From: Josh Poimboeuf <jpoimboe@kernel.org>

Cache the results of the unwind to ensure the unwind is only performed
once, even when called by multiple tracers.

The cache nr_entries gets cleared every time the task enters the kernel.
When a stacktrace is requested, nr_entries gets set to the number of
entries in the stacktrace. If another stacktrace is requested, if
nr_entries is not zero, then it contains the same stacktrace that would be
retrieved so it is not processed again and the entries is given to the
caller.

Co-developed-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Josh Poimboeuf <jpoimboe@kernel.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
---
Changes since v4: https://lore.kernel.org/all/51855c0902486060cd6e1ccc6b22fd092a2e676d.1737511963.git.jpoimboe@kernel.org/

- Instead of using a cookie to determine if the cached stacktrace is the
  same, use nr_entries. By clearing it every time the task enters the
  kernel, if it's not zero, then that means the stacktrace is still in the
  cache. (Peter Zijlstra)

 include/linux/entry-common.h          |  2 ++
 include/linux/unwind_deferred.h       |  7 +++++++
 include/linux/unwind_deferred_types.h |  7 ++++++-
 kernel/unwind/deferred.c              | 27 ++++++++++++++++++++-------
 4 files changed, 35 insertions(+), 8 deletions(-)

diff --git a/include/linux/entry-common.h b/include/linux/entry-common.h
index fc61d0205c97..fb2b27154fee 100644
--- a/include/linux/entry-common.h
+++ b/include/linux/entry-common.h
@@ -12,6 +12,7 @@
 #include <linux/resume_user_mode.h>
 #include <linux/tick.h>
 #include <linux/kmsan.h>
+#include <linux/unwind_deferred.h>
 
 #include <asm/entry-common.h>
 
@@ -111,6 +112,7 @@ static __always_inline void enter_from_user_mode(struct pt_regs *regs)
 
 	CT_WARN_ON(__ct_state() != CT_STATE_USER);
 	user_exit_irqoff();
+	unwind_enter_from_user_mode();
 
 	instrumentation_begin();
 	kmsan_unpoison_entry_regs(regs);
diff --git a/include/linux/unwind_deferred.h b/include/linux/unwind_deferred.h
index 5064ebe38c4f..54f1aa6caf29 100644
--- a/include/linux/unwind_deferred.h
+++ b/include/linux/unwind_deferred.h
@@ -12,6 +12,11 @@ void unwind_task_free(struct task_struct *task);
 
 int unwind_deferred_trace(struct unwind_stacktrace *trace);
 
+static __always_inline void unwind_enter_from_user_mode(void)
+{
+	current->unwind_info.cache.nr_entries = 0;
+}
+
 #else /* !CONFIG_UNWIND_USER */
 
 static inline void unwind_task_init(struct task_struct *task) {}
@@ -19,6 +24,8 @@ static inline void unwind_task_free(struct task_struct *task) {}
 
 static inline int unwind_deferred_trace(struct unwind_stacktrace *trace) { return -ENOSYS; }
 
+static inline void unwind_enter_from_user_mode(void) {}
+
 #endif /* !CONFIG_UNWIND_USER */
 
 #endif /* _LINUX_UNWIND_USER_DEFERRED_H */
diff --git a/include/linux/unwind_deferred_types.h b/include/linux/unwind_deferred_types.h
index aa32db574e43..b3b7389ee6eb 100644
--- a/include/linux/unwind_deferred_types.h
+++ b/include/linux/unwind_deferred_types.h
@@ -2,8 +2,13 @@
 #ifndef _LINUX_UNWIND_USER_DEFERRED_TYPES_H
 #define _LINUX_UNWIND_USER_DEFERRED_TYPES_H
 
-struct unwind_task_info {
+struct unwind_cache {
 	unsigned long		*entries;
+	unsigned int		nr_entries;
+};
+
+struct unwind_task_info {
+	struct unwind_cache	cache;
 };
 
 #endif /* _LINUX_UNWIND_USER_DEFERRED_TYPES_H */
diff --git a/kernel/unwind/deferred.c b/kernel/unwind/deferred.c
index 6a4cec4877c5..99d4d9e049cd 100644
--- a/kernel/unwind/deferred.c
+++ b/kernel/unwind/deferred.c
@@ -12,6 +12,7 @@
 int unwind_deferred_trace(struct unwind_stacktrace *trace)
 {
 	struct unwind_task_info *info = &current->unwind_info;
+	struct unwind_cache *cache = &info->cache;
 
 	/* Should always be called from faultable context */
 	might_fault();
@@ -20,17 +21,29 @@ int unwind_deferred_trace(struct unwind_stacktrace *trace)
 	if (!current->mm)
 		return -EINVAL;
 
-       if (!info->entries) {
-               info->entries = kmalloc_array(UNWIND_MAX_ENTRIES, sizeof(long),
-                                             GFP_KERNEL);
-               if (!info->entries)
-		       return -ENOMEM;
+	if (!cache->entries) {
+		cache->entries = kmalloc_array(UNWIND_MAX_ENTRIES, sizeof(long),
+					       GFP_KERNEL);
+		if (!cache->entries)
+			return -ENOMEM;
+        }
+
+	trace->entries = cache->entries;
+
+	if (cache->nr_entries) {
+               /*
+                * The user stack has already been previously unwound in this
+                * entry context.  Skip the unwind and use the cache.
+                */
+               trace->nr = cache->nr_entries;
+               return 0;
        }
 
 	trace->nr = 0;
-	trace->entries = info->entries;
 	unwind_user(trace, UNWIND_MAX_ENTRIES);
 
+	cache->nr_entries = trace->nr;
+
 	return 0;
 }
 
@@ -45,5 +58,5 @@ void unwind_task_free(struct task_struct *task)
 {
 	struct unwind_task_info *info = &task->unwind_info;
 
-	kfree(info->entries);
+	kfree(info->cache.entries);
 }
-- 
2.47.2

On Thu, 24 Apr 2025 12:25:37 -0400
Steven Rostedt <rostedt@goodmis.org> wrote:

> --- a/include/linux/unwind_deferred.h
> +++ b/include/linux/unwind_deferred.h
> @@ -12,6 +12,11 @@ void unwind_task_free(struct task_struct *task);
>  
>  int unwind_deferred_trace(struct unwind_stacktrace *trace);
>  
> +static __always_inline void unwind_enter_from_user_mode(void)
> +{
> +	current->unwind_info.cache.nr_entries = 0;
> +}
> +

Hmm, the first patches just had unwind_enter_from_user_mode(), but the
later patches (that I'm currently working on) have unwind_exit_to_user_mode().

I think the exit is better than the entry, and I will change this to just
use the exit_to_user_mode and removed the enter_from_user_mode.

-- Steve