On Tue, 2016-03-15 at 14:07 +0100, Christophe Leroy wrote:I mean it doesn't fail with 4.8.3
Some versions of GCC, reportedly before 4.8, fail with"before 4.8" means "< 4.8", not "<= 4.8" -- did you mean "before 4.9"?
arch/powerpc/mm/8xx_mmu.c:139:2: error: memory input 1 is not directly
addressable
Change the one-element array into a simple variable to avoid this.I thought I tried this with 4.7.2 without success, but now it seems to be
Signed-off-by: Christophe Leroy <christophe.leroy@xxxxxx>
Cc: Scott Wood <oss@xxxxxxxxxxxx>
---
Verified with GCC 4.4.4 and GCC 4.8.3
v2: compilation result is compliant with CPU6 ERRATA
(using stw/lwz and not stwx/lwzx)
arch/powerpc/include/asm/reg_8xx.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/powerpc/include/asm/reg_8xx.h
b/arch/powerpc/include/asm/reg_8xx.h
index d41412c..94d01f8 100644
--- a/arch/powerpc/include/asm/reg_8xx.h
+++ b/arch/powerpc/include/asm/reg_8xx.h
@@ -53,7 +53,7 @@
#ifdef CONFIG_8xx_CPU6
#define do_mtspr_cpu6(rn, rn_addr, v) \
do {
\
- int _reg_cpu6 = rn_addr, _tmp_cpu6[1]; \
+ int _reg_cpu6 = rn_addr, _tmp_cpu6; \
asm volatile("stw %0, %1;" \
"lwz %0, %1;" \
"mtspr " __stringify(rn) ",%2" :
\
working...
-Scott