Skip to contentSkip to conversation

USB: serial: kl5kusb105: fix bulk-out buffer overflow

PATCH Superseded HyeongJun An 2 participants 7 messages Jun 7 – Jun 8

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

PATCH 1/1 USB: serial: kl5kusb105: fix bulk-out buffer overflow

klsi_105_prepare_write_buffer() is called by the generic write path with the bulk-out buffer and its size (bulk_out_size, 64 bytes). It stores a two-byte length header at the start of the buffer and copies the payload from the write fifo starting at buf + KLSI_HDR_LEN, but passes the full buffer size as the number of bytes to copy:

count = kfifo_out_locked(&port->write_fifo, buf + KLSI_HDR_LEN,

                           size, &port->lock);

When the fifo holds at least size bytes, size bytes are copied starting two bytes into the size-byte buffer, writing KLSI_HDR_LEN bytes past its end. Copy at most size - KLSI_HDR_LEN bytes instead, leaving room for the header as safe_serial already does.

Writing bulk_out_size or more bytes to the tty triggers a slab out-of-bounds write, observed with KASAN by emulating the device with dummy_hcd and raw-gadget:

  BUG: KASAN: slab-out-of-bounds in kfifo_copy_out+0x83/0xc0
  Write of size 64 at addr ffff888112c62202 by task python3
   kfifo_copy_out
   klsi_105_prepare_write_buffer [kl5kusb105]
   usb_serial_generic_write_start [usbserial]
  Allocated by task 139:
   usb_serial_probe [usbserial]
  The buggy address is located 2 bytes inside of allocated 64-byte region

The out-of-bounds write no longer occurs with this change applied.

Fixes 60b3013cdaf3 ("USB: kl5usb105: reimplement using generic framework")
---
 drivers/usb/serial/kl5kusb105.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/usb/serial/kl5kusb105.c +2 -2
index ed8531a64768..e72a0b45a707 100644
@@ -330,8 +330,8 @@ static int klsi_105_prepare_write_buffer(struct usb_serial_port *port,
 	unsigned char *buf = dest;
 	int count;
 
-	count = kfifo_out_locked(&port->write_fifo, buf + KLSI_HDR_LEN, size,
-								&port->lock);
+	count = kfifo_out_locked(&port->write_fifo, buf + KLSI_HDR_LEN,
+				 size - KLSI_HDR_LEN, &port->lock);
 	put_unaligned_le16(count, buf);
 
 	return count + KLSI_HDR_LEN;
Signature · 1 line
2.43.0
maintainer
#1 lore raw
Quoted 6 lines · HyeongJun An

klsi_105_prepare_write_buffer() is called by the generic write path with the bulk-out buffer and its size (bulk_out_size, 64 bytes). It stores a two-byte length header at the start of the buffer and copies the payload from the write fifo starting at buf + KLSI_HDR_LEN, but passes the full buffer size as the number of bytes to copy:

Quoted 7 lines

count = kfifo_out_locked(&port->write_fifo, buf + KLSI_HDR_LEN,

                           size, &port->lock);

When the fifo holds at least size bytes, size bytes are copied starting two bytes into the size-byte buffer, writing KLSI_HDR_LEN bytes past its end. Copy at most size - KLSI_HDR_LEN bytes instead, leaving room for the header as safe_serial already does.

Good catch!

How was this found? Did you use some kind of static checker or LLM?

Johan

#2 lore raw

Hi Johan,

Thanks a lot!

Yes, I used an LLM to compare the custom prepare_write_buffer() handlers in drivers/usb/serial/. kl5kusb105 passes the full "size" to the fifo copy, while the ones with a header or trailer, like safe_serial, reserve that space first.

Thanks,
HyeongJun

maintainer
#3 lore raw
Quoted 4 lines · HyeongJun An

Yes, I used an LLM to compare the custom prepare_write_buffer() handlers in drivers/usb/serial/. kl5kusb105 passes the full "size" to the fifo copy, while the ones with a header or trailer, like safe_serial, reserve that space first.

Thanks for confirming. This needs to be documented in the commit message, see:

Documentation/process/submitting-patches.rst ["Using Assisted-by"] Documentation/process/coding-assistants.rst

Can you send a v2 with the missing tag?

Johan

author
#4 lore raw

Hi Johan,

Sure, thanks! I've just sent a v2 with the "Assisted-by" tag.

Thanks,
HyeongJun

Quoted 17 lines · Johan Hovold
Quoted 4 lines · HyeongJun An

Yes, I used an LLM to compare the custom prepare_write_buffer() handlers in drivers/usb/serial/. kl5kusb105 passes the full "size" to the fifo copy, while the ones with a header or trailer, like safe_serial, reserve that space first.

Thanks for confirming. This needs to be documented in the commit message, see:

Documentation/process/submitting-patches.rst ["Using Assisted-by"] Documentation/process/coding-assistants.rst

Can you send a v2 with the missing tag?

Johan

#5 lore raw

PATCH v2 1/1 USB: serial: kl5kusb105: fix bulk-out buffer overflow

In reply to HyeongJun An

klsi_105_prepare_write_buffer() is called by the generic write path with the bulk-out buffer and its size (bulk_out_size, 64 bytes). It stores a two-byte length header at the start of the buffer and copies the payload from the write fifo starting at buf + KLSI_HDR_LEN, but passes the full buffer size as the number of bytes to copy:

count = kfifo_out_locked(&port->write_fifo, buf + KLSI_HDR_LEN,

                           size, &port->lock);

When the fifo holds at least size bytes, size bytes are copied starting two bytes into the size-byte buffer, writing KLSI_HDR_LEN bytes past its end. Copy at most size - KLSI_HDR_LEN bytes instead, leaving room for the header as safe_serial already does.

Writing bulk_out_size or more bytes to the tty triggers a slab out-of-bounds write, observed with KASAN by emulating the device with dummy_hcd and raw-gadget:

  BUG: KASAN: slab-out-of-bounds in kfifo_copy_out+0x83/0xc0
  Write of size 64 at addr ffff888112c62202 by task python3
   kfifo_copy_out
   klsi_105_prepare_write_buffer [kl5kusb105]
   usb_serial_generic_write_start [usbserial]
  Allocated by task 139:
   usb_serial_probe [usbserial]
  The buggy address is located 2 bytes inside of allocated 64-byte region

The out-of-bounds write no longer occurs with this change applied.

Fixes 60b3013cdaf3 ("USB: kl5usb105: reimplement using generic framework")
Assisted-by Claude:claude-opus-4-8

---
v2:
- Add Assisted-by tag as requested by Johan.

 drivers/usb/serial/kl5kusb105.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/usb/serial/kl5kusb105.c +2 -2
index ed8531a64768..e72a0b45a707 100644
@@ -330,8 +330,8 @@ static int klsi_105_prepare_write_buffer(struct usb_serial_port *port,
 	unsigned char *buf = dest;
 	int count;
 
-	count = kfifo_out_locked(&port->write_fifo, buf + KLSI_HDR_LEN, size,
-								&port->lock);
+	count = kfifo_out_locked(&port->write_fifo, buf + KLSI_HDR_LEN,
+				 size - KLSI_HDR_LEN, &port->lock);
 	put_unaligned_le16(count, buf);
 
 	return count + KLSI_HDR_LEN;
Signature · 1 line
2.43.0
maintainer
#6 lore raw
Quoted 14 lines · HyeongJun An

klsi_105_prepare_write_buffer() is called by the generic write path with the bulk-out buffer and its size (bulk_out_size, 64 bytes). It stores a two-byte length header at the start of the buffer and copies the payload from the write fifo starting at buf + KLSI_HDR_LEN, but passes the full buffer size as the number of bytes to copy:

count = kfifo_out_locked(&port->write_fifo, buf + KLSI_HDR_LEN,

                           size, &port->lock);

When the fifo holds at least size bytes, size bytes are copied starting two bytes into the size-byte buffer, writing KLSI_HDR_LEN bytes past its end. Copy at most size - KLSI_HDR_LEN bytes instead, leaving room for the header as safe_serial already does.

Quoted 7 lines
Fixes 60b3013cdaf3 ("USB: kl5usb105: reimplement using generic framework")
Assisted-by Claude:claude-opus-4-8

---
v2:
- Add Assisted-by tag as requested by Johan.

Now applied, thanks.

Johan