[PATCH v3 0/2] Add basic tracing support for m68k

[PATCH v3 0/2] Add basic tracing support for m68k

[Jean-Michel Hautbois]

In order to debug latency issues, I wanted to use ftrace on my M54418
coldfire. Sadly, it is not supported yet.
Thanks to Steven [1] it does not sound too difficult.

This small series adds basic functions to make it work, and for the few
tests I could do, it seems to be working well.

Here is a simple output I get:

```

            bash-232     [000] d..3.   947.629000: thread_noise:     bash:232 start 947.629000000 duration 0 ns
      timerlat/0-274     [000] .....   947.629000: #51598 context thread timer_latency    409280 ns
            bash-232     [000] d.h..   947.630000: #51599 context    irq timer_latency    110720 ns
            bash-232     [000] dnh1.   947.630000: irq_noise: timer:206 start 947.629000000 duration 1000000 ns
            bash-232     [000] d..3.   947.630000: thread_noise:     bash:232 start 947.630000000 duration 0 ns
      timerlat/0-274     [000] .....   947.630000: #51599 context thread timer_latency    407168 ns
            bash-232     [000] d.h..   947.631000: #51600 context    irq timer_latency    108608 ns
            bash-232     [000] dnh1.   947.631000: irq_noise: timer:206 start 947.630000000 duration 1000000 ns
            bash-232     [000] d..3.   947.631000: thread_noise:     bash:232 start 947.631000000 duration 0 ns
      timerlat/0-274     [000] .....   947.631000: #51600 context thread timer_latency    401472 ns
```

I used this on my board and afaik, it works quite well.

```

 telnetd-229       0Dnh4.   13us+:      229:120:R   + [000]     118: 49:R irq/100-enet-fe
 telnetd-229       0Dnh4.   53us+: <stack trace>
 => __trace_stack
 => probe_wakeup
 => try_to_wake_up
 => wake_up_process
 => __handle_irq_event_percpu
 => handle_irq_event
 => handle_level_irq
 => generic_handle_irq
 => do_IRQ
 => inthandler
 => ip_finish_output2
 => __ip_finish_output
 => ip_output
 => __ip_queue_xmit
 => ip_queue_xmit
 => __tcp_transmit_skb
 => tcp_write_xmit
 => __tcp_push_pending_frames
 => tcp_push
 => tcp_sendmsg_locked
 => tcp_sendmsg
 => inet_sendmsg
 => sock_write_iter
 => vfs_write
 => ksys_write
 => sys_write
 => system_call
 telnetd-229       0Dnh4.   68us!: wake_up_process <-__handle_irq_event_percpu
 telnetd-229       0D..3.  348us+: __schedule <-preempt_schedule_irq
 telnetd-229       0D..3.  368us+:      229:120:R ==> [000]     118: 49:R irq/100-enet-fe
 telnetd-229       0D..3.  395us : <stack trace>
 => __trace_stack
 => probe_wakeup_sched_switch
 => __schedule
 => preempt_schedule_irq
 => Lkernel_return
 => ip_finish_output2
 => __ip_finish_output
 => ip_output
 => __ip_queue_xmit
 => ip_queue_xmit
 => __tcp_transmit_skb
 => tcp_write_xmit
 => __tcp_push_pending_frames
 => tcp_push
 => tcp_sendmsg_locked
 => tcp_sendmsg
 => inet_sendmsg
 => sock_write_iter
 => vfs_write
 => ksys_write
 => sys_write
 => system_call
```

[1]: https://lore.kernel.org/linux-m68k/20241018124511.70d29198@gandalf.local.home

Signed-off-by: Jean-Michel Hautbois <jeanmichel.hautbois@yoseli.org>
---
Changes in v3:
- Do not unconditionaly select the frame pointers
- Link to v2: https://lore.kernel.org/r/20241203-add-m68k-tracing-support-v2-0-77302097a9f4@yoseli.org

Changes in v2:
- Use the frame pointer and declare it
- Comments on trace_irq have been tested, but adding the trace_on/off in
  entry.S fails badly
- Link to v1: https://lore.kernel.org/r/20241021-add-m68k-tracing-support-v1-0-0883d704525b@yoseli.org

---
Jean-Michel Hautbois (2):
      m68k: Add tracirqs
      arch: m68k: Add STACKTRACE support

 arch/m68k/Kconfig             |  6 ++++
 arch/m68k/kernel/Makefile     |  1 +
 arch/m68k/kernel/irq.c        |  2 ++
 arch/m68k/kernel/stacktrace.c | 70 +++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 79 insertions(+)
---
base-commit: e70140ba0d2b1a30467d4af6bcfe761327b9ec95
change-id: 20241021-add-m68k-tracing-support-36c18d2233d8

Best regards,
-- 
Jean-Michel Hautbois <jeanmichel.hautbois@yoseli.org>

[PATCH v3 1/2] m68k: Add tracirqs

[Jean-Michel Hautbois]

The TRACE_IRQFLAGS_SUPPORT requires the architecture to call
trace_hardirqs_off() when interrupts are disabled and
trace_hardirqs_on() when they are enabled.
Add those calls to do_IRQ function.

Signed-off-by: Jean-Michel Hautbois <jeanmichel.hautbois@yoseli.org>
---
 arch/m68k/Kconfig      | 1 +
 arch/m68k/kernel/irq.c | 2 ++
 2 files changed, 3 insertions(+)

diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig
index 7c4f7bcc89d76326b9d7280a3bcba79412ead7d2..793ab1e2762609725bbf793f6dffecfa3ecfff0f 100644
--- a/arch/m68k/Kconfig
+++ b/arch/m68k/Kconfig
@@ -39,6 +39,7 @@ config M68K
 	select OLD_SIGSUSPEND3
 	select UACCESS_MEMCPY if !MMU
 	select ZONE_DMA
+	select TRACE_IRQFLAGS_SUPPORT
 
 config CPU_BIG_ENDIAN
 	def_bool y
diff --git a/arch/m68k/kernel/irq.c b/arch/m68k/kernel/irq.c
index 9ab4f550342e5de11c528f55781432675ffd66bf..74cf60ebbc4bca51f3caa4046dbd2bdb02355711 100644
--- a/arch/m68k/kernel/irq.c
+++ b/arch/m68k/kernel/irq.c
@@ -21,9 +21,11 @@ asmlinkage void do_IRQ(int irq, struct pt_regs *regs)
 {
 	struct pt_regs *oldregs = set_irq_regs(regs);
 
+	trace_hardirqs_off();
 	irq_enter();
 	generic_handle_irq(irq);
 	irq_exit();
+	trace_hardirqs_on();
 
 	set_irq_regs(oldregs);
 }

-- 
2.39.5

[PATCH v3 2/2] arch: m68k: Add STACKTRACE support

[Jean-Michel Hautbois]

In order to use tracing, implement a basic arch_stack_walk() based on
the one in PowerPC.
Tested on a M54418 coldfire.

Signed-off-by: Jean-Michel Hautbois <jeanmichel.hautbois@yoseli.org>
---
 arch/m68k/Kconfig             |  5 ++++
 arch/m68k/kernel/Makefile     |  1 +
 arch/m68k/kernel/stacktrace.c | 70 +++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 76 insertions(+)

diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig
index 793ab1e2762609725bbf793f6dffecfa3ecfff0f..3ad8596aab71190807f8c11dd5876aa1198760f3 100644
--- a/arch/m68k/Kconfig
+++ b/arch/m68k/Kconfig
@@ -106,6 +106,11 @@ config BOOTINFO_PROC
 	  Say Y to export the bootinfo used to boot the kernel in a
 	  "bootinfo" file in procfs.  This is useful with kexec.
 
+config STACKTRACE_SUPPORT
+	def_bool MMU_COLDFIRE
+	select ARCH_STACKWALK
+	select ARCH_WANT_FRAME_POINTERS
+
 menu "Platform setup"
 
 source "arch/m68k/Kconfig.cpu"
diff --git a/arch/m68k/kernel/Makefile b/arch/m68k/kernel/Makefile
index 6c732ed3998b714a4842ee29c977550a61979779..cb02bcfe04c6b265fa97db9237395a262e649989 100644
--- a/arch/m68k/kernel/Makefile
+++ b/arch/m68k/kernel/Makefile
@@ -23,3 +23,4 @@ obj-$(CONFIG_UBOOT)		+= uboot.o
 
 obj-$(CONFIG_EARLY_PRINTK)	+= early_printk.o
 
+obj-y	+= stacktrace.o
diff --git a/arch/m68k/kernel/stacktrace.c b/arch/m68k/kernel/stacktrace.c
new file mode 100644
index 0000000000000000000000000000000000000000..4c2fb6b0cf675ee5a3a21393a50603fea98a1b03
--- /dev/null
+++ b/arch/m68k/kernel/stacktrace.c
@@ -0,0 +1,70 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/*
+ * Stack trace utility functions etc.
+ *
+ * Copyright 2024 Jean-Michel Hautbois, Yoseli SAS.
+ */
+
+#include <asm/current.h>
+#include <asm/ptrace.h>
+#include <linux/sched.h>
+#include <linux/sched/task_stack.h>
+#include <linux/stacktrace.h>
+
+static inline unsigned long current_stack_frame(void)
+{
+	unsigned long sp;
+
+	asm volatile("movl %%fp, %0" : "=r"(sp));
+	return sp;
+}
+
+static inline int validate_sp(unsigned long sp, struct task_struct *task)
+{
+	unsigned long stack_start, stack_end;
+
+	if (task == current)
+		stack_start = (unsigned long)task_stack_page(task);
+	else
+		stack_start = (unsigned long)task->thread.esp0;
+
+	stack_end = stack_start + THREAD_SIZE;
+
+	if (sp < stack_start || sp >= stack_end)
+		return 0;
+
+	return 1;
+}
+
+void __no_sanitize_address arch_stack_walk(stack_trace_consume_fn consume_entry, void *cookie,
+					   struct task_struct *task, struct pt_regs *regs)
+{
+	unsigned long sp;
+
+	if (regs && !consume_entry(cookie, regs->pc))
+		return;
+
+	if (regs)
+		sp = (unsigned long) regs;
+	else if (task == current)
+		sp = current_stack_frame();
+	else
+		sp = task->thread.ksp;
+
+	for (;;) {
+		unsigned long *stack = (unsigned long *) sp;
+		unsigned long newsp, ip;
+
+		if (!validate_sp(sp, task))
+			return;
+
+		newsp = stack[0];
+		ip = stack[1];
+
+		if (!consume_entry(cookie, ip))
+			return;
+
+		sp = newsp;
+	}
+}

-- 
2.39.5

Re: [PATCH v3 2/2] arch: m68k: Add STACKTRACE support

[kernel test robot]

Hi Jean-Michel,

kernel test robot noticed the following build errors:

[auto build test ERROR on e70140ba0d2b1a30467d4af6bcfe761327b9ec95]

url:    https://github.com/intel-lab-lkp/linux/commits/Jean-Michel-Hautbois/m68k-Add-tracirqs/20241204-121927
base:   e70140ba0d2b1a30467d4af6bcfe761327b9ec95
patch link:    https://lore.kernel.org/r/20241203-add-m68k-tracing-support-v3-2-4de93f6cae55%40yoseli.org
patch subject: [PATCH v3 2/2] arch: m68k: Add STACKTRACE support
config: m68k-allnoconfig (https://download.01.org/0day-ci/archive/20241204/202412041842.wqwewnMc-lkp@intel.com/config)
compiler: m68k-linux-gcc (GCC) 14.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20241204/202412041842.wqwewnMc-lkp@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202412041842.wqwewnMc-lkp@intel.com/

All errors (new ones prefixed by >>):

>> arch/m68k/kernel/stacktrace.c:40:44: error: unknown type name 'stack_trace_consume_fn'
      40 | void __no_sanitize_address arch_stack_walk(stack_trace_consume_fn consume_entry, void *cookie,
         |                                            ^~~~~~~~~~~~~~~~~~~~~~


vim +/stack_trace_consume_fn +40 arch/m68k/kernel/stacktrace.c

    39	
  > 40	void __no_sanitize_address arch_stack_walk(stack_trace_consume_fn consume_entry, void *cookie,

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki