From: jonathan.austin@arm.com (Jonathan Austin)
Date: Tue, 25 Jun 2013 12:30:08 +0100
Subject: amba-pl08x and 'get_signal' namespace collision/build error
Message-ID: <51C97F40.5030101@arm.com>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

Hi all,

There's a patch making its way to mainline via Russell's tree
(8d96250700: ARM: mm: Transparent huge page support for LPAE systems)
that breaks the build of the amba-pl08x driver (drivers/dma/amba-pl08x.c) 
because the 'get_signal' macro from include/linux/signal.h is now in the
driver's scope and it clobbers a (previously) valid function call.

The error:
drivers/dma/amba-pl08x.c: In function ?pl08x_request_mux?:
drivers/dma/amba-pl08x.c:303:13: error: expected identifier before ?(? token

Here's the problematic include:
In file included from include/linux/sched.h:32:0,
--new include--> from /data/build/a32/linux/arch/arm/include/asm/tlbflush.h:204,
                 from /data/build/a32/linux/arch/arm/include/asm/pgtable.h:28,
                 from include/linux/mm.h:44,
                 from include/linux/scatterlist.h:6,
                 from include/linux/dmaengine.h:27,
                 from include/linux/amba/pl08x.h:21,
                 from drivers/dma/amba-pl08x.c:74:
include/linux/signal.h:298:2: error: #error get_signal defined here

Below is a 'fix' that *doesn't* require any renaming of either the
get_signal function or macro, but there's nothing to say that driver
shouldn't be able to use the pl08x_platform_data struct *and* do
scheduling so this doesn't seem like the best fix really.

So can we do something better?

Al: the commit where you add the get_signal() helper suggests you want
to make it a function not a macro in the future - what needs to be done
before that can happen? I think that would be the best/easiest fix.


Jonny

----------8<-----------
diff --git a/arch/arm/include/asm/tlbflush.h b/arch/arm/include/asm/tlbflush.h
index c78efbc..ce1185c 100644
--- a/arch/arm/include/asm/tlbflush.h
+++ b/arch/arm/include/asm/tlbflush.h
@@ -201,7 +201,10 @@
 
 #ifndef __ASSEMBLY__
 
-#include <linux/sched.h>
+#include <linux/smp.h>
+#include <linux/mm_types.h>
+
+#include <asm/barrier.h>
 
 struct cpu_tlb_fns {
        void (*flush_user_range)(unsigned long, unsigned long, struct vm_area_struct *);
diff --git a/arch/arm/kernel/suspend.c b/arch/arm/kernel/suspend.c
index 38a5067..3100983 100644
--- a/arch/arm/kernel/suspend.c
+++ b/arch/arm/kernel/suspend.c
@@ -1,4 +1,5 @@
 #include <linux/init.h>
+#include <linux/sched.h>
 
 #include <asm/idmap.h>
 #include <asm/pgalloc.h>
diff --git a/arch/arm/mm/idmap.c b/arch/arm/mm/idmap.c
index 83cb3ac..304e8ce 100644
--- a/arch/arm/mm/idmap.c
+++ b/arch/arm/mm/idmap.c
@@ -1,6 +1,7 @@
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/slab.h>
+#include <linux/sched.h>
 
 #include <asm/cputype.h>
 #include <asm/idmap.h>
diff --git a/arch/arm/mm/mm.h b/arch/arm/mm/mm.h
index d5a4e9a..6eeaf3d 100644
--- a/arch/arm/mm/mm.h
+++ b/arch/arm/mm/mm.h
@@ -1,6 +1,7 @@
 #ifdef CONFIG_MMU
 #include <linux/list.h>
 #include <linux/vmalloc.h>
+#include <linux/sched.h>
 
 /* the upper-most page table pointer */
 extern pmd_t *top_pmd;