Skip to contentSkip to conversation

bpf: Better build_id caching in stack_map_get_build_id_offset()

PATCH Applied Ihor Solodrai 3 participants 4 messages Jun 15 – Jun 22

Subsystems (2)
lore
Share & tools
lore

Ihor Solodrai

author reviewer
#0 lore raw

PATCH v3 1/1 bpf: Better build_id caching in stack_map_get_build_id_offset()

This patch is a follow up to recent implementation of stack_map_get_build_id_offset_sleepable() [1].

stack_map_get_build_id_offset() and its sleepable variant each cached only the last successfully resolved VMA, with separate bookkeeping in each function. A run of IPs in a VMA with no usable build ID will repeat the lookup for every frame: find_vma() in the non-sleepable path, a VMA lock and a blocking build_id_parse_file() in the sleepable.

Factor the per-call cache into a shared struct stack_map_build_id_cache with two independent slots [2][3], used by both functions:

* resolved - last VMA that produced a build ID (file, build_id and

                 range), reused to skip the lookup and the parse;

* unresolved - last VMA with no usable build ID (range only), reused to emit a raw IP without another lookup or parse.

Keeping the slots independent means a build-ID-less VMA no longer evicts the last resolved build ID, so a trace alternating between a binary and a region without one stops re-resolving the binary on every return.

The shared lookup tests [vm_start, vm_end), matching the sleepable path; the non-sleepable path previously reused a build ID for ip == vm_end (range_in_vma() is inclusive) and now re-resolves it correctly.

[1] https://lore.kernel.org/bpf/[email protected]/
[2] https://lore.kernel.org/bpf/CAEf4Bza2fRDGhLQoPE-EzM7F34xaEJfi5Exmxb-iWVUN3F06=g@mail.gmail.com/
[3] https://lore.kernel.org/bpf/CAEf4BzZXJFr=1iiVx937ht=4PYQkQHg=eFk810zhMDzXQG3ihw@mail.gmail.com/

---

v2->v3:
* more refactoring, no functional changes
* define and use struct stack_map_cached_vma as a cache slot type
* use compound literals where appropriate
* add res = &cache.resolved local in stack_map_get_build_id_offset_sleepable()
* minor comment fixups
v2: https://lore.kernel.org/bpf/[email protected]/

v1->v2:
* nits in attempt to de-claudify the diff (Alexei)
* factor out local variables in stack_map_build_id_set_from_cache() to make the conditions more readable
* make comments more concise
* I am ignoring the issue flagged by sashiko, since it's unrelated to the patch
v1: https://lore.kernel.org/bpf/[email protected]/

---
 kernel/bpf/stackmap.c | 183 ++++++++++++++++++++++++++++++------------
 1 file changed, 130 insertions(+), 53 deletions(-)
kernel/bpf/stackmap.c +130 -53
index 77ba03216c09..41fe87d7302f 100644
@@ -175,6 +175,95 @@ static inline void stack_map_build_id_set_valid(struct bpf_stack_build_id *id,
 		memcpy(id->build_id, build_id, BUILD_ID_SIZE_MAX);
 }
 
+/*
+ * A cached VMA lookup result. The range [vm_start, vm_end) is always set.
+ * vm_pgoff, file, build_id are set only when the build ID was resolved.
+ * Zero vm_end marks the slot empty. build_id aliases the id_offs[] entry.
+ */
+struct stack_map_cached_vma {
+	unsigned long vm_start;
+	unsigned long vm_end;
+	unsigned long vm_pgoff;
+	struct file *file; /* pinned in the sleepable path; NULL otherwise */
+	const unsigned char *build_id;
+};
+
+/*
+ * Per stack_map_get_build_id_offset() call cache of the last VMA with a build ID
+ * resolved and the last VMA with no usable build ID. Adjacent stack frames tend
+ * to land in the same VMA or the same backing file, so caching the last result
+ * of each kind lets us skip unnecessary VMA lookups and build ID parse calls.
+ * Keeping the two slots independent means a build-ID-less VMA doesn't evict the
+ * last resolved build ID.
+ */
+struct stack_map_build_id_cache {
+	struct stack_map_cached_vma resolved;
+	struct stack_map_cached_vma unresolved;
+};
+
+/*
+ * Fill @id from a cached range covering @ip. On a hit this writes @id (resolved
+ * range -> build ID + offset, unresolved range -> raw ip) and returns 0; on a
+ * miss it leaves @id untouched and returns -ENOENT.
+ */
+static int stack_map_build_id_set_from_cache(struct stack_map_build_id_cache *cache,
+					     struct bpf_stack_build_id *id, u64 ip)
+{
+	unsigned long vm_start, vm_end, vm_pgoff;
+	u64 offset;
+
+	vm_start = cache->resolved.vm_start;
+	vm_end = cache->resolved.vm_end;
+	if (vm_end && ip >= vm_start && ip < vm_end) {
+		vm_pgoff = cache->resolved.vm_pgoff;
+		offset = stack_map_build_id_offset(vm_pgoff, vm_start, ip);
+		stack_map_build_id_set_valid(id, offset, cache->resolved.build_id);
+		return 0;
+	}
+
+	vm_start = cache->unresolved.vm_start;
+	vm_end = cache->unresolved.vm_end;
+	if (vm_end && ip >= vm_start && ip < vm_end) {
+		stack_map_build_id_set_ip(id);
+		return 0;
+	}
+
+	return -ENOENT;
+}
+
+/*
+ * Record @vma's build ID as the last resolved one. @file is the pinned backing
+ * file in the sleepable path (released when evicted), or NULL otherwise.
+ */
+static void stack_map_build_id_cache_set_resolved(struct stack_map_build_id_cache *cache,
+						  struct file *file,
+						  const unsigned char *build_id,
+						  unsigned long vm_start,
+						  unsigned long vm_end,
+						  unsigned long vm_pgoff)
+{
+	if (cache->resolved.file)
+		fput(cache->resolved.file);
+	cache->resolved = (struct stack_map_cached_vma){
+		.vm_start = vm_start,
+		.vm_end = vm_end,
+		.vm_pgoff = vm_pgoff,
+		.file = file,
+		.build_id = build_id,
+	};
+}
+
+/* Record [vm_start, vm_end) as a range with no usable build ID. */
+static void stack_map_build_id_cache_set_unresolved(struct stack_map_build_id_cache *cache,
+						    unsigned long vm_start,
+						    unsigned long vm_end)
+{
+	cache->unresolved = (struct stack_map_cached_vma){
+		.vm_start = vm_start,
+		.vm_end = vm_end,
+	};
+}
+
 struct stack_map_vma_lock {
 	struct vm_area_struct *vma;
 	struct mm_struct *mm;
@@ -244,15 +333,9 @@ static void stack_map_unlock_vma(struct stack_map_vma_lock *lock)
 static void stack_map_get_build_id_offset_sleepable(struct bpf_stack_build_id *id_offs,
 						    u32 trace_nr)
 {
-	struct mm_struct *mm = current->mm;
-	struct stack_map_vma_lock lock = { .mm = mm };
-	struct {
-		struct file *file;
-		const unsigned char *build_id;
-		unsigned long vm_start;
-		unsigned long vm_end;
-		unsigned long vm_pgoff;
-	} cache = {};
+	struct stack_map_vma_lock lock = { .mm = current->mm };
+	struct stack_map_build_id_cache cache = {};
+	struct stack_map_cached_vma *res = &cache.resolved;
 	unsigned long vm_pgoff, vm_start, vm_end;
 	struct vm_area_struct *vma;
 	struct file *file;
@@ -262,44 +345,39 @@ static void stack_map_get_build_id_offset_sleepable(struct bpf_stack_build_id *i
 	for (u32 i = 0; i < trace_nr; i++) {
 		ip = READ_ONCE(id_offs[i].ip);
 
-		/*
-		 * Range cache fast path: if ip falls within the previously
-		 * resolved VMA range, reuse the cache build_id without
-		 * re-acquiring the VMA lock.
-		 */
-		if (cache.build_id && ip >= cache.vm_start && ip < cache.vm_end) {
-			offset = stack_map_build_id_offset(cache.vm_pgoff, cache.vm_start, ip);
-			stack_map_build_id_set_valid(&id_offs[i], offset, cache.build_id);
+		if (!stack_map_build_id_set_from_cache(&cache, &id_offs[i], ip))
 			continue;
-		}
 
 		vma = stack_map_lock_vma(&lock, ip);
 		if (!vma) {
 			stack_map_build_id_set_ip(&id_offs[i]);
 			continue;
 		}
+
+		vm_pgoff = vma->vm_pgoff;
+		vm_start = vma->vm_start;
+		vm_end = vma->vm_end;
+
 		if (vma_is_anonymous(vma) || !vma->vm_file) {
-			stack_map_build_id_set_ip(&id_offs[i]);
 			stack_map_unlock_vma(&lock);
+			stack_map_build_id_set_ip(&id_offs[i]);
+			stack_map_build_id_cache_set_unresolved(&cache, vm_start, vm_end);
 			continue;
 		}
 
 		file = vma->vm_file;
-		vm_pgoff = vma->vm_pgoff;
-		vm_start = vma->vm_start;
-		vm_end = vma->vm_end;
 		offset = stack_map_build_id_offset(vm_pgoff, vm_start, ip);
 
 		/*
-		 * Same backing file as previous (e.g. different VMAs
-		 * of the same ELF binary). Reuse the cache build_id.
+		 * Same backing file as the last resolved VMA (another mapping
+		 * of the same ELF binary): reuse its build_id without re-parsing.
 		 */
-		if (file == cache.file) {
+		if (file == res->file) {
 			stack_map_unlock_vma(&lock);
-			stack_map_build_id_set_valid(&id_offs[i], offset, cache.build_id);
-			cache.vm_start = vm_start;
-			cache.vm_end = vm_end;
-			cache.vm_pgoff = vm_pgoff;
+			stack_map_build_id_set_valid(&id_offs[i], offset, res->build_id);
+			res->vm_start = vm_start;
+			res->vm_end = vm_end;
+			res->vm_pgoff = vm_pgoff;
 			continue;
 		}
 
@@ -310,21 +388,17 @@ static void stack_map_get_build_id_offset_sleepable(struct bpf_stack_build_id *i
 		if (build_id_parse_file(file, id_offs[i].build_id, NULL)) {
 			stack_map_build_id_set_ip(&id_offs[i]);
 			fput(file);
+			stack_map_build_id_cache_set_unresolved(&cache, vm_start, vm_end);
 			continue;
 		}
 
 		stack_map_build_id_set_valid(&id_offs[i], offset, id_offs[i].build_id);
-		if (cache.file)
-			fput(cache.file);
-		cache.file = file;
-		cache.build_id = id_offs[i].build_id;
-		cache.vm_start = vm_start;
-		cache.vm_end = vm_end;
-		cache.vm_pgoff = vm_pgoff;
+		stack_map_build_id_cache_set_resolved(&cache, file, id_offs[i].build_id,
+						      vm_start, vm_end, vm_pgoff);
 	}
 
-	if (cache.file)
-		fput(cache.file);
+	if (res->file)
+		fput(res->file);
 }
 
 /*
@@ -343,8 +417,8 @@ static void stack_map_get_build_id_offset(struct bpf_stack_build_id *id_offs,
 	struct mmap_unlock_irq_work *work = NULL;
 	bool irq_work_busy = bpf_mmap_unlock_get_irq_work(&work);
 	bool has_user_ctx = user && current && current->mm;
-	struct vm_area_struct *vma, *prev_vma = NULL;
-	const unsigned char *prev_build_id = NULL;
+	struct stack_map_build_id_cache cache = {};
+	struct vm_area_struct *vma;
 	int i;
 
 	if (may_fault && has_user_ctx) {
@@ -365,27 +439,30 @@ static void stack_map_get_build_id_offset(struct bpf_stack_build_id *id_offs,
 
 	for (i = 0; i < trace_nr; i++) {
 		u64 ip = READ_ONCE(id_offs[i].ip);
-		u64 offset;
 
-		if (prev_build_id && range_in_vma(prev_vma, ip, ip)) {
-			vma = prev_vma;
-			offset = stack_map_build_id_offset(vma->vm_pgoff, vma->vm_start, ip);
-			stack_map_build_id_set_valid(&id_offs[i], offset, prev_build_id);
+		if (!stack_map_build_id_set_from_cache(&cache, &id_offs[i], ip))
 			continue;
-		}
+
 		vma = find_vma(current->mm, ip);
 		if (!vma || vma_is_anonymous(vma) ||
 		    fetch_build_id(vma, id_offs[i].build_id, may_fault)) {
-			/* per entry fall back to ips */
+			/* per entry fall back to ips; cache build-ID-less range */
 			stack_map_build_id_set_ip(&id_offs[i]);
-			prev_vma = vma;
-			prev_build_id = NULL;
+			if (vma)
+				stack_map_build_id_cache_set_unresolved(&cache,
+						vma->vm_start, vma->vm_end);
 			continue;
 		}
-		offset = stack_map_build_id_offset(vma->vm_pgoff, vma->vm_start, ip);
-		stack_map_build_id_set_valid(&id_offs[i], offset, id_offs[i].build_id);
-		prev_vma = vma;
-		prev_build_id = id_offs[i].build_id;
+		/*
+		 * mmap_lock is held for the whole loop, so the cached VMA
+		 * fields stay valid; no file pinning is needed here.
+		 */
+		stack_map_build_id_set_valid(&id_offs[i],
+			stack_map_build_id_offset(vma->vm_pgoff, vma->vm_start, ip),
+			id_offs[i].build_id);
+		stack_map_build_id_cache_set_resolved(&cache, NULL, id_offs[i].build_id,
+						      vma->vm_start, vma->vm_end,
+						      vma->vm_pgoff);
 	}
 	bpf_mmap_unlock_mm(work, current->mm);
 }
Signature · 1 line
2.54.0
#1 lore raw
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] The non-sleepable path uses find_vma() without verifying if the requested ip actually falls within the returned VMA, leading to bogus file offsets and leaking the build ID of adjacent mappings. --

commit 61bb9d39eaa120b04f78291638b9b3dcc5b364d9
Author: Ihor Solodrai <[email protected]>

bpf: Better build_id caching in stack_map_get_build_id_offset()

This patch refactors the build ID caching logic in stack_map_get_build_id_offset and its sleepable variant into a shared structure. It uses two independent slots to cache the last resolved and unresolved VMAs, avoiding cache eviction when alternating between regions.

Quoted 3 lines
kernel/bpf/stackmap.c +0 -0

[ ... ]

Quoted 6 lines
kernel/bpf/stackmap.c +0 -1
@@ -365,27 +439,30 @@ static void stack_map_get_build_id_offset(struct bpf_stack_build_id *id_offs,
 
 	for (i = 0; i < trace_nr; i++) {
 		u64 ip = READ_ONCE(id_offs[i].ip);
 
-		if (prev_build_id && range_in_vma(prev_vma, ip, ip)) {

kernel/bpf/stackmap.c static void stack_map_get_build_id_offset(struct bpf_stack_build_id *id_offs, · line 448

Quoted 10 lines
kernel/bpf/stackmap.c +2 -4
-			vma = prev_vma;
-			offset = stack_map_build_id_offset(vma->vm_pgoff, vma->vm_start, ip);
-			stack_map_build_id_set_valid(&id_offs[i], offset, prev_build_id);
+		if (!stack_map_build_id_set_from_cache(&cache, &id_offs[i], ip))
 			continue;
-		}
+
 		vma = find_vma(current->mm, ip);
 		if (!vma || vma_is_anonymous(vma) ||
 		    fetch_build_id(vma, id_offs[i].build_id, may_fault)) {

[Severity: High]
This isn't a bug introduced by this patch, but does the use of find_vma() here without a bounds check allow an unmapped ip to underflow the offset?

find_vma() returns the first VMA whose vm_end > ip, but it does not guarantee that ip >= vma->vm_start.

If an unmapped ip is queried, could find_vma() return the next adjacent VMA, causing the offset calculation ip - vma->vm_start to underflow? This appears like it would report a bogus file offset and leak the build ID of the adjacent mapping.

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=1
author reviewer
#2 lore raw
Quoted 47 lines · On 6/15/26 1:12 PM, [email protected] wrote:

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] The non-sleepable path uses find_vma() without verifying if the requested ip actually falls within the returned VMA, leading to bogus file offsets and leaking the build ID of adjacent mappings. --

commit 61bb9d39eaa120b04f78291638b9b3dcc5b364d9
Author: Ihor Solodrai <[email protected]>

bpf: Better build_id caching in stack_map_get_build_id_offset()

This patch refactors the build ID caching logic in stack_map_get_build_id_offset and its sleepable variant into a shared structure. It uses two independent slots to cache the last resolved and unresolved VMAs, avoiding cache eviction when alternating between regions.

Quoted 3 lines
kernel/bpf/stackmap.c +0 -0

[ ... ]

Quoted 6 lines
kernel/bpf/stackmap.c +0 -1
@@ -365,27 +439,30 @@ static void stack_map_get_build_id_offset(struct bpf_stack_build_id *id_offs,
 
 	for (i = 0; i < trace_nr; i++) {
 		u64 ip = READ_ONCE(id_offs[i].ip);
 
-		if (prev_build_id && range_in_vma(prev_vma, ip, ip)) {

kernel/bpf/stackmap.c static void stack_map_get_build_id_offset(struct bpf_stack_build_id *id_offs, · line 448

Quoted 10 lines
kernel/bpf/stackmap.c +2 -4
-			vma = prev_vma;
-			offset = stack_map_build_id_offset(vma->vm_pgoff, vma->vm_start, ip);
-			stack_map_build_id_set_valid(&id_offs[i], offset, prev_build_id);
+		if (!stack_map_build_id_set_from_cache(&cache, &id_offs[i], ip))
 			continue;
-		}
+
 		vma = find_vma(current->mm, ip);
 		if (!vma || vma_is_anonymous(vma) ||
 		    fetch_build_id(vma, id_offs[i].build_id, may_fault)) {

[Severity: High]
This isn't a bug introduced by this patch, but does the use of find_vma() here without a bounds check allow an unmapped ip to underflow the offset?

Quoted 8 lines

find_vma() returns the first VMA whose vm_end > ip, but it does not guarantee that ip >= vma->vm_start.

If an unmapped ip is queried, could find_vma() return the next adjacent VMA, causing the offset calculation ip - vma->vm_start to underflow? This appears like it would report a bogus file offset and leak the build ID of the adjacent mapping.

This was raised in v1, ignoring as it's not relevant to the patch.

#3 lore raw
In reply to Ihor Solodrai
This patch was applied to bpf/bpf.git (master) by Alexei Starovoitov <[email protected]>:

Hello:

This patch was applied to bpf/bpf.git (master)
by Alexei Starovoitov <[email protected]>:

Quoted 10 lines · On Mon, 15 Jun 2026 12:55:36 -0700 you wrote:

This patch is a follow up to recent implementation of stack_map_get_build_id_offset_sleepable() [1].

stack_map_get_build_id_offset() and its sleepable variant each cached only the last successfully resolved VMA, with separate bookkeeping in each function. A run of IPs in a VMA with no usable build ID will repeat the lookup for every frame: find_vma() in the non-sleepable path, a VMA lock and a blocking build_id_parse_file() in the sleepable.

[...]

Here is the summary with links:
- [bpf-next,v3] bpf: Better build_id caching in stack_map_get_build_id_offset()

    https://git.kernel.org/bpf/bpf/c/5e72b5b15729

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html