After a call to console_unlock() in vcs_write() the vc_data struct can be
freed by vc_port_destruct(). Because of that, the struct vc_data pointer
must be reloaded in the while loop in vcs_write() after console_lock() to
avoid a UAF when vcs_size() is called.
Syzkaller reported a UAF in vcs_size().
BUG: KASAN: slab-use-after-free in vcs_size (drivers/tty/vt/vc_screen.c:215)
Read of size 4 at addr ffff8880beab89a8 by task repro_vcs_size/4119
Call Trace:
<TASK>
__asan_report_load4_noabort (mm/kasan/report_generic.c:380)
vcs_size (drivers/tty/vt/vc_screen.c:215)
vcs_write (drivers/tty/vt/vc_screen.c:664)
vfs_write (fs/read_write.c:582 fs/read_write.c:564)
...
<TASK>
Allocated by task 1213:
kmalloc_trace (mm/slab_common.c:1064)
vc_allocate (./include/linux/slab.h:559 ./include/linux/slab.h:680
drivers/tty/vt/vt.c:1078 drivers/tty/vt/vt.c:1058)
con_install (drivers/tty/vt/vt.c:3334)
tty_init_dev (drivers/tty/tty_io.c:1303 drivers/tty/tty_io.c:1415
drivers/tty/tty_io.c:1392)
tty_open (drivers/tty/tty_io.c:2082 drivers/tty/tty_io.c:2128)
chrdev_open (fs/char_dev.c:415)
do_dentry_open (fs/open.c:921)
vfs_open (fs/open.c:1052)
...
Freed by task 4116:
kfree (mm/slab_common.c:1016)
vc_port_destruct (drivers/tty/vt/vt.c:1044)
tty_port_destructor (drivers/tty/tty_port.c:296)
tty_port_put (drivers/tty/tty_port.c:312)
vt_disallocate_all (drivers/tty/vt/vt_ioctl.c:662 (discriminator 2))
vt_ioctl (drivers/tty/vt/vt_ioctl.c:903)
tty_ioctl (drivers/tty/tty_io.c:2778)
...
The buggy address belongs to the object at ffff8880beab8800
which belongs to the cache kmalloc-1k of size 1024
The buggy address is located 424 bytes inside of
freed 1024-byte region [ffff8880beab8800, ffff8880beab8c00)
The buggy address belongs to the physical page:
page:00000000afc77580 refcount:1 mapcount:0 mapping:0000000000000000
index:0x0 pfn:0xbeab8
head:00000000afc77580 order:3 entire_mapcount:0 nr_pages_mapped:0
pincount:0
flags: 0xfffffc0010200(slab|head|node=0|zone=1|lastcpupid=0x1fffff)
page_type: 0xffffffff()
raw: 000fffffc0010200 ffff888100042dc0 ffffea000426de00 dead000000000002
raw: 0000000000000000 0000000000100010 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected
Memory state around the buggy address:
ffff8880beab8880: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
ffff8880beab8900: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>ffff8880beab8980: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
^
ffff8880beab8a00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
ffff8880beab8a80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
==================================================================
Disabling lock debugging due to kernel taint
Fixes: ac751efa6a0d ("console: rename acquire/release_console_sem() to console_lock/unlock()")
Reported-by: syzkaller <[email protected]>
Signed-off-by: George Kennedy <[email protected]>
---
This patch is similar to the recent UAF vcs_read() patches (226fae124b2d & 46d733d0efc7).
vcs_write() should have been checked for the same UAF issue at the same time the fixes went
into vcs_read().
drivers/tty/vt/vc_screen.c | 10 ++++++++--
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/drivers/tty/vt/vc_screen.c b/drivers/tty/vt/vc_screen.c
index 498ba9c0ee93..cbd534a5cead 100644
--- a/drivers/tty/vt/vc_screen.c
+++ b/drivers/tty/vt/vc_screen.c
@@ -656,10 +656,16 @@ vcs_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos)
}
}
- /* The vcs_size might have changed while we slept to grab
- * the user buffer, so recheck.
+ /* The vc might have been freed or vcs_size might have changed
+ * while we slept to grab the user buffer, so recheck.
* Return data written up to now on failure.
*/
+ vc = vcs_vc(inode, &viewed);
+ if (!vc) {
+ if (written)
+ break;
+ goto unlock_out;
+ }
size = vcs_size(vc, attr, false);
if (size < 0) {
if (written)
--
2.31.1
Hi George,
kernel test robot noticed the following build warnings:
https://git-scm.com/docs/git-format-patch#_base_tree_information]
url: https://github.com/intel-lab-lkp/linux/commits/George-Kennedy/vc_screen-reload-load-of-struct-vc_data-pointer-in-vcs_write-to-avoid-UAF/20230510-031202
base: https://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty.git tty-testing
patch link: https://lore.kernel.org/r/1683659319-29701-1-git-send-email-george.kennedy%40oracle.com
patch subject: [PATCH] vc_screen: reload load of struct vc_data pointer in vcs_write() to avoid UAF
config: ia64-randconfig-m031-20230509 (https://download.01.org/0day-ci/archive/20230510/[email protected]/config)
compiler: ia64-linux-gcc (GCC) 12.1.0
If you fix the issue, kindly add following tag where applicable
| Reported-by: kernel test robot <[email protected]>
| Reported-by: Dan Carpenter <[email protected]>
| Link: https://lore.kernel.org/r/[email protected]/
smatch warnings:
drivers/tty/vt/vc_screen.c:707 vcs_write() warn: maybe return -EFAULT instead of the bytes remaining?
vim +707 drivers/tty/vt/vc_screen.c
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 589 static ssize_t
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 590 vcs_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos)
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 591 {
496ad9aa8ef448 drivers/tty/vt/vc_screen.c Al Viro 2013-01-23 592 struct inode *inode = file_inode(file);
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 593 struct vc_data *vc;
95e0d57fd49780 drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 594 char *con_buf;
95e0d57fd49780 drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 595 u16 *org0, *org;
95e0d57fd49780 drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 596 unsigned int written;
2665bef49f96df drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 597 int size;
2665bef49f96df drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 598 ssize_t ret;
2665bef49f96df drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 599 loff_t pos;
2665bef49f96df drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 600 bool viewed, attr;
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 601
0c9acb1af77a3c drivers/tty/vt/vc_screen.c Nicolas Pitre 2019-11-05 602 if (use_unicode(inode))
0c9acb1af77a3c drivers/tty/vt/vc_screen.c Nicolas Pitre 2019-11-05 603 return -EOPNOTSUPP;
0c9acb1af77a3c drivers/tty/vt/vc_screen.c Nicolas Pitre 2019-11-05 604
fcdba07ee390d9 drivers/tty/vt/vc_screen.c Jiri Olsa 2011-02-07 605 con_buf = (char *) __get_free_page(GFP_KERNEL);
fcdba07ee390d9 drivers/tty/vt/vc_screen.c Jiri Olsa 2011-02-07 606 if (!con_buf)
fcdba07ee390d9 drivers/tty/vt/vc_screen.c Jiri Olsa 2011-02-07 607 return -ENOMEM;
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 608
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 609 pos = *ppos;
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 610
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 611 /* Select the proper current console and verify
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 612 * sanity of the situation under the console lock.
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 613 */
ac751efa6a0d70 drivers/tty/vt/vc_screen.c Torben Hohn 2011-01-25 614 console_lock();
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 615
d21b0be246bf3b drivers/tty/vt/vc_screen.c Nicolas Pitre 2018-06-26 616 attr = use_attributes(inode);
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 617 ret = -ENXIO;
fcdba07ee390d9 drivers/tty/vt/vc_screen.c Jiri Olsa 2011-02-07 618 vc = vcs_vc(inode, &viewed);
fcdba07ee390d9 drivers/tty/vt/vc_screen.c Jiri Olsa 2011-02-07 619 if (!vc)
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 620 goto unlock_out;
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 621
71d4abfab322e8 drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 622 size = vcs_size(vc, attr, false);
71d4abfab322e8 drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 623 if (size < 0) {
71d4abfab322e8 drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 624 ret = size;
71d4abfab322e8 drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 625 goto unlock_out;
71d4abfab322e8 drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 626 }
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 627 ret = -EINVAL;
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 628 if (pos < 0 || pos > size)
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 629 goto unlock_out;
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 630 if (count > size - pos)
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 631 count = size - pos;
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 632 written = 0;
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 633 while (count) {
95e0d57fd49780 drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 634 unsigned int this_round = count;
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 635
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 636 if (this_round > CON_BUF_SIZE)
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 637 this_round = CON_BUF_SIZE;
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 638
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 639 /* Temporarily drop the console lock so that we can read
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 640 * in the write data from userspace safely.
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 641 */
ac751efa6a0d70 drivers/tty/vt/vc_screen.c Torben Hohn 2011-01-25 642 console_unlock();
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 643 ret = copy_from_user(con_buf, buf, this_round);
ac751efa6a0d70 drivers/tty/vt/vc_screen.c Torben Hohn 2011-01-25 644 console_lock();
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 645
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 646 if (ret) {
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 647 this_round -= ret;
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 648 if (!this_round) {
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 649 /* Abort loop if no data were copied. Otherwise
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 650 * fail with -EFAULT.
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 651 */
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 652 if (written)
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 653 break;
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 654 ret = -EFAULT;
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 655 goto unlock_out;
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 656 }
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 657 }
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 658
50f4095ab8ede7 drivers/tty/vt/vc_screen.c George Kennedy 2023-05-09 659 /* The vc might have been freed or vcs_size might have changed
50f4095ab8ede7 drivers/tty/vt/vc_screen.c George Kennedy 2023-05-09 660 * while we slept to grab the user buffer, so recheck.
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 661 * Return data written up to now on failure.
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 662 */
50f4095ab8ede7 drivers/tty/vt/vc_screen.c George Kennedy 2023-05-09 663 vc = vcs_vc(inode, &viewed);
50f4095ab8ede7 drivers/tty/vt/vc_screen.c George Kennedy 2023-05-09 664 if (!vc) {
50f4095ab8ede7 drivers/tty/vt/vc_screen.c George Kennedy 2023-05-09 665 if (written)
50f4095ab8ede7 drivers/tty/vt/vc_screen.c George Kennedy 2023-05-09 666 break;
50f4095ab8ede7 drivers/tty/vt/vc_screen.c George Kennedy 2023-05-09 667 goto unlock_out;
The warning message is wrong, but there does need to be a
"ret = something" here. Either an error code or ret = size on success.
50f4095ab8ede7 drivers/tty/vt/vc_screen.c George Kennedy 2023-05-09 668 }
71d4abfab322e8 drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 669 size = vcs_size(vc, attr, false);
dc1892c4bc6960 drivers/tty/vt/vc_screen.c Jiri Olsa 2011-02-07 670 if (size < 0) {
dc1892c4bc6960 drivers/tty/vt/vc_screen.c Jiri Olsa 2011-02-07 671 if (written)
dc1892c4bc6960 drivers/tty/vt/vc_screen.c Jiri Olsa 2011-02-07 672 break;
dc1892c4bc6960 drivers/tty/vt/vc_screen.c Jiri Olsa 2011-02-07 673 ret = size;
dc1892c4bc6960 drivers/tty/vt/vc_screen.c Jiri Olsa 2011-02-07 674 goto unlock_out;
dc1892c4bc6960 drivers/tty/vt/vc_screen.c Jiri Olsa 2011-02-07 675 }
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 676 if (pos >= size)
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 677 break;
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 678 if (this_round > size - pos)
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 679 this_round = size - pos;
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 680
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 681 /* OK, now actually push the write to the console
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 682 * under the lock using the local kernel buffer.
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 683 */
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 684
95e0d57fd49780 drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 685 if (attr)
95e0d57fd49780 drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 686 org = vcs_write_buf(vc, con_buf, pos, this_round,
95e0d57fd49780 drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 687 viewed, &org0);
95e0d57fd49780 drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 688 else
9e6363784e3f1e drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 689 org = vcs_write_buf_noattr(vc, con_buf, pos, this_round,
9e6363784e3f1e drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 690 viewed, &org0);
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 691
95e0d57fd49780 drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 692 count -= this_round;
95e0d57fd49780 drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 693 written += this_round;
95e0d57fd49780 drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 694 buf += this_round;
95e0d57fd49780 drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 695 pos += this_round;
95e0d57fd49780 drivers/tty/vt/vc_screen.c Jiri Slaby 2020-08-18 696 if (org)
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 697 update_region(vc, (unsigned long)(org0), org - org0);
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 698 }
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 699 *ppos += written;
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 700 ret = written;
432c9ed22aff64 drivers/char/vc_screen.c Nicolas Pitre 2010-10-01 701 if (written)
432c9ed22aff64 drivers/char/vc_screen.c Nicolas Pitre 2010-10-01 702 vcs_scr_updated(vc);
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 703
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 704 unlock_out:
ac751efa6a0d70 drivers/tty/vt/vc_screen.c Torben Hohn 2011-01-25 705 console_unlock();
fcdba07ee390d9 drivers/tty/vt/vc_screen.c Jiri Olsa 2011-02-07 706 free_page((unsigned long) con_buf);
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 @707 return ret;
^1da177e4c3f41 drivers/char/vc_screen.c Linus Torvalds 2005-04-16 708 }
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests