Skip to contentSkip to conversation

af_unix: Remove io_uring dead code in GC.

PATCH Applied Kuniyuki Iwashima 4 participants 9 messages Jan 29, 2024 – Feb 14, 2024

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

cover af_unix: Remove io_uring dead code in GC.

I will post another series that rewrites the garbage collector for AF_UNIX socket.

This is a prep series to clean up changes to GC made by io_uring but now not necessary.

Kuniyuki Iwashima (3):
af_unix: Replace BUG_ON() with WARN_ON_ONCE().
af_unix: Remove io_uring code for GC.
af_unix: Remove CONFIG_UNIX_SCM.

 include/net/af_unix.h |   8 +--
 net/Makefile          |   2 +-
 net/unix/Kconfig      |   5 --
 net/unix/Makefile     |   2 -
 net/unix/af_unix.c    |  63 ++++++++++++++++-
 net/unix/garbage.c    | 106 ++++++++++++++++++++--------
 net/unix/scm.c        | 156 ------------------------------------------
 net/unix/scm.h        |  10 ---
 8 files changed, 143 insertions(+), 209 deletions(-)
 delete mode 100644 net/unix/scm.c
 delete mode 100644 net/unix/scm.h
Signature · 1 line
2.30.2
#1 lore raw

PATCH 1/3 af_unix: Replace BUG_ON() with WARN_ON_ONCE().

This is a prep patch for the last patch in this series so that checkpatch will not warn about BUG_ON().

---
 net/unix/garbage.c | 8 ++++----
 net/unix/scm.c     | 8 ++++----
 2 files changed, 8 insertions(+), 8 deletions(-)
net/unix/garbage.c +4 -4
index 4046c606f0e6..af676bb8fb67 100644
@@ -145,7 +145,7 @@ static void scan_children(struct sock *x, void (*func)(struct unix_sock *),
 			/* An embryo cannot be in-flight, so it's safe
 			 * to use the list link.
 			 */
-			BUG_ON(!list_empty(&u->link));
+			WARN_ON_ONCE(!list_empty(&u->link));
 			list_add_tail(&u->link, &embryos);
 		}
 		spin_unlock(&x->sk_receive_queue.lock);
@@ -213,8 +213,8 @@ static void __unix_gc(struct work_struct *work)
 
 		total_refs = file_count(u->sk.sk_socket->file);
 
-		BUG_ON(!u->inflight);
-		BUG_ON(total_refs < u->inflight);
+		WARN_ON_ONCE(!u->inflight);
+		WARN_ON_ONCE(total_refs < u->inflight);
 		if (total_refs == u->inflight) {
 			list_move_tail(&u->link, &gc_candidates);
 			__set_bit(UNIX_GC_CANDIDATE, &u->gc_flags);
@@ -294,7 +294,7 @@ static void __unix_gc(struct work_struct *work)
 		list_move_tail(&u->link, &gc_inflight_list);
 
 	/* All candidates should have been detached by now. */
-	BUG_ON(!list_empty(&gc_candidates));
+	WARN_ON_ONCE(!list_empty(&gc_candidates));
 
 	/* Paired with READ_ONCE() in wait_for_unix_gc(). */
 	WRITE_ONCE(gc_in_progress, false);
net/unix/scm.c +4 -4
index b5ae5ab16777..505e56cf02a2 100644
@@ -51,10 +51,10 @@ void unix_inflight(struct user_struct *user, struct file *fp)
 
 	if (u) {
 		if (!u->inflight) {
-			BUG_ON(!list_empty(&u->link));
+			WARN_ON_ONCE(!list_empty(&u->link));
 			list_add_tail(&u->link, &gc_inflight_list);
 		} else {
-			BUG_ON(list_empty(&u->link));
+			WARN_ON_ONCE(list_empty(&u->link));
 		}
 		u->inflight++;
 		/* Paired with READ_ONCE() in wait_for_unix_gc() */
@@ -71,8 +71,8 @@ void unix_notinflight(struct user_struct *user, struct file *fp)
 	spin_lock(&unix_gc_lock);
 
 	if (u) {
-		BUG_ON(!u->inflight);
-		BUG_ON(list_empty(&u->link));
+		WARN_ON_ONCE(!u->inflight);
+		WARN_ON_ONCE(list_empty(&u->link));
 
 		u->inflight--;
 		if (!u->inflight)
Signature · 1 line
2.30.2
#2 lore raw

PATCH 2/3 af_unix: Remove io_uring code for GC.

In reply to Kuniyuki Iwashima

Since commit 705318a99a13 ("io_uring/af_unix: disable sending io_uring over sockets"), io_uring's unix socket cannot be passed via SCM_RIGHTS, so it does not contribute to cyclic reference and no longer be candidate for garbage collection.

Also, commit 6e5e6d274956 ("io_uring: drop any code related to SCM_RIGHTS") cleaned up SCM_RIGHTS code in io_uring.

Let's do it in AF_UNIX as well by reverting commit 0091bfc81741 ("io_uring/af_unix: defer registered files gc to io_uring release") and commit 10369080454d ("net: reclaim skb->scm_io_uring bit").

---
 include/net/af_unix.h |  1 -
 net/unix/garbage.c    | 25 ++-----------------------
 net/unix/scm.c        |  6 ------
 3 files changed, 2 insertions(+), 30 deletions(-)
include/net/af_unix.h +0 -1
index f045bbd9017d..9e39b2ec4524 100644
@@ -20,7 +20,6 @@ static inline struct unix_sock *unix_get_socket(struct file *filp)
 void unix_inflight(struct user_struct *user, struct file *fp);
 void unix_notinflight(struct user_struct *user, struct file *fp);
 void unix_destruct_scm(struct sk_buff *skb);
-void io_uring_destruct_scm(struct sk_buff *skb);
 void unix_gc(void);
 void wait_for_unix_gc(struct scm_fp_list *fpl);
 struct sock *unix_peer_get(struct sock *sk);
net/unix/garbage.c +2 -23
index af676bb8fb67..ce5b5f87b16e 100644
@@ -184,12 +184,10 @@ static bool gc_in_progress;
 
 static void __unix_gc(struct work_struct *work)
 {
-	struct sk_buff *next_skb, *skb;
-	struct unix_sock *u;
-	struct unix_sock *next;
 	struct sk_buff_head hitlist;
-	struct list_head cursor;
+	struct unix_sock *u, *next;
 	LIST_HEAD(not_cycle_list);
+	struct list_head cursor;
 
 	spin_lock(&unix_gc_lock);
 
@@ -269,30 +267,11 @@ static void __unix_gc(struct work_struct *work)
 
 	spin_unlock(&unix_gc_lock);
 
-	/* We need io_uring to clean its registered files, ignore all io_uring
-	 * originated skbs. It's fine as io_uring doesn't keep references to
-	 * other io_uring instances and so killing all other files in the cycle
-	 * will put all io_uring references forcing it to go through normal
-	 * release.path eventually putting registered files.
-	 */
-	skb_queue_walk_safe(&hitlist, skb, next_skb) {
-		if (skb->destructor == io_uring_destruct_scm) {
-			__skb_unlink(skb, &hitlist);
-			skb_queue_tail(&skb->sk->sk_receive_queue, skb);
-		}
-	}
-
 	/* Here we are. Hitlist is filled. Die. */
 	__skb_queue_purge(&hitlist);
 
 	spin_lock(&unix_gc_lock);
 
-	/* There could be io_uring registered files, just push them back to
-	 * the inflight list
-	 */
-	list_for_each_entry_safe(u, next, &gc_candidates, link)
-		list_move_tail(&u->link, &gc_inflight_list);
-
 	/* All candidates should have been detached by now. */
 	WARN_ON_ONCE(!list_empty(&gc_candidates));
 
net/unix/scm.c +0 -6
index 505e56cf02a2..db65b0ab5947 100644
@@ -148,9 +148,3 @@ void unix_destruct_scm(struct sk_buff *skb)
 	sock_wfree(skb);
 }
 EXPORT_SYMBOL(unix_destruct_scm);
-
-void io_uring_destruct_scm(struct sk_buff *skb)
-{
-	unix_destruct_scm(skb);
-}
-EXPORT_SYMBOL(io_uring_destruct_scm);
Signature · 1 line
2.30.2
#3 lore raw

Re: [PATCH v1 net-next 2/3] af_unix: Remove io_uring code for GC.

Hi,

Quoted 86 lines · Kuniyuki Iwashima

Since commit 705318a99a13 ("io_uring/af_unix: disable sending io_uring over sockets"), io_uring's unix socket cannot be passed via SCM_RIGHTS, so it does not contribute to cyclic reference and no longer be candidate for garbage collection.

Also, commit 6e5e6d274956 ("io_uring: drop any code related to SCM_RIGHTS") cleaned up SCM_RIGHTS code in io_uring.

Let's do it in AF_UNIX as well by reverting commit 0091bfc81741 ("io_uring/af_unix: defer registered files gc to io_uring release") and commit 10369080454d ("net: reclaim skb->scm_io_uring bit").

---
 include/net/af_unix.h |  1 -
 net/unix/garbage.c    | 25 ++-----------------------
 net/unix/scm.c        |  6 ------
 3 files changed, 2 insertions(+), 30 deletions(-)
include/net/af_unix.h +0 -1
index f045bbd9017d..9e39b2ec4524 100644
@@ -20,7 +20,6 @@ static inline struct unix_sock *unix_get_socket(struct file *filp)
 void unix_inflight(struct user_struct *user, struct file *fp);
 void unix_notinflight(struct user_struct *user, struct file *fp);
 void unix_destruct_scm(struct sk_buff *skb);
-void io_uring_destruct_scm(struct sk_buff *skb);
 void unix_gc(void);
 void wait_for_unix_gc(struct scm_fp_list *fpl);
 struct sock *unix_peer_get(struct sock *sk);
net/unix/garbage.c +2 -23
index af676bb8fb67..ce5b5f87b16e 100644
@@ -184,12 +184,10 @@ static bool gc_in_progress;
 
 static void __unix_gc(struct work_struct *work)
 {
-	struct sk_buff *next_skb, *skb;
-	struct unix_sock *u;
-	struct unix_sock *next;
 	struct sk_buff_head hitlist;
-	struct list_head cursor;
+	struct unix_sock *u, *next;
 	LIST_HEAD(not_cycle_list);
+	struct list_head cursor;
 
 	spin_lock(&unix_gc_lock);
 
@@ -269,30 +267,11 @@ static void __unix_gc(struct work_struct *work)
 
 	spin_unlock(&unix_gc_lock);
 
-	/* We need io_uring to clean its registered files, ignore all io_uring
-	 * originated skbs. It's fine as io_uring doesn't keep references to
-	 * other io_uring instances and so killing all other files in the cycle
-	 * will put all io_uring references forcing it to go through normal
-	 * release.path eventually putting registered files.
-	 */
-	skb_queue_walk_safe(&hitlist, skb, next_skb) {
-		if (skb->destructor == io_uring_destruct_scm) {
-			__skb_unlink(skb, &hitlist);
-			skb_queue_tail(&skb->sk->sk_receive_queue, skb);
-		}
-	}
-
 	/* Here we are. Hitlist is filled. Die. */
 	__skb_queue_purge(&hitlist);
 
 	spin_lock(&unix_gc_lock);
 
-	/* There could be io_uring registered files, just push them back to
-	 * the inflight list
-	 */
-	list_for_each_entry_safe(u, next, &gc_candidates, link)
-		list_move_tail(&u->link, &gc_inflight_list);
-
 	/* All candidates should have been detached by now. */
 	WARN_ON_ONCE(!list_empty(&gc_candidates));
 
net/unix/scm.c +0 -0
index 505e56cf02a2..db65b0ab5947 100644

net/unix/scm.c void unix_destruct_scm(struct sk_buff *skb) · line 156

Quoted 9 lines
net/unix/scm.c +0 -6
@@ -148,9 +148,3 @@ void unix_destruct_scm(struct sk_buff *skb)
 	sock_wfree(skb);
 }
 EXPORT_SYMBOL(unix_destruct_scm);
-
-void io_uring_destruct_scm(struct sk_buff *skb)
-{
-	unix_destruct_scm(skb);
-}
-EXPORT_SYMBOL(io_uring_destruct_scm);

Syzkaller found below issue.
There is WARNING in __unix_gc in v6.8-rc3_internal-devel_hourly-20240205-094544, the kernel contains kernel-next patches.

Bisected and found first bad commit:
"
11498715f266 af_unix: Remove io_uring code for GC. "
It's the same patch as above.

All detailed info: https://github.com/xupengfe/syzkaller_logs/tree/main/240211_144134___unix_gc
Syzkaller reproduced code: https://github.com/xupengfe/syzkaller_logs/blob/main/240211_144134___unix_gc/repro.c
Syzkaller repro syscall steps: https://github.com/xupengfe/syzkaller_logs/blob/main/240211_144134___unix_gc/repro.prog
Kconfig(make olddefconfig): https://github.com/xupengfe/syzkaller_logs/blob/main/240211_144134___unix_gc/kconfig_origin
Bisect info: https://github.com/xupengfe/syzkaller_logs/blob/main/240211_144134___unix_gc/bisect_info.log
Issue dmesg: https://github.com/xupengfe/syzkaller_logs/blob/main/240211_144134___unix_gc/3561c4956a5c9e7f995ae47d4ef703eb9c6a93cd_dmesg.log
bzImage: https://github.com/xupengfe/syzkaller_logs/raw/main/240211_144134___unix_gc/bzImage_3561c4956a5c.tar.gz
repro.report: https://github.com/xupengfe/syzkaller_logs/blob/main/240211_144134___unix_gc/repro.report

"

[   27.629798] ------------[ cut here ]------------
[   27.630447] WARNING: CPU: 0 PID: 52 at net/unix/garbage.c:345 __unix_gc+0x99e/0xb50
[   27.631312] Modules linked in:
[   27.631671] CPU: 0 PID: 52 Comm: kworker/u4:3 Not tainted 6.8.0-rc3-3561c4956a5c+ #1
[   27.632787] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014
[   27.634018] Workqueue: events_unbound __unix_gc
[   27.634544] RIP: 0010:__unix_gc+0x99e/0xb50
[   27.635026] Code: b2 4f fc 0f 0b e9 6c f8 ff ff e8 0d b2 4f fc 31 d2 48 c7 c6 e0 8f 12 85 4c 89 e7 e8 ec f1 ff ff e9 32 fb ff ff e8 f2 b1 4f fc <0f> 0b e9 7e fe ff ff 4c 89 e7 e8 c3 dd b0 fc e9 9c fa ff ff e8 b9
[   27.637177] RSP: 0018:ffff88800c677b90 EFLAGS: 00010293
[   27.637768] RAX: 0000000000000000 RBX: dffffc0000000000 RCX: ffffffff8140b3c2
[   27.638544] RDX: ffff88800bfbca00 RSI: ffffffff8512a5be RDI: ffff88800c677af8
[   27.639329] RBP: ffff88800c677cc8 R08: 0000000000000001 R09: ffffed10018cef5f
[   27.640112] R10: 0000000000000003 R11: 0000000000000001 R12: ffff88800c677c00
[   27.640992] R13: ffff88800c677c00 R14: ffff88800c677c00 R15: ffff88800c677c00
[   27.641768] FS:  0000000000000000(0000) GS:ffff88806cc00000(0000) knlGS:0000000000000000
[   27.642646] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   27.643285] CR2: 00007f6063373fa8 CR3: 0000000006a7e004 CR4: 0000000000770ef0
[   27.644069] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[   27.644876] DR3: 0000000000000000 DR6: 00000000ffff07f0 DR7: 0000000000000400
[   27.645658] PKRU: 55555554
[   27.645974] Call Trace:
[   27.646266]  <TASK>
[   27.646524]  ? show_regs+0xa9/0xc0
[   27.646933]  ? __warn+0xef/0x340
[   27.647317]  ? report_bug+0x25e/0x4b0
[   27.647748]  ? __unix_gc+0x99e/0xb50
[   27.648190]  ? report_bug+0x2cb/0x4b0
[   27.648630]  ? __unix_gc+0x99e/0xb50
[   27.649049]  ? handle_bug+0xa2/0x130
[   27.649470]  ? exc_invalid_op+0x3c/0x80
[   27.649922]  ? asm_exc_invalid_op+0x1f/0x30
[   27.650416]  ? do_raw_spin_lock+0x142/0x290
[   27.650892]  ? __unix_gc+0x99e/0xb50
[   27.651315]  ? __unix_gc+0x99e/0xb50
[   27.651742]  ? __pfx___unix_gc+0x10/0x10
[   27.652209]  ? __this_cpu_preempt_check+0x21/0x30
[   27.652757]  ? lock_acquire+0x1d9/0x530
[   27.653219]  ? __this_cpu_preempt_check+0x21/0x30
[   27.653754]  ? _raw_spin_unlock_irq+0x2c/0x60
[   27.654267]  process_one_work+0x813/0x15a0
[   27.654757]  ? __pfx_process_one_work+0x10/0x10
[   27.655271]  ? move_linked_works+0x1bf/0x2c0
[   27.655767]  ? __this_cpu_preempt_check+0x21/0x30
[   27.656346]  ? assign_work+0x19f/0x250
[   27.656780]  ? lock_is_held_type+0xf0/0x150
[   27.657267]  worker_thread+0x823/0x11a0
[   27.657710]  ? _raw_spin_unlock_irqrestore+0x35/0x70
[   27.658269]  ? trace_hardirqs_on+0x26/0x120
[   27.658771]  kthread+0x35f/0x470
[   27.659153]  ? __pfx_worker_thread+0x10/0x10
[   27.659647]  ? __pfx_kthread+0x10/0x10
[   27.660089]  ret_from_fork+0x56/0x90
[   27.660535]  ? __pfx_kthread+0x10/0x10
[   27.660973]  ret_from_fork_asm+0x1b/0x30
[   27.661451]  </TASK>
[   27.661715] irq event stamp: 12659
[   27.662104] hardirqs last  enabled at (12667): [<ffffffff814359a5>] console_unlock+0x2d5/0x310
[   27.663049] hardirqs last disabled at (12674): [<ffffffff8143598a>] console_unlock+0x2ba/0x310
[   27.663991] softirqs last  enabled at (12306): [<ffffffff8126fcf8>] __irq_exit_rcu+0xa8/0x110
[   27.664946] softirqs last disabled at (12291): [<ffffffff8126fcf8>] __irq_exit_rcu+0xa8/0x110
[   27.665878] ---[ end trace 0000000000000000 ]---

"

As above WARNING and bisect info, do you mind to take a look?

Thanks!

---

If you don't need the following environment to reproduce the problem or if you already have one reproduced environment, please ignore the following information.

How to reproduce:
git clone https://gitlab.com/xupengfe/repro_vm_env.git
cd repro_vm_env
tar -xvf repro_vm_env.tar.gz
cd repro_vm_env; ./start3.sh // it needs qemu-system-x86_64 and I used v7.1.0
// start3.sh will load bzImage_2241ab53cbb5cdb08a6b2d4688feb13971058f65 v6.2-rc5 kernel
// You could change the bzImage_xxx as you want
// Maybe you need to remove line "-drive if=pflash,format=raw,readonly=on,file=./OVMF_CODE.fd \" for different qemu version
You could use below command to log in, there is no password for root.
ssh -p 10023 root@localhost

After login vm(virtual machine) successfully, you could transfer reproduced binary to the vm by below way, and reproduce the problem in vm:
gcc -pthread -o repro repro.c
scp -P 10023 repro root@localhost:/root/

Get the bzImage for target kernel:
Please use target kconfig and copy it to kernel_src/.config make olddefconfig
make -jx bzImage //x should equal or less than cpu num your pc has

Fill the bzImage file into above start3.sh to load the target kernel in vm.

Tips:
If you already have qemu-system-x86_64, please ignore below info.
If you want to install qemu v7.1.0 version:
git clone https://github.com/qemu/qemu.git
cd qemu
git checkout -f v7.1.0
mkdir build
cd build
yum install -y ninja-build.x86_64
yum -y install libslirp-devel.x86_64
../configure --target-list=x86_64-softmmu --enable-kvm --enable-vnc --enable-gtk --enable-sdl --enable-usb-redir --enable-slirp make
make install

Best Regards,
Thanks!

Quoted 3 lines
Signature · 1 line
2.30.2
#4 lore raw

Re: [PATCH v1 net-next 2/3] af_unix: Remove io_uring code for GC.

Quoted 103 lines · Pengfei Xu

Hi,

Quoted 86 lines · Kuniyuki Iwashima

Since commit 705318a99a13 ("io_uring/af_unix: disable sending io_uring over sockets"), io_uring's unix socket cannot be passed via SCM_RIGHTS, so it does not contribute to cyclic reference and no longer be candidate for garbage collection.

Also, commit 6e5e6d274956 ("io_uring: drop any code related to SCM_RIGHTS") cleaned up SCM_RIGHTS code in io_uring.

Let's do it in AF_UNIX as well by reverting commit 0091bfc81741 ("io_uring/af_unix: defer registered files gc to io_uring release") and commit 10369080454d ("net: reclaim skb->scm_io_uring bit").

---
 include/net/af_unix.h |  1 -
 net/unix/garbage.c    | 25 ++-----------------------
 net/unix/scm.c        |  6 ------
 3 files changed, 2 insertions(+), 30 deletions(-)
include/net/af_unix.h +0 -1
index f045bbd9017d..9e39b2ec4524 100644
@@ -20,7 +20,6 @@ static inline struct unix_sock *unix_get_socket(struct file *filp)
 void unix_inflight(struct user_struct *user, struct file *fp);
 void unix_notinflight(struct user_struct *user, struct file *fp);
 void unix_destruct_scm(struct sk_buff *skb);
-void io_uring_destruct_scm(struct sk_buff *skb);
 void unix_gc(void);
 void wait_for_unix_gc(struct scm_fp_list *fpl);
 struct sock *unix_peer_get(struct sock *sk);
net/unix/garbage.c +2 -23
index af676bb8fb67..ce5b5f87b16e 100644
@@ -184,12 +184,10 @@ static bool gc_in_progress;
 
 static void __unix_gc(struct work_struct *work)
 {
-	struct sk_buff *next_skb, *skb;
-	struct unix_sock *u;
-	struct unix_sock *next;
 	struct sk_buff_head hitlist;
-	struct list_head cursor;
+	struct unix_sock *u, *next;
 	LIST_HEAD(not_cycle_list);
+	struct list_head cursor;
 
 	spin_lock(&unix_gc_lock);
 
@@ -269,30 +267,11 @@ static void __unix_gc(struct work_struct *work)
 
 	spin_unlock(&unix_gc_lock);
 
-	/* We need io_uring to clean its registered files, ignore all io_uring
-	 * originated skbs. It's fine as io_uring doesn't keep references to
-	 * other io_uring instances and so killing all other files in the cycle
-	 * will put all io_uring references forcing it to go through normal
-	 * release.path eventually putting registered files.
-	 */
-	skb_queue_walk_safe(&hitlist, skb, next_skb) {
-		if (skb->destructor == io_uring_destruct_scm) {
-			__skb_unlink(skb, &hitlist);
-			skb_queue_tail(&skb->sk->sk_receive_queue, skb);
-		}
-	}
-
 	/* Here we are. Hitlist is filled. Die. */
 	__skb_queue_purge(&hitlist);
 
 	spin_lock(&unix_gc_lock);
 
-	/* There could be io_uring registered files, just push them back to
-	 * the inflight list
-	 */
-	list_for_each_entry_safe(u, next, &gc_candidates, link)
-		list_move_tail(&u->link, &gc_inflight_list);
-
 	/* All candidates should have been detached by now. */
 	WARN_ON_ONCE(!list_empty(&gc_candidates));
 
net/unix/scm.c +0 -0
index 505e56cf02a2..db65b0ab5947 100644

net/unix/scm.c void unix_destruct_scm(struct sk_buff *skb) · line 156

Quoted 9 lines
net/unix/scm.c +0 -6
@@ -148,9 +148,3 @@ void unix_destruct_scm(struct sk_buff *skb)
 	sock_wfree(skb);
 }
 EXPORT_SYMBOL(unix_destruct_scm);
-
-void io_uring_destruct_scm(struct sk_buff *skb)
-{
-	unix_destruct_scm(skb);
-}
-EXPORT_SYMBOL(io_uring_destruct_scm);

Syzkaller found below issue.
There is WARNING in __unix_gc in v6.8-rc3_internal-devel_hourly-20240205-094544, the kernel contains kernel-next patches.

Quoted 5 lines

Bisected and found first bad commit:
"
11498715f266 af_unix: Remove io_uring code for GC. "
It's the same patch as above.

It should be fixed by:

commit 1279f9d9dec2d7462823a18c29ad61359e0a007d

Author: Kuniyuki Iwashima <[email protected]>
Date:   Sat Feb 3 10:31:49 2024 -0800
    af_unix: Call kfree_skb() for dead unix_(sk)->oob_skb in GC.

which is in Linus's tree.

Signature · 1 line
Jens Axboe
#5 lore raw

Re: [PATCH v1 net-next 2/3] af_unix: Remove io_uring code for GC.

Hi Jens Axboe,

Quoted 112 lines · Jens Axboe
Quoted 103 lines · Pengfei Xu

Hi,

Quoted 86 lines · Kuniyuki Iwashima

Since commit 705318a99a13 ("io_uring/af_unix: disable sending io_uring over sockets"), io_uring's unix socket cannot be passed via SCM_RIGHTS, so it does not contribute to cyclic reference and no longer be candidate for garbage collection.

Also, commit 6e5e6d274956 ("io_uring: drop any code related to SCM_RIGHTS") cleaned up SCM_RIGHTS code in io_uring.

Let's do it in AF_UNIX as well by reverting commit 0091bfc81741 ("io_uring/af_unix: defer registered files gc to io_uring release") and commit 10369080454d ("net: reclaim skb->scm_io_uring bit").

---
 include/net/af_unix.h |  1 -
 net/unix/garbage.c    | 25 ++-----------------------
 net/unix/scm.c        |  6 ------
 3 files changed, 2 insertions(+), 30 deletions(-)
include/net/af_unix.h +0 -1
index f045bbd9017d..9e39b2ec4524 100644
@@ -20,7 +20,6 @@ static inline struct unix_sock *unix_get_socket(struct file *filp)
 void unix_inflight(struct user_struct *user, struct file *fp);
 void unix_notinflight(struct user_struct *user, struct file *fp);
 void unix_destruct_scm(struct sk_buff *skb);
-void io_uring_destruct_scm(struct sk_buff *skb);
 void unix_gc(void);
 void wait_for_unix_gc(struct scm_fp_list *fpl);
 struct sock *unix_peer_get(struct sock *sk);
net/unix/garbage.c +2 -23
index af676bb8fb67..ce5b5f87b16e 100644
@@ -184,12 +184,10 @@ static bool gc_in_progress;
 
 static void __unix_gc(struct work_struct *work)
 {
-	struct sk_buff *next_skb, *skb;
-	struct unix_sock *u;
-	struct unix_sock *next;
 	struct sk_buff_head hitlist;
-	struct list_head cursor;
+	struct unix_sock *u, *next;
 	LIST_HEAD(not_cycle_list);
+	struct list_head cursor;
 
 	spin_lock(&unix_gc_lock);
 
@@ -269,30 +267,11 @@ static void __unix_gc(struct work_struct *work)
 
 	spin_unlock(&unix_gc_lock);
 
-	/* We need io_uring to clean its registered files, ignore all io_uring
-	 * originated skbs. It's fine as io_uring doesn't keep references to
-	 * other io_uring instances and so killing all other files in the cycle
-	 * will put all io_uring references forcing it to go through normal
-	 * release.path eventually putting registered files.
-	 */
-	skb_queue_walk_safe(&hitlist, skb, next_skb) {
-		if (skb->destructor == io_uring_destruct_scm) {
-			__skb_unlink(skb, &hitlist);
-			skb_queue_tail(&skb->sk->sk_receive_queue, skb);
-		}
-	}
-
 	/* Here we are. Hitlist is filled. Die. */
 	__skb_queue_purge(&hitlist);
 
 	spin_lock(&unix_gc_lock);
 
-	/* There could be io_uring registered files, just push them back to
-	 * the inflight list
-	 */
-	list_for_each_entry_safe(u, next, &gc_candidates, link)
-		list_move_tail(&u->link, &gc_inflight_list);
-
 	/* All candidates should have been detached by now. */
 	WARN_ON_ONCE(!list_empty(&gc_candidates));
 
net/unix/scm.c +0 -0
index 505e56cf02a2..db65b0ab5947 100644

net/unix/scm.c void unix_destruct_scm(struct sk_buff *skb) · line 156

Quoted 9 lines
net/unix/scm.c +0 -6
@@ -148,9 +148,3 @@ void unix_destruct_scm(struct sk_buff *skb)
 	sock_wfree(skb);
 }
 EXPORT_SYMBOL(unix_destruct_scm);
-
-void io_uring_destruct_scm(struct sk_buff *skb)
-{
-	unix_destruct_scm(skb);
-}
-EXPORT_SYMBOL(io_uring_destruct_scm);

Syzkaller found below issue.
There is WARNING in __unix_gc in v6.8-rc3_internal-devel_hourly-20240205-094544, the kernel contains kernel-next patches.

Quoted 5 lines

Bisected and found first bad commit:
"
11498715f266 af_unix: Remove io_uring code for GC. "
It's the same patch as above.

It should be fixed by:

Quoted 7 lines

commit 1279f9d9dec2d7462823a18c29ad61359e0a007d

Author: Kuniyuki Iwashima <[email protected]>
Date:   Sat Feb 3 10:31:49 2024 -0800
    af_unix: Call kfree_skb() for dead unix_(sk)->oob_skb in GC.

which is in Linus's tree.

Thank you for the commit tip for the fix. This is indeed the same problem and has been fixed.
I will check the community mail carefully next time to avoid reporting duplicate problems.

Best Regards,
Thanks!

Quoted 4 lines
Signature · 2 lines
-- 
Jens Axboe
#6 lore raw

PATCH 3/3 af_unix: Remove CONFIG_UNIX_SCM.

In reply to Kuniyuki Iwashima

Originally, the code related to garbage collection was all in garbage.c.

Commit f4e65870e5ce ("net: split out functions related to registering inflight socket files") moved some functions to scm.c for io_uring and added CONFIG_UNIX_SCM just in case AF_UNIX was built as module.

However, since commit 97154bcf4d1b ("af_unix: Kconfig: make CONFIG_UNIX bool"), AF_UNIX is no longer built separately. Also, io_uring does not support SCM_RIGHTS now.

Let's move the functions back to garbage.c

---
 include/net/af_unix.h |   7 +-
 net/Makefile          |   2 +-
 net/unix/Kconfig      |   5 --
 net/unix/Makefile     |   2 -
 net/unix/af_unix.c    |  63 +++++++++++++++++-
 net/unix/garbage.c    |  73 +++++++++++++++++++-
 net/unix/scm.c        | 150 ------------------------------------------
 net/unix/scm.h        |  10 ---
 8 files changed, 137 insertions(+), 175 deletions(-)
 delete mode 100644 net/unix/scm.c
 delete mode 100644 net/unix/scm.h
include/net/af_unix.h +4 -3
index 9e39b2ec4524..54e346152eb1 100644
@@ -17,19 +17,20 @@ static inline struct unix_sock *unix_get_socket(struct file *filp)
 }
 #endif
 
+extern spinlock_t unix_gc_lock;
+extern unsigned int unix_tot_inflight;
+
 void unix_inflight(struct user_struct *user, struct file *fp);
 void unix_notinflight(struct user_struct *user, struct file *fp);
-void unix_destruct_scm(struct sk_buff *skb);
 void unix_gc(void);
 void wait_for_unix_gc(struct scm_fp_list *fpl);
+
 struct sock *unix_peer_get(struct sock *sk);
 
 #define UNIX_HASH_MOD	(256 - 1)
 #define UNIX_HASH_SIZE	(256 * 2)
 #define UNIX_HASH_BITS	8
 
-extern unsigned int unix_tot_inflight;
-
 struct unix_address {
 	refcount_t	refcnt;
 	int		len;
net/Makefile +1 -1
index b06b5539e7a6..65bb8c72a35e 100644
@@ -17,7 +17,7 @@ obj-$(CONFIG_NETFILTER)		+= netfilter/
 obj-$(CONFIG_INET)		+= ipv4/
 obj-$(CONFIG_TLS)		+= tls/
 obj-$(CONFIG_XFRM)		+= xfrm/
-obj-$(CONFIG_UNIX_SCM)		+= unix/
+obj-$(CONFIG_UNIX)		+= unix/
 obj-y				+= ipv6/
 obj-$(CONFIG_PACKET)		+= packet/
 obj-$(CONFIG_NET_KEY)		+= key/
net/unix/Kconfig +0 -5
index 28b232f281ab..8b5d04210d7c 100644
@@ -16,11 +16,6 @@ config UNIX
 
 	  Say Y unless you know what you are doing.
 
-config UNIX_SCM
-	bool
-	depends on UNIX
-	default y
-
 config	AF_UNIX_OOB
 	bool
 	depends on UNIX
net/unix/Makefile +0 -2
index 20491825b4d0..4ddd125c4642 100644
@@ -11,5 +11,3 @@ unix-$(CONFIG_BPF_SYSCALL) += unix_bpf.o
 
 obj-$(CONFIG_UNIX_DIAG)	+= unix_diag.o
 unix_diag-y		:= diag.o
-
-obj-$(CONFIG_UNIX_SCM)	+= scm.o
net/unix/af_unix.c +61 -2

81 lines not on this page · raw message

net/unix/garbage.c +71 -2

82 lines not on this page · raw message

net/unix/scm.c deleted +0 -150

150 lines not on this page · raw message

net/unix/scm.h deleted +0 -10
deleted file mode 100644
index 5a255a477f16..000000000000
@@ -1,10 +0,0 @@
-#ifndef NET_UNIX_SCM_H
-#define NET_UNIX_SCM_H
-
-extern struct list_head gc_inflight_list;
-extern spinlock_t unix_gc_lock;
-
-int unix_attach_fds(struct scm_cookie *scm, struct sk_buff *skb);
-void unix_detach_fds(struct scm_cookie *scm, struct sk_buff *skb);
-
-#endif
Signature · 1 line
2.30.2
Acked-by on the v1 series
#7 lore raw
In reply to Kuniyuki Iwashima
Quoted 5 lines · Kuniyuki Iwashima

I will post another series that rewrites the garbage collector for AF_UNIX socket.

This is a prep series to clean up changes to GC made by io_uring but now not necessary.

Didn't look at this in detail, but it looks like it's just reverting the changes we had to make there originally to accommodate io_uring. Hence they are fine to get reverted:

Signature · 1 line
Jens Axboe
#8 lore raw
In reply to Kuniyuki Iwashima
This series was applied to netdev/net-next.git (main) by Jakub Kicinski <[email protected]>:

Hello:

This series was applied to netdev/net-next.git (main) by Jakub Kicinski <[email protected]>:

Quoted 8 lines · On Mon, 29 Jan 2024 11:04:32 -0800 you wrote:

I will post another series that rewrites the garbage collector for AF_UNIX socket.

This is a prep series to clean up changes to GC made by io_uring but now not necessary.

[...]

Here is the summary with links:
- [v1,net-next,1/3] af_unix: Replace BUG_ON() with WARN_ON_ONCE().

    https://git.kernel.org/netdev/net-next/c/d0f6dc263468
  - [v1,net-next,2/3] af_unix: Remove io_uring code for GC.
    https://git.kernel.org/netdev/net-next/c/11498715f266
  - [v1,net-next,3/3] af_unix: Remove CONFIG_UNIX_SCM.
    https://git.kernel.org/netdev/net-next/c/99a7a5b9943e

You are awesome, thank you!

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