[PATCH 0/2] ARM: 9259/1: stacktrace: Add USER_STACKTRACE support

[PATCH 0/2] ARM: 9259/1: stacktrace: Add USER_STACKTRACE support

[Jinjie Ruan]

Add USER_STACKTRACE support and Fix callchain_trace() return value bug
by the way.

Jinjie Ruan (2):
  ARM: 9258/1: Fix callchain_trace() return value
  ARM: 9259/1: stacktrace: Add USER_STACKTRACE support

 arch/arm/Kconfig                 |  1 +
 arch/arm/kernel/perf_callchain.c | 73 +++-----------------------------
 arch/arm/kernel/stacktrace.c     | 65 ++++++++++++++++++++++++++++
 3 files changed, 73 insertions(+), 66 deletions(-)

-- 
2.34.1

[PATCH 1/2] ARM: 9258/1: Fix callchain_trace() return value

[Jinjie Ruan]

perf_callchain_store() return 0 on success, -1 otherwise, fix
callchain_trace() to return correct bool value. So walk_stackframe() can
have a chance to stop walking the stack ahead.

Fixes: 70ccc7c0667b ("ARM: 9258/1: stacktrace: Make stack walk callback consistent with generic code")
Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com>
---
 arch/arm/kernel/perf_callchain.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/arch/arm/kernel/perf_callchain.c b/arch/arm/kernel/perf_callchain.c
index 7147edbe56c6..1d230ac9d0eb 100644
--- a/arch/arm/kernel/perf_callchain.c
+++ b/arch/arm/kernel/perf_callchain.c
@@ -85,8 +85,7 @@ static bool
 callchain_trace(void *data, unsigned long pc)
 {
 	struct perf_callchain_entry_ctx *entry = data;
-	perf_callchain_store(entry, pc);
-	return true;
+	return perf_callchain_store(entry, pc) == 0;
 }
 
 void
-- 
2.34.1

[PATCH 2/2] ARM: 9259/1: stacktrace: Add USER_STACKTRACE support

[Jinjie Ruan]

Currently, userstacktrace is unsupported for ARM. So use the
perf_callchain_user() code as blueprint to implement the
arch_stack_walk_user() which add userstacktrace support on ARM.
Meanwhile, we can use arch_stack_walk_user() to simplify the implementation
of perf_callchain_user().

A ftrace test case is shown as below:
	# cd /sys/kernel/debug/tracing
	# echo 1 > options/userstacktrace
	# echo 1 > options/sym-userobj
	# echo 1 > events/sched/sched_process_fork/enable
	# cat trace

	......
	              sh-100     [000] .....    51.779261: sched_process_fork: comm=sh pid=100 child_comm=sh child_pid=108
	              sh-100     [000] .....    51.779285: <user stack trace>
	 => /lib/libc.so.6[+0xb3c8c]
	 => /bin/busybox[+0xffb901f1]

Also a simple perf test is ok as below:
	# perf record -e cpu-clock --call-graph fp top
	# perf report --call-graph

	.....
	  [[31m  65.00%[[m     0.00%  top      [kernel.kallsyms]  [k] __ret_fast_syscall

	            |
	            ---__ret_fast_syscall
	               |
	               |--[[31m30.00%[[m--__se_sys_getdents64
	               |          iterate_dir
	               |          |
	               |          |--[[31m25.00%[[m--proc_pid_readdir

Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com>
Tested-by: Jinjie Ruan <ruanjinjie@huawei.com>
---
 arch/arm/Kconfig                 |  1 +
 arch/arm/kernel/perf_callchain.c | 70 +++-----------------------------
 arch/arm/kernel/stacktrace.c     | 65 +++++++++++++++++++++++++++++
 3 files changed, 72 insertions(+), 64 deletions(-)

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index ee5115252aac..9f09a16338e3 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -149,6 +149,7 @@ config ARM
 	select HAVE_ARCH_VMAP_STACK if MMU && ARM_HAS_GROUP_RELOCS
 	select TRACE_IRQFLAGS_SUPPORT if !CPU_V7M
 	select USE_OF if !(ARCH_FOOTBRIDGE || ARCH_RPC || ARCH_SA1100)
+	select USER_STACKTRACE_SUPPORT
 	# Above selects are sorted alphabetically; please add new ones
 	# according to that.  Thanks.
 	help
diff --git a/arch/arm/kernel/perf_callchain.c b/arch/arm/kernel/perf_callchain.c
index 1d230ac9d0eb..cdb7aa31c6ec 100644
--- a/arch/arm/kernel/perf_callchain.c
+++ b/arch/arm/kernel/perf_callchain.c
@@ -12,70 +12,6 @@
 
 #include <asm/stacktrace.h>
 
-/*
- * The registers we're interested in are at the end of the variable
- * length saved register structure. The fp points at the end of this
- * structure so the address of this struct is:
- * (struct frame_tail *)(xxx->fp)-1
- *
- * This code has been adapted from the ARM OProfile support.
- */
-struct frame_tail {
-	struct frame_tail __user *fp;
-	unsigned long sp;
-	unsigned long lr;
-} __attribute__((packed));
-
-/*
- * Get the return address for a single stackframe and return a pointer to the
- * next frame tail.
- */
-static struct frame_tail __user *
-user_backtrace(struct frame_tail __user *tail,
-	       struct perf_callchain_entry_ctx *entry)
-{
-	struct frame_tail buftail;
-	unsigned long err;
-
-	if (!access_ok(tail, sizeof(buftail)))
-		return NULL;
-
-	pagefault_disable();
-	err = __copy_from_user_inatomic(&buftail, tail, sizeof(buftail));
-	pagefault_enable();
-
-	if (err)
-		return NULL;
-
-	perf_callchain_store(entry, buftail.lr);
-
-	/*
-	 * Frame pointers should strictly progress back up the stack
-	 * (towards higher addresses).
-	 */
-	if (tail + 1 >= buftail.fp)
-		return NULL;
-
-	return buftail.fp - 1;
-}
-
-void
-perf_callchain_user(struct perf_callchain_entry_ctx *entry, struct pt_regs *regs)
-{
-	struct frame_tail __user *tail;
-
-	perf_callchain_store(entry, regs->ARM_pc);
-
-	if (!current->mm)
-		return;
-
-	tail = (struct frame_tail __user *)regs->ARM_fp - 1;
-
-	while ((entry->nr < entry->max_stack) &&
-	       tail && !((unsigned long)tail & 0x3))
-		tail = user_backtrace(tail, entry);
-}
-
 /*
  * Gets called by walk_stackframe() for every stackframe. This will be called
  * whist unwinding the stackframe and is like a subroutine return so we use
@@ -88,6 +24,12 @@ callchain_trace(void *data, unsigned long pc)
 	return perf_callchain_store(entry, pc) == 0;
 }
 
+void
+perf_callchain_user(struct perf_callchain_entry_ctx *entry, struct pt_regs *regs)
+{
+	arch_stack_walk_user(callchain_trace, entry, regs);
+}
+
 void
 perf_callchain_kernel(struct perf_callchain_entry_ctx *entry, struct pt_regs *regs)
 {
diff --git a/arch/arm/kernel/stacktrace.c b/arch/arm/kernel/stacktrace.c
index 620aa82e3bdd..b744792755b5 100644
--- a/arch/arm/kernel/stacktrace.c
+++ b/arch/arm/kernel/stacktrace.c
@@ -194,4 +194,69 @@ void arch_stack_walk(stack_trace_consume_fn consume_entry, void *cookie,
 
 	walk_stackframe(&frame, consume_entry, cookie);
 }
+
+/*
+ * The registers we're interested in are at the end of the variable
+ * length saved register structure. The fp points at the end of this
+ * structure so the address of this struct is:
+ * (struct frame_tail *)(xxx->fp)-1
+ *
+ * This code has been adapted from the ARM OProfile support.
+ */
+struct frame_tail {
+	struct frame_tail __user *fp;
+	unsigned long sp;
+	unsigned long lr;
+} __packed;
+
+/*
+ * Get the return address for a single stackframe and return a pointer to the
+ * next frame tail.
+ */
+static struct frame_tail __user *
+unwind_user_frame(struct frame_tail __user *tail, void *cookie,
+		  stack_trace_consume_fn consume_entry)
+{
+	struct frame_tail buftail;
+	unsigned long err;
+
+	if (!access_ok(tail, sizeof(buftail)))
+		return NULL;
+
+	pagefault_disable();
+	err = __copy_from_user_inatomic(&buftail, tail, sizeof(buftail));
+	pagefault_enable();
+
+	if (err)
+		return NULL;
+
+	if (!consume_entry(cookie, buftail.lr))
+		return NULL;
+
+	/*
+	 * Frame pointers should strictly progress back up the stack
+	 * (towards higher addresses).
+	 */
+	if (tail + 1 >= buftail.fp)
+		return NULL;
+
+	return buftail.fp - 1;
+}
+
+void arch_stack_walk_user(stack_trace_consume_fn consume_entry, void *cookie,
+			  const struct pt_regs *regs)
+{
+	struct frame_tail __user *tail;
+
+	if (!consume_entry(cookie, regs->ARM_pc))
+		return;
+
+	if (!current->mm)
+		return;
+
+	tail = (struct frame_tail __user *)regs->ARM_fp - 1;
+
+	while (tail && !((unsigned long)tail & 0x3))
+		tail = unwind_user_frame(tail, cookie, consume_entry);
+}
 #endif
-- 
2.34.1

Re: [PATCH 1/2] ARM: 9258/1: Fix callchain_trace() return value

[Linus Walleij]

On Tue, Jun 18, 2024 at 5:23 AM Jinjie Ruan <ruanjinjie@huawei.com> wrote:

> perf_callchain_store() return 0 on success, -1 otherwise, fix
> callchain_trace() to return correct bool value. So walk_stackframe() can
> have a chance to stop walking the stack ahead.
>
> Fixes: 70ccc7c0667b ("ARM: 9258/1: stacktrace: Make stack walk callback consistent with generic code")
> Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com>

Reviewed-by: Linus Walleij <linus.walleij@linaro.org>

Please put this into Russell's patch tracker.

Yours,
Linus Walleij

Re: [PATCH 1/2] ARM: 9258/1: Fix callchain_trace() return value

[Jinjie Ruan]

On 2024/6/26 17:50, Linus Walleij wrote:
> On Tue, Jun 18, 2024 at 5:23 AM Jinjie Ruan <ruanjinjie@huawei.com> wrote:
> 
>> perf_callchain_store() return 0 on success, -1 otherwise, fix
>> callchain_trace() to return correct bool value. So walk_stackframe() can
>> have a chance to stop walking the stack ahead.
>>
>> Fixes: 70ccc7c0667b ("ARM: 9258/1: stacktrace: Make stack walk callback consistent with generic code")
>> Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com>
> 
> Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
> 
> Please put this into Russell's patch tracker.

Thank you, I'll put it, and could you please help to review another patch?

> 
> Yours,
> Linus Walleij