linux-next: voyager tree build failure

linux-next: voyager tree build failure

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 557 bytes --]

Hi James,

Today's linux-next build (x86_64 allmodconfig) failed like this:

arch/x86/xen/smp.c:478: error: 'apic_safe_smp_processor_id' undeclared here (not in a function)
arch/x86/kernel/smp.c:219: error: 'apic_safe_smp_processor_id' undeclared here (not in a function)

Caused by commit 6a6df48477c1959b1baa78ecabba72548d362f60 ("[VOYAGER]
x86: add {safe,hard}_smp_processor_id to smp_ops").

I have dropped the voyager tree for today.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]

Re: linux-next: voyager tree build failure

[James Bottomley]

On Tue, 2009-06-09 at 19:40 +1000, Stephen Rothwell wrote:
> Hi James,
> 
> Today's linux-next build (x86_64 allmodconfig) failed like this:
> 
> arch/x86/xen/smp.c:478: error: 'apic_safe_smp_processor_id' undeclared here (not in a function)
> arch/x86/kernel/smp.c:219: error: 'apic_safe_smp_processor_id' undeclared here (not in a function)
> 
> Caused by commit 6a6df48477c1959b1baa78ecabba72548d362f60 ("[VOYAGER]
> x86: add {safe,hard}_smp_processor_id to smp_ops").
> 
> I have dropped the voyager tree for today.

Thanks for finding this.  The fix looks to be a dummy definition of this
function for x86_64.  The final fix (which has been under discussion)
will be the elimination of safe_smp_processor_id() altogether.

I've merged this into the

[VOYAGER] x86: add {safe,hard}_smp_processor_id to smp_ops

patch and respun the tree (and built it with an x86-64 cross compiler),
so it should be safe to include next time around

Thanks,

James

---

diff -u b/arch/x86/include/asm/smp.h b/arch/x86/include/asm/smp.h
--- b/arch/x86/include/asm/smp.h
+++ b/arch/x86/include/asm/smp.h
@@ -177,6 +177,15 @@
 	ti->cpu;							\
 })
 #define safe_smp_processor_id()		smp_processor_id()
+/*
+ * dummy function for smp_ops.safe_smp_processor_id() which is unused
+ * on x86_64.  This will also break the build if the definition leaks
+ * outside of the guards in kernel/apic/ipi.c on x86_64
+ */
+static inline int apic_safe_smp_processor_id(void)
+{
+	return 0;
+}
 
 #endif
 

Re: linux-next: voyager tree build failure

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 727 bytes --]

Hi James,

On Tue, 09 Jun 2009 10:18:38 -0500 James Bottomley <James.Bottomley@HansenPartnership.com> wrote:
>
> Thanks for finding this.  The fix looks to be a dummy definition of this
> function for x86_64.  The final fix (which has been under discussion)
> will be the elimination of safe_smp_processor_id() altogether.
> 
> I've merged this into the
> 
> [VOYAGER] x86: add {safe,hard}_smp_processor_id to smp_ops
> 
> patch and respun the tree (and built it with an x86-64 cross compiler),
> so it should be safe to include next time around

Thanks.  We will see how we do (later) in the morning.

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

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]