of 'pte_mkwrite' was here

Caused by commit 41aed1ec2e09 ("lru: add an element to a memcg list")
from the akpm tree.  This patch added an include of linux/mm.h to
linux/memcontrol.h, but that was unneeded as the rest of the patch only
added dependencies on pointers to struct page and struct mem_cgroup -
both of which are already declared in this header.

I have applied the following patch for today:

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Thu, 6 Jun 2013 16:52:42 +1000
Subject: [PATCH] lru: memcontrol.h does not need mm.h

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 include/linux/memcontrol.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index c8b1412..57b8c2b 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -24,7 +24,6 @@
 #include <linux/hardirq.h>
 #include <linux/jump_label.h>
 #include <linux/list_lru.h>
-#include <linux/mm.h>
 
 struct mem_cgroup;
 struct page_cgroup;
-- 
1.8.1

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 459 bytes --]

Hi all,

After merging the final tree, today's linux-next build (sparc64 defconfig)
failed like this:

drivers/rtc/class.c: In function 'rtc_init':
drivers/rtc/class.c:339:18: error: lvalue required as unary '&' operand

Caused by commit c5a32d622261 ("drivers/rtc/class: convert from Legacy pm
ops to dev_pm_ops") from the akpm tree.

I have reverted that commit for today.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 574 bytes --]

Hi all,

After merging the final tree, today's linux-next build (arm defconfig)
failed like this:

kernel/built-in.o: In function `console_unlock':
cpu_pm.c:(.text+0x4418): undefined reference to `__bad_udelay'

Caused by commit 4df7323a125f ("printk: avoid softlockups in
console_unlock()") which added

	ndelay(max_interrupt_disabled_duration() / 2);

to console_unlock() which can be a 2.5 second delay ... and the maximum
delayed allowed by udelay (at least in this build) is 2ms.

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[Jan Kara]

On Mon 04-03-13 14:28:21, Stephen Rothwell wrote:
> Hi all,
> 
> After merging the final tree, today's linux-next build (arm defconfig)
> failed like this:
> 
> kernel/built-in.o: In function `console_unlock':
> cpu_pm.c:(.text+0x4418): undefined reference to `__bad_udelay'
> 
> Caused by commit 4df7323a125f ("printk: avoid softlockups in
> console_unlock()") which added
> 
> 	ndelay(max_interrupt_disabled_duration() / 2);
> 
> to console_unlock() which can be a 2.5 second delay ... and the maximum
> delayed allowed by udelay (at least in this build) is 2ms.
  Bah, one always learns. Didn't know there's some max limit on ndelay().
I'll replace that with a while loop or something like that. Thanks for
report!

								Honza
-- 
Jan Kara <jack@suse.cz>
SUSE Labs, CR

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 594 bytes --]

Hi all,

After merging the final tree, today's linux-next build (sparc64 defconfig)
failed like this:

arch/sparc/built-in.o: In function `setup_arch':
(.init.text+0xc38): undefined reference to `early_console'
arch/sparc/built-in.o: In function `setup_arch':
(.init.text+0xc44): undefined reference to `early_console'

Caused by commit 16ac99bd7acd ("early_printk: consolidate random copies
of identical code").  This build does not have CONFIG_EARLY_PRINTK defined.

I have reverted that commit for today.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[Andrew Morton]

On Mon, 4 Mar 2013 14:10:51 +1100 Stephen Rothwell <sfr@canb.auug.org.au> wrote:

> Hi all,
> 
> After merging the final tree, today's linux-next build (sparc64 defconfig)
> failed like this:
> 
> arch/sparc/built-in.o: In function `setup_arch':
> (.init.text+0xc38): undefined reference to `early_console'
> arch/sparc/built-in.o: In function `setup_arch':
> (.init.text+0xc44): undefined reference to `early_console'
> 
> Caused by commit 16ac99bd7acd ("early_printk: consolidate random copies
> of identical code").  This build does not have CONFIG_EARLY_PRINTK defined.

Yes, it looks like some ifdeffing is needed in
arch/sparc/kernel/setup_32.c and in arch/sparc/kernel/setup_64.c.  And
I suspect arch/powerpc/kernel/udbg.c will break with
CONFIG_EARLY_PRINTK=n.

> I have reverted that commit for today.

I had a shot at fixing it all but I'm not very confident in the result,
so I guess I'll drop it as well.

diff -puN arch/powerpc/kernel/udbg.c~early_printk-consolidate-random-copies-of-identical-code-fix arch/powerpc/kernel/udbg.c
--- a/arch/powerpc/kernel/udbg.c~early_printk-consolidate-random-copies-of-identical-code-fix
+++ a/arch/powerpc/kernel/udbg.c
@@ -162,9 +162,10 @@ static struct console udbg_console = {
  */
 void __init register_early_udbg_console(void)
 {
+#ifdef CONFIG_EARLY_PRINTK
 	if (early_console)
 		return;
-
+#endif
 	if (!udbg_putc)
 		return;
 
diff -puN arch/sparc/kernel/setup_32.c~early_printk-consolidate-random-copies-of-identical-code-fix arch/sparc/kernel/setup_32.c
--- a/arch/sparc/kernel/setup_32.c~early_printk-consolidate-random-copies-of-identical-code-fix
+++ a/arch/sparc/kernel/setup_32.c
@@ -108,6 +108,7 @@ unsigned long cmdline_memory_size __init
 /* which CPU booted us (0xff = not set) */
 unsigned char boot_cpu_id = 0xff; /* 0xff will make it into DATA section... */
 
+#ifdef CONFIG_EARLY_PRINTK
 static void
 prom_console_write(struct console *con, const char *s, unsigned n)
 {
@@ -120,6 +121,7 @@ static struct console prom_early_console
 	.flags =	CON_PRINTBUFFER | CON_BOOT,
 	.index =	-1,
 };
+#endif
 
 /* 
  * Process kernel command line switches that are specific to the
@@ -138,7 +140,9 @@ static void __init process_switch(char c
 		prom_halt();
 		break;
 	case 'p':
+#ifdef CONFIG_EARLY_PRINTK
 		prom_early_console.flags &= ~CON_BOOT;
+#endif
 		break;
 	default:
 		printk("Unknown boot switch (-%c)\n", c);
@@ -309,8 +313,10 @@ void __init setup_arch(char **cmdline_p)
 
 	boot_flags_init(*cmdline_p);
 
+#ifdef CONFIG_EARLY_PRINTK
 	early_console = &prom_early_console;
 	register_console(&prom_early_console);
+#endif
 
 	printk("ARCH: ");
 	switch(sparc_cpu_model) {
diff -puN arch/sparc/kernel/setup_64.c~early_printk-consolidate-random-copies-of-identical-code-fix arch/sparc/kernel/setup_64.c
--- a/arch/sparc/kernel/setup_64.c~early_printk-consolidate-random-copies-of-identical-code-fix
+++ a/arch/sparc/kernel/setup_64.c
@@ -75,21 +75,23 @@ struct screen_info screen_info = {
 	16                      /* orig-video-points */
 };
 
+/* Exported for mm/init.c:paging_init. */
+unsigned long cmdline_memory_size = 0;
+
+#ifdef CONFIG_EARLY_PRINTK
 static void
 prom_console_write(struct console *con, const char *s, unsigned n)
 {
 	prom_write(s, n);
 }
 
-/* Exported for mm/init.c:paging_init. */
-unsigned long cmdline_memory_size = 0;
-
 static struct console prom_early_console = {
 	.name =		"earlyprom",
 	.write =	prom_console_write,
 	.flags =	CON_PRINTBUFFER | CON_BOOT | CON_ANYTIME,
 	.index =	-1,
 };
+#endif
 
 /* 
  * Process kernel command line switches that are specific to the
@@ -106,7 +108,9 @@ static void __init process_switch(char c
 		prom_halt();
 		break;
 	case 'p':
+#ifdef CONFIG_EARLY_PRINTK
 		prom_early_console.flags &= ~CON_BOOT;
+#endif
 		break;
 	case 'P':
 		/* Force UltraSPARC-III P-Cache on. */
@@ -551,11 +555,17 @@ static void __init init_sparc64_elf_hwca
 		pause_patch();
 }
 
+#ifdef CONFIG_EARLY_PRINTK
 static inline void register_prom_console(void)
 {
 	early_console = &prom_early_console;
 	register_console(&prom_early_console);
 }
+#else
+static inline void register_prom_console(void)
+{
+}
+#endif
 
 void __init setup_arch(char **cmdline_p)
 {
_

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 459 bytes --]

Hi all,

After merging the final tree, today's linux-next build (sparc32 defconfig)
failed like this:

arch/sparc/mm/init_32.c: In function 'show_mem':
arch/sparc/mm/init_32.c:60:23: error: invalid operands to binary << (have 'atomic_long_t' and 'int')

Caused by commit "swap: add per-partition lock for swapfile" from the
akpm tree.

I have reverted that commit for today.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[Shaohua Li]

On Thu, Jan 24, 2013 at 04:54:48PM +1100, Stephen Rothwell wrote:
> Hi all,
> 
> After merging the final tree, today's linux-next build (sparc32 defconfig)
> failed like this:
> 
> arch/sparc/mm/init_32.c: In function 'show_mem':
> arch/sparc/mm/init_32.c:60:23: error: invalid operands to binary << (have 'atomic_long_t' and 'int')
> 
> Caused by commit "swap: add per-partition lock for swapfile" from the
> akpm tree.
> 
> I have reverted that commit for today.

can you please apply this:


---
 arch/sparc/mm/init_32.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Index: linux/arch/sparc/mm/init_32.c
===================================================================
--- linux.orig/arch/sparc/mm/init_32.c	2012-08-06 16:00:44.133458270 +0800
+++ linux/arch/sparc/mm/init_32.c	2013-01-24 18:28:23.242077473 +0800
@@ -57,7 +57,7 @@ void show_mem(unsigned int filter)
 	printk("Mem-info:\n");
 	show_free_areas(filter);
 	printk("Free swap:       %6ldkB\n",
-	       nr_swap_pages << (PAGE_SHIFT-10));
+	       get_nr_swap_pages() << (PAGE_SHIFT-10));
 	printk("%ld pages of RAM\n", totalram_pages);
 	printk("%ld free pages\n", nr_free_pages());
 }

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1526 bytes --]

Hi,

On Thu, 24 Jan 2013 18:30:40 +0800 Shaohua Li <shli@kernel.org> wrote:
>
> On Thu, Jan 24, 2013 at 04:54:48PM +1100, Stephen Rothwell wrote:
> > 
> > After merging the final tree, today's linux-next build (sparc32 defconfig)
> > failed like this:
> > 
> > arch/sparc/mm/init_32.c: In function 'show_mem':
> > arch/sparc/mm/init_32.c:60:23: error: invalid operands to binary << (have 'atomic_long_t' and 'int')
> > 
> > Caused by commit "swap: add per-partition lock for swapfile" from the
> > akpm tree.
> > 
> > I have reverted that commit for today.
> 
> can you please apply this:
> 
> 
> ---
>  arch/sparc/mm/init_32.c |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> Index: linux/arch/sparc/mm/init_32.c
> ===================================================================
> --- linux.orig/arch/sparc/mm/init_32.c	2012-08-06 16:00:44.133458270 +0800
> +++ linux/arch/sparc/mm/init_32.c	2013-01-24 18:28:23.242077473 +0800
> @@ -57,7 +57,7 @@ void show_mem(unsigned int filter)
>  	printk("Mem-info:\n");
>  	show_free_areas(filter);
>  	printk("Free swap:       %6ldkB\n",
> -	       nr_swap_pages << (PAGE_SHIFT-10));
> +	       get_nr_swap_pages() << (PAGE_SHIFT-10));
>  	printk("%ld pages of RAM\n", totalram_pages);
>  	printk("%ld free pages\n", nr_free_pages());
>  }

Please provide a proper patch to Andrew including a commit message and a
Signed-off-by and Reported-by lines.

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 812 bytes --]

Hi all,

After merging the final tree, today's linux-next build (arm defconfig)
failed like this:

mm/memblock.c: In function 'memblock_find_in_range_node':
mm/memblock.c:104:2: error: invalid use of undefined type 'struct movablecore_map'
mm/memblock.c:123:4: error: invalid use of undefined type 'struct movablecore_map'
mm/memblock.c:130:7: error: invalid use of undefined type 'struct movablecore_map'
mm/memblock.c:131:4: error: invalid use of undefined type 'struct movablecore_map'

Caused by commit "page_alloc: bootmem limit with movablecore_map" from
the akpm tree.  The definition of struct movablecore_map is protected by
CONFIG_HAVE_MEMBLOCK_NODE_MAP but its use is not.

I have reverted that commit for today.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[Tang Chen]

Hi Stephen,

On 01/21/2013 02:08 PM, Stephen Rothwell wrote:
> Hi all,
>
> After merging the final tree, today's linux-next build (arm defconfig)
> failed like this:
>
> mm/memblock.c: In function 'memblock_find_in_range_node':
> mm/memblock.c:104:2: error: invalid use of undefined type 'struct movablecore_map'
> mm/memblock.c:123:4: error: invalid use of undefined type 'struct movablecore_map'
> mm/memblock.c:130:7: error: invalid use of undefined type 'struct movablecore_map'
> mm/memblock.c:131:4: error: invalid use of undefined type 'struct movablecore_map'
>
> Caused by commit "page_alloc: bootmem limit with movablecore_map" from
> the akpm tree.  The definition of struct movablecore_map is protected by
> CONFIG_HAVE_MEMBLOCK_NODE_MAP but its use is not.
>
> I have reverted that commit for today.

Thank you very much for reporting this. It was my mistake to miss this 
definition.

I will post a new version of "page_alloc: bootmem limit with 
movablecore_map" since
you have reverted it.


CONFIG_HAVE_MEMBLOCK_NODE_MAP is selected by x86=y, but I don't have any 
non-x86 box.
So I didn't test it. Please tell me if you have any problem with it on 
other platforms.

Thanks. :)

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 465 bytes --]

Hi all,

After merging the final tree, today's linux-next build (sparc32 defconfig)
failed like this:

arch/sparc/kernel/sys_sparc_32.c: In function 'arch_get_unmapped_area':
arch/sparc/kernel/sys_sparc_32.c:41:26: error: unused variable 'vmm' [-Werror=unused-variable]

Caused by commit "mm: use vm_unmapped_area() on sparc32 architecture".

I have reverted that commit for today.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1226 bytes --]

Hi all,

After merging the final tree, today's linux-next build (sparc64 defconfig)
failed like this:

arch/sparc/kernel/sys_sparc_64.c: In function 'arch_get_unmapped_area':
arch/sparc/kernel/sys_sparc_64.c:92:16: error: unused variable 'start_addr' [-Werror=unused-variable]
arch/sparc/kernel/sys_sparc_64.c: In function 'arch_get_unmapped_area_topdown':
arch/sparc/kernel/sys_sparc_64.c:194:20: error: 'do_colour_align' undeclared (first use in this function)
arch/sparc/kernel/sys_sparc_64.c:194:20: note: each undeclared identifier is reported only once for each function it appears in
arch/sparc/kernel/sys_sparc_64.c:194:51: error: 'shm_align_mask' undeclared (first use in this function)

Caused by commit "mm: use vm_unmapped_area() on sparc64 architecture".  I
have reverted that commit for today (and the following fix patch).

arch/sparc/mm/hugetlbpage.c: In function 'hugetlb_get_unmapped_area_topdown':
arch/sparc/mm/hugetlbpage.c:63:25: error: unused variable 'vma' [-Werror=unused-variable]

Caused by commit "mm: use vm_unmapped_area() in hugetlbfs on sparc64
architecture".  I have reverted that commit for today.

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[Andrew Morton]

On Fri, 9 Nov 2012 15:09:57 +1100
Stephen Rothwell <sfr@canb.auug.org.au> wrote:

> Hi all,
> 
> After merging the final tree, today's linux-next build (sparc64 defconfig)
> failed like this:
> 
> arch/sparc/kernel/sys_sparc_64.c: In function 'arch_get_unmapped_area':
> arch/sparc/kernel/sys_sparc_64.c:92:16: error: unused variable 'start_addr' [-Werror=unused-variable]
> arch/sparc/kernel/sys_sparc_64.c: In function 'arch_get_unmapped_area_topdown':
> arch/sparc/kernel/sys_sparc_64.c:194:20: error: 'do_colour_align' undeclared (first use in this function)

s/colour/color/

> arch/sparc/kernel/sys_sparc_64.c:194:20: note: each undeclared identifier is reported only once for each function it appears in
> arch/sparc/kernel/sys_sparc_64.c:194:51: error: 'shm_align_mask' undeclared (first use in this function)

This one has me puzzled.  I *think* we should be using

info.align_mask = do_colour_align ? (PAGE_MASK & SHMLBA) : 0;
info.align_offset = (pgoff << PAGE_SHIFT) & (SHMLBA-1);

or similar.   It needs more work.
		
> Caused by commit "mm: use vm_unmapped_area() on sparc64 architecture".  I
> have reverted that commit for today (and the following fix patch).
> 
> arch/sparc/mm/hugetlbpage.c: In function 'hugetlb_get_unmapped_area_topdown':
> arch/sparc/mm/hugetlbpage.c:63:25: error: unused variable 'vma' [-Werror=unused-variable]
> 
> Caused by commit "mm: use vm_unmapped_area() in hugetlbfs on sparc64
> architecture".  I have reverted that commit for today.

I'll drop

mm-use-vm_unmapped_area-on-sparc64-architecture.patch
mm-use-vm_unmapped_area-on-sparc64-architecture-fix.patch
mm-use-vm_unmapped_area-in-hugetlbfs-on-sparc64-architecture.patch

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[David Miller]

From: Andrew Morton <akpm@linux-foundation.org>
Date: Wed, 14 Nov 2012 14:18:47 -0800

> s/colour/color/

The whole file uses the COLOUR spelling in it's macros, for
consistency I see no reason not to use the same spelling
for local variables.

Else, fix the whole file instead of leaving an inconsistent
mess around.

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[Andrew Morton]

On Wed, 14 Nov 2012 17:30:59 -0500 (EST)
David Miller <davem@davemloft.net> wrote:

> From: Andrew Morton <akpm@linux-foundation.org>
> Date: Wed, 14 Nov 2012 14:18:47 -0800
> 
> > s/colour/color/
> 
> The whole file uses the COLOUR spelling in it's macros, for
> consistency I see no reason not to use the same spelling
> for local variables.
> 
> Else, fix the whole file instead of leaving an inconsistent
> mess around.

Sure, but which way do you want it?

Current mainline:

akpm:/usr/src/linux-3.7-rc5> grep -i colour arch/sparc/kernel/*.c arch/sparc/mm/*.c
arch/sparc/kernel/sys_sparc_32.c:#define COLOUR_ALIGN(addr)      (((addr)+SHMLBA-1)&~(SHMLBA-1))
arch/sparc/kernel/sys_sparc_32.c:               addr = COLOUR_ALIGN(addr);
arch/sparc/kernel/sys_sparc_32.c:                       addr = COLOUR_ALIGN(addr);
arch/sparc/kernel/sys_sparc_64.c:static inline unsigned long COLOUR_ALIGN(unsigned long addr,
arch/sparc/kernel/sys_sparc_64.c:static inline unsigned long COLOUR_ALIGN_DOWN(unsigned long addr,
arch/sparc/kernel/sys_sparc_64.c:                       addr = COLOUR_ALIGN(addr, pgoff);
arch/sparc/kernel/sys_sparc_64.c:               addr = COLOUR_ALIGN(addr, pgoff);
arch/sparc/kernel/sys_sparc_64.c:                       addr = COLOUR_ALIGN(addr, pgoff);
arch/sparc/kernel/sys_sparc_64.c:                       addr = COLOUR_ALIGN(addr, pgoff);
arch/sparc/kernel/sys_sparc_64.c:               unsigned long base = COLOUR_ALIGN_DOWN(addr-len, pgoff);
arch/sparc/kernel/sys_sparc_64.c:               addr = COLOUR_ALIGN_DOWN(addr, pgoff);
arch/sparc/kernel/sys_sparc_64.c:                       addr = COLOUR_ALIGN_DOWN(addr, pgoff);


arch/sparc/kernel/sys_sparc_64.c: * 1) For file backed MAP_SHARED mmap()'s we D-cache color align,
arch/sparc/kernel/sys_sparc_64.c:	int do_color_align;
arch/sparc/kernel/sys_sparc_64.c:	do_color_align = 0;
arch/sparc/kernel/sys_sparc_64.c:		do_color_align = 1;
arch/sparc/kernel/sys_sparc_64.c:		if (do_color_align)
arch/sparc/kernel/sys_sparc_64.c:	if (do_color_align)
arch/sparc/kernel/sys_sparc_64.c:		if (do_color_align)
arch/sparc/kernel/sys_sparc_64.c:	int do_color_align;
arch/sparc/kernel/sys_sparc_64.c:	do_color_align = 0;
arch/sparc/kernel/sys_sparc_64.c:		do_color_align = 1;
arch/sparc/kernel/sys_sparc_64.c:		if (do_color_align)
arch/sparc/kernel/sys_sparc_64.c:	if (do_color_align) {
arch/sparc/kernel/sys_sparc_64.c:	if (do_color_align)
arch/sparc/kernel/sys_sparc_64.c:		if (do_color_align)
arch/sparc/mm/hugetlbpage.c: * definition we don't have to worry about any page coloring stuff
arch/sparc/mm/iommu.c:	/* To be coherent on HyperSparc, the page color of DVMA
arch/sparc/mm/iommu.c:		iommu->usemap.num_colors = vac_cache_size >> PAGE_SHIFT;
arch/sparc/mm/iommu.c:		iommu->usemap.num_colors = 1;
arch/sparc/mm/iommu.c:	/* page color = pfn of page */
arch/sparc/mm/iommu.c:	/* page color = physical address */


I suspect that most people expect "color".  And given that this
patchset removes COLOUR_ALIGN_DOWN (and should have removed
arch/sparc/kernel/sys_sparc_32.c:COLOR_ALIGN), we end up with

--- a/arch/sparc/kernel/sys_sparc_64.c~a
+++ a/arch/sparc/kernel/sys_sparc_64.c
@@ -75,7 +75,7 @@ static inline int invalid_64bit_range(un
  *    the spitfire/niagara VA-hole.
  */
 
-static inline unsigned long COLOUR_ALIGN(unsigned long addr,
+static inline unsigned long COLOR_ALIGN(unsigned long addr,
 					 unsigned long pgoff)
 {
 	unsigned long base = (addr+SHMLBA-1)&~(SHMLBA-1);
@@ -113,7 +113,7 @@ unsigned long arch_get_unmapped_area(str
 
 	if (addr) {
 		if (do_color_align)
-			addr = COLOUR_ALIGN(addr, pgoff);
+			addr = COLOR_ALIGN(addr, pgoff);
 		else
 			addr = PAGE_ALIGN(addr);
 
@@ -176,7 +176,7 @@ arch_get_unmapped_area_topdown(struct fi
 	/* requesting a specific address */
 	if (addr) {
 		if (do_color_align)
-			addr = COLOUR_ALIGN(addr, pgoff);
+			addr = COLOR_ALIGN(addr, pgoff);
 		else
 			addr = PAGE_ALIGN(addr);
 
_

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[David Miller]

From: Andrew Morton <akpm@linux-foundation.org>
Date: Wed, 14 Nov 2012 15:09:19 -0800

> I suspect that most people expect "color".  And given that this
> patchset removes COLOUR_ALIGN_DOWN (and should have removed
> arch/sparc/kernel/sys_sparc_32.c:COLOR_ALIGN), we end up with

That's fine.

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 590 bytes --]

Hi all,

After merging the final tree, today's linux-next build (i386 defconfig)
failed like this:

arch/x86/mm/hugetlbpage.c: In function 'hugetlb_get_unmapped_area_topdown':
arch/x86/mm/hugetlbpage.c:299:20: error: 'mm' undeclared (first use in this function)
arch/x86/mm/hugetlbpage.c:299:20: note: each undeclared identifier is reported only once for each function it appears in

Caused by commit "mm: use vm_unmapped_area() in hugetlbfs on i386
architecture".

I have reverted that commit for today.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[Andrew Morton]

On Fri, 9 Nov 2012 14:58:32 +1100 Stephen Rothwell <sfr@canb.auug.org.au> wrote:

> Hi all,
> 
> After merging the final tree, today's linux-next build (i386 defconfig)
> failed like this:
> 
> arch/x86/mm/hugetlbpage.c: In function 'hugetlb_get_unmapped_area_topdown':
> arch/x86/mm/hugetlbpage.c:299:20: error: 'mm' undeclared (first use in this function)
> arch/x86/mm/hugetlbpage.c:299:20: note: each undeclared identifier is reported only once for each function it appears in
> 
> Caused by commit "mm: use vm_unmapped_area() in hugetlbfs on i386
> architecture".
> 

oop, I fixed that but forgot to commit the changes, sorry.

	s/mm/current->mm/

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1262 bytes --]

Hi Andrew,

On Thu, 8 Nov 2012 20:01:54 -0800 Andrew Morton <akpm@linux-foundation.org> wrote:
>
> oop, I fixed that but forgot to commit the changes, sorry.
> 
> 	s/mm/current->mm/

I have applied this to the akpm tree today:

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Mon, 12 Nov 2012 10:51:18 +1100
Subject: [PATCH] mm: fix for use vm_unmapped_area() in hugetlbfs on i386 architecture

Fix-from: Adnrew Morton <akpm@linux-foundation.org>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 arch/x86/mm/hugetlbpage.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/mm/hugetlbpage.c b/arch/x86/mm/hugetlbpage.c
index c00c4a4..ae1aa71 100644
--- a/arch/x86/mm/hugetlbpage.c
+++ b/arch/x86/mm/hugetlbpage.c
@@ -296,7 +296,7 @@ static unsigned long hugetlb_get_unmapped_area_topdown(struct file *file,
 	info.flags = VM_UNMAPPED_AREA_TOPDOWN;
 	info.length = len;
 	info.low_limit = PAGE_SIZE;
-	info.high_limit = mm->mmap_base;
+	info.high_limit = current->mm->mmap_base;
 	info.align_mask = PAGE_MASK & ~huge_page_mask(h);
 	info.align_offset = 0;
 	addr = vm_unmapped_area(&info);
-- 
1.7.10.280.gaa39

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1687 bytes --]

Hi all,

After merging the final tree, today's linux-next build (i386 defconfig)
failed like this:

drivers/rtc/class.c: In function 'rtc_init':
drivers/rtc/class.c:261:23: error: 'rtc_suspend' undeclared (first use in this function)
drivers/rtc/class.c:261:23: note: each undeclared identifier is reported only once for each function it appears in
drivers/rtc/class.c:262:22: error: 'rtc_resume' undeclared (first use in this function)

Caused by commit "rtc_sysfs_show_hctosys-return-0-if-resume-failed-fix"
from the akpm tree.

I have added the following patch for today:

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Mon, 17 Sep 2012 21:43:56 +1000
Subject: [PATCH] rtc_sysfs_show_hctosys-return-0-if-resume-failed-fix-2

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 drivers/rtc/class.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/rtc/class.c b/drivers/rtc/class.c
index ca58ea9..f8a0aab 100644
--- a/drivers/rtc/class.c
+++ b/drivers/rtc/class.c
@@ -34,8 +34,9 @@ static void rtc_device_release(struct device *dev)
 #ifdef CONFIG_RTC_HCTOSYS_DEVICE
 /* Result of the last RTC to system clock attempt. */
 int rtc_hctosys_ret = -ENODEV;
+#endif
 
-#ifdef CONFIG_PM
+#if defined(CONFIG_PM) && defined(CONFIG_RTC_HCTOSYS_DEVICE)
 /*
  * On suspend(), measure the delta between one RTC and the
  * system's wall clock; restore it on resume().
@@ -129,7 +130,6 @@ static int rtc_resume(struct device *dev)
 #define rtc_suspend	NULL
 #define rtc_resume	NULL
 #endif
-#endif
 
 
 /**
-- 
1.7.10.280.gaa39

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 745 bytes --]

Hi all,

After merging the final tree, today's linux-next build (sparc64 defconfig)
failed like this:

drivers/built-in.o: In function `rtc_hctosys':
hctosys.c:(.init.text+0x4a98): undefined reference to `rtc_hctosys_ret'
hctosys.c:(.init.text+0x4b54): undefined reference to `rtc_hctosys_ret'
hctosys.c:(.init.text+0x4b58): undefined reference to `rtc_hctosys_ret'

Caused by commit "rtc_sysfs_show_hctosys(): display 0 if resume failed"
from the akpm tree.  rtc_hctosys_ret was moved into class.c but protected
by defined(CONFIG_PM) && defined(CONFIG_RTC_HCTOSYS_DEVICE) and this
build does not have CONFIG_PM set.

I have reverted that commit for today.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[David Fries]

[-- Attachment #1: Type: text/plain, Size: 1343 bytes --]

On Thu, Sep 13, 2012 at 06:11:27PM +1000, Stephen Rothwell wrote:
> Hi all,
> 
> After merging the final tree, today's linux-next build (sparc64 defconfig)
> failed like this:
> 
> drivers/built-in.o: In function `rtc_hctosys':
> hctosys.c:(.init.text+0x4a98): undefined reference to `rtc_hctosys_ret'
> hctosys.c:(.init.text+0x4b54): undefined reference to `rtc_hctosys_ret'
> hctosys.c:(.init.text+0x4b58): undefined reference to `rtc_hctosys_ret'

Can you post your .config?  `grep RTC .config`
It would seem to me that drivers/rtc/hctosys.c is being compiled
without drivers/rtc/class.c which now holds the global variable, but
Kconfig says that RTC_HCTOSYS 'depends on RTC_CLASS = y'.

config RTC_HCTOSYS
	depends on RTC_HCTOSYS_DEVICE
Or, how do you depend on a string value being set?

config RTC_HCTOSYS_DEVICE
	depends on RTC_CLASS = y

> Caused by commit "rtc_sysfs_show_hctosys(): display 0 if resume failed"
> from the akpm tree.  rtc_hctosys_ret was moved into class.c but protected
> by defined(CONFIG_PM) && defined(CONFIG_RTC_HCTOSYS_DEVICE) and this
> build does not have CONFIG_PM set.
> 
> I have reverted that commit for today.
> -- 
> Cheers,
> Stephen Rothwell                    sfr@canb.auug.org.au



-- 
David Fries <david@fries.net>    PGP pub CB1EE8F0
http://fries.net/~david/

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1331 bytes --]

Hi David,

On Thu, 13 Sep 2012 08:24:08 -0500 David Fries <david@fries.net> wrote:
>
> On Thu, Sep 13, 2012 at 06:11:27PM +1000, Stephen Rothwell wrote:
> > Hi all,
> > 
> > After merging the final tree, today's linux-next build (sparc64 defconfig)
> > failed like this:
> > 
> > drivers/built-in.o: In function `rtc_hctosys':
> > hctosys.c:(.init.text+0x4a98): undefined reference to `rtc_hctosys_ret'
> > hctosys.c:(.init.text+0x4b54): undefined reference to `rtc_hctosys_ret'
> > hctosys.c:(.init.text+0x4b58): undefined reference to `rtc_hctosys_ret'
> 
> Can you post your .config?  `grep RTC .config`
> It would seem to me that drivers/rtc/hctosys.c is being compiled
> without drivers/rtc/class.c which now holds the global variable, but
> Kconfig says that RTC_HCTOSYS 'depends on RTC_CLASS = y'.

Did you read the bit you quoted below?  CONFIG_PM is not set, so that the
part of class.c that contains the definition of rtc_hctosys_ret is not
compiled ...

> > Caused by commit "rtc_sysfs_show_hctosys(): display 0 if resume failed"
> > from the akpm tree.  rtc_hctosys_ret was moved into class.c but protected
> > by defined(CONFIG_PM) && defined(CONFIG_RTC_HCTOSYS_DEVICE) and this
> > build does not have CONFIG_PM set.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[David Fries]

On Thu, Sep 13, 2012 at 11:34:29PM +1000, Stephen Rothwell wrote:
> Hi David,
> 
> On Thu, 13 Sep 2012 08:24:08 -0500 David Fries <david@fries.net> wrote:
> >
> > On Thu, Sep 13, 2012 at 06:11:27PM +1000, Stephen Rothwell wrote:
> > > Hi all,
> > > 
> > > After merging the final tree, today's linux-next build (sparc64 defconfig)
> > > failed like this:
> > > 
> > > drivers/built-in.o: In function `rtc_hctosys':
> > > hctosys.c:(.init.text+0x4a98): undefined reference to `rtc_hctosys_ret'
> > > hctosys.c:(.init.text+0x4b54): undefined reference to `rtc_hctosys_ret'
> > > hctosys.c:(.init.text+0x4b58): undefined reference to `rtc_hctosys_ret'
> > 
> > Can you post your .config?  `grep RTC .config`
> > It would seem to me that drivers/rtc/hctosys.c is being compiled
> > without drivers/rtc/class.c which now holds the global variable, but
> > Kconfig says that RTC_HCTOSYS 'depends on RTC_CLASS = y'.
> 
> Did you read the bit you quoted below?

Oops, sorry, I had just got out of bed and was trying to hurry to
work.  The following version defines rtc_hctosys_ret in
CONFIG_RTC_HCTOSYS_DEVICE and CONFIG_PM is checked later.

> CONFIG_PM is not set, so that the
> part of class.c that contains the definition of rtc_hctosys_ret is not
> compiled ...
> 
> > > Caused by commit "rtc_sysfs_show_hctosys(): display 0 if resume failed"
> > > from the akpm tree.  rtc_hctosys_ret was moved into class.c but protected
> > > by defined(CONFIG_PM) && defined(CONFIG_RTC_HCTOSYS_DEVICE) and this
> > > build does not have CONFIG_PM set.



From 7c5aac626f9f64b9a4cc8c1cca1e4cb7cfefb819 Mon Sep 17 00:00:00 2001
From: David Fries <David@Fries.net>
Date: Sat, 25 Aug 2012 13:48:08 -0500
Subject: [PATCH] rtc_sysfs_show_hctosys(): display 0 if resume failed
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Without this patch /sys/class/rtc/$CONFIG_RTC_HCTOSYS_DEVICE/hctosys
contains a 1 (meaning "This rtc was used to initialize the system
clock") even if setting the time by do_settimeofday() at bootup failed.
The RTC can also be used to set the clock on resume, if it did 1,
otherwise 0.  Previously there was no indication if the RTC was used
to set the clock in resume.

This uses only CONFIG_RTC_HCTOSYS_DEVICE for conditional compilation
instead of it and CONFIG_RTC_HCTOSYS to be more consistent.
rtc_hctosys_ret was moved to class.c so class.c no longer depends on
hctosys.c.

Signed-off-by: David Fries <David@Fries.net>
Cc: Matthew Garrett <mjg@redhat.com>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Cc: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Cc: Andrew Morton <akpm@linux-foundation.org>
---
 drivers/rtc/class.c     |    8 +++++++-
 drivers/rtc/hctosys.c   |    4 +---
 drivers/rtc/rtc-sysfs.c |    6 ++++++
 include/linux/rtc.h     |    2 +-
 4 files changed, 15 insertions(+), 5 deletions(-)

diff --git a/drivers/rtc/class.c b/drivers/rtc/class.c
index dc4c274..a022b1c 100644
--- a/drivers/rtc/class.c
+++ b/drivers/rtc/class.c
@@ -31,8 +31,11 @@ static void rtc_device_release(struct device *dev)
 	kfree(rtc);
 }
 
-#if defined(CONFIG_PM) && defined(CONFIG_RTC_HCTOSYS_DEVICE)
+#ifdef CONFIG_RTC_HCTOSYS_DEVICE
+/* Result of the last RTC to system clock attempt. */
+int rtc_hctosys_ret = -ENODEV;
 
+#ifdef CONFIG_PM
 /*
  * On suspend(), measure the delta between one RTC and the
  * system's wall clock; restore it on resume().
@@ -84,6 +87,7 @@ static int rtc_resume(struct device *dev)
 	struct timespec		new_system, new_rtc;
 	struct timespec		sleep_time;
 
+	rtc_hctosys_ret = -ENODEV;
 	if (strcmp(dev_name(&rtc->dev), CONFIG_RTC_HCTOSYS_DEVICE) != 0)
 		return 0;
 
@@ -117,6 +121,7 @@ static int rtc_resume(struct device *dev)
 
 	if (sleep_time.tv_sec >= 0)
 		timekeeping_inject_sleeptime(&sleep_time);
+	rtc_hctosys_ret = 0;
 	return 0;
 }
 
@@ -124,6 +129,7 @@ static int rtc_resume(struct device *dev)
 #define rtc_suspend	NULL
 #define rtc_resume	NULL
 #endif
+#endif
 
 
 /**
diff --git a/drivers/rtc/hctosys.c b/drivers/rtc/hctosys.c
index bc90b09..4aa60d7 100644
--- a/drivers/rtc/hctosys.c
+++ b/drivers/rtc/hctosys.c
@@ -22,8 +22,6 @@
  * the best guess is to add 0.5s.
  */
 
-int rtc_hctosys_ret = -ENODEV;
-
 static int __init rtc_hctosys(void)
 {
 	int err = -ENODEV;
@@ -56,7 +54,7 @@ static int __init rtc_hctosys(void)
 
 	rtc_tm_to_time(&tm, &tv.tv_sec);
 
-	do_settimeofday(&tv);
+	err = do_settimeofday(&tv);
 
 	dev_info(rtc->dev.parent,
 		"setting system clock to "
diff --git a/drivers/rtc/rtc-sysfs.c b/drivers/rtc/rtc-sysfs.c
index 380083c..b70e2bb 100644
--- a/drivers/rtc/rtc-sysfs.c
+++ b/drivers/rtc/rtc-sysfs.c
@@ -102,6 +102,12 @@ rtc_sysfs_set_max_user_freq(struct device *dev, struct device_attribute *attr,
 	return n;
 }
 
+/**
+ * rtc_sysfs_show_hctosys - indicate if the given RTC set the system time
+ *
+ * Returns 1 if the system clock was set by this RTC at the last
+ * boot or resume event.
+ */
 static ssize_t
 rtc_sysfs_show_hctosys(struct device *dev, struct device_attribute *attr,
 		char *buf)
diff --git a/include/linux/rtc.h b/include/linux/rtc.h
index f071b39..20ec4d3 100644
--- a/include/linux/rtc.h
+++ b/include/linux/rtc.h
@@ -276,7 +276,7 @@ static inline bool is_leap_year(unsigned int year)
 	return (!(year % 4) && (year % 100)) || !(year % 400);
 }
 
-#ifdef CONFIG_RTC_HCTOSYS
+#ifdef CONFIG_RTC_HCTOSYS_DEVICE
 extern int rtc_hctosys_ret;
 #else
 #define rtc_hctosys_ret -ENODEV
-- 
1.7.2.5

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 640 bytes --]

Hi all,

After merging the final tree, today's linux-next build (sparc64 defconfig)
failed like this:

mm/internal.h: In function 'swap_cache_hit':
mm/internal.h:377:3: error: implicit declaration of function 'atomic_dec_if_positive' [-Werror=implicit-function-declaration]

Caused by commit "swap: add a simple detector for inappropriate swapin
readahead" from the akmp tree.  atomic_dec_if_positive() appears to only
exist on avr32, microblaze, mips, powerpc and x86 ...

I have reverted that commit (and the following trivial fix commit) for
today.

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[Shaohua Li]

On Thu, Sep 13, 2012 at 06:01:21PM +1000, Stephen Rothwell wrote:
> Hi all,
> 
> After merging the final tree, today's linux-next build (sparc64 defconfig)
> failed like this:
> 
> mm/internal.h: In function 'swap_cache_hit':
> mm/internal.h:377:3: error: implicit declaration of function 'atomic_dec_if_positive' [-Werror=implicit-function-declaration]
> 
> Caused by commit "swap: add a simple detector for inappropriate swapin
> readahead" from the akmp tree.  atomic_dec_if_positive() appears to only
> exist on avr32, microblaze, mips, powerpc and x86 ...
> 
> I have reverted that commit (and the following trivial fix commit) for
> today.

The x86 implementation of atomic_dec_if_positive is quite generic, how about
add this one?


Signed-off-by: Shaohua Li <shli@fusionio.com>
---
 arch/microblaze/include/asm/atomic.h |    3 ++-
 arch/powerpc/include/asm/atomic.h    |    3 ++-
 arch/x86/include/asm/atomic.h        |   24 ------------------------
 include/linux/atomic.h               |   25 +++++++++++++++++++++++++
 4 files changed, 29 insertions(+), 26 deletions(-)

Index: linux/arch/microblaze/include/asm/atomic.h
===================================================================
--- linux.orig/arch/microblaze/include/asm/atomic.h	2012-09-13 17:56:54.206807900 +0800
+++ linux/arch/microblaze/include/asm/atomic.h	2012-09-13 17:57:18.202506238 +0800
@@ -9,7 +9,7 @@
  * Atomically test *v and decrement if it is greater than 0.
  * The function returns the old value of *v minus 1.
  */
-static inline int atomic_dec_if_positive(atomic_t *v)
+static inline int __atomic_dec_if_positive(atomic_t *v)
 {
 	unsigned long flags;
 	int res;
@@ -22,5 +22,6 @@ static inline int atomic_dec_if_positive
 
 	return res;
 }
+#define atomic_dec_if_positive __atomic_dec_if_positive
 
 #endif /* _ASM_MICROBLAZE_ATOMIC_H */
Index: linux/arch/powerpc/include/asm/atomic.h
===================================================================
--- linux.orig/arch/powerpc/include/asm/atomic.h	2012-09-13 17:56:54.210807849 +0800
+++ linux/arch/powerpc/include/asm/atomic.h	2012-09-13 17:57:18.202506238 +0800
@@ -247,7 +247,7 @@ static __inline__ int atomic_inc_not_zer
  * The function returns the old value of *v minus 1, even if
  * the atomic variable, v, was not decremented.
  */
-static __inline__ int atomic_dec_if_positive(atomic_t *v)
+static __inline__ int __atomic_dec_if_positive(atomic_t *v)
 {
 	int t;
 
@@ -268,6 +268,7 @@ static __inline__ int atomic_dec_if_posi
 
 	return t;
 }
+#define atomic_dec_if_positive __atomic_dec_if_positive
 
 #define smp_mb__before_atomic_dec()     smp_mb()
 #define smp_mb__after_atomic_dec()      smp_mb()
Index: linux/arch/x86/include/asm/atomic.h
===================================================================
--- linux.orig/arch/x86/include/asm/atomic.h	2012-09-13 17:56:54.210807849 +0800
+++ linux/arch/x86/include/asm/atomic.h	2012-09-13 17:57:18.202506238 +0800
@@ -240,30 +240,6 @@ static inline int __atomic_add_unless(at
 	return c;
 }
 
-
-/*
- * atomic_dec_if_positive - decrement by 1 if old value positive
- * @v: pointer of type atomic_t
- *
- * The function returns the old value of *v minus 1, even if
- * the atomic variable, v, was not decremented.
- */
-static inline int atomic_dec_if_positive(atomic_t *v)
-{
-	int c, old, dec;
-	c = atomic_read(v);
-	for (;;) {
-		dec = c - 1;
-		if (unlikely(dec < 0))
-			break;
-		old = atomic_cmpxchg((v), c, dec);
-		if (likely(old == c))
-			break;
-		c = old;
-	}
-	return dec;
-}
-
 /**
  * atomic_inc_short - increment of a short integer
  * @v: pointer to type int
Index: linux/include/linux/atomic.h
===================================================================
--- linux.orig/include/linux/atomic.h	2012-09-13 17:56:54.210807849 +0800
+++ linux/include/linux/atomic.h	2012-09-13 17:57:18.202506238 +0800
@@ -86,6 +86,31 @@ static inline int atomic_dec_unless_posi
 }
 #endif
 
+/*
+ * atomic_dec_if_positive - decrement by 1 if old value positive
+ * @v: pointer of type atomic_t
+ *
+ * The function returns the old value of *v minus 1, even if
+ * the atomic variable, v, was not decremented.
+ */
+#ifndef atomic_dec_if_positive
+static inline int atomic_dec_if_positive(atomic_t *v)
+{
+	int c, old, dec;
+	c = atomic_read(v);
+	for (;;) {
+		dec = c - 1;
+		if (unlikely(dec < 0))
+			break;
+		old = atomic_cmpxchg((v), c, dec);
+		if (likely(old == c))
+			break;
+		c = old;
+	}
+	return dec;
+}
+#endif
+
 #ifndef CONFIG_ARCH_HAS_ATOMIC_OR
 static inline void atomic_or(int i, atomic_t *v)
 {

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1032 bytes --]

Hi,

On Thu, 13 Sep 2012 18:01:36 +0800 Shaohua Li <shli@kernel.org> wrote:
>
> Index: linux/arch/microblaze/include/asm/atomic.h
> ===================================================================
> --- linux.orig/arch/microblaze/include/asm/atomic.h	2012-09-13 17:56:54.206807900 +0800
> +++ linux/arch/microblaze/include/asm/atomic.h	2012-09-13 17:57:18.202506238 +0800
> @@ -9,7 +9,7 @@
>   * Atomically test *v and decrement if it is greater than 0.
>   * The function returns the old value of *v minus 1.
>   */
> -static inline int atomic_dec_if_positive(atomic_t *v)
> +static inline int __atomic_dec_if_positive(atomic_t *v)
>  {
>  	unsigned long flags;
>  	int res;
> @@ -22,5 +22,6 @@ static inline int atomic_dec_if_positive
>  
>  	return res;
>  }
> +#define atomic_dec_if_positive __atomic_dec_if_positive

You could just do

#define atomic_dec_if_positive	atomic_dec_if_positive

without renaming the function.

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 691 bytes --]

Hi all,

After merging the final tree, today's linux-next build (powerpc
allyesconfig) failed like this:

drivers/isdn/i4l/isdn_common.c: In function 'isdn_init':
drivers/isdn/i4l/isdn_common.c:2365:9: error: 'KERN_CONG' undeclared (first use in this function)
drivers/isdn/i4l/isdn_common.c:2365:9: note: each undeclared identifier is reported only once for each function it appears in
drivers/isdn/i4l/isdn_common.c:2365:19: error: expected ')' before string constant

Caused by commit 87577ec95e0c ("ISDN: add missing KERN_CONT").  Please
build test this stuff.  :-(

I reverted that commit for today.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 440 bytes --]

Hi Andrew,

After merging the akpm tree, today's linux-next build (any without
CONFIG_MMU e.g. bfin BF537-STAMP_defconfig) failed like this:

fs/proc/task_nommu.c:173: undefined reference to `vm_is_stack'

Caused by commit "procfs: mark thread stack correctly in proc/<pid>/maps"
from the akpm tree which only defines that function when CONFIG_MMU is
set.

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 813 bytes --]

Hi Andrew,

Today's linux-next fails to build (s390 allnoconfig) like this:

In file included from arch/s390/mm/maccess.c:15:
include/linux/gfp.h: In function 'gfp_zonelist':
include/linux/gfp.h:267: error: '__enabled_CONFIG_NUMA' undeclared (first use in this function)
include/linux/gfp.h:267: error: (Each undeclared identifier is reported only once
include/linux/gfp.h:267: error: for each function it appears in.)
include/linux/gfp.h:267: error: '__enabled_CONFIG_NUMA_MODULE' undeclared (first use in this function)

and many more.  This also happens in other architectures and configs but
(obviously) not all builds.

Presumably caused by commit 053361e1a072 ("mm: replace NUMA_BUILD with
IS_ENABLED(CONFIG_NUMA)").

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[Konstantin Khlebnikov]

Stephen Rothwell wrote:
> Hi Andrew,
>
> Today's linux-next fails to build (s390 allnoconfig) like this:
>
> In file included from arch/s390/mm/maccess.c:15:
> include/linux/gfp.h: In function 'gfp_zonelist':
> include/linux/gfp.h:267: error: '__enabled_CONFIG_NUMA' undeclared (first use in this function)
> include/linux/gfp.h:267: error: (Each undeclared identifier is reported only once
> include/linux/gfp.h:267: error: for each function it appears in.)
> include/linux/gfp.h:267: error: '__enabled_CONFIG_NUMA_MODULE' undeclared (first use in this function)
>
> and many more.  This also happens in other architectures and configs but
> (obviously) not all builds.
>
> Presumably caused by commit 053361e1a072 ("mm: replace NUMA_BUILD with
> IS_ENABLED(CONFIG_NUMA)").
>

Oops, yes, my fault.

'config NUMA' defined in arch-specific Kconfig, some of them does not have it.
Thus Kconfig generates these constants not for all architectures.

config MIGRATION and config COMPACTION already in generic mm/Kconfig

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1099 bytes --]

Hi Andrew,

On Fri, 17 Feb 2012 21:20:37 +1100 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>
> Today's linux-next fails to build (s390 allnoconfig) like this:
> 
> In file included from arch/s390/mm/maccess.c:15:
> include/linux/gfp.h: In function 'gfp_zonelist':
> include/linux/gfp.h:267: error: '__enabled_CONFIG_NUMA' undeclared (first use in this function)
> include/linux/gfp.h:267: error: (Each undeclared identifier is reported only once
> include/linux/gfp.h:267: error: for each function it appears in.)
> include/linux/gfp.h:267: error: '__enabled_CONFIG_NUMA_MODULE' undeclared (first use in this function)
> 
> and many more.  This also happens in other architectures and configs but
> (obviously) not all builds.
> 
> Presumably caused by commit 053361e1a072 ("mm: replace NUMA_BUILD with
> IS_ENABLED(CONFIG_NUMA)").

I have removed that patch from the akpm tree until a solution is worked out.

Everything seemed to rebase ok with out it and I can't see much actually
depending on it.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[Andrew Morton]

On Mon, 20 Feb 2012 10:04:36 +1100 Stephen Rothwell <sfr@canb.auug.org.au> wrote:

> Hi Andrew,
> 
> On Fri, 17 Feb 2012 21:20:37 +1100 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> >
> > Today's linux-next fails to build (s390 allnoconfig) like this:
> > 
> > In file included from arch/s390/mm/maccess.c:15:
> > include/linux/gfp.h: In function 'gfp_zonelist':
> > include/linux/gfp.h:267: error: '__enabled_CONFIG_NUMA' undeclared (first use in this function)
> > include/linux/gfp.h:267: error: (Each undeclared identifier is reported only once
> > include/linux/gfp.h:267: error: for each function it appears in.)
> > include/linux/gfp.h:267: error: '__enabled_CONFIG_NUMA_MODULE' undeclared (first use in this function)
> > 
> > and many more.  This also happens in other architectures and configs but
> > (obviously) not all builds.
> > 
> > Presumably caused by commit 053361e1a072 ("mm: replace NUMA_BUILD with
> > IS_ENABLED(CONFIG_NUMA)").
> 
> I have removed that patch from the akpm tree until a solution is worked out.
> 
> Everything seemed to rebase ok with out it and I can't see much actually
> depending on it.

yup, I've removed it from the for-next sections and will probably drop
it altogether.

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1816 bytes --]

Hi Andrew,

After merging the final tree, today's linux-next build (powerpc
allnoconfig) failed like this:

In file included from include/linux/posix_types.h:47:0,
                 from include/linux/types.h:17,
                 from include/linux/page-flags.h:8,
                 from kernel/bounds.c:9:
arch/powerpc/include/asm/posix_types.h:15:14: error: conflicting types for '__kernel_size_t'
arch/powerpc/include/asm/posix_types.h:14:22: note: previous declaration of '__kernel_size_t' was here
In file included from include/linux/page-flags.h:8:0,
                 from kernel/bounds.c:9:
include/linux/types.h:68:1: error: unknown type name '__kernel_ssize_t'

Caused by commit a9dbe86e5995 ("powerpc: use generic posix_types.h").

This think was pointed out a couple of days ago, so maybe Andrew needs a
newer version of the patch.

I added this fix patch for today:

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Fri, 17 Feb 2012 16:25:48 +1100
Subject: [PATCH] powerpc: use generic posix_types.h fix

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 arch/powerpc/include/asm/posix_types.h |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/arch/powerpc/include/asm/posix_types.h b/arch/powerpc/include/asm/posix_types.h
index 1fbe027f..f139325 100644
--- a/arch/powerpc/include/asm/posix_types.h
+++ b/arch/powerpc/include/asm/posix_types.h
@@ -12,7 +12,7 @@ typedef unsigned long	__kernel_old_dev_t;
 #define __kernel_old_dev_t __kernel_old_dev_t
 #else
 typedef unsigned int	__kernel_size_t;
-typedef int		__kernel_size_t;
+typedef int		__kernel_ssize_t;
 typedef long		__kernel_ptrdiff_t;
 #define __kernel_size_t __kernel_size_t
 
-- 
1.7.9

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[Benjamin Herrenschmidt]

On Fri, 2012-02-17 at 16:30 +1100, Stephen Rothwell wrote:
> Hi Andrew,
> 
> After merging the final tree, today's linux-next build (powerpc
> allnoconfig) failed like this:
> 
> In file included from include/linux/posix_types.h:47:0,
>                  from include/linux/types.h:17,
>                  from include/linux/page-flags.h:8,
>                  from kernel/bounds.c:9:
> arch/powerpc/include/asm/posix_types.h:15:14: error: conflicting types for '__kernel_size_t'
> arch/powerpc/include/asm/posix_types.h:14:22: note: previous declaration of '__kernel_size_t' was here
> In file included from include/linux/page-flags.h:8:0,
>                  from kernel/bounds.c:9:
> include/linux/types.h:68:1: error: unknown type name '__kernel_ssize_t'
> 
> Caused by commit a9dbe86e5995 ("powerpc: use generic posix_types.h").
> 
> This think was pointed out a couple of days ago, so maybe Andrew needs a
> newer version of the patch.

Yes, I pointed that out to Peter a while back, maybe it got lost ...

Cheers,
Ben.

> I added this fix patch for today:
> 
> From: Stephen Rothwell <sfr@canb.auug.org.au>
> Date: Fri, 17 Feb 2012 16:25:48 +1100
> Subject: [PATCH] powerpc: use generic posix_types.h fix
> 
> Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
> ---
>  arch/powerpc/include/asm/posix_types.h |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/arch/powerpc/include/asm/posix_types.h b/arch/powerpc/include/asm/posix_types.h
> index 1fbe027f..f139325 100644
> --- a/arch/powerpc/include/asm/posix_types.h
> +++ b/arch/powerpc/include/asm/posix_types.h
> @@ -12,7 +12,7 @@ typedef unsigned long	__kernel_old_dev_t;
>  #define __kernel_old_dev_t __kernel_old_dev_t
>  #else
>  typedef unsigned int	__kernel_size_t;
> -typedef int		__kernel_size_t;
> +typedef int		__kernel_ssize_t;
>  typedef long		__kernel_ptrdiff_t;
>  #define __kernel_size_t __kernel_size_t
>  
> -- 
> 1.7.9
> 

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1621 bytes --]

Hi all,

After merging the final tree, today's linux-next build (i386 defconfig)
failed like this:

In file included from arch/x86/include/asm/elf.h:90:0,
                 from include/linux/elf.h:7,
                 from arch/x86/kernel/cpu/amd.c:4:
arch/x86/include/asm/desc.h: In function 'native_set_ldt':
arch/x86/include/asm/desc.h:199:3: error: implicit declaration of function 'this_cpu_read' [-Werror=implicit-function-declaration]

Caused by commit a65a3870c856 ("x86: use this_cpu_xxx to replace percpu_xxx funcs").

I tried changing the include of asm/percpu.h to linux/percpu.h in
arch/x86/include/asm/smp.h (the above is caused by smp_processor_id()
using this_cpu_read()), but that made things worse.  So I added an
include of linux/percpu.h to arch/x86/include/asm/desc.h (but this seems
a bit fragile).

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Fri, 20 Jan 2012 12:51:40 +1100
Subject: [PATCH] percpu: x86: fix smp_processor_id's need for this_cpu_read

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 arch/x86/include/asm/desc.h |    2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)

diff --git a/arch/x86/include/asm/desc.h b/arch/x86/include/asm/desc.h
index e95822d..3072a2d 100644
--- a/arch/x86/include/asm/desc.h
+++ b/arch/x86/include/asm/desc.h
@@ -6,6 +6,7 @@
 #include <asm/mmu.h>
 
 #include <linux/smp.h>
+#include <linux/percpu.h>
 
 static inline void fill_ldt(struct desc_struct *desc, const struct user_desc *info)
 {
-- 
1.7.9.rc0.23.g7e521

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 970 bytes --]

Hi all,

After merging the final tree, today's linux-next build (m68k allmodconfig)
failed like this:

drivers/leds/leds-tca6507.c:158: error: field 'gpio' has incomplete type
drivers/leds/leds-tca6507.c: In function 'tca6507_gpio_set_value':
drivers/leds/leds-tca6507.c:534: warning: type defaults to 'int' in declaration of '__mptr'
drivers/leds/leds-tca6507.c:534: warning: initialization from incompatible pointer type
drivers/leds/leds-tca6507.c: In function 'tca6507_probe':
drivers/leds/leds-tca6507.c:618: error: implicit declaration of function 'gpiochip_add'
drivers/leds/leds-tca6507.c: In function 'tca6507_remove':
drivers/leds/leds-tca6507.c:651: error: implicit declaration of function 'gpiochip_remove'

Caused by commit 7a9244d89523 ("leds: add driver for TCA6507 LED controller").

Maybe this driver needs some dependencies ...
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[NeilBrown]

[-- Attachment #1: Type: text/plain, Size: 1170 bytes --]

On Sat, 17 Dec 2011 15:42:19 +1100 Stephen Rothwell <sfr@canb.auug.org.au>
wrote:

> Hi all,
> 
> After merging the final tree, today's linux-next build (m68k allmodconfig)
> failed like this:
> 
> drivers/leds/leds-tca6507.c:158: error: field 'gpio' has incomplete type
> drivers/leds/leds-tca6507.c: In function 'tca6507_gpio_set_value':
> drivers/leds/leds-tca6507.c:534: warning: type defaults to 'int' in declaration of '__mptr'
> drivers/leds/leds-tca6507.c:534: warning: initialization from incompatible pointer type
> drivers/leds/leds-tca6507.c: In function 'tca6507_probe':
> drivers/leds/leds-tca6507.c:618: error: implicit declaration of function 'gpiochip_add'
> drivers/leds/leds-tca6507.c: In function 'tca6507_remove':
> drivers/leds/leds-tca6507.c:651: error: implicit declaration of function 'gpiochip_remove'
> 
> Caused by commit 7a9244d89523 ("leds: add driver for TCA6507 LED controller").
> 
> Maybe this driver needs some dependencies ...

Yes... I think it actually needs the 'gpio' aspect of the driver to be
dependant of GPIOLIB being available.

I'll get a patch or two off to Andrew on Monday.

Thanks,
NeilBrown

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 828 bytes --]

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 604 bytes --]

Hi Andrew,

After merging the final tree, today's linux-next build (i386 defconfig)
failed like this:

kernel/cpuset.c: In function 'cpuset_change_task_nodemask':
kernel/cpuset.c:971:17: error: 'struct task_struct' has no member named 'mempolicy'

Caused by commit abf5d6d23d83 ("cpusets: stall when updating mems_allowed
for mempolicy or disjoint nodemask") from the akpm tree.  The mempolicy
member is only available when CONFIG_NUMA is set.

I have reverted that commit for today.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

Re: linux-next: build failure after merge of the final tree (akpm tree related)

[Andrew Morton]

On Wed, 30 Nov 2011 15:42:31 +1100 Stephen Rothwell <sfr@canb.auug.org.au> wrote:

> Hi Andrew,
> 
> After merging the final tree, today's linux-next build (i386 defconfig)
> failed like this:
> 
> kernel/cpuset.c: In function 'cpuset_change_task_nodemask':
> kernel/cpuset.c:971:17: error: 'struct task_struct' has no member named 'mempolicy'
> 
> Caused by commit abf5d6d23d83 ("cpusets: stall when updating mems_allowed
> for mempolicy or disjoint nodemask") from the akpm tree.  The mempolicy
> member is only available when CONFIG_NUMA is set.
> 

This obviously can be hacked around, but one wonders whether we really
need to include things like cpuset_change_task_nodemask() in a
CONFIG_NUMA=n vmlinux?

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1307 bytes --]

Hi all,

After merging the final tree, today's linux-next build (powerpc allyesconfig)
failed like this:

kernel/cgroup_task_counter.c: In function 'task_counter_create':
kernel/cgroup_task_counter.c:60:3: error: implicit declaration of function 'ERR_PTR' [-Werror=implicit-function-declaration]
kernel/cgroup_task_counter.c:60:3: warning: return makes pointer from integer without a cast [enabled by default]

Caused by commit 7d8662e11b18 ("cgroups: add a task counter subsystem")
from the akpm tree.

I added the following patch for today:

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Wed, 5 Oct 2011 19:20:15 +1100
Subject: [PATCH] cgroups: ERR_PTR needs err.h

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 kernel/cgroup_task_counter.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/kernel/cgroup_task_counter.c b/kernel/cgroup_task_counter.c
index 2374905..201d8e3 100644
--- a/kernel/cgroup_task_counter.c
+++ b/kernel/cgroup_task_counter.c
@@ -11,6 +11,7 @@
 #include <linux/cgroup.h>
 #include <linux/slab.h>
 #include <linux/res_counter.h>
+#include <linux/err.h>
 
 
 struct task_counter {
-- 
1.7.6.3

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1162 bytes --]

Hi all,

After merging the final tree, the next-20110929 build (um defconfig)
failed like this:

In file included from arch/um/os-Linux/sys-i386/registers.c:13:
arch/um/include/shared/user.h:26: error: expected declaration specifiers or '...' before numeric constant
arch/um/include/shared/user.h:26: error: expected declaration specifiers or '...' before numeric constant
arch/um/include/shared/user.h:26: warning: return type defaults to 'int'
arch/um/include/shared/user.h:26: warning: function declaration isn't a prototype
arch/um/include/shared/user.h: In function '__printf':
arch/um/include/shared/user.h:29: error: expected declaration specifiers or '...' before numeric constant
arch/um/include/shared/user.h:29: error: expected declaration specifiers or '...' before numeric constant
arch/um/include/shared/user.h:29: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'int'

And it went downhill from here ...

Caused by commit 820b1a71d7a7 ("treewide: use __printf not __attribute__
((format(printf,...)))").

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 490 bytes --]

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

Hi all,

After merging the final tree, today's linux-next build (sparc32 defconfig)
failed like this:

fs/direct-io.c: In function '__blockdev_direct_IO':
fs/direct-io.c:1327:2: error: implicit declaration of function 'prefetch'

Caused by commit 30ddb4d48213 ("dio: optimize cache misses in the
submission path").

I have added this patch for today:

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Wed, 28 Sep 2011 19:58:28 +1000
Subject: [PATCH] dio: using prefetch requires including prefetch.h

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 fs/direct-io.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/fs/direct-io.c b/fs/direct-io.c
index a05ca05..8d22849 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -36,6 +36,7 @@
 #include <linux/rwsem.h>
 #include <linux/uio.h>
 #include <linux/atomic.h>
+#include <linux/prefetch.h>
 
 /*
  * How many user pages to map in one call to get_user_pages().  This determines
-- 
1.7.6.3

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

Hi all,

After merging the final tree, today's linux-next build (sparc32 defconfig)
failed like this:

arch/sparc/lib/atomic32.c:58:5: error: redefinition of 'atomic_add_unless'
include/linux/atomic.h:15:19: note: previous definition of 'atomic_add_unless' was here

Presumably caused by commit 1af08a1407f4 ("This is in preparation for
more generic atomic").

That commit is now in Linus' tree as commit f24219b4e90c ("atomic: move
atomic_add_unless to generic code"), so this patch is probably applicable
there, now.

I have applied this patch for today:

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Wed, 27 Jul 2011 13:48:55 +1000
Subject: [PATCH] sparc: rename atomic_add_unless

Should have been done in commit 1af08a1407f4 ("This is in preparation
for more generic atomic").

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 arch/sparc/lib/atomic32.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/sparc/lib/atomic32.c b/arch/sparc/lib/atomic32.c
index 1a371f8..8600eb2 100644
--- a/arch/sparc/lib/atomic32.c
+++ b/arch/sparc/lib/atomic32.c
@@ -55,7 +55,7 @@ int atomic_cmpxchg(atomic_t *v, int old, int new)
 }
 EXPORT_SYMBOL(atomic_cmpxchg);
 
-int atomic_add_unless(atomic_t *v, int a, int u)
+int __atomic_add_unless(atomic_t *v, int a, int u)
 {
 	int ret;
 	unsigned long flags;
@@ -67,7 +67,7 @@ int atomic_add_unless(atomic_t *v, int a, int u)
 	spin_unlock_irqrestore(ATOMIC_HASH(v), flags);
 	return ret != u;
 }
-EXPORT_SYMBOL(atomic_add_unless);
+EXPORT_SYMBOL(__atomic_add_unless);
 
 /* Atomic operations are already serializing */
 void atomic_set(atomic_t *v, int i)
-- 
1.7.5.4

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

linux-next: build failure after merge of the final tree (akpm tree related)

[Stephen Rothwell]

Hi all,

After merging the final tree, today's linux-next build (powerpc
allyesconfig) failed like this:

drivers/video/backlight/ams369fg06.o:(.bss+0x510): multiple definition of `before_power'
drivers/video/backlight/s6e63m0.o:(.bss+0x5e0): first defined here

Caused by commit 45193c8486ef ("Add the ams369fg06 amoled panel driver.
The ams369fg06 amoled panel (480").  The variable should probably be
static in both files.

I have applied the following patch for today:

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Fri, 1 Jul 2011 15:34:22 +1000
Subject: [PATCH] backlight: make some varibales static

fixes this build error:

drivers/video/backlight/ams369fg06.o:(.bss+0x510): multiple definition of `before_power'
drivers/video/backlight/s6e63m0.o:(.bss+0x5e0): first defined here

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 drivers/video/backlight/ams369fg06.c |    2 +-
 drivers/video/backlight/s6e63m0.c    |    2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/video/backlight/ams369fg06.c b/drivers/video/backlight/ams369fg06.c
index 86a7daf..d1d8237 100644
--- a/drivers/video/backlight/ams369fg06.c
+++ b/drivers/video/backlight/ams369fg06.c
@@ -560,7 +560,7 @@ static int __devexit ams369fg06_remove(struct spi_device *spi)
 }
 
 #if defined(CONFIG_PM)
-unsigned int before_power;
+static unsigned int before_power;
 
 static int ams369fg06_suspend(struct spi_device *spi, pm_message_t mesg)
 {
diff --git a/drivers/video/backlight/s6e63m0.c b/drivers/video/backlight/s6e63m0.c
index 322040f..e1803ba 100644
--- a/drivers/video/backlight/s6e63m0.c
+++ b/drivers/video/backlight/s6e63m0.c
@@ -840,7 +840,7 @@ static int __devexit s6e63m0_remove(struct spi_device *spi)
 }
 
 #if defined(CONFIG_PM)
-unsigned int before_power;
+static unsigned int before_power;
 
 static int s6e63m0_suspend(struct spi_device *spi, pm_message_t mesg)
 {
-- 
1.7.5.4


-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/