From 77ec8a545136272ca3056ca3759836df0cfee5e5 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Mon, 1 Feb 2016 14:12:08 +0100
Subject: s390/stacktrace: use nosched instead of savesched parameter

Use the inversed "nosched" logic like all other architectures.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Tested-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/stacktrace.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

(limited to 'arch/s390/kernel')

diff --git a/arch/s390/kernel/stacktrace.c b/arch/s390/kernel/stacktrace.c
index 8f64ebd..3872d14 100644
--- a/arch/s390/kernel/stacktrace.c
+++ b/arch/s390/kernel/stacktrace.c
@@ -14,7 +14,7 @@ static unsigned long save_context_stack(struct stack_trace *trace,
 					unsigned long sp,
 					unsigned long low,
 					unsigned long high,
-					int savesched)
+					int nosched)
 {
 	struct stack_frame *sf;
 	struct pt_regs *regs;
@@ -46,7 +46,7 @@ static unsigned long save_context_stack(struct stack_trace *trace,
 			return sp;
 		regs = (struct pt_regs *)sp;
 		addr = regs->psw.addr;
-		if (savesched || !in_sched_functions(addr)) {
+		if (!nosched || !in_sched_functions(addr)) {
 			if (!trace->skip)
 				trace->entries[trace->nr_entries++] = addr;
 			else
@@ -66,13 +66,13 @@ static void __save_stack_trace(struct stack_trace *trace, unsigned long sp)
 	frame_size = STACK_FRAME_OVERHEAD + sizeof(struct pt_regs);
 	new_sp = save_context_stack(trace, sp,
 			S390_lowcore.panic_stack + frame_size - PAGE_SIZE,
-			S390_lowcore.panic_stack + frame_size, 1);
+			S390_lowcore.panic_stack + frame_size, 0);
 	new_sp = save_context_stack(trace, new_sp,
 			S390_lowcore.async_stack + frame_size - ASYNC_SIZE,
-			S390_lowcore.async_stack + frame_size, 1);
+			S390_lowcore.async_stack + frame_size, 0);
 	save_context_stack(trace, new_sp,
 			   S390_lowcore.thread_info,
-			   S390_lowcore.thread_info + THREAD_SIZE, 1);
+			   S390_lowcore.thread_info + THREAD_SIZE, 0);
 }
 
 void save_stack_trace(struct stack_trace *trace)
@@ -98,7 +98,7 @@ void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
 	}
 	low = (unsigned long) task_stack_page(tsk);
 	high = (unsigned long) task_pt_regs(tsk);
-	save_context_stack(trace, sp, low, high, 0);
+	save_context_stack(trace, sp, low, high, 1);
 	if (trace->nr_entries < trace->max_entries)
 		trace->entries[trace->nr_entries++] = ULONG_MAX;
 }
-- 
cgit v1.1


From 76737ce17ab4c88f14e4452076610474108246d7 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Sun, 31 Jan 2016 17:06:16 +0100
Subject: s390: add current_stack_pointer() helper function

Implement current_stack_pointer() helper function and use it
everywhere, instead of having several different inline assembly
variants.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Tested-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/dumpstack.c  | 16 ++++++++--------
 arch/s390/kernel/irq.c        |  3 +--
 arch/s390/kernel/stacktrace.c |  9 +++------
 3 files changed, 12 insertions(+), 16 deletions(-)

(limited to 'arch/s390/kernel')

diff --git a/arch/s390/kernel/dumpstack.c b/arch/s390/kernel/dumpstack.c
index 02bd02f..feda48f 100644
--- a/arch/s390/kernel/dumpstack.c
+++ b/arch/s390/kernel/dumpstack.c
@@ -67,12 +67,11 @@ static void show_trace(struct task_struct *task, unsigned long *stack)
 {
 	const unsigned long frame_size =
 		STACK_FRAME_OVERHEAD + sizeof(struct pt_regs);
-	register unsigned long __r15 asm ("15");
 	unsigned long sp;
 
 	sp = (unsigned long) stack;
 	if (!sp)
-		sp = task ? task->thread.ksp : __r15;
+		sp = task ? task->thread.ksp : current_stack_pointer();
 	printk("Call Trace:\n");
 #ifdef CONFIG_CHECK_STACK
 	sp = __show_trace(sp,
@@ -95,15 +94,16 @@ static void show_trace(struct task_struct *task, unsigned long *stack)
 
 void show_stack(struct task_struct *task, unsigned long *sp)
 {
-	register unsigned long *__r15 asm ("15");
 	unsigned long *stack;
 	int i;
 
-	if (!sp)
-		stack = task ? (unsigned long *) task->thread.ksp : __r15;
-	else
-		stack = sp;
-
+	stack = sp;
+	if (!stack) {
+		if (!task)
+			stack = (unsigned long *)current_stack_pointer();
+		else
+			stack = (unsigned long *)task->thread.ksp;
+	}
 	for (i = 0; i < 20; i++) {
 		if (((addr_t) stack & (THREAD_SIZE-1)) == 0)
 			break;
diff --git a/arch/s390/kernel/irq.c b/arch/s390/kernel/irq.c
index f41d520..c373a1d 100644
--- a/arch/s390/kernel/irq.c
+++ b/arch/s390/kernel/irq.c
@@ -164,8 +164,7 @@ void do_softirq_own_stack(void)
 {
 	unsigned long old, new;
 
-	/* Get current stack pointer. */
-	asm volatile("la %0,0(15)" : "=a" (old));
+	old = current_stack_pointer();
 	/* Check against async. stack address range. */
 	new = S390_lowcore.async_stack;
 	if (((new - old) >> (PAGE_SHIFT + THREAD_ORDER)) != 0) {
diff --git a/arch/s390/kernel/stacktrace.c b/arch/s390/kernel/stacktrace.c
index 3872d14..7dd8360 100644
--- a/arch/s390/kernel/stacktrace.c
+++ b/arch/s390/kernel/stacktrace.c
@@ -77,10 +77,9 @@ static void __save_stack_trace(struct stack_trace *trace, unsigned long sp)
 
 void save_stack_trace(struct stack_trace *trace)
 {
-	register unsigned long r15 asm ("15");
 	unsigned long sp;
 
-	sp = r15;
+	sp = current_stack_pointer();
 	__save_stack_trace(trace, sp);
 	if (trace->nr_entries < trace->max_entries)
 		trace->entries[trace->nr_entries++] = ULONG_MAX;
@@ -92,10 +91,8 @@ void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
 	unsigned long sp, low, high;
 
 	sp = tsk->thread.ksp;
-	if (tsk == current) {
-		/* Get current stack pointer. */
-		asm volatile("la %0,0(15)" : "=a" (sp));
-	}
+	if (tsk == current)
+		sp = current_stack_pointer();
 	low = (unsigned long) task_stack_page(tsk);
 	high = (unsigned long) task_pt_regs(tsk);
 	save_context_stack(trace, sp, low, high, 1);
-- 
cgit v1.1


From ca0fd75a54cd4b1b1957bfa5611644f5b8b8465e Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Sun, 31 Jan 2016 17:15:56 +0100
Subject: s390/dumpstack: add missing ri bit to show_registers() output

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/dumpstack.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'arch/s390/kernel')

diff --git a/arch/s390/kernel/dumpstack.c b/arch/s390/kernel/dumpstack.c
index feda48f..9a4852b 100644
--- a/arch/s390/kernel/dumpstack.c
+++ b/arch/s390/kernel/dumpstack.c
@@ -142,7 +142,8 @@ void show_registers(struct pt_regs *regs)
 	       mask_bits(regs, PSW_MASK_MCHECK), mask_bits(regs, PSW_MASK_WAIT),
 	       mask_bits(regs, PSW_MASK_PSTATE), mask_bits(regs, PSW_MASK_ASC),
 	       mask_bits(regs, PSW_MASK_CC), mask_bits(regs, PSW_MASK_PM));
-	printk(" EA:%x", mask_bits(regs, PSW_MASK_EA | PSW_MASK_BA));
+	printk(" RI:%x EA:%x", mask_bits(regs, PSW_MASK_RI),
+	       mask_bits(regs, PSW_MASK_EA | PSW_MASK_BA));
 	printk("\n%s GPRS: %016lx %016lx %016lx %016lx\n", mode,
 	       regs->gprs[0], regs->gprs[1], regs->gprs[2], regs->gprs[3]);
 	printk("           %016lx %016lx %016lx %016lx\n",
-- 
cgit v1.1


From 1f021ea0da3bc499387d87e31aeb1e36d2db052b Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Sun, 31 Jan 2016 17:36:59 +0100
Subject: s390/dumpstack: use bit fields to decode psw mask in show_registers()

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/dumpstack.c | 17 ++++-------------
 1 file changed, 4 insertions(+), 13 deletions(-)

(limited to 'arch/s390/kernel')

diff --git a/arch/s390/kernel/dumpstack.c b/arch/s390/kernel/dumpstack.c
index 9a4852b..e1c786d 100644
--- a/arch/s390/kernel/dumpstack.c
+++ b/arch/s390/kernel/dumpstack.c
@@ -121,13 +121,9 @@ static void show_last_breaking_event(struct pt_regs *regs)
 	printk(" [<%016lx>] %pSR\n", regs->args[0], (void *)regs->args[0]);
 }
 
-static inline int mask_bits(struct pt_regs *regs, unsigned long bits)
-{
-	return (regs->psw.mask & bits) / ((~bits + 1) & bits);
-}
-
 void show_registers(struct pt_regs *regs)
 {
+	struct psw_bits *psw = &psw_bits(regs->psw);
 	char *mode;
 
 	mode = user_mode(regs) ? "User" : "Krnl";
@@ -136,14 +132,9 @@ void show_registers(struct pt_regs *regs)
 		printk(" (%pSR)", (void *)regs->psw.addr);
 	printk("\n");
 	printk("           R:%x T:%x IO:%x EX:%x Key:%x M:%x W:%x "
-	       "P:%x AS:%x CC:%x PM:%x", mask_bits(regs, PSW_MASK_PER),
-	       mask_bits(regs, PSW_MASK_DAT), mask_bits(regs, PSW_MASK_IO),
-	       mask_bits(regs, PSW_MASK_EXT), mask_bits(regs, PSW_MASK_KEY),
-	       mask_bits(regs, PSW_MASK_MCHECK), mask_bits(regs, PSW_MASK_WAIT),
-	       mask_bits(regs, PSW_MASK_PSTATE), mask_bits(regs, PSW_MASK_ASC),
-	       mask_bits(regs, PSW_MASK_CC), mask_bits(regs, PSW_MASK_PM));
-	printk(" RI:%x EA:%x", mask_bits(regs, PSW_MASK_RI),
-	       mask_bits(regs, PSW_MASK_EA | PSW_MASK_BA));
+	       "P:%x AS:%x CC:%x PM:%x", psw->r, psw->t, psw->i, psw->e,
+	       psw->key, psw->m, psw->w, psw->p, psw->as, psw->cc, psw->pm);
+	printk(" RI:%x EA:%x", psw->ri, psw->eaba);
 	printk("\n%s GPRS: %016lx %016lx %016lx %016lx\n", mode,
 	       regs->gprs[0], regs->gprs[1], regs->gprs[2], regs->gprs[3]);
 	printk("           %016lx %016lx %016lx %016lx\n",
-- 
cgit v1.1


From 758d39ebd3d5666edb3b1c339f7f138c349ff8bf Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Tue, 9 Feb 2016 12:58:54 +0100
Subject: s390/dumpstack: merge all four stack tracers

We have four different stack tracers of which three had bugs. So it's
time to merge them to a single stack tracer which allows to specify a
call back function which will be called for each step.

This patch changes behavior a bit:

- the "nosched" and "in_sched_functions" check within
  save_stack_trace_tsk did work only for the last stack frame within a
  context. Now it considers the check for each stack frame like it
  should.

- both the oprofile variant and the perf_events variant did save a
  return address twice if a zero back chain was detected, which
  indicates an interrupt frame. The new dump_trace function will call
  the oprofile and perf_events backends with the psw address that is
  contained within the corresponding pt_regs structure instead.

- the original show_trace and save_context_stack functions did already
  use the psw address of the pt_regs structure if a zero back chain
  was detected. However now we ignore the psw address if it is a user
  space address. After all we trace the kernel stack and not the user
  space stack. This way we also get rid of the garbage user space
  address in case of warnings and / or panic call traces.

So this should make life easier since now there is only one stack
tracer left which we can break.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/dumpstack.c  | 63 +++++++++++++++++++-------------
 arch/s390/kernel/perf_event.c | 56 ++++-------------------------
 arch/s390/kernel/stacktrace.c | 84 ++++++++++++-------------------------------
 3 files changed, 66 insertions(+), 137 deletions(-)

(limited to 'arch/s390/kernel')

diff --git a/arch/s390/kernel/dumpstack.c b/arch/s390/kernel/dumpstack.c
index e1c786d..2150b01 100644
--- a/arch/s390/kernel/dumpstack.c
+++ b/arch/s390/kernel/dumpstack.c
@@ -19,28 +19,28 @@
 #include <asm/ipl.h>
 
 /*
- * For show_trace we have tree different stack to consider:
+ * For dump_trace we have tree different stack to consider:
  *   - the panic stack which is used if the kernel stack has overflown
  *   - the asynchronous interrupt stack (cpu related)
  *   - the synchronous kernel stack (process related)
- * The stack trace can start at any of the three stack and can potentially
+ * The stack trace can start at any of the three stacks and can potentially
  * touch all of them. The order is: panic stack, async stack, sync stack.
  */
 static unsigned long
-__show_trace(unsigned long sp, unsigned long low, unsigned long high)
+__dump_trace(dump_trace_func_t func, void *data, unsigned long sp,
+	     unsigned long low, unsigned long high)
 {
 	struct stack_frame *sf;
 	struct pt_regs *regs;
-	unsigned long addr;
 
 	while (1) {
 		if (sp < low || sp > high - sizeof(*sf))
 			return sp;
 		sf = (struct stack_frame *) sp;
-		addr = sf->gprs[8];
-		printk("([<%016lx>] %pSR)\n", addr, (void *)addr);
 		/* Follow the backchain. */
 		while (1) {
+			if (func(data, sf->gprs[8]))
+				return sp;
 			low = sp;
 			sp = sf->back_chain;
 			if (!sp)
@@ -48,45 +48,58 @@ __show_trace(unsigned long sp, unsigned long low, unsigned long high)
 			if (sp <= low || sp > high - sizeof(*sf))
 				return sp;
 			sf = (struct stack_frame *) sp;
-			addr = sf->gprs[8];
-			printk(" [<%016lx>] %pSR\n", addr, (void *)addr);
 		}
 		/* Zero backchain detected, check for interrupt frame. */
 		sp = (unsigned long) (sf + 1);
 		if (sp <= low || sp > high - sizeof(*regs))
 			return sp;
 		regs = (struct pt_regs *) sp;
-		addr = regs->psw.addr;
-		printk(" [<%016lx>] %pSR\n", addr, (void *)addr);
+		if (!user_mode(regs)) {
+			if (func(data, regs->psw.addr))
+				return sp;
+		}
 		low = sp;
 		sp = regs->gprs[15];
 	}
 }
 
-static void show_trace(struct task_struct *task, unsigned long *stack)
+void dump_trace(dump_trace_func_t func, void *data, struct task_struct *task,
+		unsigned long sp)
 {
-	const unsigned long frame_size =
-		STACK_FRAME_OVERHEAD + sizeof(struct pt_regs);
-	unsigned long sp;
+	unsigned long frame_size;
 
-	sp = (unsigned long) stack;
-	if (!sp)
-		sp = task ? task->thread.ksp : current_stack_pointer();
-	printk("Call Trace:\n");
+	frame_size = STACK_FRAME_OVERHEAD + sizeof(struct pt_regs);
 #ifdef CONFIG_CHECK_STACK
-	sp = __show_trace(sp,
+	sp = __dump_trace(func, data, sp,
 			  S390_lowcore.panic_stack + frame_size - 4096,
 			  S390_lowcore.panic_stack + frame_size);
 #endif
-	sp = __show_trace(sp,
+	sp = __dump_trace(func, data, sp,
 			  S390_lowcore.async_stack + frame_size - ASYNC_SIZE,
 			  S390_lowcore.async_stack + frame_size);
 	if (task)
-		__show_trace(sp, (unsigned long) task_stack_page(task),
-			     (unsigned long) task_stack_page(task) + THREAD_SIZE);
+		__dump_trace(func, data, sp,
+			     (unsigned long)task_stack_page(task),
+			     (unsigned long)task_stack_page(task) + THREAD_SIZE);
 	else
-		__show_trace(sp, S390_lowcore.thread_info,
+		__dump_trace(func, data, sp,
+			     S390_lowcore.thread_info,
 			     S390_lowcore.thread_info + THREAD_SIZE);
+}
+EXPORT_SYMBOL_GPL(dump_trace);
+
+static int show_address(void *data, unsigned long address)
+{
+	printk("([<%016lx>] %pSR)\n", address, (void *)address);
+	return 0;
+}
+
+static void show_trace(struct task_struct *task, unsigned long sp)
+{
+	if (!sp)
+		sp = task ? task->thread.ksp : current_stack_pointer();
+	printk("Call Trace:\n");
+	dump_trace(show_address, NULL, task, sp);
 	if (!task)
 		task = current;
 	debug_show_held_locks(task);
@@ -112,7 +125,7 @@ void show_stack(struct task_struct *task, unsigned long *sp)
 		printk("%016lx ", *stack++);
 	}
 	printk("\n");
-	show_trace(task, sp);
+	show_trace(task, (unsigned long)sp);
 }
 
 static void show_last_breaking_event(struct pt_regs *regs)
@@ -152,7 +165,7 @@ void show_regs(struct pt_regs *regs)
 	show_registers(regs);
 	/* Show stack backtrace if pt_regs is from kernel mode */
 	if (!user_mode(regs))
-		show_trace(NULL, (unsigned long *) regs->gprs[15]);
+		show_trace(NULL, regs->gprs[15]);
 	show_last_breaking_event(regs);
 }
 
diff --git a/arch/s390/kernel/perf_event.c b/arch/s390/kernel/perf_event.c
index 0943b11..a1b7086 100644
--- a/arch/s390/kernel/perf_event.c
+++ b/arch/s390/kernel/perf_event.c
@@ -222,64 +222,20 @@ static int __init service_level_perf_register(void)
 }
 arch_initcall(service_level_perf_register);
 
-/* See also arch/s390/kernel/traps.c */
-static unsigned long __store_trace(struct perf_callchain_entry *entry,
-				   unsigned long sp,
-				   unsigned long low, unsigned long high)
+static int __perf_callchain_kernel(void *data, unsigned long address)
 {
-	struct stack_frame *sf;
-	struct pt_regs *regs;
-
-	while (1) {
-		if (sp < low || sp > high - sizeof(*sf))
-			return sp;
-		sf = (struct stack_frame *) sp;
-		perf_callchain_store(entry, sf->gprs[8]);
-		/* Follow the backchain. */
-		while (1) {
-			low = sp;
-			sp = sf->back_chain;
-			if (!sp)
-				break;
-			if (sp <= low || sp > high - sizeof(*sf))
-				return sp;
-			sf = (struct stack_frame *) sp;
-			perf_callchain_store(entry, sf->gprs[8]);
-		}
-		/* Zero backchain detected, check for interrupt frame. */
-		sp = (unsigned long) (sf + 1);
-		if (sp <= low || sp > high - sizeof(*regs))
-			return sp;
-		regs = (struct pt_regs *) sp;
-		perf_callchain_store(entry, sf->gprs[8]);
-		low = sp;
-		sp = regs->gprs[15];
-	}
+	struct perf_callchain_entry *entry = data;
+
+	perf_callchain_store(entry, address);
+	return 0;
 }
 
 void perf_callchain_kernel(struct perf_callchain_entry *entry,
 			   struct pt_regs *regs)
 {
-	unsigned long head, frame_size;
-	struct stack_frame *head_sf;
-
 	if (user_mode(regs))
 		return;
-
-	frame_size = STACK_FRAME_OVERHEAD + sizeof(struct pt_regs);
-	head = regs->gprs[15];
-	head_sf = (struct stack_frame *) head;
-
-	if (!head_sf || !head_sf->back_chain)
-		return;
-
-	head = head_sf->back_chain;
-	head = __store_trace(entry, head,
-			     S390_lowcore.async_stack + frame_size - ASYNC_SIZE,
-			     S390_lowcore.async_stack + frame_size);
-
-	__store_trace(entry, head, S390_lowcore.thread_info,
-		      S390_lowcore.thread_info + THREAD_SIZE);
+	dump_trace(__perf_callchain_kernel, entry, NULL, regs->gprs[15]);
 }
 
 /* Perf defintions for PMU event attributes in sysfs */
diff --git a/arch/s390/kernel/stacktrace.c b/arch/s390/kernel/stacktrace.c
index 7dd8360..44f84b2 100644
--- a/arch/s390/kernel/stacktrace.c
+++ b/arch/s390/kernel/stacktrace.c
@@ -10,69 +10,31 @@
 #include <linux/kallsyms.h>
 #include <linux/module.h>
 
-static unsigned long save_context_stack(struct stack_trace *trace,
-					unsigned long sp,
-					unsigned long low,
-					unsigned long high,
-					int nosched)
+static int __save_address(void *data, unsigned long address, int nosched)
 {
-	struct stack_frame *sf;
-	struct pt_regs *regs;
-	unsigned long addr;
+	struct stack_trace *trace = data;
 
-	while(1) {
-		if (sp < low || sp > high)
-			return sp;
-		sf = (struct stack_frame *)sp;
-		while(1) {
-			addr = sf->gprs[8];
-			if (!trace->skip)
-				trace->entries[trace->nr_entries++] = addr;
-			else
-				trace->skip--;
-			if (trace->nr_entries >= trace->max_entries)
-				return sp;
-			low = sp;
-			sp = sf->back_chain;
-			if (!sp)
-				break;
-			if (sp <= low || sp > high - sizeof(*sf))
-				return sp;
-			sf = (struct stack_frame *)sp;
-		}
-		/* Zero backchain detected, check for interrupt frame. */
-		sp = (unsigned long)(sf + 1);
-		if (sp <= low || sp > high - sizeof(*regs))
-			return sp;
-		regs = (struct pt_regs *)sp;
-		addr = regs->psw.addr;
-		if (!nosched || !in_sched_functions(addr)) {
-			if (!trace->skip)
-				trace->entries[trace->nr_entries++] = addr;
-			else
-				trace->skip--;
-		}
-		if (trace->nr_entries >= trace->max_entries)
-			return sp;
-		low = sp;
-		sp = regs->gprs[15];
+	if (nosched && in_sched_functions(address))
+		return 0;
+	if (trace->skip > 0) {
+		trace->skip--;
+		return 0;
 	}
+	if (trace->nr_entries < trace->max_entries) {
+		trace->entries[trace->nr_entries++] = address;
+		return 0;
+	}
+	return 1;
 }
 
-static void __save_stack_trace(struct stack_trace *trace, unsigned long sp)
+static int save_address(void *data, unsigned long address)
 {
-	unsigned long new_sp, frame_size;
+	return __save_address(data, address, 0);
+}
 
-	frame_size = STACK_FRAME_OVERHEAD + sizeof(struct pt_regs);
-	new_sp = save_context_stack(trace, sp,
-			S390_lowcore.panic_stack + frame_size - PAGE_SIZE,
-			S390_lowcore.panic_stack + frame_size, 0);
-	new_sp = save_context_stack(trace, new_sp,
-			S390_lowcore.async_stack + frame_size - ASYNC_SIZE,
-			S390_lowcore.async_stack + frame_size, 0);
-	save_context_stack(trace, new_sp,
-			   S390_lowcore.thread_info,
-			   S390_lowcore.thread_info + THREAD_SIZE, 0);
+static int save_address_nosched(void *data, unsigned long address)
+{
+	return __save_address(data, address, 1);
 }
 
 void save_stack_trace(struct stack_trace *trace)
@@ -80,7 +42,7 @@ void save_stack_trace(struct stack_trace *trace)
 	unsigned long sp;
 
 	sp = current_stack_pointer();
-	__save_stack_trace(trace, sp);
+	dump_trace(save_address, trace, NULL, sp);
 	if (trace->nr_entries < trace->max_entries)
 		trace->entries[trace->nr_entries++] = ULONG_MAX;
 }
@@ -88,14 +50,12 @@ EXPORT_SYMBOL_GPL(save_stack_trace);
 
 void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
 {
-	unsigned long sp, low, high;
+	unsigned long sp;
 
 	sp = tsk->thread.ksp;
 	if (tsk == current)
 		sp = current_stack_pointer();
-	low = (unsigned long) task_stack_page(tsk);
-	high = (unsigned long) task_pt_regs(tsk);
-	save_context_stack(trace, sp, low, high, 1);
+	dump_trace(save_address_nosched, trace, tsk, sp);
 	if (trace->nr_entries < trace->max_entries)
 		trace->entries[trace->nr_entries++] = ULONG_MAX;
 }
@@ -106,7 +66,7 @@ void save_stack_trace_regs(struct pt_regs *regs, struct stack_trace *trace)
 	unsigned long sp;
 
 	sp = kernel_stack_pointer(regs);
-	__save_stack_trace(trace, sp);
+	dump_trace(save_address, trace, NULL, sp);
 	if (trace->nr_entries < trace->max_entries)
 		trace->entries[trace->nr_entries++] = ULONG_MAX;
 }
-- 
cgit v1.1


From bb3aa614014e11aea2c1aab528752e12b54df62d Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Fri, 19 Feb 2016 10:46:08 +0100
Subject: s390: add z13s model number to z13 elf platform

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/setup.c | 1 +
 1 file changed, 1 insertion(+)

(limited to 'arch/s390/kernel')

diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 9220db5..b1fbcc0 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -779,6 +779,7 @@ static int __init setup_hwcaps(void)
 		strcpy(elf_platform, "zEC12");
 		break;
 	case 0x2964:
+	case 0x2965:
 		strcpy(elf_platform, "z13");
 		break;
 	}
-- 
cgit v1.1


From ee8479bb97dc790904e9afad503ae833e28ed8d3 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Thu, 25 Feb 2016 10:28:49 +0100
Subject: s390/dis: use correct escape sequence for '%' character

The double escape character sequence introduced with commit
272fa59ccb4f ("s390/dis: Fix handling of format specifiers") is not
necessary anymore since commit 561e10300269 ("s390/dis: Fix printing
of the register numbers").

Instead this now generates an extra '%' character:

lg      %%r1,160(%%r11)

So fix this and basically revert 272fa59ccb4f.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/dis.c | 17 +++++------------
 1 file changed, 5 insertions(+), 12 deletions(-)

(limited to 'arch/s390/kernel')

diff --git a/arch/s390/kernel/dis.c b/arch/s390/kernel/dis.c
index 62973ef..8cb9bfd 100644
--- a/arch/s390/kernel/dis.c
+++ b/arch/s390/kernel/dis.c
@@ -1920,23 +1920,16 @@ static int print_insn(char *buffer, unsigned char *code, unsigned long addr)
 			}
 			if (separator)
 				ptr += sprintf(ptr, "%c", separator);
-			/*
-			 * Use four '%' characters below because of the
-			 * following two conversions:
-			 *
-			 *  1) sprintf: %%%%r -> %%r
-			 *  2) printk : %%r   -> %r
-			 */
 			if (operand->flags & OPERAND_GPR)
-				ptr += sprintf(ptr, "%%%%r%i", value);
+				ptr += sprintf(ptr, "%%r%i", value);
 			else if (operand->flags & OPERAND_FPR)
-				ptr += sprintf(ptr, "%%%%f%i", value);
+				ptr += sprintf(ptr, "%%f%i", value);
 			else if (operand->flags & OPERAND_AR)
-				ptr += sprintf(ptr, "%%%%a%i", value);
+				ptr += sprintf(ptr, "%%a%i", value);
 			else if (operand->flags & OPERAND_CR)
-				ptr += sprintf(ptr, "%%%%c%i", value);
+				ptr += sprintf(ptr, "%%c%i", value);
 			else if (operand->flags & OPERAND_VR)
-				ptr += sprintf(ptr, "%%%%v%i", value);
+				ptr += sprintf(ptr, "%%v%i", value);
 			else if (operand->flags & OPERAND_PCREL)
 				ptr += sprintf(ptr, "%lx", (signed int) value
 								      + addr);
-- 
cgit v1.1


From 5d7eccecf8621e3cf5adcec9cf80aa444b4610d4 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Wed, 24 Feb 2016 14:27:46 +0100
Subject: s390/fault: merge report_user_fault implementations

We have two close to identical report_user_fault functions.
Add a parameter to one and get rid of the other one in order
to reduce code duplication.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/traps.c | 21 ++-------------------
 1 file changed, 2 insertions(+), 19 deletions(-)

(limited to 'arch/s390/kernel')

diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c
index 017eb03d..dd97a3e 100644
--- a/arch/s390/kernel/traps.c
+++ b/arch/s390/kernel/traps.c
@@ -22,8 +22,6 @@
 #include <asm/fpu/api.h>
 #include "entry.h"
 
-int show_unhandled_signals = 1;
-
 static inline void __user *get_trap_ip(struct pt_regs *regs)
 {
 	unsigned long address;
@@ -35,21 +33,6 @@ static inline void __user *get_trap_ip(struct pt_regs *regs)
 	return (void __user *) (address - (regs->int_code >> 16));
 }
 
-static inline void report_user_fault(struct pt_regs *regs, int signr)
-{
-	if ((task_pid_nr(current) > 1) && !show_unhandled_signals)
-		return;
-	if (!unhandled_signal(current, signr))
-		return;
-	if (!printk_ratelimit())
-		return;
-	printk("User process fault: interruption code %04x ilc:%d ",
-	       regs->int_code & 0xffff, regs->int_code >> 17);
-	print_vma_addr("in ", regs->psw.addr);
-	printk("\n");
-	show_regs(regs);
-}
-
 int is_valid_bugaddr(unsigned long addr)
 {
 	return 1;
@@ -65,7 +48,7 @@ void do_report_trap(struct pt_regs *regs, int si_signo, int si_code, char *str)
 		info.si_code = si_code;
 		info.si_addr = get_trap_ip(regs);
 		force_sig_info(si_signo, &info, current);
-		report_user_fault(regs, si_signo);
+		report_user_fault(regs, si_signo, 0);
         } else {
                 const struct exception_table_entry *fixup;
 		fixup = search_exception_tables(regs->psw.addr);
@@ -111,7 +94,7 @@ NOKPROBE_SYMBOL(do_per_trap);
 void default_trap_handler(struct pt_regs *regs)
 {
 	if (user_mode(regs)) {
-		report_user_fault(regs, SIGSEGV);
+		report_user_fault(regs, SIGSEGV, 0);
 		do_exit(SIGSEGV);
 	} else
 		die(regs, "Unknown program exception");
-- 
cgit v1.1


From b1685ab9bd3ae14830acac8ffdc7aafc0fb416e3 Mon Sep 17 00:00:00 2001
From: Christian Borntraeger <borntraeger@de.ibm.com>
Date: Mon, 29 Feb 2016 16:05:35 +0100
Subject: s390/cpumf: Improve guest detection heuristics

commit e22cf8ca6f75 ("s390/cpumf: rework program parameter setting
to detect guest samples") requires guest changes to get proper
guest/host. We can do better: We can use the primary asn value,
which is set on all Linux variants to compare this with the host
pp value.
We now have the following cases:
1. Guest using PP
host sample:  gpp == 0, asn == hpp --> host
guest sample: gpp != 0 --> guest
2. Guest not using PP
host sample:  gpp == 0, asn == hpp --> host
guest sample: gpp == 0, asn != hpp --> guest

As soon as the host no longer sets CR4, we must back out
this heuristics - let's add a comment in switch_to.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/entry.S        | 1 +
 arch/s390/kernel/perf_cpum_sf.c | 9 ++++++---
 2 files changed, 7 insertions(+), 3 deletions(-)

(limited to 'arch/s390/kernel')

diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S
index cd5a191..4ba688c 100644
--- a/arch/s390/kernel/entry.S
+++ b/arch/s390/kernel/entry.S
@@ -186,6 +186,7 @@ ENTRY(__switch_to)
 	stg	%r5,__LC_THREAD_INFO		# store thread info of next
 	stg	%r15,__LC_KERNEL_STACK		# store end of kernel stack
 	lg	%r15,__THREAD_ksp(%r1)		# load kernel stack of next
+	/* c4 is used in guest detection: arch/s390/kernel/perf_cpum_sf.c */
 	lctl	%c4,%c4,__TASK_pid(%r3)		# load pid to control reg. 4
 	mvc	__LC_CURRENT_PID(4,%r0),__TASK_pid(%r3) # store pid of next
 	lmg	%r6,%r15,__SF_GPRS(%r15)	# load gprs of next task
diff --git a/arch/s390/kernel/perf_cpum_sf.c b/arch/s390/kernel/perf_cpum_sf.c
index 3d8da1e..1a43474 100644
--- a/arch/s390/kernel/perf_cpum_sf.c
+++ b/arch/s390/kernel/perf_cpum_sf.c
@@ -1022,10 +1022,13 @@ static int perf_push_sample(struct perf_event *event, struct sf_raw_sample *sfr)
 	/*
 	 * A non-zero guest program parameter indicates a guest
 	 * sample.
-	 * Note that some early samples might be misaccounted to
-	 * the host.
+	 * Note that some early samples or samples from guests without
+	 * lpp usage would be misaccounted to the host. We use the asn
+	 * value as a heuristic to detect most of these guest samples.
+	 * If the value differs from the host hpp value, we assume
+	 * it to be a KVM guest.
 	 */
-	if (sfr->basic.gpp)
+	if (sfr->basic.gpp || sfr->basic.prim_asn != (u16) sfr->basic.hpp)
 		sde_regs->in_guest = 1;
 
 	overflow = 0;
-- 
cgit v1.1


From baebc70a4db86515d55ff1f226088a8e7f5821a0 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Thu, 3 Mar 2016 20:49:57 -0800
Subject: s390: Use pr_warn instead of pr_warning

Convert the uses of pr_warning to pr_warn so there are fewer
uses of the old pr_warning.

Miscellanea:

o Align arguments
o Coalesce formats

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/cpcmd.c | 3 +--
 arch/s390/kernel/debug.c | 6 ++----
 arch/s390/kernel/time.c  | 6 ++----
 3 files changed, 5 insertions(+), 10 deletions(-)

(limited to 'arch/s390/kernel')

diff --git a/arch/s390/kernel/cpcmd.c b/arch/s390/kernel/cpcmd.c
index 7f76891..7f48e56 100644
--- a/arch/s390/kernel/cpcmd.c
+++ b/arch/s390/kernel/cpcmd.c
@@ -96,8 +96,7 @@ int cpcmd(const char *cmd, char *response, int rlen, int *response_code)
 			(((unsigned long)response + rlen) >> 31)) {
 		lowbuf = kmalloc(rlen, GFP_KERNEL | GFP_DMA);
 		if (!lowbuf) {
-			pr_warning("The cpcmd kernel function failed to "
-				   "allocate a response buffer\n");
+			pr_warn("The cpcmd kernel function failed to allocate a response buffer\n");
 			return -ENOMEM;
 		}
 		spin_lock_irqsave(&cpcmd_lock, flags);
diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c
index c890a55..aa12de7 100644
--- a/arch/s390/kernel/debug.c
+++ b/arch/s390/kernel/debug.c
@@ -699,8 +699,7 @@ debug_info_t *debug_register_mode(const char *name, int pages_per_area,
 	/* Since debugfs currently does not support uid/gid other than root, */
 	/* we do not allow gid/uid != 0 until we get support for that. */
 	if ((uid != 0) || (gid != 0))
-		pr_warning("Root becomes the owner of all s390dbf files "
-			   "in sysfs\n");
+		pr_warn("Root becomes the owner of all s390dbf files in sysfs\n");
 	BUG_ON(!initialized);
 	mutex_lock(&debug_mutex);
 
@@ -1307,8 +1306,7 @@ debug_input_level_fn(debug_info_t * id, struct debug_view *view,
 		new_level = debug_get_uint(str);
 	}
 	if(new_level < 0) {
-		pr_warning("%s is not a valid level for a debug "
-			   "feature\n", str);
+		pr_warn("%s is not a valid level for a debug feature\n", str);
 		rc = -EINVAL;
 	} else {
 		debug_set_level(id, new_level);
diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index 99f84ac31..c4e5f18 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -499,8 +499,7 @@ static void etr_reset(void)
 		if (etr_port0_online && etr_port1_online)
 			set_bit(CLOCK_SYNC_ETR, &clock_sync_flags);
 	} else if (etr_port0_online || etr_port1_online) {
-		pr_warning("The real or virtual hardware system does "
-			   "not provide an ETR interface\n");
+		pr_warn("The real or virtual hardware system does not provide an ETR interface\n");
 		etr_port0_online = etr_port1_online = 0;
 	}
 }
@@ -1464,8 +1463,7 @@ static void __init stp_reset(void)
 	if (rc == 0)
 		set_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags);
 	else if (stp_online) {
-		pr_warning("The real or virtual hardware system does "
-			   "not provide an STP interface\n");
+		pr_warn("The real or virtual hardware system does not provide an STP interface\n");
 		free_page((unsigned long) stp_page);
 		stp_page = NULL;
 		stp_online = 0;
-- 
cgit v1.1


From 1e133ab296f3ff8d9e58a5e758291ed39ba72ad7 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Tue, 8 Mar 2016 11:49:57 +0100
Subject: s390/mm: split arch/s390/mm/pgtable.c

The pgtable.c file is quite big, before it grows any larger split it
into pgtable.c, pgalloc.c and gmap.c. In addition move the gmap related
header definitions into the new gmap.h header and all of the pgste
helpers from pgtable.h to pgtable.c.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/asm-offsets.c | 1 +
 1 file changed, 1 insertion(+)

(limited to 'arch/s390/kernel')

diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c
index 53bbc9e..1f95cc1 100644
--- a/arch/s390/kernel/asm-offsets.c
+++ b/arch/s390/kernel/asm-offsets.c
@@ -12,6 +12,7 @@
 #include <asm/idle.h>
 #include <asm/vdso.h>
 #include <asm/pgtable.h>
+#include <asm/gmap.h>
 
 /*
  * Make sure that the compiler is new enough. We want a compiler that
-- 
cgit v1.1


From 7eb792bf7c253cb63202aae72885f5f7abdd8668 Mon Sep 17 00:00:00 2001
From: Adam Buchbinder <adam.buchbinder@gmail.com>
Date: Fri, 4 Mar 2016 11:20:04 -0800
Subject: s390: Fix misspellings in comments

Signed-off-by: Adam Buchbinder <adam.buchbinder@gmail.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/perf_cpum_cf.c | 2 +-
 arch/s390/kernel/perf_event.c   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch/s390/kernel')

diff --git a/arch/s390/kernel/perf_cpum_cf.c b/arch/s390/kernel/perf_cpum_cf.c
index 929c147..58bf457 100644
--- a/arch/s390/kernel/perf_cpum_cf.c
+++ b/arch/s390/kernel/perf_cpum_cf.c
@@ -383,7 +383,7 @@ static int __hw_perf_event_init(struct perf_event *event)
 
 	/* Validate the counter that is assigned to this event.
 	 * Because the counter facility can use numerous counters at the
-	 * same time without constraints, it is not necessary to explicity
+	 * same time without constraints, it is not necessary to explicitly
 	 * validate event groups (event->group_leader != event).
 	 */
 	err = validate_event(hwc);
diff --git a/arch/s390/kernel/perf_event.c b/arch/s390/kernel/perf_event.c
index a1b7086..c3e4099 100644
--- a/arch/s390/kernel/perf_event.c
+++ b/arch/s390/kernel/perf_event.c
@@ -238,7 +238,7 @@ void perf_callchain_kernel(struct perf_callchain_entry *entry,
 	dump_trace(__perf_callchain_kernel, entry, NULL, regs->gprs[15]);
 }
 
-/* Perf defintions for PMU event attributes in sysfs */
+/* Perf definitions for PMU event attributes in sysfs */
 ssize_t cpumf_events_sysfs_show(struct device *dev,
 				struct device_attribute *attr, char *page)
 {
-- 
cgit v1.1


From 8f100bb1ff27873dd71f636da670e503b9ade3c6 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Thu, 10 Mar 2016 10:32:21 +0100
Subject: s390/cpumf: add missing lpp magic initialization

Add the missing lpp magic initialization for cpu 0. Without this all
samples on cpu 0 do not have the most significant bit set in the
program parameter field, which we use to distinguish between guest and
host samples if the pid is also 0.

We did initialize the lpp magic in the absolute zero lowcore but
forgot that when switching to the allocated lowcore on cpu 0 only.

Reported-by: Shu Juan Zhang <zhshuj@cn.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: stable@vger.kernel.org # v4.4+
Fixes: e22cf8ca6f75 ("s390/cpumf: rework program parameter setting to detect guest samples")
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/setup.c | 1 +
 1 file changed, 1 insertion(+)

(limited to 'arch/s390/kernel')

diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index b1fbcc0..cc46767 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -327,6 +327,7 @@ static void __init setup_lowcore(void)
 		+ PAGE_SIZE - STACK_FRAME_OVERHEAD - sizeof(struct pt_regs);
 	lc->current_task = (unsigned long) init_thread_union.thread_info.task;
 	lc->thread_info = (unsigned long) &init_thread_union;
+	lc->lpp = LPP_MAGIC;
 	lc->machine_flags = S390_lowcore.machine_flags;
 	lc->stfl_fac_list = S390_lowcore.stfl_fac_list;
 	memcpy(lc->stfle_fac_list, S390_lowcore.stfle_fac_list,
-- 
cgit v1.1


From e370e4769463a65dcf8806fa26d2874e0542ac41 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Thu, 10 Mar 2016 09:52:55 +0100
Subject: s390: fix floating pointer register corruption (again)

There is a tricky interaction between the machine check handler
and the critical sections of load_fpu_regs and save_fpu_regs
functions. If the machine check interrupts one of the two
functions the critical section cleanup will complete the function
before the machine check handler s390_do_machine_check is called.
Trouble is that the machine check handler needs to validate the
floating point registers *before* and not *after* the completion
of load_fpu_regs/save_fpu_regs.

The simplest solution is to rewind the PSW to the start of the
load_fpu_regs/save_fpu_regs and retry the function after the
return from the machine check handler.

Tested-by: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: <stable@vger.kernel.org> # 4.3+
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/entry.S | 106 +----------------------------------------------
 1 file changed, 2 insertions(+), 104 deletions(-)

(limited to 'arch/s390/kernel')

diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S
index 4ba688c..2d47f9c 100644
--- a/arch/s390/kernel/entry.S
+++ b/arch/s390/kernel/entry.S
@@ -1200,114 +1200,12 @@ cleanup_critical:
 	.quad	.Lpsw_idle_lpsw
 
 .Lcleanup_save_fpu_regs:
-	TSTMSK	__LC_CPU_FLAGS,_CIF_FPU
-	bor	%r14
-	clg	%r9,BASED(.Lcleanup_save_fpu_regs_done)
-	jhe	5f
-	clg	%r9,BASED(.Lcleanup_save_fpu_regs_fp)
-	jhe	4f
-	clg	%r9,BASED(.Lcleanup_save_fpu_regs_vx_high)
-	jhe	3f
-	clg	%r9,BASED(.Lcleanup_save_fpu_regs_vx_low)
-	jhe	2f
-	clg	%r9,BASED(.Lcleanup_save_fpu_fpc_end)
-	jhe	1f
-	lg	%r2,__LC_CURRENT
-	aghi	%r2,__TASK_thread
-0:	# Store floating-point controls
-	stfpc	__THREAD_FPU_fpc(%r2)
-1:	# Load register save area and check if VX is active
-	lg	%r3,__THREAD_FPU_regs(%r2)
-	TSTMSK	__LC_MACHINE_FLAGS,MACHINE_FLAG_VX
-	jz	4f			  # no VX -> store FP regs
-2:	# Store vector registers (V0-V15)
-	VSTM	%v0,%v15,0,%r3		  # vstm 0,15,0(3)
-3:	# Store vector registers (V16-V31)
-	VSTM	%v16,%v31,256,%r3	  # vstm 16,31,256(3)
-	j	5f			  # -> done, set CIF_FPU flag
-4:	# Store floating-point registers
-	std	0,0(%r3)
-	std	1,8(%r3)
-	std	2,16(%r3)
-	std	3,24(%r3)
-	std	4,32(%r3)
-	std	5,40(%r3)
-	std	6,48(%r3)
-	std	7,56(%r3)
-	std	8,64(%r3)
-	std	9,72(%r3)
-	std	10,80(%r3)
-	std	11,88(%r3)
-	std	12,96(%r3)
-	std	13,104(%r3)
-	std	14,112(%r3)
-	std	15,120(%r3)
-5:	# Set CIF_FPU flag
-	oi	__LC_CPU_FLAGS+7,_CIF_FPU
-	lg	%r9,48(%r11)		# return from save_fpu_regs
+	larl	%r9,save_fpu_regs
 	br	%r14
-.Lcleanup_save_fpu_fpc_end:
-	.quad	.Lsave_fpu_regs_fpc_end
-.Lcleanup_save_fpu_regs_vx_low:
-	.quad	.Lsave_fpu_regs_vx_low
-.Lcleanup_save_fpu_regs_vx_high:
-	.quad	.Lsave_fpu_regs_vx_high
-.Lcleanup_save_fpu_regs_fp:
-	.quad	.Lsave_fpu_regs_fp
-.Lcleanup_save_fpu_regs_done:
-	.quad	.Lsave_fpu_regs_done
 
 .Lcleanup_load_fpu_regs:
-	TSTMSK	__LC_CPU_FLAGS,_CIF_FPU
-	bnor	%r14
-	clg	%r9,BASED(.Lcleanup_load_fpu_regs_done)
-	jhe	1f
-	clg	%r9,BASED(.Lcleanup_load_fpu_regs_fp)
-	jhe	2f
-	clg	%r9,BASED(.Lcleanup_load_fpu_regs_vx_high)
-	jhe	3f
-	clg	%r9,BASED(.Lcleanup_load_fpu_regs_vx)
-	jhe	4f
-	lg	%r4,__LC_CURRENT
-	aghi	%r4,__TASK_thread
-	lfpc	__THREAD_FPU_fpc(%r4)
-	TSTMSK	__LC_MACHINE_FLAGS,MACHINE_FLAG_VX
-	lg	%r4,__THREAD_FPU_regs(%r4)	# %r4 <- reg save area
-	jz	2f				# -> no VX, load FP regs
-4:	# Load V0 ..V15 registers
-	VLM	%v0,%v15,0,%r4
-3:	# Load V16..V31 registers
-	VLM	%v16,%v31,256,%r4
-	j	1f
-2:	# Load floating-point registers
-	ld	0,0(%r4)
-	ld	1,8(%r4)
-	ld	2,16(%r4)
-	ld	3,24(%r4)
-	ld	4,32(%r4)
-	ld	5,40(%r4)
-	ld	6,48(%r4)
-	ld	7,56(%r4)
-	ld	8,64(%r4)
-	ld	9,72(%r4)
-	ld	10,80(%r4)
-	ld	11,88(%r4)
-	ld	12,96(%r4)
-	ld	13,104(%r4)
-	ld	14,112(%r4)
-	ld	15,120(%r4)
-1:	# Clear CIF_FPU bit
-	ni	__LC_CPU_FLAGS+7,255-_CIF_FPU
-	lg	%r9,48(%r11)		# return from load_fpu_regs
+	larl	%r9,load_fpu_regs
 	br	%r14
-.Lcleanup_load_fpu_regs_vx:
-	.quad	.Lload_fpu_regs_vx
-.Lcleanup_load_fpu_regs_vx_high:
-	.quad	.Lload_fpu_regs_vx_high
-.Lcleanup_load_fpu_regs_fp:
-	.quad	.Lload_fpu_regs_fp
-.Lcleanup_load_fpu_regs_done:
-	.quad	.Lload_fpu_regs_done
 
 /*
  * Integer constants
-- 
cgit v1.1