While trying to see how bad a "allyesconfig" looked for sparc32
I noticed a lot of build errors were related to missing
definition of atomic64.
A few samples:
kernel/trace/trace_clock.c:117:19: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'trace_counter'
kernel/trace/trace_clock.c: In function 'trace_clock_counter':
kernel/trace/trace_clock.c:126:2: error: implicit declaration of function 'atomic64_add_return'
In file included from net/mac80211/ieee80211_i.h:30:0,
from net/mac80211/main.c:27:
net/mac80211/key.h:88:4: error: expected specifier-qualifier-list before 'atomic64_t'
net/mac80211/key.h:100:4: error: expected specifier-qualifier-list before 'atomic64_t'
make[3]: *** [net/mac80211/main.o] Error 1
In file included from net/mac80211/ieee80211_i.h:30:0,
from net/mac80211/chan.c:6:
net/mac80211/key.h:88:4: error: expected specifier-qualifier-list before 'atomic64_t'
net/mac80211/key.h:100:4: error: expected specifier-qualifier-list before 'atomic64_t'
To add generic support for atomic64_t looks to be trivial.
But do we want to add this - when it is not really required?
We could of course hide it behind a
CONFIG_SPARC32_ATOMIC64 option..
I do not have the footprint numbers yet, will post them in ~24 hours.
Sam
diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index f92602e..27060b2 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -31,6 +31,7 @@ config SPARC
config SPARC32
def_bool !64BIT
+ select GENERIC_ATOMIC64
config SPARC64
def_bool 64BIT
diff --git a/arch/sparc/include/asm/atomic_32.h b/arch/sparc/include/asm/atomic_32.h
index 03fd0c5..9dd0a76 100644
--- a/arch/sparc/include/asm/atomic_32.h
+++ b/arch/sparc/include/asm/atomic_32.h
@@ -13,6 +13,8 @@
#include <linux/types.h>
+#include <asm-generic/atomic64.h>
+
#include <asm/system.h>
#define ATOMIC_INIT(i) { (i) }
--
To unsubscribe from this list: send the line "unsubscribe sparclinux" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
