[PATCH 01/14] csky: fixup abiv2 mmap(... O_SYNC) failed.

[PATCH 01/14] csky: fixup abiv2 mmap(... O_SYNC) failed.

[guoren]

From: Guo Ren <ren_guo@c-sky.com>

Glibc function mmap(... O_SYNC) will make page to _PAGE_UNCACHE +
_PAGE_SO and strong-order page couldn't support unalignment access.
So remove _PAGE_SO from _PAGE_UNCACHE, also sync abiv1 with the macro
of _PAGE_SO.

Signed-off-by: Guo Ren <ren_guo@c-sky.com>
Reported-by: Liu Renwei <Renwei.Liu@verisilicon.com>
Tested-by: Yuan Qiyun <qiyun_yuan@c-sky.com>
---
 arch/csky/abiv1/inc/abi/pgtable-bits.h | 1 +
 arch/csky/abiv2/inc/abi/pgtable-bits.h | 2 +-
 arch/csky/mm/ioremap.c                 | 2 +-
 3 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/arch/csky/abiv1/inc/abi/pgtable-bits.h b/arch/csky/abiv1/inc/abi/pgtable-bits.h
index 455075b..d605445 100644
--- a/arch/csky/abiv1/inc/abi/pgtable-bits.h
+++ b/arch/csky/abiv1/inc/abi/pgtable-bits.h
@@ -26,6 +26,7 @@
 
 #define _PAGE_CACHE		(3<<9)
 #define _PAGE_UNCACHE		(2<<9)
+#define _PAGE_SO		_PAGE_UNCACHE
 
 #define _CACHE_MASK		(7<<9)
 
diff --git a/arch/csky/abiv2/inc/abi/pgtable-bits.h b/arch/csky/abiv2/inc/abi/pgtable-bits.h
index b20ae19..137f793 100644
--- a/arch/csky/abiv2/inc/abi/pgtable-bits.h
+++ b/arch/csky/abiv2/inc/abi/pgtable-bits.h
@@ -32,6 +32,6 @@
 #define _CACHE_MASK		_PAGE_CACHE
 
 #define _CACHE_CACHED		(_PAGE_VALID | _PAGE_CACHE | _PAGE_BUF)
-#define _CACHE_UNCACHED		(_PAGE_VALID | _PAGE_SO)
+#define _CACHE_UNCACHED		(_PAGE_VALID)
 
 #endif /* __ASM_CSKY_PGTABLE_BITS_H */
diff --git a/arch/csky/mm/ioremap.c b/arch/csky/mm/ioremap.c
index 7ad3ff1..cb7c03e 100644
--- a/arch/csky/mm/ioremap.c
+++ b/arch/csky/mm/ioremap.c
@@ -30,7 +30,7 @@ void __iomem *ioremap(phys_addr_t addr, size_t size)
 	vaddr = (unsigned long)area->addr;
 
 	prot = __pgprot(_PAGE_PRESENT | __READABLE | __WRITEABLE |
-			_PAGE_GLOBAL | _CACHE_UNCACHED);
+			_PAGE_GLOBAL | _CACHE_UNCACHED | _PAGE_SO);
 
 	if (ioremap_page_range(vaddr, vaddr + size, addr, prot)) {
 		free_vm_area(area);
-- 
2.7.4

[PATCH 02/14] csky: bugfix gdb coredump error.

[guoren]

From: Guo Ren <ren_guo@c-sky.com>

In gdb/bfd elf32-csky.c, csky_elf_grok_prstatus() use fixed size of
elf_prstatus. It's 148 for abiv1 and 220 for abiv2, the size is enough
for coredump and no need full sizeof(struct pt_regs).

Signed-off-by: Guo Ren <ren_guo@c-sky.com>
Reported-by: Lu Baoquan <lu.baoquan@intellif.com>
Reported-by: Liu Mao <liu.mao@intellif.com>
---
 arch/csky/include/asm/elf.h | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/arch/csky/include/asm/elf.h b/arch/csky/include/asm/elf.h
index 48b5366..d6dbc00 100644
--- a/arch/csky/include/asm/elf.h
+++ b/arch/csky/include/asm/elf.h
@@ -31,7 +31,12 @@ typedef unsigned long elf_greg_t;
 
 typedef struct user_fp elf_fpregset_t;
 
-#define ELF_NGREG (sizeof(struct pt_regs) / sizeof(elf_greg_t))
+/*
+ * In gdb/bfd elf32-csky.c, csky_elf_grok_prstatus() use fixed size of
+ * elf_prstatus. It's 148 for abiv1 and 220 for abiv2, the size is enough
+ * for coredump and no need full sizeof(struct pt_regs).
+ */
+#define ELF_NGREG ((sizeof(struct pt_regs) / sizeof(elf_greg_t)) - 2)
 
 typedef elf_greg_t elf_gregset_t[ELF_NGREG];
 
-- 
2.7.4

[PATCH 03/14] csky: fixup remove vdsp implement for kernel.

[guoren]

From: Guo Ren <ren_guo@c-sky.com>

The vr regs for vdsp only saved in task_switch not in every
exception trap-in. The memcpy with vdsp instructions will
destroy the vr regs for user space applications.

Signed-off-by: Guo Ren <ren_guo@c-sky.com>
---
 arch/csky/abiv2/memcpy.S | 8 +-------
 1 file changed, 1 insertion(+), 7 deletions(-)

diff --git a/arch/csky/abiv2/memcpy.S b/arch/csky/abiv2/memcpy.S
index 987fec6..145bf3a 100644
--- a/arch/csky/abiv2/memcpy.S
+++ b/arch/csky/abiv2/memcpy.S
@@ -27,13 +27,7 @@ ENTRY(memcpy)
 
 	LABLE_ALIGN
 .L_len_larger_16bytes:
-#if defined(__CSKY_VDSPV2__)
-	vldx.8	vr0, (r1), r19
-	PRE_BNEZAD (r18)
-	addi	r1, 16
-	vstx.8	vr0, (r0), r19
-	addi	r0, 16
-#elif defined(__CK860__)
+#if defined(__CK860__)
 	ldw	r3, (r1, 0)
 	stw	r3, (r0, 0)
 	ldw	r3, (r1, 4)
-- 
2.7.4

[PATCH 04/14] csky: remove syscall_exit_work

[guoren]

From: Guo Ren <ren_guo@c-sky.com>

Remove syscall_exit_work and union all to ret_from_exception.

Signed-off-by: Guo Ren <ren_guo@c-sky.com>
---
 arch/csky/kernel/entry.S | 16 ++++------------
 1 file changed, 4 insertions(+), 12 deletions(-)

diff --git a/arch/csky/kernel/entry.S b/arch/csky/kernel/entry.S
index 79f92b8..2b4a851 100644
--- a/arch/csky/kernel/entry.S
+++ b/arch/csky/kernel/entry.S
@@ -183,18 +183,10 @@ ENTRY(csky_systemcall)
 #endif
 	stw	a0, (sp, LSAVE_A0)	/* Save return value */
 
-	movi	a0, 1			/* leave system call */
-	mov	a1, sp			/* sp = pt_regs pointer */
-	jbsr	syscall_trace
-
-syscall_exit_work:
-	ld	syscallid, (sp, LSAVE_PSR)
-	btsti	syscallid, 31
-	bt	2f
-
-	jmpi	resume_userspace
-
-2:      RESTORE_ALL
+	movi    a0, 1                   /* leave system call */
+	mov     a1, sp                  /* right now, sp --> pt_regs */
+	jbsr    syscall_trace
+	br	ret_from_exception
 
 ENTRY(ret_from_kernel_thread)
 	jbsr	schedule_tail
-- 
2.7.4

[PATCH 05/14] csky: fixup save hi,lo,dspcr regs in switch_stack.

[guoren]

From: Guo Ren <ren_guo@c-sky.com>

HI, LO, DSPCR registers are 807/810 related regs and no need for 610/860.
All of the regs must be saved in pt_regs and switch_stack. This patch
fixup saving dspcr reg in switch_stack and pt_regs.

Signed-off-by: Guo Ren <ren_guo@c-sky.com>
---
 arch/csky/abiv1/inc/abi/switch_context.h | 17 ++++++++++++++
 arch/csky/abiv2/inc/abi/entry.h          | 28 ++++++++++++++++++++---
 arch/csky/abiv2/inc/abi/switch_context.h | 32 ++++++++++++++++++++++++++
 arch/csky/include/asm/processor.h        |  3 +--
 arch/csky/include/uapi/asm/ptrace.h      | 39 +-------------------------------
 arch/csky/kernel/asm-offsets.c           |  2 --
 arch/csky/kernel/entry.S                 | 22 ------------------
 arch/csky/kernel/ptrace.c                |  4 +++-
 8 files changed, 79 insertions(+), 68 deletions(-)
 create mode 100644 arch/csky/abiv1/inc/abi/switch_context.h
 create mode 100644 arch/csky/abiv2/inc/abi/switch_context.h

diff --git a/arch/csky/abiv1/inc/abi/switch_context.h b/arch/csky/abiv1/inc/abi/switch_context.h
new file mode 100644
index 0000000..17c8268
--- /dev/null
+++ b/arch/csky/abiv1/inc/abi/switch_context.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd.
+
+#ifndef __ABI_CSKY_PTRACE_H
+#define __ABI_CSKY_PTRACE_H
+
+struct switch_stack {
+	unsigned long r8;
+	unsigned long r9;
+	unsigned long r10;
+	unsigned long r11;
+	unsigned long r12;
+	unsigned long r13;
+	unsigned long r14;
+	unsigned long r15;
+};
+#endif /* __ABI_CSKY_PTRACE_H */
diff --git a/arch/csky/abiv2/inc/abi/entry.h b/arch/csky/abiv2/inc/abi/entry.h
index acd0521..edc5cc0 100644
--- a/arch/csky/abiv2/inc/abi/entry.h
+++ b/arch/csky/abiv2/inc/abi/entry.h
@@ -57,6 +57,8 @@
 	stw	lr, (sp, 60)
 	mflo	lr
 	stw	lr, (sp, 64)
+	mfcr	lr, cr14
+	stw	lr, (sp, 68)
 #endif
 	subi	sp, 80
 .endm
@@ -77,6 +79,8 @@
 	mthi	a0
 	ldw	a0, (sp, 144)
 	mtlo	a0
+	ldw	a0, (sp, 148)
+	mtcr	a0, cr14
 #endif
 
 	ldw     a0, (sp, 24)
@@ -93,9 +97,9 @@
 .endm
 
 .macro SAVE_SWITCH_STACK
-	subi	sp, 64
+	subi    sp, 64
 	stm	r4-r11, (sp)
-	stw	r15, (sp, 32)
+	stw	lr,  (sp, 32)
 	stw	r16, (sp, 36)
 	stw	r17, (sp, 40)
 	stw	r26, (sp, 44)
@@ -103,11 +107,29 @@
 	stw	r28, (sp, 52)
 	stw	r29, (sp, 56)
 	stw	r30, (sp, 60)
+#ifdef CONFIG_CPU_HAS_HILO
+	subi	sp, 16
+	mfhi	lr
+	stw	lr, (sp, 0)
+	mflo	lr
+	stw	lr, (sp, 4)
+	mfcr	lr, cr14
+	stw	lr, (sp, 8)
+#endif
 .endm
 
 .macro RESTORE_SWITCH_STACK
+#ifdef CONFIG_CPU_HAS_HILO
+	ldw	lr, (sp, 0)
+	mthi	lr
+	ldw	lr, (sp, 4)
+	mtlo	lr
+	ldw	lr, (sp, 8)
+	mtcr	lr, cr14
+	addi	sp, 16
+#endif
 	ldm	r4-r11, (sp)
-	ldw	r15, (sp, 32)
+	ldw	lr,  (sp, 32)
 	ldw	r16, (sp, 36)
 	ldw	r17, (sp, 40)
 	ldw	r26, (sp, 44)
diff --git a/arch/csky/abiv2/inc/abi/switch_context.h b/arch/csky/abiv2/inc/abi/switch_context.h
new file mode 100644
index 0000000..73a8124
--- /dev/null
+++ b/arch/csky/abiv2/inc/abi/switch_context.h
@@ -0,0 +1,32 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd.
+
+#ifndef __ABI_CSKY_PTRACE_H
+#define __ABI_CSKY_PTRACE_H
+
+struct switch_stack {
+#ifdef CONFIG_CPU_HAS_HILO
+	unsigned long rhi;
+	unsigned long rlo;
+	unsigned long cr14;
+	unsigned long pad;
+#endif
+	unsigned long r4;
+	unsigned long r5;
+	unsigned long r6;
+	unsigned long r7;
+	unsigned long r8;
+	unsigned long r9;
+	unsigned long r10;
+	unsigned long r11;
+
+	unsigned long r15;
+	unsigned long r16;
+	unsigned long r17;
+	unsigned long r26;
+	unsigned long r27;
+	unsigned long r28;
+	unsigned long r29;
+	unsigned long r30;
+};
+#endif /* __ABI_CSKY_PTRACE_H */
diff --git a/arch/csky/include/asm/processor.h b/arch/csky/include/asm/processor.h
index b174865..5ac3f32 100644
--- a/arch/csky/include/asm/processor.h
+++ b/arch/csky/include/asm/processor.h
@@ -11,6 +11,7 @@
 #include <asm/cache.h>
 #include <abi/reg_ops.h>
 #include <abi/regdef.h>
+#include <abi/switch_context.h>
 #ifdef CONFIG_CPU_HAS_FPU
 #include <abi/fpu.h>
 #endif
@@ -50,8 +51,6 @@ struct thread_struct {
 	unsigned long  ksp;       /* kernel stack pointer */
 	unsigned long  sr;        /* saved status register */
 	unsigned long  esp0;      /* points to SR of stack frame */
-	unsigned long  hi;
-	unsigned long  lo;
 
 	/* Other stuff associated with the thread. */
 	unsigned long address;      /* Last user fault */
diff --git a/arch/csky/include/uapi/asm/ptrace.h b/arch/csky/include/uapi/asm/ptrace.h
index f10d02c..a4eaa8d 100644
--- a/arch/csky/include/uapi/asm/ptrace.h
+++ b/arch/csky/include/uapi/asm/ptrace.h
@@ -36,7 +36,7 @@ struct pt_regs {
 
 	unsigned long	rhi;
 	unsigned long	rlo;
-	unsigned long	pad; /* reserved */
+	unsigned long	dcsr;
 #endif
 };
 
@@ -48,43 +48,6 @@ struct user_fp {
 	unsigned long	reserved;
 };
 
-/*
- * Switch stack for switch_to after push pt_regs.
- *
- * ABI_CSKYV2: r4 ~ r11, r15 ~ r17, r26 ~ r30;
- * ABI_CSKYV1: r8 ~ r14, r15;
- */
-struct  switch_stack {
-#if defined(__CSKYABIV2__)
-	unsigned long   r4;
-	unsigned long   r5;
-	unsigned long   r6;
-	unsigned long   r7;
-	unsigned long   r8;
-	unsigned long   r9;
-	unsigned long   r10;
-	unsigned long   r11;
-#else
-	unsigned long   r8;
-	unsigned long   r9;
-	unsigned long   r10;
-	unsigned long   r11;
-	unsigned long   r12;
-	unsigned long   r13;
-	unsigned long   r14;
-#endif
-	unsigned long   r15;
-#if defined(__CSKYABIV2__)
-	unsigned long   r16;
-	unsigned long   r17;
-	unsigned long   r26;
-	unsigned long   r27;
-	unsigned long   r28;
-	unsigned long   r29;
-	unsigned long   r30;
-#endif
-};
-
 #ifdef __KERNEL__
 
 #define PS_S	0x80000000 /* Supervisor Mode */
diff --git a/arch/csky/kernel/asm-offsets.c b/arch/csky/kernel/asm-offsets.c
index 8d3ed81..2f07670 100644
--- a/arch/csky/kernel/asm-offsets.c
+++ b/arch/csky/kernel/asm-offsets.c
@@ -24,8 +24,6 @@ int main(void)
 	DEFINE(THREAD_FESR,       offsetof(struct thread_struct, user_fp.fesr));
 	DEFINE(THREAD_FCR,        offsetof(struct thread_struct, user_fp.fcr));
 	DEFINE(THREAD_FPREG,      offsetof(struct thread_struct, user_fp.vr));
-	DEFINE(THREAD_DSPHI,      offsetof(struct thread_struct, hi));
-	DEFINE(THREAD_DSPLO,      offsetof(struct thread_struct, lo));
 
 	/* offsets into the thread_info struct */
 	DEFINE(TINFO_FLAGS,       offsetof(struct thread_info, flags));
diff --git a/arch/csky/kernel/entry.S b/arch/csky/kernel/entry.S
index 2b4a851..de378e4 100644
--- a/arch/csky/kernel/entry.S
+++ b/arch/csky/kernel/entry.S
@@ -346,34 +346,12 @@ ENTRY(__switch_to)
 
 	stw	sp, (a3, THREAD_KSP)
 
-#ifdef CONFIG_CPU_HAS_HILO
-	lrw	r10, THREAD_DSPHI
-	add	r10, a3
-	mfhi	r6
-	mflo	r7
-	stw	r6, (r10, 0)		/* THREAD_DSPHI */
-	stw	r7, (r10, 4)		/* THREAD_DSPLO */
-	mfcr	r6, cr14
-	stw	r6, (r10, 8)		/* THREAD_DSPCSR */
-#endif
-
 	/* Set up next process to run */
 	lrw	a3, TASK_THREAD
 	addu	a3, a1
 
 	ldw	sp, (a3, THREAD_KSP)	/* Set next kernel sp */
 
-#ifdef CONFIG_CPU_HAS_HILO
-	lrw	r10, THREAD_DSPHI
-	add	r10, a3
-	ldw	r6, (r10, 8)		/* THREAD_DSPCSR */
-	mtcr	r6, cr14
-	ldw	r6, (r10, 0)		/* THREAD_DSPHI */
-	ldw	r7, (r10, 4)		/* THREAD_DSPLO */
-	mthi	r6
-	mtlo	r7
-#endif
-
 	ldw	a2, (a3, THREAD_SR)	/* Set next PSR */
 	mtcr	a2, psr
 
diff --git a/arch/csky/kernel/ptrace.c b/arch/csky/kernel/ptrace.c
index 34b3025..a76d466 100644
--- a/arch/csky/kernel/ptrace.c
+++ b/arch/csky/kernel/ptrace.c
@@ -95,7 +95,9 @@ static int gpr_set(struct task_struct *target,
 		return ret;
 
 	regs.sr = task_pt_regs(target)->sr;
-
+#ifdef CONFIG_CPU_HAS_HILO
+	regs.dcsr = task_pt_regs(target)->dcsr;
+#endif
 	task_thread_info(target)->tp_value = regs.tls;
 
 	*task_pt_regs(target) = regs;
-- 
2.7.4

[PATCH 07/14] csky: CPU-hotplug supported for SMP

[guoren]

From: Guo Ren <ren_guo@c-sky.com>

This is a simple implement of CPU-hotplug for power saving. CPU use
wait instruction to enter power saving mode and waiting for IPI wakeup
signal.

Signed-off-by: Guo Ren <ren_guo@c-sky.com>
---
 arch/csky/Kconfig           |  9 ++++++
 arch/csky/include/asm/smp.h |  4 +++
 arch/csky/kernel/smp.c      | 71 +++++++++++++++++++++++++++++++++++----------
 3 files changed, 69 insertions(+), 15 deletions(-)

diff --git a/arch/csky/Kconfig b/arch/csky/Kconfig
index cb64f8d..8bdbe92 100644
--- a/arch/csky/Kconfig
+++ b/arch/csky/Kconfig
@@ -198,6 +198,15 @@ config RAM_BASE
 	hex "DRAM start addr (the same with memory-section in dts)"
 	default 0x0
 
+config HOTPLUG_CPU
+	bool "Support for hot-pluggable CPUs"
+	select GENERIC_IRQ_MIGRATION
+	depends on SMP
+	help
+	  Say Y here to allow turning CPUs off and on. CPUs can be
+	  controlled through /sys/devices/system/cpu/cpu1/hotplug/target.
+
+	  Say N if you want to disable CPU hotplug.
 endmenu
 
 source "kernel/Kconfig.hz"
diff --git a/arch/csky/include/asm/smp.h b/arch/csky/include/asm/smp.h
index 4a929c4..668b79c 100644
--- a/arch/csky/include/asm/smp.h
+++ b/arch/csky/include/asm/smp.h
@@ -21,6 +21,10 @@ void __init set_send_ipi(void (*func)(const struct cpumask *mask), int irq);
 
 #define raw_smp_processor_id()	(current_thread_info()->cpu)
 
+int __cpu_disable(void);
+
+void __cpu_die(unsigned int cpu);
+
 #endif /* CONFIG_SMP */
 
 #endif /* __ASM_CSKY_SMP_H */
diff --git a/arch/csky/kernel/smp.c b/arch/csky/kernel/smp.c
index 74d6273..ddc4dd7 100644
--- a/arch/csky/kernel/smp.c
+++ b/arch/csky/kernel/smp.c
@@ -16,6 +16,7 @@
 #include <linux/of.h>
 #include <linux/sched/task_stack.h>
 #include <linux/sched/mm.h>
+#include <linux/sched/hotplug.h>
 #include <asm/irq.h>
 #include <asm/traps.h>
 #include <asm/sections.h>
@@ -112,12 +113,8 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 {
 }
 
-static void __init enable_smp_ipi(void)
-{
-	enable_percpu_irq(ipi_irq, 0);
-}
-
 static int ipi_dummy_dev;
+
 void __init setup_smp_ipi(void)
 {
 	int rc;
@@ -130,7 +127,7 @@ void __init setup_smp_ipi(void)
 	if (rc)
 		panic("%s IRQ request failed\n", __func__);
 
-	enable_smp_ipi();
+	enable_percpu_irq(ipi_irq, 0);
 }
 
 void __init setup_smp(void)
@@ -161,12 +158,10 @@ volatile unsigned int secondary_stack;
 
 int __cpu_up(unsigned int cpu, struct task_struct *tidle)
 {
-	unsigned int tmp;
-
-	secondary_stack = (unsigned int)tidle->stack + THREAD_SIZE;
+	unsigned long mask = 1 << cpu;
 
+	secondary_stack = (unsigned int)tidle->stack + THREAD_SIZE - 8;
 	secondary_hint = mfcr("cr31");
-
 	secondary_ccr  = mfcr("cr18");
 
 	/*
@@ -176,10 +171,13 @@ int __cpu_up(unsigned int cpu, struct task_struct *tidle)
 	 */
 	mtcr("cr17", 0x22);
 
-	/* Enable cpu in SMP reset ctrl reg */
-	tmp = mfcr("cr<29, 0>");
-	tmp |= 1 << cpu;
-	mtcr("cr<29, 0>", tmp);
+	if (mask & mfcr("cr<29, 0>")) {
+		send_arch_ipi(cpumask_of(cpu));
+	} else {
+		/* Enable cpu in SMP reset ctrl reg */
+		mask |= mfcr("cr<29, 0>");
+		mtcr("cr<29, 0>", mask);
+	}
 
 	/* Wait for the cpu online */
 	while (!cpu_online(cpu));
@@ -219,7 +217,7 @@ void csky_start_secondary(void)
 	init_fpu();
 #endif
 
-	enable_smp_ipi();
+	enable_percpu_irq(ipi_irq, 0);
 
 	mmget(mm);
 	mmgrab(mm);
@@ -235,3 +233,46 @@ void csky_start_secondary(void)
 	preempt_disable();
 	cpu_startup_entry(CPUHP_AP_ONLINE_IDLE);
 }
+
+#ifdef CONFIG_HOTPLUG_CPU
+int __cpu_disable(void)
+{
+	unsigned int cpu = smp_processor_id();
+
+	set_cpu_online(cpu, false);
+
+	irq_migrate_all_off_this_cpu();
+
+	clear_tasks_mm_cpumask(cpu);
+
+	return 0;
+}
+
+void __cpu_die(unsigned int cpu)
+{
+	if (!cpu_wait_death(cpu, 5)) {
+		pr_crit("CPU%u: shutdown failed\n", cpu);
+		return;
+	}
+	pr_notice("CPU%u: shutdown\n", cpu);
+}
+
+void arch_cpu_idle_dead(void)
+{
+	idle_task_exit();
+
+	cpu_report_death();
+
+	while (!secondary_stack)
+		arch_cpu_idle();
+
+	local_irq_disable();
+
+	asm volatile(
+		"mov	sp, %0\n"
+		"mov	r8, %0\n"
+		"jmpi	csky_start_secondary"
+		:
+		: "r" (secondary_stack));
+}
+#endif
-- 
2.7.4

[PATCH 08/14] csky: stacktrace supported.

[guoren]

From: Guo Ren <ren_guo@c-sky.com>

The gcc option "-mbacktrace" will push fp(r8),lr into stack and we could
unwind the stack with:
	fp = *fp
	lr = (unsigned int *)fp[1]

Signed-off-by: Guo Ren <ren_guo@c-sky.com>
---
 arch/csky/Kconfig                   |  3 ++
 arch/csky/Makefile                  |  4 +++
 arch/csky/include/asm/thread_info.h |  4 +++
 arch/csky/kernel/Makefile           |  1 +
 arch/csky/kernel/process.c          | 29 +++++++++++--------
 arch/csky/kernel/stacktrace.c       | 57 +++++++++++++++++++++++++++++++++++++
 6 files changed, 86 insertions(+), 12 deletions(-)
 create mode 100644 arch/csky/kernel/stacktrace.c

diff --git a/arch/csky/Kconfig b/arch/csky/Kconfig
index 8bdbe92..65804d1 100644
--- a/arch/csky/Kconfig
+++ b/arch/csky/Kconfig
@@ -94,6 +94,9 @@ config MMU
 config RWSEM_GENERIC_SPINLOCK
 	def_bool y
 
+config STACKTRACE_SUPPORT
+	def_bool y
+
 config TIME_LOW_RES
 	def_bool y
 
diff --git a/arch/csky/Makefile b/arch/csky/Makefile
index c639fc1..3607a6e 100644
--- a/arch/csky/Makefile
+++ b/arch/csky/Makefile
@@ -47,6 +47,10 @@ ifeq ($(CSKYABI),abiv2)
 KBUILD_CFLAGS += -mno-stack-size
 endif
 
+ifdef CONFIG_STACKTRACE
+KBUILD_CFLAGS += -mbacktrace
+endif
+
 abidirs := $(patsubst %,arch/csky/%/,$(CSKYABI))
 KBUILD_CFLAGS += $(patsubst %,-I$(srctree)/%inc,$(abidirs))
 
diff --git a/arch/csky/include/asm/thread_info.h b/arch/csky/include/asm/thread_info.h
index a2c69a7..0e9d035 100644
--- a/arch/csky/include/asm/thread_info.h
+++ b/arch/csky/include/asm/thread_info.h
@@ -10,6 +10,7 @@
 #include <asm/types.h>
 #include <asm/page.h>
 #include <asm/processor.h>
+#include <abi/switch_context.h>
 
 struct thread_info {
 	struct task_struct	*task;
@@ -36,6 +37,9 @@ struct thread_info {
 
 #define THREAD_SIZE_ORDER (THREAD_SHIFT - PAGE_SHIFT)
 
+#define thread_saved_fp(tsk) \
+	((unsigned long)(((struct switch_stack *)(tsk->thread.ksp))->r8))
+
 static inline struct thread_info *current_thread_info(void)
 {
 	unsigned long sp;
diff --git a/arch/csky/kernel/Makefile b/arch/csky/kernel/Makefile
index 4422de7..ba5ca48 100644
--- a/arch/csky/kernel/Makefile
+++ b/arch/csky/kernel/Makefile
@@ -6,3 +6,4 @@ obj-y += process.o cpu-probe.o ptrace.o dumpstack.o
 
 obj-$(CONFIG_MODULES)			+= module.o
 obj-$(CONFIG_SMP)			+= smp.o
+obj-$(CONFIG_STACKTRACE)		+= stacktrace.o
diff --git a/arch/csky/kernel/process.c b/arch/csky/kernel/process.c
index 8ed2002..e555740 100644
--- a/arch/csky/kernel/process.c
+++ b/arch/csky/kernel/process.c
@@ -93,26 +93,31 @@ int dump_task_regs(struct task_struct *tsk, elf_gregset_t *pr_regs)
 
 unsigned long get_wchan(struct task_struct *p)
 {
-	unsigned long esp, pc;
-	unsigned long stack_page;
+	unsigned long lr;
+	unsigned long *fp, *stack_start, *stack_end;
 	int count = 0;
 
 	if (!p || p == current || p->state == TASK_RUNNING)
 		return 0;
 
-	stack_page = (unsigned long)p;
-	esp = p->thread.esp0;
+	stack_start = (unsigned long *)end_of_stack(p);
+	stack_end = (unsigned long *)(task_stack_page(p) + THREAD_SIZE);
+
+	fp = (unsigned long *) thread_saved_fp(p);
 	do {
-		if (esp < stack_page+sizeof(struct task_struct) ||
-		    esp >= 8184+stack_page)
+		if (fp < stack_start || fp > stack_end)
 			return 0;
-		/*FIXME: There's may be error here!*/
-		pc = ((unsigned long *)esp)[1];
-		/* FIXME: This depends on the order of these functions. */
-		if (!in_sched_functions(pc))
-			return pc;
-		esp = *(unsigned long *) esp;
+#ifdef CONFIG_STACKTRACE
+		lr = fp[1];
+		fp = (unsigned long *)fp[0];
+#else
+		lr = *fp++;
+#endif
+		if (!in_sched_functions(lr) &&
+		    __kernel_text_address(lr))
+			return lr;
 	} while (count++ < 16);
+
 	return 0;
 }
 EXPORT_SYMBOL(get_wchan);
diff --git a/arch/csky/kernel/stacktrace.c b/arch/csky/kernel/stacktrace.c
new file mode 100644
index 0000000..fec777a
--- /dev/null
+++ b/arch/csky/kernel/stacktrace.c
@@ -0,0 +1,57 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. */
+
+#include <linux/sched/debug.h>
+#include <linux/sched/task_stack.h>
+#include <linux/stacktrace.h>
+#include <linux/ftrace.h>
+
+void save_stack_trace(struct stack_trace *trace)
+{
+	save_stack_trace_tsk(current, trace);
+}
+EXPORT_SYMBOL_GPL(save_stack_trace);
+
+void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
+{
+	unsigned long *fp, *stack_start, *stack_end;
+	unsigned long addr;
+	int skip = trace->skip;
+	int savesched;
+	int graph_idx = 0;
+
+	if (tsk == current) {
+		asm volatile("mov %0, r8\n":"=r"(fp));
+		savesched = 1;
+	} else {
+		fp = (unsigned long *)thread_saved_fp(tsk);
+		savesched = 0;
+	}
+
+	addr = (unsigned long) fp & THREAD_MASK;
+	stack_start = (unsigned long *) addr;
+	stack_end = (unsigned long *) (addr + THREAD_SIZE);
+
+	while (fp > stack_start && fp < stack_end) {
+		unsigned long lpp, fpp;
+
+		fpp = fp[0];
+		lpp = fp[1];
+		if (!__kernel_text_address(lpp))
+			break;
+		else
+			lpp = ftrace_graph_ret_addr(tsk, &graph_idx, lpp, NULL);
+
+		if (savesched || !in_sched_functions(lpp)) {
+			if (skip) {
+				skip--;
+			} else {
+				trace->entries[trace->nr_entries++] = lpp;
+				if (trace->nr_entries >= trace->max_entries)
+					break;
+			}
+		}
+		fp = (unsigned long *)fpp;
+	}
+}
+EXPORT_SYMBOL_GPL(save_stack_trace_tsk);
-- 
2.7.4

[PATCH 09/14] csky: optimize kernel panic print.

[guoren]

From: Guo Ren <ren_guo@c-sky.com>

Use STACKTRACE to optimize panic print more pretty and align registers
printing.

Signed-off-by: Guo Ren <ren_guo@c-sky.com>
---
 arch/csky/kernel/dumpstack.c | 59 ++++++++++++++------------------------------
 arch/csky/kernel/ptrace.c    | 34 ++++++++++++++-----------
 arch/csky/mm/fault.c         |  4 +--
 3 files changed, 41 insertions(+), 56 deletions(-)

diff --git a/arch/csky/kernel/dumpstack.c b/arch/csky/kernel/dumpstack.c
index a9a03ac..659253e 100644
--- a/arch/csky/kernel/dumpstack.c
+++ b/arch/csky/kernel/dumpstack.c
@@ -7,60 +7,39 @@ int kstack_depth_to_print = 48;
 
 void show_trace(unsigned long *stack)
 {
-	unsigned long *endstack;
+	unsigned long *stack_end;
+	unsigned long *stack_start;
+	unsigned long *fp;
 	unsigned long addr;
-	int i;
 
-	pr_info("Call Trace:\n");
-	addr = (unsigned long)stack + THREAD_SIZE - 1;
-	endstack = (unsigned long *)(addr & -THREAD_SIZE);
-	i = 0;
-	while (stack + 1 <= endstack) {
-		addr = *stack++;
-		/*
-		 * If the address is either in the text segment of the
-		 * kernel, or in the region which contains vmalloc'ed
-		 * memory, it *may* be the address of a calling
-		 * routine; if so, print it so that someone tracing
-		 * down the cause of the crash will be able to figure
-		 * out the call path that was taken.
-		 */
-		if (__kernel_text_address(addr)) {
-#ifndef CONFIG_KALLSYMS
-			if (i % 5 == 0)
-				pr_cont("\n       ");
+	addr = (unsigned long) stack & THREAD_MASK;
+	stack_start = (unsigned long *) addr;
+	stack_end = (unsigned long *) (addr + THREAD_SIZE);
+
+	fp = stack;
+	pr_info("\nCall Trace:");
+
+	while (fp > stack_start && fp < stack_end) {
+#ifdef CONFIG_STACKTRACE
+		addr	= fp[1];
+		fp	= (unsigned long *) fp[0];
+#else
+		addr	= *fp++;
 #endif
-			pr_cont(" [<%08lx>] %pS\n", addr, (void *)addr);
-			i++;
-		}
+		if (__kernel_text_address(addr))
+			pr_cont("\n[<%08lx>] %pS", addr, (void *)addr);
 	}
 	pr_cont("\n");
 }
 
 void show_stack(struct task_struct *task, unsigned long *stack)
 {
-	unsigned long *p;
-	unsigned long *endstack;
-	int i;
-
 	if (!stack) {
 		if (task)
-			stack = (unsigned long *)task->thread.esp0;
+			stack = (unsigned long *)thread_saved_fp(task);
 		else
 			stack = (unsigned long *)&stack;
 	}
-	endstack = (unsigned long *)
-		(((unsigned long)stack + THREAD_SIZE - 1) & -THREAD_SIZE);
 
-	pr_info("Stack from %08lx:", (unsigned long)stack);
-	p = stack;
-	for (i = 0; i < kstack_depth_to_print; i++) {
-		if (p + 1 > endstack)
-			break;
-		if (i % 8 == 0)
-			pr_cont("\n       ");
-		pr_cont(" %08lx", *p++);
-	}
-	pr_cont("\n");
 	show_trace(stack);
 }
diff --git a/arch/csky/kernel/ptrace.c b/arch/csky/kernel/ptrace.c
index a76d466..eb62e07 100644
--- a/arch/csky/kernel/ptrace.c
+++ b/arch/csky/kernel/ptrace.c
@@ -241,6 +241,7 @@ asmlinkage void syscall_trace(int why, struct pt_regs *regs)
 	regs->regs[SYSTRACE_SAVENUM] = saved_why;
 }
 
+extern void show_stack(struct task_struct *task, unsigned long *stack);
 void show_regs(struct pt_regs *fp)
 {
 	unsigned long   *sp;
@@ -263,35 +264,37 @@ void show_regs(struct pt_regs *fp)
 		       (int) (((unsigned long) current) + 2 * PAGE_SIZE));
 	}
 
-	pr_info("PC: 0x%08lx\n", (long)fp->pc);
+	pr_info("PC: 0x%08lx (%pS)\n", (long)fp->pc, (void *)fp->pc);
+	pr_info("LR: 0x%08lx (%pS)\n", (long)fp->lr, (void *)fp->lr);
+	pr_info("SP: 0x%08lx\n", (long)fp);
 	pr_info("orig_a0: 0x%08lx\n", fp->orig_a0);
 	pr_info("PSR: 0x%08lx\n", (long)fp->sr);
 
-	pr_info("a0: 0x%08lx  a1: 0x%08lx  a2: 0x%08lx  a3: 0x%08lx\n",
-	       fp->a0, fp->a1, fp->a2, fp->a3);
+	pr_info(" a0: 0x%08lx   a1: 0x%08lx   a2: 0x%08lx   a3: 0x%08lx\n",
+		fp->a0, fp->a1, fp->a2, fp->a3);
 #if defined(__CSKYABIV2__)
-	pr_info("r4: 0x%08lx  r5: 0x%08lx    r6: 0x%08lx    r7: 0x%08lx\n",
+	pr_info(" r4: 0x%08lx   r5: 0x%08lx   r6: 0x%08lx   r7: 0x%08lx\n",
 		fp->regs[0], fp->regs[1], fp->regs[2], fp->regs[3]);
-	pr_info("r8: 0x%08lx  r9: 0x%08lx   r10: 0x%08lx   r11: 0x%08lx\n",
+	pr_info(" r8: 0x%08lx   r9: 0x%08lx  r10: 0x%08lx  r11: 0x%08lx\n",
 		fp->regs[4], fp->regs[5], fp->regs[6], fp->regs[7]);
-	pr_info("r12 0x%08lx  r13: 0x%08lx   r15: 0x%08lx\n",
+	pr_info("r12: 0x%08lx  r13: 0x%08lx  r15: 0x%08lx\n",
 		fp->regs[8], fp->regs[9], fp->lr);
-	pr_info("r16:0x%08lx   r17: 0x%08lx   r18: 0x%08lx    r19: 0x%08lx\n",
+	pr_info("r16: 0x%08lx  r17: 0x%08lx  r18: 0x%08lx  r19: 0x%08lx\n",
 		fp->exregs[0], fp->exregs[1], fp->exregs[2], fp->exregs[3]);
-	pr_info("r20 0x%08lx   r21: 0x%08lx   r22: 0x%08lx    r23: 0x%08lx\n",
+	pr_info("r20: 0x%08lx  r21: 0x%08lx  r22: 0x%08lx  r23: 0x%08lx\n",
 		fp->exregs[4], fp->exregs[5], fp->exregs[6], fp->exregs[7]);
-	pr_info("r24 0x%08lx   r25: 0x%08lx   r26: 0x%08lx    r27: 0x%08lx\n",
+	pr_info("r24: 0x%08lx  r25: 0x%08lx  r26: 0x%08lx  r27: 0x%08lx\n",
 		fp->exregs[8], fp->exregs[9], fp->exregs[10], fp->exregs[11]);
-	pr_info("r28 0x%08lx   r29: 0x%08lx   r30: 0x%08lx    tls: 0x%08lx\n",
+	pr_info("r28: 0x%08lx  r29: 0x%08lx  r30: 0x%08lx  tls: 0x%08lx\n",
 		fp->exregs[12], fp->exregs[13], fp->exregs[14], fp->tls);
-	pr_info("hi 0x%08lx    lo: 0x%08lx\n",
+	pr_info(" hi: 0x%08lx   lo: 0x%08lx\n",
 		fp->rhi, fp->rlo);
 #else
-	pr_info("r6: 0x%08lx   r7: 0x%08lx   r8: 0x%08lx   r9: 0x%08lx\n",
+	pr_info(" r6: 0x%08lx   r7: 0x%08lx   r8: 0x%08lx   r9: 0x%08lx\n",
 		fp->regs[0], fp->regs[1], fp->regs[2], fp->regs[3]);
-	pr_info("r10: 0x%08lx   r11: 0x%08lx   r12: 0x%08lx   r13: 0x%08lx\n",
+	pr_info("r10: 0x%08lx  r11: 0x%08lx  r12: 0x%08lx  r13: 0x%08lx\n",
 		fp->regs[4], fp->regs[5], fp->regs[6], fp->regs[7]);
-	pr_info("r14 0x%08lx   r1: 0x%08lx   r15: 0x%08lx\n",
+	pr_info("r14: 0x%08lx   r1: 0x%08lx  r15: 0x%08lx\n",
 		fp->regs[8], fp->regs[9], fp->lr);
 #endif
 
@@ -313,4 +316,7 @@ void show_regs(struct pt_regs *fp)
 		pr_cont("%08x ", (int) *sp++);
 	}
 	pr_cont("\n");
+
+	show_stack(NULL, (unsigned long *)fp->regs[4]);
+	return;
 }
diff --git a/arch/csky/mm/fault.c b/arch/csky/mm/fault.c
index 7df57f9..6a94661 100644
--- a/arch/csky/mm/fault.c
+++ b/arch/csky/mm/fault.c
@@ -188,8 +188,8 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long write,
 	 * terminate things with extreme prejudice.
 	 */
 	bust_spinlocks(1);
-	pr_alert("Unable to %s at vaddr: %08lx, epc: %08lx\n",
-		 __func__, address, regs->pc);
+	pr_alert("Unable to handle kernel paging request at virtual "
+		 "address 0x%08lx, pc: 0x%08lx\n", address, regs->pc);
 	die_if_kernel("Oops", regs, write);
 
 out_of_memory:
-- 
2.7.4

[PATCH 10/14] csky: remove unused members in processor.h

[guoren]

From: Guo Ren <ren_guo@c-sky.com>

Cleanup struct cpuinfo_csky and struct thread_struct, remove all esp0
related code. We could get pt_regs from sp and backtrace could use fp
in switch_stack.

Signed-off-by: Guo Ren <ren_guo@c-sky.com>
---
 arch/csky/include/asm/processor.h | 12 ------------
 arch/csky/kernel/asm-offsets.c    |  1 -
 arch/csky/kernel/entry.S          | 12 ------------
 arch/csky/kernel/ptrace.c         |  4 ----
 arch/csky/kernel/signal.c         |  2 --
 arch/csky/kernel/traps.c          |  6 ------
 arch/csky/mm/fault.c              |  3 ---
 7 files changed, 40 deletions(-)

diff --git a/arch/csky/include/asm/processor.h b/arch/csky/include/asm/processor.h
index 5ac3f32..8f45481 100644
--- a/arch/csky/include/asm/processor.h
+++ b/arch/csky/include/asm/processor.h
@@ -17,14 +17,7 @@
 #endif
 
 struct cpuinfo_csky {
-	unsigned long udelay_val;
 	unsigned long asid_cache;
-	/*
-	 * Capability and feature descriptor structure for CSKY CPU
-	 */
-	unsigned long options;
-	unsigned int processor_id[4];
-	unsigned int fpu_id;
 } __aligned(SMP_CACHE_BYTES);
 
 extern struct cpuinfo_csky cpu_data[];
@@ -50,11 +43,6 @@ extern struct cpuinfo_csky cpu_data[];
 struct thread_struct {
 	unsigned long  ksp;       /* kernel stack pointer */
 	unsigned long  sr;        /* saved status register */
-	unsigned long  esp0;      /* points to SR of stack frame */
-
-	/* Other stuff associated with the thread. */
-	unsigned long address;      /* Last user fault */
-	unsigned long error_code;
 
 	/* FPU regs */
 	struct user_fp __aligned(16) user_fp;
diff --git a/arch/csky/kernel/asm-offsets.c b/arch/csky/kernel/asm-offsets.c
index 2f07670..9b48b1b 100644
--- a/arch/csky/kernel/asm-offsets.c
+++ b/arch/csky/kernel/asm-offsets.c
@@ -20,7 +20,6 @@ int main(void)
 	/* offsets into the thread struct */
 	DEFINE(THREAD_KSP,        offsetof(struct thread_struct, ksp));
 	DEFINE(THREAD_SR,         offsetof(struct thread_struct, sr));
-	DEFINE(THREAD_ESP0,       offsetof(struct thread_struct, esp0));
 	DEFINE(THREAD_FESR,       offsetof(struct thread_struct, user_fp.fesr));
 	DEFINE(THREAD_FCR,        offsetof(struct thread_struct, user_fp.fcr));
 	DEFINE(THREAD_FPREG,      offsetof(struct thread_struct, user_fp.vr));
diff --git a/arch/csky/kernel/entry.S b/arch/csky/kernel/entry.S
index de378e4..5137ed9 100644
--- a/arch/csky/kernel/entry.S
+++ b/arch/csky/kernel/entry.S
@@ -122,16 +122,6 @@ ENTRY(csky_systemcall)
 
 	psrset  ee, ie
 
-	/* Stack frame for syscall, origin call set_esp0 */
-	mov     r12, sp
-
-	bmaski  r11, 13
-	andn    r12, r11
-	bgeni   r11, 9
-	addi    r11, 32
-	addu    r12, r11
-	st      sp, (r12, 0)
-
 	lrw     r11, __NR_syscalls
 	cmphs   syscallid, r11		/* Check nr of syscall */
 	bt      ret_from_exception
@@ -230,8 +220,6 @@ resume_userspace:
 1:  RESTORE_ALL
 
 exit_work:
-	mov	a0, sp			/* Stack address is arg[0] */
-	jbsr	set_esp0		/* Call C level */
 	btsti	r8, TIF_NEED_RESCHED
 	bt	work_resched
 	/* If thread_info->flag is empty, RESTORE_ALL */
diff --git a/arch/csky/kernel/ptrace.c b/arch/csky/kernel/ptrace.c
index eb62e07..57f1afe 100644
--- a/arch/csky/kernel/ptrace.c
+++ b/arch/csky/kernel/ptrace.c
@@ -50,15 +50,11 @@ static void singlestep_enable(struct task_struct *tsk)
  */
 void user_enable_single_step(struct task_struct *child)
 {
-	if (child->thread.esp0 == 0)
-		return;
 	singlestep_enable(child);
 }
 
 void user_disable_single_step(struct task_struct *child)
 {
-	if (child->thread.esp0 == 0)
-		return;
 	singlestep_disable(child);
 }
 
diff --git a/arch/csky/kernel/signal.c b/arch/csky/kernel/signal.c
index 66e1b72..e680dc1 100644
--- a/arch/csky/kernel/signal.c
+++ b/arch/csky/kernel/signal.c
@@ -238,8 +238,6 @@ static void do_signal(struct pt_regs *regs, int syscall)
 	if (!user_mode(regs))
 		return;
 
-	current->thread.esp0 = (unsigned long)regs;
-
 	/*
 	 * If we were from a system call, check for system call restarting...
 	 */
diff --git a/arch/csky/kernel/traps.c b/arch/csky/kernel/traps.c
index a8368ed..f487a9b 100644
--- a/arch/csky/kernel/traps.c
+++ b/arch/csky/kernel/traps.c
@@ -106,7 +106,6 @@ void buserr(struct pt_regs *regs)
 	pr_err("User mode Bus Error\n");
 	show_regs(regs);
 
-	current->thread.esp0 = (unsigned long) regs;
 	force_sig_fault(SIGSEGV, 0, (void __user *)regs->pc, current);
 }
 
@@ -162,8 +161,3 @@ asmlinkage void trap_c(struct pt_regs *regs)
 	}
 	send_sig(sig, current, 0);
 }
-
-asmlinkage void set_esp0(unsigned long ssp)
-{
-	current->thread.esp0 = ssp;
-}
diff --git a/arch/csky/mm/fault.c b/arch/csky/mm/fault.c
index 6a94661..d6f4b66 100644
--- a/arch/csky/mm/fault.c
+++ b/arch/csky/mm/fault.c
@@ -172,8 +172,6 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long write,
 bad_area_nosemaphore:
 	/* User mode accesses just cause a SIGSEGV */
 	if (user_mode(regs)) {
-		tsk->thread.address = address;
-		tsk->thread.error_code = write;
 		force_sig_fault(SIGSEGV, si_code, (void __user *)address, current);
 		return;
 	}
@@ -207,6 +205,5 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long write,
 	if (!user_mode(regs))
 		goto no_context;
 
-	tsk->thread.address = address;
 	force_sig_fault(SIGBUS, BUS_ADRERR, (void __user *)address, current);
 }
-- 
2.7.4

[PATCH 11/14] csky: basic ftrace supported

[guoren]

From: Guo Ren <ren_guo@c-sky.com>

When gcc with -pg, it'll add _mcount stub in every function. We need
implement the _mcount in kernel and ftrace depends on stackstrace.

To do: call-graph, dynamic ftrace

Signed-off-by: Guo Ren <ren_guo@c-sky.com>
---
 arch/csky/Kconfig              |  1 +
 arch/csky/abiv2/Makefile       |  1 +
 arch/csky/abiv2/mcount.S       | 24 ++++++++++++++++++++++++
 arch/csky/include/asm/ftrace.h |  9 +++++++++
 arch/csky/kernel/Makefile      |  5 +++++
 arch/csky/kernel/ftrace.c      | 24 ++++++++++++++++++++++++
 6 files changed, 64 insertions(+)
 create mode 100644 arch/csky/abiv2/mcount.S
 create mode 100644 arch/csky/include/asm/ftrace.h
 create mode 100644 arch/csky/kernel/ftrace.c

diff --git a/arch/csky/Kconfig b/arch/csky/Kconfig
index 65804d1..0b9a290 100644
--- a/arch/csky/Kconfig
+++ b/arch/csky/Kconfig
@@ -29,6 +29,7 @@ config CSKY
 	select GENERIC_SCHED_CLOCK
 	select GENERIC_SMP_IDLE_THREAD
 	select HAVE_ARCH_TRACEHOOK
+	select HAVE_FUNCTION_TRACER
 	select HAVE_GENERIC_DMA_COHERENT
 	select HAVE_KERNEL_GZIP
 	select HAVE_KERNEL_LZO
diff --git a/arch/csky/abiv2/Makefile b/arch/csky/abiv2/Makefile
index 069ca72..b1d44f6 100644
--- a/arch/csky/abiv2/Makefile
+++ b/arch/csky/abiv2/Makefile
@@ -8,3 +8,4 @@ obj-y				+= strcmp.o
 obj-y				+= strcpy.o
 obj-y				+= strlen.o
 obj-y				+= strksyms.o
+obj-$(CONFIG_FUNCTION_TRACER)	+= mcount.o
diff --git a/arch/csky/abiv2/mcount.S b/arch/csky/abiv2/mcount.S
new file mode 100644
index 0000000..73377d5
--- /dev/null
+++ b/arch/csky/abiv2/mcount.S
@@ -0,0 +1,24 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd.
+
+#include <linux/linkage.h>
+
+ENTRY (_mcount)
+	subi	sp, 20
+	stw	a0, (sp, 0)
+	stw	a1, (sp, 4)
+	stw	a2, (sp, 8)
+	stw	a3, (sp, 12)
+	stw	lr, (sp, 16)
+	mov	a1, lr
+	ldw	a0, (sp, 20)
+	jsri    csky_mcount
+	ldw	a0, (sp, 0)
+	ldw	a1, (sp, 4)
+	ldw	a2, (sp, 8)
+	ldw	a3, (sp, 12)
+	ldw	t1, (sp, 16)
+	ldw	lr, (sp, 20)
+	addi	sp, 24
+	jmp	t1
+END (_mcount)
diff --git a/arch/csky/include/asm/ftrace.h b/arch/csky/include/asm/ftrace.h
new file mode 100644
index 0000000..1d22a17
--- /dev/null
+++ b/arch/csky/include/asm/ftrace.h
@@ -0,0 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd.
+
+#ifndef __ASM_CSKY_FTRACE_H
+#define __ASM_CSKY_FTRACE_H
+
+extern void _mcount(unsigned long from_pc);
+
+#endif /* __ASM_CSKY_FTRACE_H */
diff --git a/arch/csky/kernel/Makefile b/arch/csky/kernel/Makefile
index ba5ca48..3c0e2d1 100644
--- a/arch/csky/kernel/Makefile
+++ b/arch/csky/kernel/Makefile
@@ -6,4 +6,9 @@ obj-y += process.o cpu-probe.o ptrace.o dumpstack.o
 
 obj-$(CONFIG_MODULES)			+= module.o
 obj-$(CONFIG_SMP)			+= smp.o
+obj-$(CONFIG_FUNCTION_TRACER)		+= ftrace.o
 obj-$(CONFIG_STACKTRACE)		+= stacktrace.o
+
+ifdef CONFIG_FUNCTION_TRACER
+CFLAGS_REMOVE_ftrace.o = $(CC_FLAGS_FTRACE)
+endif
diff --git a/arch/csky/kernel/ftrace.c b/arch/csky/kernel/ftrace.c
new file mode 100644
index 0000000..ad054f719
--- /dev/null
+++ b/arch/csky/kernel/ftrace.c
@@ -0,0 +1,24 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd.
+
+#include <linux/ftrace.h>
+#include <linux/uaccess.h>
+
+extern void (*ftrace_trace_function)(unsigned long, unsigned long,
+				     struct ftrace_ops*, struct pt_regs*);
+
+
+noinline void __naked ftrace_stub(unsigned long ip, unsigned long parent_ip,
+				  struct ftrace_ops *op, struct pt_regs *regs)
+{
+	asm volatile ("\n");
+}
+
+noinline void csky_mcount(unsigned long from_pc, unsigned long self_pc)
+{
+	if (ftrace_trace_function != ftrace_stub)
+		ftrace_trace_function(self_pc, from_pc, NULL, NULL);
+}
+
+/* _mcount is defined in abi's mcount.S */
+EXPORT_SYMBOL(_mcount);
-- 
2.7.4

[PATCH 12/14] csky: ftrace call graph supported.

[guoren]

From: Guo Ren <ren_guo@c-sky.com>

With csky-gcc -pg -mbacktrace, ftrace call graph supported.

Signed-off-by: Guo Ren <ren_guo@c-sky.com>
---
 arch/csky/Kconfig              |   1 +
 arch/csky/abiv2/mcount.S       | 116 ++++++++++++++++++++++++++++++++++++++---
 arch/csky/include/asm/ftrace.h |   6 ++-
 arch/csky/kernel/ftrace.c      |  48 +++++++++++++----
 4 files changed, 151 insertions(+), 20 deletions(-)

diff --git a/arch/csky/Kconfig b/arch/csky/Kconfig
index 0b9a290..6e81477 100644
--- a/arch/csky/Kconfig
+++ b/arch/csky/Kconfig
@@ -30,6 +30,7 @@ config CSKY
 	select GENERIC_SMP_IDLE_THREAD
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_FUNCTION_TRACER
+	select HAVE_FUNCTION_GRAPH_TRACER
 	select HAVE_GENERIC_DMA_COHERENT
 	select HAVE_KERNEL_GZIP
 	select HAVE_KERNEL_LZO
diff --git a/arch/csky/abiv2/mcount.S b/arch/csky/abiv2/mcount.S
index 73377d5..c633379 100644
--- a/arch/csky/abiv2/mcount.S
+++ b/arch/csky/abiv2/mcount.S
@@ -2,23 +2,123 @@
 // Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd.
 
 #include <linux/linkage.h>
+#include <asm/ftrace.h>
 
-ENTRY (_mcount)
-	subi	sp, 20
+/*
+ * csky-gcc with -pg will put the following asm after prologue:
+ *      push	r15
+ *      jsri	_mcount
+ *
+ * stack layout after mcount_enter in _mcount():
+ *
+ * current sp => 0:+-------+
+ *                 | a0-a3 | -> must save all argument regs
+ *             +16:+-------+
+ *                 | lr    | -> _mcount lr (instrumente function's pc)
+ *             +20:+-------+
+ *                 | fp=r8 | -> instrumented function fp
+ *             +24:+-------+
+ *                 | plr   | -> instrumented function lr (parent's pc)
+ *                 +-------+
+ */
+
+.macro mcount_enter
+	subi	sp, 24
 	stw	a0, (sp, 0)
 	stw	a1, (sp, 4)
 	stw	a2, (sp, 8)
 	stw	a3, (sp, 12)
 	stw	lr, (sp, 16)
-	mov	a1, lr
-	ldw	a0, (sp, 20)
-	jsri    csky_mcount
+	stw	r8, (sp, 20)
+.endm
+
+.macro mcount_exit
 	ldw	a0, (sp, 0)
 	ldw	a1, (sp, 4)
 	ldw	a2, (sp, 8)
 	ldw	a3, (sp, 12)
 	ldw	t1, (sp, 16)
-	ldw	lr, (sp, 20)
-	addi	sp, 24
+	ldw	r8, (sp, 20)
+	ldw	lr, (sp, 24)
+	addi	sp, 28
 	jmp	t1
-END (_mcount)
+.endm
+
+.macro save_return_regs
+	subi	sp, 16
+	stw	a0, (sp, 0)
+	stw	a1, (sp, 4)
+	stw	a2, (sp, 8)
+	stw	a3, (sp, 12)
+.endm
+
+.macro restore_return_regs
+	mov	lr, a0
+	ldw	a0, (sp, 0)
+	ldw	a1, (sp, 4)
+	ldw	a2, (sp, 8)
+	ldw	a3, (sp, 12)
+	addi	sp, 16
+.endm
+
+ENTRY(ftrace_stub)
+	jmp	lr
+END(ftrace_stub)
+
+ENTRY(_mcount)
+	mcount_enter
+
+	/* r26 is link register, only used with jsri translation */
+	lrw	r26, ftrace_trace_function
+	ldw	r26, (r26, 0)
+	lrw	a1, ftrace_stub
+	cmpne	r26, a1
+	bf	skip_ftrace
+
+	mov	a0, lr
+	subi	a0, MCOUNT_INSN_SIZE
+	ldw	a1, (sp, 24)
+
+	jsr	r26
+
+#ifndef CONFIG_FUNCTION_GRAPH_TRACER
+skip_ftrace:
+	mcount_exit
+#else
+skip_ftrace:
+	lrw	a0, ftrace_graph_return
+	ldw	a0, (a0, 0)
+	lrw	a1, ftrace_stub
+	cmpne	a0, a1
+	bt	ftrace_graph_caller
+
+	lrw	a0, ftrace_graph_entry
+	ldw	a0, (a0, 0)
+	lrw	a1, ftrace_graph_entry_stub
+	cmpne	a0, a1
+	bt	ftrace_graph_caller
+
+	mcount_exit
+#endif
+END(_mcount)
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+ENTRY(ftrace_graph_caller)
+	mov	a0, sp
+	addi	a0, 24
+	ldw	a1, (sp, 16)
+	subi	a1, MCOUNT_INSN_SIZE
+	mov	a2, r8
+	lrw	r26, prepare_ftrace_return
+	jsr	r26
+	mcount_exit
+END(ftrace_graph_caller)
+
+ENTRY(return_to_handler)
+	save_return_regs
+	mov	a0, r8
+	jsri	ftrace_return_to_handler
+	restore_return_regs
+	jmp	lr
+END(return_to_handler)
+#endif
diff --git a/arch/csky/include/asm/ftrace.h b/arch/csky/include/asm/ftrace.h
index 1d22a17..7547c45 100644
--- a/arch/csky/include/asm/ftrace.h
+++ b/arch/csky/include/asm/ftrace.h
@@ -4,6 +4,10 @@
 #ifndef __ASM_CSKY_FTRACE_H
 #define __ASM_CSKY_FTRACE_H
 
-extern void _mcount(unsigned long from_pc);
+#define MCOUNT_INSN_SIZE 4
+
+#define HAVE_FUNCTION_GRAPH_FP_TEST
+
+#define HAVE_FUNCTION_GRAPH_RET_ADDR_PTR
 
 #endif /* __ASM_CSKY_FTRACE_H */
diff --git a/arch/csky/kernel/ftrace.c b/arch/csky/kernel/ftrace.c
index ad054f719..274c431 100644
--- a/arch/csky/kernel/ftrace.c
+++ b/arch/csky/kernel/ftrace.c
@@ -4,21 +4,47 @@
 #include <linux/ftrace.h>
 #include <linux/uaccess.h>
 
-extern void (*ftrace_trace_function)(unsigned long, unsigned long,
-				     struct ftrace_ops*, struct pt_regs*);
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr,
+			   unsigned long frame_pointer)
+{
+	unsigned long return_hooker = (unsigned long)&return_to_handler;
+	unsigned long old;
 
+	if (unlikely(atomic_read(&current->tracing_graph_pause)))
+		return;
 
-noinline void __naked ftrace_stub(unsigned long ip, unsigned long parent_ip,
-				  struct ftrace_ops *op, struct pt_regs *regs)
-{
-	asm volatile ("\n");
-}
+	old = *parent;
 
-noinline void csky_mcount(unsigned long from_pc, unsigned long self_pc)
-{
-	if (ftrace_trace_function != ftrace_stub)
-		ftrace_trace_function(self_pc, from_pc, NULL, NULL);
+	if (!function_graph_enter(old, self_addr,
+			*(unsigned long *)frame_pointer, parent)) {
+		/*
+		 * For csky-gcc function has sub-call:
+		 * subi	sp,	sp, 8
+		 * stw	r8,	(sp, 0)
+		 * mov	r8,	sp
+		 * st.w r15,	(sp, 0x4)
+		 * push	r15
+		 * jl	_mcount
+		 * We only need set *parent for resume
+		 *
+		 * For csky-gcc function has no sub-call:
+		 * subi	sp,	sp, 4
+		 * stw	r8,	(sp, 0)
+		 * mov	r8,	sp
+		 * push	r15
+		 * jl	_mcount
+		 * We need set *parent and *(frame_pointer + 4) for resume,
+		 * because lr is resumed twice.
+		 */
+		*parent = return_hooker;
+		frame_pointer += 4;
+		if (*(unsigned long *)frame_pointer == old)
+			*(unsigned long *)frame_pointer = return_hooker;
+	}
 }
+#endif
 
 /* _mcount is defined in abi's mcount.S */
+extern void _mcount(void);
 EXPORT_SYMBOL(_mcount);
-- 
2.7.4

[PATCH 14/14] csky: Add EM_CSKY_OLD 39

[guoren]

From: Guo Ren <ren_guo@c-sky.com>

C-SKY historically used 39, the same value as MCORE, from which the
architecture was derived.

C-SKY binutils support both EM_CSKY and EM_CSKY_OLD, confirmed by
binutils:include/elf/common.h

Signed-off-by: Guo Ren <ren_guo@c-sky.com>
---
 arch/csky/include/asm/elf.h | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/arch/csky/include/asm/elf.h b/arch/csky/include/asm/elf.h
index d6dbc00..e1ec558 100644
--- a/arch/csky/include/asm/elf.h
+++ b/arch/csky/include/asm/elf.h
@@ -8,6 +8,7 @@
 #include <abi/regdef.h>
 
 #define ELF_ARCH EM_CSKY
+#define EM_CSKY_OLD 39
 
 /* CSKY Relocations */
 #define R_CSKY_NONE               0
@@ -43,7 +44,8 @@ typedef elf_greg_t elf_gregset_t[ELF_NGREG];
 /*
  * This is used to ensure we don't load something for the wrong architecture.
  */
-#define elf_check_arch(x) ((x)->e_machine == ELF_ARCH)
+#define elf_check_arch(x) (((x)->e_machine == ELF_ARCH) || \
+			   ((x)->e_machine == EM_CSKY_OLD))
 
 /*
  * These are used to set parameters in the core dumps.
-- 
2.7.4