Fix Thumb-2 undef handling for mixed-arch kernels

PATCH New Dave Martin 4 participants 18 messages Aug 16, 2011 – Aug 17, 2011 Arm (32-bit)

Share & tools

Dave Martin

author

2011-08-16 14:14 UTC #0 lore raw

cover Fix Thumb-2 undef handling for mixed-arch kernels

In kernels which support v6 and v7 platforms in a single binary such as omap2plus_defconfig for example, undefined instruction exceptions taken on Thumb instructions are not processed correctly, because the kernel assumes at build-time that Thumb-2 won't be supported in userspace.

This series implements a workaround, by allowing the __und_usr handler to check the CPU architecture at runtime, in affected kernels.

Changes since RFC:
* Renamed the global variable to __cpu_architecture and provided cpu_architecture() as in inline function in <asm/system.h>. This brings back the old API, but more efficiently.
* Dropped all the patches required to handle the reverted API change.
* cpu_architecture is now declared __pure, which may lead to slightly more efficient code in some cases.
* Added BUG_ON() to cpu_architecture() so that if __cpu_architecture is not set early enough, it gets noticed.

Thanks to Tixy and Nicolas Pitre for their helpful feedback which helped slim down this series.

Dave Martin (3):
ARM: Make cpu_alignment into a global variable
ARM: entry: Remove unnecessary masking when decoding Thumb-2
instructions
ARM: entry: Fix Thumb-2 undef handling for multi-CPU kernels

 arch/arm/include/asm/system.h |   11 +++++++++-
 arch/arm/kernel/entry-armv.S  |   44 ++++++++++++++++++++++++++++++++++++----
 arch/arm/kernel/setup.c       |    5 +++-
 3 files changed, 53 insertions(+), 7 deletions(-)

--
1.7.4.1

Dave Martin

author

2011-08-16 14:14 UTC #1 lore raw

PATCH 1/3 ARM: Make cpu_alignment into a global variable

The CPU architecture really should not be changing at runtime, so make it a global variable instead of a function.

The cpu_architecture() function in <asm/system.h> remains the correct way to read this variable from C code.

Signed-off-by Dave Martin <[email protected]>

---
 arch/arm/include/asm/system.h |   11 ++++++++++-
 arch/arm/kernel/setup.c       |    5 ++++-
 2 files changed, 14 insertions(+), 2 deletions(-)

arch/arm/include/asm/system.h +10 -1

index 832888d..b0445f7 100644
@@ -57,6 +57,8 @@
 
 #ifndef __ASSEMBLY__
 
+#include <linux/bug.h>
+#include <linux/compiler.h>
 #include <linux/linkage.h>
 #include <linux/irqflags.h>

@@ -104,7 +106,14 @@ struct mm_struct;
 extern void show_pte(struct mm_struct *mm, unsigned long addr);
 extern void __show_regs(struct pt_regs *);
 
-extern int cpu_architecture(void);
+extern int __cpu_architecture;
+
+static inline int __pure cpu_architecture(void)
+{
+	BUG_ON(__cpu_architecture == CPU_ARCH_UNKNOWN);
+	return __cpu_architecture;
+}
+
 extern void cpu_init(void);
 
 void arm_machine_restart(char mode, const char *cmd);

arch/arm/kernel/setup.c +4 -1

index 70bca64..1e0c1b3 100644
@@ -42,6 +42,7 @@
 #include <asm/cacheflush.h>
 #include <asm/cachetype.h>
 #include <asm/tlbflush.h>
+#include <asm/system.h>
 
 #include <asm/prom.h>
 #include <asm/mach/arch.h>

@@ -114,6 +115,7 @@ struct cpu_cache_fns cpu_cache __read_mostly;
 struct outer_cache_fns outer_cache __read_mostly;
 EXPORT_SYMBOL(outer_cache);
 #endif
+int __cpu_architecture __read_mostly = CPU_ARCH_UNKNOWN;
 
 struct stack {
 	u32 irq[3];

@@ -210,7 +212,7 @@ static const char *proc_arch[] = {
 	"?(17)",
 };
 
-int cpu_architecture(void)
+static int __init __get_cpu_architecture(void)
 {
 	int cpu_arch;

@@ -413,6 +415,7 @@ static void __init setup_processor(void)
 	}
 
 	cpu_name = list->cpu_name;
+	__cpu_architecture = __get_cpu_architecture();
 
 #ifdef MULTI_CPU
 	processor = *list->proc;

Signature · 1 line

1.7.4.1

Nicolas Pitre

2011-08-16 16:15 UTC #2 lore raw

PATCH 1/3 ARM: Make cpu_alignment into a global variable

Quoted 7 lines · Dave Martin

The CPU architecture really should not be changing at runtime, so make it a global variable instead of a function.
The cpu_architecture() function in <asm/system.h> remains the correct way to read this variable from C code.
Signed-off-by Dave Martin <[email protected]>

Acked-by Nicolas Pitre <[email protected]>

Quoted 74 lines

---
 arch/arm/include/asm/system.h |   11 ++++++++++-
 arch/arm/kernel/setup.c       |    5 ++++-
 2 files changed, 14 insertions(+), 2 deletions(-)

arch/arm/include/asm/system.h +10 -1

index 832888d..b0445f7 100644
@@ -57,6 +57,8 @@
 
 #ifndef __ASSEMBLY__
 
+#include <linux/bug.h>
+#include <linux/compiler.h>
 #include <linux/linkage.h>
 #include <linux/irqflags.h>

@@ -104,7 +106,14 @@ struct mm_struct;
 extern void show_pte(struct mm_struct *mm, unsigned long addr);
 extern void __show_regs(struct pt_regs *);
 
-extern int cpu_architecture(void);
+extern int __cpu_architecture;
+
+static inline int __pure cpu_architecture(void)
+{
+	BUG_ON(__cpu_architecture == CPU_ARCH_UNKNOWN);
+	return __cpu_architecture;
+}
+
 extern void cpu_init(void);
 
 void arm_machine_restart(char mode, const char *cmd);

arch/arm/kernel/setup.c +4 -1

index 70bca64..1e0c1b3 100644
@@ -42,6 +42,7 @@
 #include <asm/cacheflush.h>
 #include <asm/cachetype.h>
 #include <asm/tlbflush.h>
+#include <asm/system.h>
 
 #include <asm/prom.h>
 #include <asm/mach/arch.h>

@@ -114,6 +115,7 @@ struct cpu_cache_fns cpu_cache __read_mostly;
 struct outer_cache_fns outer_cache __read_mostly;
 EXPORT_SYMBOL(outer_cache);
 #endif
+int __cpu_architecture __read_mostly = CPU_ARCH_UNKNOWN;
 
 struct stack {
 	u32 irq[3];

@@ -210,7 +212,7 @@ static const char *proc_arch[] = {
 	"?(17)",
 };
 
-int cpu_architecture(void)
+static int __init __get_cpu_architecture(void)
 {
 	int cpu_arch;

@@ -413,6 +415,7 @@ static void __init setup_processor(void)
 	}
 
 	cpu_name = list->cpu_name;
+	__cpu_architecture = __get_cpu_architecture();
 
 #ifdef MULTI_CPU
 	processor = *list->proc;

Signature · 1 line

1.7.4.1

Dave Martin

author

2011-08-16 16:21 UTC #3 lore raw

PATCH 1/3 ARM: Make cpu_alignment into a global variable

Quoted 11 lines · Nicolas Pitre

Quoted 7 lines · Dave Martin
The CPU architecture really should not be changing at runtime, so make it a global variable instead of a function.
The cpu_architecture() function in <asm/system.h> remains the correct way to read this variable from C code.
Signed-off-by Dave Martin <[email protected]>
Acked-by Nicolas Pitre <[email protected]>

Thanks
---Dave

Quoted 75 lines

---
 arch/arm/include/asm/system.h |   11 ++++++++++-
 arch/arm/kernel/setup.c       |    5 ++++-
 2 files changed, 14 insertions(+), 2 deletions(-)

arch/arm/include/asm/system.h +10 -1

index 832888d..b0445f7 100644
@@ -57,6 +57,8 @@
 
 #ifndef __ASSEMBLY__
 
+#include <linux/bug.h>
+#include <linux/compiler.h>
 #include <linux/linkage.h>
 #include <linux/irqflags.h>

@@ -104,7 +106,14 @@ struct mm_struct;
 extern void show_pte(struct mm_struct *mm, unsigned long addr);
 extern void __show_regs(struct pt_regs *);
 
-extern int cpu_architecture(void);
+extern int __cpu_architecture;
+
+static inline int __pure cpu_architecture(void)
+{
+	BUG_ON(__cpu_architecture == CPU_ARCH_UNKNOWN);
+	return __cpu_architecture;
+}
+
 extern void cpu_init(void);
 
 void arm_machine_restart(char mode, const char *cmd);

arch/arm/kernel/setup.c +4 -1

index 70bca64..1e0c1b3 100644
@@ -42,6 +42,7 @@
 #include <asm/cacheflush.h>
 #include <asm/cachetype.h>
 #include <asm/tlbflush.h>
+#include <asm/system.h>
 
 #include <asm/prom.h>
 #include <asm/mach/arch.h>

@@ -114,6 +115,7 @@ struct cpu_cache_fns cpu_cache __read_mostly;
 struct outer_cache_fns outer_cache __read_mostly;
 EXPORT_SYMBOL(outer_cache);
 #endif
+int __cpu_architecture __read_mostly = CPU_ARCH_UNKNOWN;
 
 struct stack {
 	u32 irq[3];

@@ -210,7 +212,7 @@ static const char *proc_arch[] = {
 	"?(17)",
 };
 
-int cpu_architecture(void)
+static int __init __get_cpu_architecture(void)
 {
 	int cpu_arch;

@@ -413,6 +415,7 @@ static void __init setup_processor(void)
 	}
 
 	cpu_name = list->cpu_name;
+	__cpu_architecture = __get_cpu_architecture();
 
 #ifdef MULTI_CPU
 	processor = *list->proc;

Signature · 1 line

1.7.4.1

Sergei Shtylyov

2011-08-17 11:31 UTC #4 lore raw

PATCH 1/3 ARM: Make cpu_alignment into a global variable

In reply to Dave Martin

Hello.

Quoted 7 lines · Dave Martin

The CPU architecture really should not be changing at runtime, so make it a global variable instead of a function.
The cpu_architecture() function in<asm/system.h> remains the correct way to read this variable from C code.
Signed-off-by Dave Martin<[email protected]>

You're talking of the CPU architecture here but the subject has
"cpu_alignment"...

WBR, Sergei

Dave Martin

author

2011-08-17 13:37 UTC #5 lore raw

PATCH 1/3 ARM: Make cpu_alignment into a global variable

Quoted 7 lines · Sergei Shtylyov

Hello.
Quoted 2 lines · Dave Martin
The CPU architecture really should not be changing at runtime, so make it a global variable instead of a function.

Quoted 7 lines

Quoted 4 lines
The cpu_architecture() function in<asm/system.h> remains the correct way to read this variable from C code.
Signed-off-by Dave Martin<[email protected]>
You're talking of the CPU architecture here but the subject has
"cpu_alignment"...

Hmmm, I guess my brain was on autopilot when I wrote that...

You're right, it should say cpu_architecture in the subject. I'll fix this before I repost.

Thanks
---Dave

Dave Martin

author

2011-08-16 14:14 UTC #6 lore raw

PATCH 2/3 ARM: entry: Remove unnecessary masking when decoding Thumb-2 instructions

In reply to Dave Martin

When testing whether a Thumb-2 instruction is 32 bits long or not, the masking done in order to test bits 11-15 of the first instruction halfword won't affect the result of the comparison, so remove it.

Signed-off-by Dave Martin <[email protected]>

---
 arch/arm/kernel/entry-armv.S |    6 ++----
 1 files changed, 2 insertions(+), 4 deletions(-)

arch/arm/kernel/entry-armv.S +2 -4

index a87cbf8..b7236d4 100644
@@ -262,8 +262,7 @@ __und_svc:
 	ldr	r0, [r4, #-4]
 #else
 	ldrh	r0, [r4, #-2]			@ Thumb instruction at LR - 2
-	and	r9, r0, #0xf800
-	cmp	r9, #0xe800			@ 32-bit instruction if xx >= 0
+	cmp	r0, #0xe800			@ 32-bit instruction if xx >= 0
 	ldrhhs	r9, [r4]			@ bottom 16 bits
 	orrhs	r0, r9, r0, lsl #16
 #endif

@@ -445,8 +444,7 @@ __und_usr:
  ARM(	ldrht	r5, [r4], #2	)
  THUMB(	ldrht	r5, [r4]	)
  THUMB(	add	r4, r4, #2	)
-	and	r0, r5, #0xf800			@ mask bits 111x x... .... ....
-	cmp	r0, #0xe800			@ 32bit instruction if xx != 0
+	cmp	r5, #0xe800			@ 32bit instruction if xx != 0
 	blo	__und_usr_unknown
 3:	ldrht	r0, [r4]
 	add	r2, r2, #2			@ r2 is PC + 2, make it PC + 4

Signature · 1 line

1.7.4.1

Nicolas Pitre

2011-08-16 16:17 UTC #7 lore raw

PATCH 2/3 ARM: entry: Remove unnecessary masking when decoding Thumb-2 instructions

Quoted 6 lines · Dave Martin

When testing whether a Thumb-2 instruction is 32 bits long or not, the masking done in order to test bits 11-15 of the first instruction halfword won't affect the result of the comparison, so remove it.
Signed-off-by Dave Martin <[email protected]>

Acked-by Nicolas Pitre <[email protected]>

Quoted 31 lines

---
 arch/arm/kernel/entry-armv.S |    6 ++----
 1 files changed, 2 insertions(+), 4 deletions(-)

arch/arm/kernel/entry-armv.S +2 -4

index a87cbf8..b7236d4 100644
@@ -262,8 +262,7 @@ __und_svc:
 	ldr	r0, [r4, #-4]
 #else
 	ldrh	r0, [r4, #-2]			@ Thumb instruction at LR - 2
-	and	r9, r0, #0xf800
-	cmp	r9, #0xe800			@ 32-bit instruction if xx >= 0
+	cmp	r0, #0xe800			@ 32-bit instruction if xx >= 0
 	ldrhhs	r9, [r4]			@ bottom 16 bits
 	orrhs	r0, r9, r0, lsl #16
 #endif

@@ -445,8 +444,7 @@ __und_usr:
  ARM(	ldrht	r5, [r4], #2	)
  THUMB(	ldrht	r5, [r4]	)
  THUMB(	add	r4, r4, #2	)
-	and	r0, r5, #0xf800			@ mask bits 111x x... .... ....
-	cmp	r0, #0xe800			@ 32bit instruction if xx != 0
+	cmp	r5, #0xe800			@ 32bit instruction if xx != 0
 	blo	__und_usr_unknown
 3:	ldrht	r0, [r4]
 	add	r2, r2, #2			@ r2 is PC + 2, make it PC + 4

Signature · 1 line

1.7.4.1

Dave Martin

author

2011-08-16 16:21 UTC #8 lore raw

PATCH 2/3 ARM: entry: Remove unnecessary masking when decoding Thumb-2 instructions

Quoted 11 lines · Nicolas Pitre

Quoted 6 lines · Dave Martin
When testing whether a Thumb-2 instruction is 32 bits long or not, the masking done in order to test bits 11-15 of the first instruction halfword won't affect the result of the comparison, so remove it.
Signed-off-by Dave Martin <[email protected]>
Acked-by Nicolas Pitre <[email protected]>

Thanks
---Dave

Quoted 32 lines

---
 arch/arm/kernel/entry-armv.S |    6 ++----
 1 files changed, 2 insertions(+), 4 deletions(-)

arch/arm/kernel/entry-armv.S +2 -4

index a87cbf8..b7236d4 100644
@@ -262,8 +262,7 @@ __und_svc:
 	ldr	r0, [r4, #-4]
 #else
 	ldrh	r0, [r4, #-2]			@ Thumb instruction at LR - 2
-	and	r9, r0, #0xf800
-	cmp	r9, #0xe800			@ 32-bit instruction if xx >= 0
+	cmp	r0, #0xe800			@ 32-bit instruction if xx >= 0
 	ldrhhs	r9, [r4]			@ bottom 16 bits
 	orrhs	r0, r9, r0, lsl #16
 #endif

@@ -445,8 +444,7 @@ __und_usr:
  ARM(	ldrht	r5, [r4], #2	)
  THUMB(	ldrht	r5, [r4]	)
  THUMB(	add	r4, r4, #2	)
-	and	r0, r5, #0xf800			@ mask bits 111x x... .... ....
-	cmp	r0, #0xe800			@ 32bit instruction if xx != 0
+	cmp	r5, #0xe800			@ 32bit instruction if xx != 0
 	blo	__und_usr_unknown
 3:	ldrht	r0, [r4]
 	add	r2, r2, #2			@ r2 is PC + 2, make it PC + 4

Signature · 1 line

1.7.4.1

Dave Martin

author

2011-08-16 14:14 UTC #9 lore raw

PATCH 3/3 ARM: entry: Fix Thumb-2 undef handling for multi-CPU kernels

In reply to Dave Martin

When v6 and >=v7 boards are supported in the same kernel, the __und_usr code currently makes a build-time assumption that Thumb-2 instructions occurring in userspace don't need to be supported. Strictly speaking this is incorrect.

This patch fixes the above case by doing a run-time check on the CPU architecture in these cases. This only affects kernels which support v6 and >=v7 CPUs together: plain v6 and plain v7 kernels are unaffected.

Signed-off-by Dave Martin <[email protected]>

---
 arch/arm/kernel/entry-armv.S |   38 +++++++++++++++++++++++++++++++++++++-
 1 files changed, 37 insertions(+), 1 deletions(-)

arch/arm/kernel/entry-armv.S +37 -1

index b7236d4..9ad50c4 100644
@@ -24,6 +24,7 @@
 #include <asm/unwind.h>
 #include <asm/unistd.h>
 #include <asm/tls.h>
+#include <asm/system.h>
 
 #include "entry-header.S"
 #include <asm/entry-macro-multi.S>

@@ -439,7 +440,27 @@ __und_usr:
 #endif
 	beq	call_fpe
 	@ Thumb instruction
-#if __LINUX_ARM_ARCH__ >= 7
+#if CONFIG_ARM_THUMB && __LINUX_ARM_ARCH__ >= 6 && CONFIG_CPU_V7
+/*
+ * Thumb-2 instruction handling.  Note that because pre-v6 and >= v6 platforms
+ * can never be supported in a single kernel, this code is not applicable at
+ * all when __LINUX_ARM_ARCH__ < 6.  This allows simplifying assumptions to be
+ * made about .arch directives.
+ */
+#if __LINUX_ARM_ARCH__ < 7
+/* If the target CPU may not be Thumb-2-capable, a run-time check is needed: */
+#define NEED_CPU_ARCHITECTURE
+	ldr	r5, .LCcpu_architecture
+	ldr	r5, [r5]
+	cmp	r5, #CPU_ARCH_ARMv7
+	blo	__und_usr_unknown
+/*
+ * The following code won't get run unless the running CPU really is v7, so
+ * coding round the lack of ldrht on older arches is pointless.  Temporarily
+ * override the assembler target arch with the minimum required instead:
+ */
+	.arch	armv6t2
+#endif
 2:
  ARM(	ldrht	r5, [r4], #2	)
  THUMB(	ldrht	r5, [r4]	)

@@ -449,7 +470,16 @@ __und_usr:
 3:	ldrht	r0, [r4]
 	add	r2, r2, #2			@ r2 is PC + 2, make it PC + 4
 	orr	r0, r0, r5, lsl #16
+
+#if __LINUX_ARM_ARCH__ < 7
+/* If the target arch was overridden, change it back: */
+#ifdef CONFIG_CPU_32v6K
+	.arch	armv6k
 #else
+	.arch	armv6
+#endif
+#endif /* __LINUX_ARM_ARCH__ < 7 */
+#else /* !(CONFIG_ARM_THUMB && __LINUX_ARM_ARCH__ >= 6 && CONFIG_CPU_V7) */
 	b	__und_usr_unknown
 #endif
  UNWIND(.fnend		)

@@ -576,6 +606,12 @@ call_fpe:
 	movw_pc	lr				@ CP#14 (Debug)
 	movw_pc	lr				@ CP#15 (Control)
 
+#ifdef NEED_CPU_ARCHITECTURE
+	.align	2
+.LCcpu_architecture:
+	.word	__cpu_architecture
+#endif
+
 #ifdef CONFIG_NEON
 	.align	6

Signature · 1 line

1.7.4.1

Tixy

2011-08-16 14:51 UTC #10 lore raw

PATCH 3/3 ARM: entry: Fix Thumb-2 undef handling for multi-CPU kernels

On Tue, 2011-08-16 at 15:14 +0100, Dave Martin wrote:
[...]

arch/arm/kernel/entry-armv.S call_fpe: · line 613

Quoted 5 lines

arch/arm/kernel/entry-armv.S +5 -0

+#ifdef NEED_CPU_ARCHITECTURE
+	.align	2
+.LCcpu_architecture:
+	.word	__cpu_architecture
+#endif

What's the convention about the prefix '.LC' as opposed to just '.L'?

Signature · 1 line

Tixy

Nicolas Pitre

2011-08-16 15:47 UTC #11 lore raw

PATCH 3/3 ARM: entry: Fix Thumb-2 undef handling for multi-CPU kernels

Quoted 9 lines · Tixy

On Tue, 2011-08-16 at 15:14 +0100, Dave Martin wrote:
[...]
arch/arm/kernel/entry-armv.S call_fpe: · line 613
Quoted 5 lines
arch/arm/kernel/entry-armv.S +5 -0
+#ifdef NEED_CPU_ARCHITECTURE
+	.align	2
+.LCcpu_architecture:
+	.word	__cpu_architecture
+#endif
What's the convention about the prefix '.LC' as opposed to just '.L'?

The .Lfoo format is more widely used than .LCfoo in the tree.

Nicolas

Dave Martin

author

2011-08-16 16:05 UTC #12 lore raw

PATCH 3/3 ARM: entry: Fix Thumb-2 undef handling for multi-CPU kernels

Quoted 12 lines · Nicolas Pitre

Quoted 9 lines · Tixy
On Tue, 2011-08-16 at 15:14 +0100, Dave Martin wrote:
[...]
arch/arm/kernel/entry-armv.S call_fpe: · line 613
Quoted 5 lines
arch/arm/kernel/entry-armv.S +5 -0
+#ifdef NEED_CPU_ARCHITECTURE
+ ? .align ?2
+.LCcpu_architecture:
+ ? .word ? __cpu_architecture
+#endif
What's the convention about the prefix '.LC' as opposed to just '.L'?

Quoted 1 line

The .Lfoo format is more widely used than .LCfoo in the tree.

...but .LCfoo is more widely used in entry-armv.S, for exactly the same kind of thing.

I thought it best to blend in... anyway, apart from the magic prefix '.L', it's just a name.

Any concerns about this, or can we leave it as-is?

Cheers
---Dave

Nicolas Pitre

2011-08-16 16:14 UTC #13 lore raw

PATCH 3/3 ARM: entry: Fix Thumb-2 undef handling for multi-CPU kernels

Quoted 13 lines · Dave Martin

Quoted 12 lines · Nicolas Pitre
Quoted 9 lines · Tixy
On Tue, 2011-08-16 at 15:14 +0100, Dave Martin wrote:
[...]
arch/arm/kernel/entry-armv.S call_fpe: · line 613
Quoted 5 lines
arch/arm/kernel/entry-armv.S +5 -0
+#ifdef NEED_CPU_ARCHITECTURE
+ ? .align ?2
+.LCcpu_architecture:
+ ? .word ? __cpu_architecture
+#endif
What's the convention about the prefix '.LC' as opposed to just '.L'?

Quoted 4 lines

Quoted 1 line
The .Lfoo format is more widely used than .LCfoo in the tree.
...but .LCfoo is more widely used in entry-armv.S, for exactly the same kind of thing.

Just use the same style then.

Nicolas

Dave Martin

author

2011-08-16 16:20 UTC #14 lore raw

PATCH 3/3 ARM: entry: Fix Thumb-2 undef handling for multi-CPU kernels

Quoted 15 lines · Nicolas Pitre

Quoted 13 lines · Dave Martin
Quoted 12 lines · Nicolas Pitre
Quoted 9 lines · Tixy
On Tue, 2011-08-16 at 15:14 +0100, Dave Martin wrote:
[...]
arch/arm/kernel/entry-armv.S call_fpe: · line 613
Quoted 5 lines
arch/arm/kernel/entry-armv.S +5 -0
+#ifdef NEED_CPU_ARCHITECTURE
+ ? .align ?2
+.LCcpu_architecture:
+ ? .word ? __cpu_architecture
+#endif
What's the convention about the prefix '.LC' as opposed to just '.L'?

Quoted 6 lines

Quoted 4 lines
Quoted 1 line
The .Lfoo format is more widely used than .LCfoo in the tree.
...but .LCfoo is more widely used in entry-armv.S, for exactly the same kind of thing.
Just use the same style then.

OK, done.

Cheers
---Dave

Tixy

2011-08-16 14:58 UTC #15 lore raw

cover Fix Thumb-2 undef handling for mixed-arch kernels

In reply to Dave Martin

Quoted 30 lines · Dave Martin

In kernels which support v6 and v7 platforms in a single binary such as omap2plus_defconfig for example, undefined instruction exceptions taken on Thumb instructions are not processed correctly, because the kernel assumes at build-time that Thumb-2 won't be supported in userspace.
This series implements a workaround, by allowing the __und_usr handler to check the CPU architecture at runtime, in affected kernels.
Changes since RFC:
* Renamed the global variable to __cpu_architecture and provided cpu_architecture() as in inline function in <asm/system.h>. This brings back the old API, but more efficiently.
* Dropped all the patches required to handle the reverted API change.
* cpu_architecture is now declared __pure, which may lead to slightly more efficient code in some cases.
* Added BUG_ON() to cpu_architecture() so that if __cpu_architecture is not set early enough, it gets noticed.
Thanks to Tixy and Nicolas Pitre for their helpful feedback which helped slim down this series.
Dave Martin (3):
ARM: Make cpu_alignment into a global variable
ARM: entry: Remove unnecessary masking when decoding Thumb-2
instructions
ARM: entry: Fix Thumb-2 undef handling for multi-CPU kernels

Quoted 5 lines

 arch/arm/include/asm/system.h |   11 +++++++++-
 arch/arm/kernel/entry-armv.S  |   44 ++++++++++++++++++++++++++++++++++++----
 arch/arm/kernel/setup.c       |    5 +++-
 3 files changed, 53 insertions(+), 7 deletions(-)

Looks good to me.

Signature · 1 line

Tixy

Dave Martin

author

2011-08-16 16:06 UTC #16 lore raw

cover Fix Thumb-2 undef handling for mixed-arch kernels

Quoted 31 lines · Tixy

Quoted 30 lines · Dave Martin
In kernels which support v6 and v7 platforms in a single binary such as omap2plus_defconfig for example, undefined instruction exceptions taken on Thumb instructions are not processed correctly, because the kernel assumes at build-time that Thumb-2 won't be supported in userspace.
This series implements a workaround, by allowing the __und_usr handler to check the CPU architecture at runtime, in affected kernels.
Changes since RFC:
? * Renamed the global variable to __cpu_architecture and provided ? ? cpu_architecture() as in inline function in <asm/system.h>. ? ? This brings back the old API, but more efficiently. ? * Dropped all the patches required to handle the reverted API ? ? change.
? * cpu_architecture is now declared __pure, which may lead to ? ? slightly more efficient code in some cases.
? * Added BUG_ON() to cpu_architecture() so that if
? ? __cpu_architecture is not set early enough, it gets noticed.
Thanks to Tixy and Nicolas Pitre for their helpful feedback which helped slim down this series.
Dave Martin (3):
? ARM: Make cpu_alignment into a global variable
? ARM: entry: Remove unnecessary masking when decoding Thumb-2 ? ? instructions
? ARM: entry: Fix Thumb-2 undef handling for multi-CPU kernels

Quoted 7 lines

Quoted 5 lines
?arch/arm/include/asm/system.h | ? 11 +++++++++-
?arch/arm/kernel/entry-armv.S ?| ? 44 ++++++++++++++++++++++++++++++++++++----
?arch/arm/kernel/setup.c ? ? ? | ? ?5 +++-
?3 files changed, 53 insertions(+), 7 deletions(-)
Looks good to me.

Thanks for looking over it.

Is that a Reviewed-by, or will you go for an Ack?

Cheers
---Dave

Tixy

2011-08-16 16:43 UTC #17 lore raw

cover Fix Thumb-2 undef handling for mixed-arch kernels

Quoted 7 lines · Dave Martin

Quoted 1 line · Tixy
Looks good to me.
Thanks for looking over it.
Is that a Reviewed-by, or will you go for an Ack?

It's certainly a Reviewed-by: Jon Medhurst <[email protected]> for all three patches.

I've been re-reading Documentation/SubmittingPatches to try and work out when it's appropriate to use Acked-by, but it seems a bit woolly - mostly for maintainers?

Signature · 1 line

Tixy