[RFC PATCH] powerpc: Fix dubious r0 usage
Michael Ellerman
mpe at ellerman.id.au
Fri Feb 17 23:08:53 AEDT 2017
Bleeding edge binutils no longer accepts r0 in places where the CPU
interprets the value as a literal 0.
This fixes the cases caught by a vanilla 64-bit and 32-bit build. There
may be others that are hidden behind other CONFIG options.
Signed-off-by: Michael Ellerman <mpe at ellerman.id.au>
---
arch/powerpc/include/asm/ppc_asm.h | 2 +-
arch/powerpc/kernel/l2cr_6xx.S | 4 +--
arch/powerpc/lib/copypage_power7.S | 14 ++++----
arch/powerpc/lib/copyuser_power7.S | 66 ++++++++++++++++++-------------------
arch/powerpc/lib/memcpy_power7.S | 66 ++++++++++++++++++-------------------
arch/powerpc/lib/string_64.S | 2 +-
arch/powerpc/purgatory/trampoline.S | 2 +-
7 files changed, 78 insertions(+), 78 deletions(-)
diff --git a/arch/powerpc/include/asm/ppc_asm.h b/arch/powerpc/include/asm/ppc_asm.h
index 025833b8df9f..c98324addc08 100644
--- a/arch/powerpc/include/asm/ppc_asm.h
+++ b/arch/powerpc/include/asm/ppc_asm.h
@@ -439,7 +439,7 @@ END_FTR_SECTION_IFCLR(CPU_FTR_601)
.machine push ; \
.machine "power4" ; \
lis scratch,0x60000000 at h; \
- dcbt r0,scratch,0b01010; \
+ dcbt 0,scratch,0b01010; \
.machine pop
/*
diff --git a/arch/powerpc/kernel/l2cr_6xx.S b/arch/powerpc/kernel/l2cr_6xx.S
index 97ec8557f974..6408f09dbbd9 100644
--- a/arch/powerpc/kernel/l2cr_6xx.S
+++ b/arch/powerpc/kernel/l2cr_6xx.S
@@ -181,7 +181,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_SPEC7450)
mtctr r4
li r4,0
1:
- lwzx r0,r0,r4
+ lwzx r0,0,r4
addi r4,r4,32 /* Go to start of next cache line */
bdnz 1b
isync
@@ -328,7 +328,7 @@ END_FTR_SECTION_IFCLR(CPU_FTR_L3CR)
mtctr r4
li r4,0
1:
- lwzx r0,r0,r4
+ lwzx r0,0,r4
dcbf 0,r4
addi r4,r4,32 /* Go to start of next cache line */
bdnz 1b
diff --git a/arch/powerpc/lib/copypage_power7.S b/arch/powerpc/lib/copypage_power7.S
index a84d333ecb09..ca5fc8fa7efc 100644
--- a/arch/powerpc/lib/copypage_power7.S
+++ b/arch/powerpc/lib/copypage_power7.S
@@ -45,13 +45,13 @@ _GLOBAL(copypage_power7)
.machine push
.machine "power4"
/* setup read stream 0 */
- dcbt r0,r4,0b01000 /* addr from */
- dcbt r0,r7,0b01010 /* length and depth from */
+ dcbt 0,r4,0b01000 /* addr from */
+ dcbt 0,r7,0b01010 /* length and depth from */
/* setup write stream 1 */
- dcbtst r0,r9,0b01000 /* addr to */
- dcbtst r0,r10,0b01010 /* length and depth to */
+ dcbtst 0,r9,0b01000 /* addr to */
+ dcbtst 0,r10,0b01010 /* length and depth to */
eieio
- dcbt r0,r8,0b01010 /* all streams GO */
+ dcbt 0,r8,0b01010 /* all streams GO */
.machine pop
#ifdef CONFIG_ALTIVEC
@@ -83,7 +83,7 @@ _GLOBAL(copypage_power7)
li r12,112
.align 5
-1: lvx v7,r0,r4
+1: lvx v7,0,r4
lvx v6,r4,r6
lvx v5,r4,r7
lvx v4,r4,r8
@@ -92,7 +92,7 @@ _GLOBAL(copypage_power7)
lvx v1,r4,r11
lvx v0,r4,r12
addi r4,r4,128
- stvx v7,r0,r3
+ stvx v7,0,r3
stvx v6,r3,r6
stvx v5,r3,r7
stvx v4,r3,r8
diff --git a/arch/powerpc/lib/copyuser_power7.S b/arch/powerpc/lib/copyuser_power7.S
index a24b4039352c..a95fef3bed10 100644
--- a/arch/powerpc/lib/copyuser_power7.S
+++ b/arch/powerpc/lib/copyuser_power7.S
@@ -315,13 +315,13 @@ err1; stb r0,0(r3)
.machine push
.machine "power4"
/* setup read stream 0 */
- dcbt r0,r6,0b01000 /* addr from */
- dcbt r0,r7,0b01010 /* length and depth from */
+ dcbt 0,r6,0b01000 /* addr from */
+ dcbt 0,r7,0b01010 /* length and depth from */
/* setup write stream 1 */
- dcbtst r0,r9,0b01000 /* addr to */
- dcbtst r0,r10,0b01010 /* length and depth to */
+ dcbtst 0,r9,0b01000 /* addr to */
+ dcbtst 0,r10,0b01010 /* length and depth to */
eieio
- dcbt r0,r8,0b01010 /* all streams GO */
+ dcbt 0,r8,0b01010 /* all streams GO */
.machine pop
beq cr1,.Lunwind_stack_nonvmx_copy
@@ -376,26 +376,26 @@ err3; std r0,0(r3)
li r11,48
bf cr7*4+3,5f
-err3; lvx v1,r0,r4
+err3; lvx v1,0,r4
addi r4,r4,16
-err3; stvx v1,r0,r3
+err3; stvx v1,0,r3
addi r3,r3,16
5: bf cr7*4+2,6f
-err3; lvx v1,r0,r4
+err3; lvx v1,0,r4
err3; lvx v0,r4,r9
addi r4,r4,32
-err3; stvx v1,r0,r3
+err3; stvx v1,0,r3
err3; stvx v0,r3,r9
addi r3,r3,32
6: bf cr7*4+1,7f
-err3; lvx v3,r0,r4
+err3; lvx v3,0,r4
err3; lvx v2,r4,r9
err3; lvx v1,r4,r10
err3; lvx v0,r4,r11
addi r4,r4,64
-err3; stvx v3,r0,r3
+err3; stvx v3,0,r3
err3; stvx v2,r3,r9
err3; stvx v1,r3,r10
err3; stvx v0,r3,r11
@@ -421,7 +421,7 @@ err3; stvx v0,r3,r11
*/
.align 5
8:
-err4; lvx v7,r0,r4
+err4; lvx v7,0,r4
err4; lvx v6,r4,r9
err4; lvx v5,r4,r10
err4; lvx v4,r4,r11
@@ -430,7 +430,7 @@ err4; lvx v2,r4,r14
err4; lvx v1,r4,r15
err4; lvx v0,r4,r16
addi r4,r4,128
-err4; stvx v7,r0,r3
+err4; stvx v7,0,r3
err4; stvx v6,r3,r9
err4; stvx v5,r3,r10
err4; stvx v4,r3,r11
@@ -451,29 +451,29 @@ err4; stvx v0,r3,r16
mtocrf 0x01,r6
bf cr7*4+1,9f
-err3; lvx v3,r0,r4
+err3; lvx v3,0,r4
err3; lvx v2,r4,r9
err3; lvx v1,r4,r10
err3; lvx v0,r4,r11
addi r4,r4,64
-err3; stvx v3,r0,r3
+err3; stvx v3,0,r3
err3; stvx v2,r3,r9
err3; stvx v1,r3,r10
err3; stvx v0,r3,r11
addi r3,r3,64
9: bf cr7*4+2,10f
-err3; lvx v1,r0,r4
+err3; lvx v1,0,r4
err3; lvx v0,r4,r9
addi r4,r4,32
-err3; stvx v1,r0,r3
+err3; stvx v1,0,r3
err3; stvx v0,r3,r9
addi r3,r3,32
10: bf cr7*4+3,11f
-err3; lvx v1,r0,r4
+err3; lvx v1,0,r4
addi r4,r4,16
-err3; stvx v1,r0,r3
+err3; stvx v1,0,r3
addi r3,r3,16
/* Up to 15B to go */
@@ -553,25 +553,25 @@ err3; lvx v0,0,r4
addi r4,r4,16
bf cr7*4+3,5f
-err3; lvx v1,r0,r4
+err3; lvx v1,0,r4
VPERM(v8,v0,v1,v16)
addi r4,r4,16
-err3; stvx v8,r0,r3
+err3; stvx v8,0,r3
addi r3,r3,16
vor v0,v1,v1
5: bf cr7*4+2,6f
-err3; lvx v1,r0,r4
+err3; lvx v1,0,r4
VPERM(v8,v0,v1,v16)
err3; lvx v0,r4,r9
VPERM(v9,v1,v0,v16)
addi r4,r4,32
-err3; stvx v8,r0,r3
+err3; stvx v8,0,r3
err3; stvx v9,r3,r9
addi r3,r3,32
6: bf cr7*4+1,7f
-err3; lvx v3,r0,r4
+err3; lvx v3,0,r4
VPERM(v8,v0,v3,v16)
err3; lvx v2,r4,r9
VPERM(v9,v3,v2,v16)
@@ -580,7 +580,7 @@ err3; lvx v1,r4,r10
err3; lvx v0,r4,r11
VPERM(v11,v1,v0,v16)
addi r4,r4,64
-err3; stvx v8,r0,r3
+err3; stvx v8,0,r3
err3; stvx v9,r3,r9
err3; stvx v10,r3,r10
err3; stvx v11,r3,r11
@@ -606,7 +606,7 @@ err3; stvx v11,r3,r11
*/
.align 5
8:
-err4; lvx v7,r0,r4
+err4; lvx v7,0,r4
VPERM(v8,v0,v7,v16)
err4; lvx v6,r4,r9
VPERM(v9,v7,v6,v16)
@@ -623,7 +623,7 @@ err4; lvx v1,r4,r15
err4; lvx v0,r4,r16
VPERM(v15,v1,v0,v16)
addi r4,r4,128
-err4; stvx v8,r0,r3
+err4; stvx v8,0,r3
err4; stvx v9,r3,r9
err4; stvx v10,r3,r10
err4; stvx v11,r3,r11
@@ -644,7 +644,7 @@ err4; stvx v15,r3,r16
mtocrf 0x01,r6
bf cr7*4+1,9f
-err3; lvx v3,r0,r4
+err3; lvx v3,0,r4
VPERM(v8,v0,v3,v16)
err3; lvx v2,r4,r9
VPERM(v9,v3,v2,v16)
@@ -653,27 +653,27 @@ err3; lvx v1,r4,r10
err3; lvx v0,r4,r11
VPERM(v11,v1,v0,v16)
addi r4,r4,64
-err3; stvx v8,r0,r3
+err3; stvx v8,0,r3
err3; stvx v9,r3,r9
err3; stvx v10,r3,r10
err3; stvx v11,r3,r11
addi r3,r3,64
9: bf cr7*4+2,10f
-err3; lvx v1,r0,r4
+err3; lvx v1,0,r4
VPERM(v8,v0,v1,v16)
err3; lvx v0,r4,r9
VPERM(v9,v1,v0,v16)
addi r4,r4,32
-err3; stvx v8,r0,r3
+err3; stvx v8,0,r3
err3; stvx v9,r3,r9
addi r3,r3,32
10: bf cr7*4+3,11f
-err3; lvx v1,r0,r4
+err3; lvx v1,0,r4
VPERM(v8,v0,v1,v16)
addi r4,r4,16
-err3; stvx v8,r0,r3
+err3; stvx v8,0,r3
addi r3,r3,16
/* Up to 15B to go */
diff --git a/arch/powerpc/lib/memcpy_power7.S b/arch/powerpc/lib/memcpy_power7.S
index 786234fd4e91..193909abd18b 100644
--- a/arch/powerpc/lib/memcpy_power7.S
+++ b/arch/powerpc/lib/memcpy_power7.S
@@ -261,12 +261,12 @@ _GLOBAL(memcpy_power7)
.machine push
.machine "power4"
- dcbt r0,r6,0b01000
- dcbt r0,r7,0b01010
- dcbtst r0,r9,0b01000
- dcbtst r0,r10,0b01010
+ dcbt 0,r6,0b01000
+ dcbt 0,r7,0b01010
+ dcbtst 0,r9,0b01000
+ dcbtst 0,r10,0b01010
eieio
- dcbt r0,r8,0b01010 /* GO */
+ dcbt 0,r8,0b01010 /* GO */
.machine pop
beq cr1,.Lunwind_stack_nonvmx_copy
@@ -321,26 +321,26 @@ _GLOBAL(memcpy_power7)
li r11,48
bf cr7*4+3,5f
- lvx v1,r0,r4
+ lvx v1,0,r4
addi r4,r4,16
- stvx v1,r0,r3
+ stvx v1,0,r3
addi r3,r3,16
5: bf cr7*4+2,6f
- lvx v1,r0,r4
+ lvx v1,0,r4
lvx v0,r4,r9
addi r4,r4,32
- stvx v1,r0,r3
+ stvx v1,0,r3
stvx v0,r3,r9
addi r3,r3,32
6: bf cr7*4+1,7f
- lvx v3,r0,r4
+ lvx v3,0,r4
lvx v2,r4,r9
lvx v1,r4,r10
lvx v0,r4,r11
addi r4,r4,64
- stvx v3,r0,r3
+ stvx v3,0,r3
stvx v2,r3,r9
stvx v1,r3,r10
stvx v0,r3,r11
@@ -366,7 +366,7 @@ _GLOBAL(memcpy_power7)
*/
.align 5
8:
- lvx v7,r0,r4
+ lvx v7,0,r4
lvx v6,r4,r9
lvx v5,r4,r10
lvx v4,r4,r11
@@ -375,7 +375,7 @@ _GLOBAL(memcpy_power7)
lvx v1,r4,r15
lvx v0,r4,r16
addi r4,r4,128
- stvx v7,r0,r3
+ stvx v7,0,r3
stvx v6,r3,r9
stvx v5,r3,r10
stvx v4,r3,r11
@@ -396,29 +396,29 @@ _GLOBAL(memcpy_power7)
mtocrf 0x01,r6
bf cr7*4+1,9f
- lvx v3,r0,r4
+ lvx v3,0,r4
lvx v2,r4,r9
lvx v1,r4,r10
lvx v0,r4,r11
addi r4,r4,64
- stvx v3,r0,r3
+ stvx v3,0,r3
stvx v2,r3,r9
stvx v1,r3,r10
stvx v0,r3,r11
addi r3,r3,64
9: bf cr7*4+2,10f
- lvx v1,r0,r4
+ lvx v1,0,r4
lvx v0,r4,r9
addi r4,r4,32
- stvx v1,r0,r3
+ stvx v1,0,r3
stvx v0,r3,r9
addi r3,r3,32
10: bf cr7*4+3,11f
- lvx v1,r0,r4
+ lvx v1,0,r4
addi r4,r4,16
- stvx v1,r0,r3
+ stvx v1,0,r3
addi r3,r3,16
/* Up to 15B to go */
@@ -499,25 +499,25 @@ _GLOBAL(memcpy_power7)
addi r4,r4,16
bf cr7*4+3,5f
- lvx v1,r0,r4
+ lvx v1,0,r4
VPERM(v8,v0,v1,v16)
addi r4,r4,16
- stvx v8,r0,r3
+ stvx v8,0,r3
addi r3,r3,16
vor v0,v1,v1
5: bf cr7*4+2,6f
- lvx v1,r0,r4
+ lvx v1,0,r4
VPERM(v8,v0,v1,v16)
lvx v0,r4,r9
VPERM(v9,v1,v0,v16)
addi r4,r4,32
- stvx v8,r0,r3
+ stvx v8,0,r3
stvx v9,r3,r9
addi r3,r3,32
6: bf cr7*4+1,7f
- lvx v3,r0,r4
+ lvx v3,0,r4
VPERM(v8,v0,v3,v16)
lvx v2,r4,r9
VPERM(v9,v3,v2,v16)
@@ -526,7 +526,7 @@ _GLOBAL(memcpy_power7)
lvx v0,r4,r11
VPERM(v11,v1,v0,v16)
addi r4,r4,64
- stvx v8,r0,r3
+ stvx v8,0,r3
stvx v9,r3,r9
stvx v10,r3,r10
stvx v11,r3,r11
@@ -552,7 +552,7 @@ _GLOBAL(memcpy_power7)
*/
.align 5
8:
- lvx v7,r0,r4
+ lvx v7,0,r4
VPERM(v8,v0,v7,v16)
lvx v6,r4,r9
VPERM(v9,v7,v6,v16)
@@ -569,7 +569,7 @@ _GLOBAL(memcpy_power7)
lvx v0,r4,r16
VPERM(v15,v1,v0,v16)
addi r4,r4,128
- stvx v8,r0,r3
+ stvx v8,0,r3
stvx v9,r3,r9
stvx v10,r3,r10
stvx v11,r3,r11
@@ -590,7 +590,7 @@ _GLOBAL(memcpy_power7)
mtocrf 0x01,r6
bf cr7*4+1,9f
- lvx v3,r0,r4
+ lvx v3,0,r4
VPERM(v8,v0,v3,v16)
lvx v2,r4,r9
VPERM(v9,v3,v2,v16)
@@ -599,27 +599,27 @@ _GLOBAL(memcpy_power7)
lvx v0,r4,r11
VPERM(v11,v1,v0,v16)
addi r4,r4,64
- stvx v8,r0,r3
+ stvx v8,0,r3
stvx v9,r3,r9
stvx v10,r3,r10
stvx v11,r3,r11
addi r3,r3,64
9: bf cr7*4+2,10f
- lvx v1,r0,r4
+ lvx v1,0,r4
VPERM(v8,v0,v1,v16)
lvx v0,r4,r9
VPERM(v9,v1,v0,v16)
addi r4,r4,32
- stvx v8,r0,r3
+ stvx v8,0,r3
stvx v9,r3,r9
addi r3,r3,32
10: bf cr7*4+3,11f
- lvx v1,r0,r4
+ lvx v1,0,r4
VPERM(v8,v0,v1,v16)
addi r4,r4,16
- stvx v8,r0,r3
+ stvx v8,0,r3
addi r3,r3,16
/* Up to 15B to go */
diff --git a/arch/powerpc/lib/string_64.S b/arch/powerpc/lib/string_64.S
index c100f4d5d5d0..7c1402e60063 100644
--- a/arch/powerpc/lib/string_64.S
+++ b/arch/powerpc/lib/string_64.S
@@ -184,7 +184,7 @@ err1; std r0,8(r3)
mtctr r6
mr r8,r3
14:
-err1; dcbz r0,r3
+err1; dcbz 0,r3
add r3,r3,r9
bdnz 14b
diff --git a/arch/powerpc/purgatory/trampoline.S b/arch/powerpc/purgatory/trampoline.S
index f9760ccf4032..0017d2bf4036 100644
--- a/arch/powerpc/purgatory/trampoline.S
+++ b/arch/powerpc/purgatory/trampoline.S
@@ -67,7 +67,7 @@ master:
mr %r16,%r3 /* save dt address in reg16 */
li %r4,20
LWZX_BE %r6,%r3,%r4 /* fetch __be32 version number at byte 20 */
- cmpwi %r0,%r6,2 /* v2 or later? */
+ cmpwi 0,%r6,2 /* v2 or later? */
blt 1f
li %r4,28
STWX_BE %r17,%r3,%r4 /* Store my cpu as __be32 at byte 28 */
--
2.7.4
More information about the Linuxppc-dev
mailing list