From: Andrii Nakryiko <andrii@kernel.org>
To: akpm@linux-foundation.org, linux-mm@vger.kernel.org
Cc: linux-fsdevel@vger.kernel.org, bpf@vger.kernel.org,
surenb@google.com, Andrii Nakryiko <andrii@kernel.org>,
syzbot+4e70c8e0a2017b432f7a@syzkaller.appspotmail.com
Subject: [PATCH mm-stable] procfs: avoid fetching build ID while holding VMA lock
Date: Wed, 28 Jan 2026 10:32:32 -0800 [thread overview]
Message-ID: <20260128183232.2854138-1-andrii@kernel.org> (raw)
Fix PROCMAP_QUERY to fetch optional build ID only after dropping mmap_lock or
per-VMA lock, whichever was used to lock VMA under question, to avoid deadlock
reported by syzbot:
-> #1 (&mm->mmap_lock){++++}-{4:4}:
__might_fault+0xed/0x170
_copy_to_iter+0x118/0x1720
copy_page_to_iter+0x12d/0x1e0
filemap_read+0x720/0x10a0
blkdev_read_iter+0x2b5/0x4e0
vfs_read+0x7f4/0xae0
ksys_read+0x12a/0x250
do_syscall_64+0xcb/0xf80
entry_SYSCALL_64_after_hwframe+0x77/0x7f
-> #0 (&sb->s_type->i_mutex_key#8){++++}-{4:4}:
__lock_acquire+0x1509/0x26d0
lock_acquire+0x185/0x340
down_read+0x98/0x490
blkdev_read_iter+0x2a7/0x4e0
__kernel_read+0x39a/0xa90
freader_fetch+0x1d5/0xa80
__build_id_parse.isra.0+0xea/0x6a0
do_procmap_query+0xd75/0x1050
procfs_procmap_ioctl+0x7a/0xb0
__x64_sys_ioctl+0x18e/0x210
do_syscall_64+0xcb/0xf80
entry_SYSCALL_64_after_hwframe+0x77/0x7f
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0 CPU1
---- ----
rlock(&mm->mmap_lock);
lock(&sb->s_type->i_mutex_key#8);
lock(&mm->mmap_lock);
rlock(&sb->s_type->i_mutex_key#8);
*** DEADLOCK ***
To make this safe, we need to grab file refcount while VMA is still locked, but
other than that everything is pretty straightforward. Internal build_id_parse()
API assumes VMA is passed, but it only needs the underlying file reference, so
just add another variant build_id_parse_file() that expects file passed
directly.
Fixes: ed5d583a88a9 ("fs/procfs: implement efficient VMA querying API for /proc/<pid>/maps")
Reported-by: syzbot+4e70c8e0a2017b432f7a@syzkaller.appspotmail.com
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
---
fs/proc/task_mmu.c | 42 ++++++++++++++++++++++++++---------------
include/linux/buildid.h | 3 +++
lib/buildid.c | 34 +++++++++++++++++++++++++--------
3 files changed, 56 insertions(+), 23 deletions(-)
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index 480db575553e..dd3b5cf9f0b7 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -656,6 +656,7 @@ static int do_procmap_query(struct mm_struct *mm, void __user *uarg)
struct proc_maps_locking_ctx lock_ctx = { .mm = mm };
struct procmap_query karg;
struct vm_area_struct *vma;
+ struct file *vm_file = NULL;
const char *name = NULL;
char build_id_buf[BUILD_ID_SIZE_MAX], *name_buf = NULL;
__u64 usize;
@@ -727,21 +728,6 @@ static int do_procmap_query(struct mm_struct *mm, void __user *uarg)
karg.inode = 0;
}
- if (karg.build_id_size) {
- __u32 build_id_sz;
-
- err = build_id_parse(vma, build_id_buf, &build_id_sz);
- if (err) {
- karg.build_id_size = 0;
- } else {
- if (karg.build_id_size < build_id_sz) {
- err = -ENAMETOOLONG;
- goto out;
- }
- karg.build_id_size = build_id_sz;
- }
- }
-
if (karg.vma_name_size) {
size_t name_buf_sz = min_t(size_t, PATH_MAX, karg.vma_name_size);
const struct path *path;
@@ -775,10 +761,34 @@ static int do_procmap_query(struct mm_struct *mm, void __user *uarg)
karg.vma_name_size = name_sz;
}
+ if (karg.build_id_size && vma->vm_file)
+ vm_file = get_file(vma->vm_file);
+
/* unlock vma or mmap_lock, and put mm_struct before copying data to user */
query_vma_teardown(&lock_ctx);
mmput(mm);
+ if (karg.build_id_size) {
+ __u32 build_id_sz;
+
+ if (vm_file)
+ err = build_id_parse_file(vm_file, build_id_buf, &build_id_sz);
+ else
+ err = -ENOENT;
+ if (err) {
+ karg.build_id_size = 0;
+ } else {
+ if (karg.build_id_size < build_id_sz) {
+ err = -ENAMETOOLONG;
+ goto out;
+ }
+ karg.build_id_size = build_id_sz;
+ }
+ }
+
+ if (vm_file)
+ fput(vm_file);
+
if (karg.vma_name_size && copy_to_user(u64_to_user_ptr(karg.vma_name_addr),
name, karg.vma_name_size)) {
kfree(name_buf);
@@ -798,6 +808,8 @@ static int do_procmap_query(struct mm_struct *mm, void __user *uarg)
out:
query_vma_teardown(&lock_ctx);
mmput(mm);
+ if (vm_file)
+ fput(vm_file);
kfree(name_buf);
return err;
}
diff --git a/include/linux/buildid.h b/include/linux/buildid.h
index 831c1b4b626c..7acc06b22fb7 100644
--- a/include/linux/buildid.h
+++ b/include/linux/buildid.h
@@ -7,7 +7,10 @@
#define BUILD_ID_SIZE_MAX 20
struct vm_area_struct;
+struct file;
+
int build_id_parse(struct vm_area_struct *vma, unsigned char *build_id, __u32 *size);
+int build_id_parse_file(struct file *file, unsigned char *build_id, __u32 *size);
int build_id_parse_nofault(struct vm_area_struct *vma, unsigned char *build_id, __u32 *size);
int build_id_parse_buf(const void *buf, unsigned char *build_id, u32 buf_size);
diff --git a/lib/buildid.c b/lib/buildid.c
index 818331051afe..dc643a6293c1 100644
--- a/lib/buildid.c
+++ b/lib/buildid.c
@@ -279,7 +279,7 @@ static int get_build_id_64(struct freader *r, unsigned char *build_id, __u32 *si
/* enough for Elf64_Ehdr, Elf64_Phdr, and all the smaller requests */
#define MAX_FREADER_BUF_SZ 64
-static int __build_id_parse(struct vm_area_struct *vma, unsigned char *build_id,
+static int __build_id_parse(struct file *file, unsigned char *build_id,
__u32 *size, bool may_fault)
{
const Elf32_Ehdr *ehdr;
@@ -287,11 +287,7 @@ static int __build_id_parse(struct vm_area_struct *vma, unsigned char *build_id,
char buf[MAX_FREADER_BUF_SZ];
int ret;
- /* only works for page backed storage */
- if (!vma->vm_file)
- return -EINVAL;
-
- freader_init_from_file(&r, buf, sizeof(buf), vma->vm_file, may_fault);
+ freader_init_from_file(&r, buf, sizeof(buf), file, may_fault);
/* fetch first 18 bytes of ELF header for checks */
ehdr = freader_fetch(&r, 0, offsetofend(Elf32_Ehdr, e_type));
@@ -332,7 +328,10 @@ static int __build_id_parse(struct vm_area_struct *vma, unsigned char *build_id,
*/
int build_id_parse_nofault(struct vm_area_struct *vma, unsigned char *build_id, __u32 *size)
{
- return __build_id_parse(vma, build_id, size, false /* !may_fault */);
+ if (!vma->vm_file)
+ return -EINVAL;
+
+ return __build_id_parse(vma->vm_file, build_id, size, false /* !may_fault */);
}
/*
@@ -348,7 +347,26 @@ int build_id_parse_nofault(struct vm_area_struct *vma, unsigned char *build_id,
*/
int build_id_parse(struct vm_area_struct *vma, unsigned char *build_id, __u32 *size)
{
- return __build_id_parse(vma, build_id, size, true /* may_fault */);
+ if (!vma->vm_file)
+ return -EINVAL;
+
+ return __build_id_parse(vma->vm_file, build_id, size, true /* may_fault */);
+}
+
+/*
+ * Parse build ID of ELF file
+ * @vma: file object
+ * @build_id: buffer to store build id, at least BUILD_ID_SIZE long
+ * @size: returns actual build id size in case of success
+ *
+ * Assumes faultable context and can cause page faults to bring in file data
+ * into page cache.
+ *
+ * Return: 0 on success; negative error, otherwise
+ */
+int build_id_parse_file(struct file *file, unsigned char *build_id, __u32 *size)
+{
+ return __build_id_parse(file, build_id, size, true /* may_fault */);
}
/**
--
2.47.3
next reply other threads:[~2026-01-28 18:32 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-01-28 18:32 Andrii Nakryiko [this message]
2026-01-28 18:50 ` [PATCH mm-stable] procfs: avoid fetching build ID while holding VMA lock Andrew Morton
2026-01-28 21:30 ` Andrii Nakryiko
2026-01-29 23:28 ` Andrew Morton
2026-01-29 5:50 ` Suren Baghdasaryan
2026-01-29 21:43 ` Andrii Nakryiko
2026-01-29 15:52 ` Yonghong Song
2026-01-29 17:14 ` Mykyta Yatsenko
2026-01-29 21:46 ` Andrii Nakryiko
2026-01-29 21:48 ` Andrii Nakryiko
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20260128183232.2854138-1-andrii@kernel.org \
--to=andrii@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=bpf@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-mm@vger.kernel.org \
--cc=surenb@google.com \
--cc=syzbot+4e70c8e0a2017b432f7a@syzkaller.appspotmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.