linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: zzoru <zzoru007@gmail.com>
To: stern@rowland.harvard.edu, gregkh@linuxfoundation.org,
	linux-usb@vger.kernel.org, usb-storage@lists.one-eyed-alien.net,
	LKML <linux-kernel@vger.kernel.org>
Cc: dokyungs@uci.edu, syzkaller <syzkaller@googlegroups.com>
Subject: WARNING in __alloc_pages_nodemask
Date: Wed, 16 Oct 2019 02:21:59 -0700	[thread overview]
Message-ID: <CALRZ7UuEBvMtL6Q3TOKjXdh9B4X5bcHUdbO7i20TmPShZHmKZg@mail.gmail.com> (raw)

We found the following crash on Linux 5.4-rc1 (Not patched in current
HEAD commit 3b1f00aceb7a67bf079a5a64aa5c6baf78a8f442)
with our customized syzkaller.

WARNING: CPU: 0 PID: 12 at linux/mm/page_alloc.c:4727
__alloc_pages_nodemask+0x333/0x760 linux/mm/page_alloc.c:4727
Kernel panic - not syncing: panic_on_warn set ...
CPU: 0 PID: 12 Comm: kworker/0:1 Not tainted 5.4.0-rc1
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.org 04/01/2014
Workqueue: usb_hub_wq hub_event
Call Trace:
 __dump_stack linux/lib/dump_stack.c:77 [inline]
 dump_stack+0x7c/0xbb linux/lib/dump_stack.c:113
 panic+0x205/0x48e linux/kernel/panic.c:220
 __warn+0x19c/0x1c0 linux/kernel/panic.c:581
 report_bug+0x1a9/0x2b0 linux/lib/bug.c:195
 fixup_bug.part.11+0x32/0x80 linux/arch/x86/kernel/traps.c:179
 fixup_bug linux/arch/x86/include/asm/irqflags.h:54 [inline]
 do_error_trap+0xf5/0x160 linux/arch/x86/kernel/traps.c:272
 do_invalid_op+0x31/0x40 linux/arch/x86/kernel/traps.c:291
 invalid_op+0x23/0x30 linux/arch/x86/entry/entry_64.S:1028
RIP: 0010:__alloc_pages_nodemask+0x333/0x760 linux/mm/page_alloc.c:4727
Code: fe ff ff 65 48 8b 04 25 00 ef 01 00 48 05 68 10 00 00 41 bc 01
00 00 00 48 89 44 24 40 e9 2a fe ff ff 81 e7 00 20 00 00 75 02 <0f> 0b
45 31 e4 e9 6e ff ff ff 41 f7 c6 00 00 40 00 0f 84 2c ff ff
RSP: 0018:ffff88801a1a6f60 EFLAGS: 00010246
RAX: 0000000000000000 RBX: 1ffff11003434def RCX: 0000000000000000
RDX: 0000000000000000 RSI: 0000000000000034 RDI: 0000000000000000
RBP: ffffffffffffff5e R08: ffffed1002d261a2 R09: dffffc0000000000
R10: 0000000000000001 R11: ffffed1002d261a1 R12: ffffffffffffff5e
R13: 0000000000000dc0 R14: ffff888017140000 R15: 0000000000000034
 alloc_pages linux/include/linux/gfp.h:509 [inline]
 kmalloc_order+0x16/0x50 linux/mm/slab_common.c:1294
 kmalloc_order_trace+0x18/0x110 linux/mm/slab_common.c:1306
 kmalloc linux/include/linux/slab.h:557 [inline]
 kzalloc linux/include/linux/slab.h:686 [inline]
 init_realtek_cr+0x2ec/0x1440 linux/drivers/usb/storage/realtek_cr.c:990
 usb_stor_acquire_resources linux/drivers/usb/storage/usb.c:784 [inline]
 usb_stor_probe2+0x738/0xbb0 linux/drivers/usb/storage/usb.c:1047
 realtek_cr_probe+0x9b/0xe0 linux/drivers/usb/storage/realtek_cr.c:1048
 usb_probe_interface+0x2b9/0x870 linux/drivers/usb/core/driver.c:361
 really_probe+0x444/0x830 linux/drivers/base/dd.c:548
 driver_probe_device+0xe3/0x250 linux/drivers/base/dd.c:721
 __device_attach_driver+0x17a/0x1e0 linux/drivers/base/dd.c:828
 bus_for_each_drv+0x143/0x1e0 linux/drivers/base/bus.c:430
 __device_attach+0x1f1/0x310 linux/drivers/base/dd.c:894
 bus_probe_device+0x1d7/0x280 linux/drivers/base/bus.c:490
 device_add+0xb72/0x1660 linux/drivers/base/core.c:2201
 usb_set_configuration+0xaca/0x1840 linux/drivers/usb/core/message.c:2023
 generic_probe+0x63/0xd0 linux/drivers/usb/core/generic.c:210
 usb_probe_device+0x89/0xd0 linux/drivers/usb/core/driver.c:266
 really_probe+0x444/0x830 linux/drivers/base/dd.c:548
 driver_probe_device+0xe3/0x250 linux/drivers/base/dd.c:721
 __device_attach_driver+0x17a/0x1e0 linux/drivers/base/dd.c:828
 bus_for_each_drv+0x143/0x1e0 linux/drivers/base/bus.c:430
 __device_attach+0x1f1/0x310 linux/drivers/base/dd.c:894
 bus_probe_device+0x1d7/0x280 linux/drivers/base/bus.c:490
 device_add+0xb72/0x1660 linux/drivers/base/core.c:2201
 usb_new_device+0x784/0x10d0 linux/drivers/usb/core/hub.c:2536
 hub_port_connect linux/drivers/usb/core/hub.c:5098 [inline]
 hub_port_connect_change linux/drivers/usb/core/hub.c:5213 [inline]
 port_event linux/drivers/usb/core/hub.c:5359 [inline]
 hub_event+0x18ed/0x3610 linux/drivers/usb/core/hub.c:5441
 process_one_work+0x8b5/0x17c0 linux/kernel/workqueue.c:2269
 worker_thread+0x82/0xb80 linux/kernel/workqueue.c:2415
 kthread+0x325/0x3e0 linux/kernel/kthread.c:255
 ret_from_fork+0x24/0x30 linux/arch/x86/entry/entry_64.S:352
Dumping ftrace buffer:
   (ftrace buffer empty)
Kernel Offset: disabled

Following is my analysis of this bug. (Just add some comments in source code):

realtek_cr.c:
struct rts51x_chip {
    u16 vendor_id;
    u16 product_id;
    char max_lun; // char -> signed 1byte
    ...
}

static int init_realtek_cr(struct us_data *us)
{
    int size, i, retval;
    ...
    us->max_lun = chip->max_lun = rts51x_get_max_lun(us); // From USB
device (-127~126) Malicious usb device can manipulate this value.

    size = (chip->max_lun + 1) * sizeof(struct rts51x_status); // Can
be negative integer or 0 (-1+1).

    chip->status = kzalloc(size, GFP_KERNEL); //  kzalloc(0) ==
ZERO_SIZE_PTR or kzalloc(negative integer) -> Warning

    ...

    for (i = 0; i <= (int)(chip->max_lun); i++) {
        retval = rts51x_check_status(us, (u8) i);
        ...
    }
}

tatic int rts51x_check_status(struct us_data *us, u8 lun)
{
    struct rts51x_chip *chip = (struct rts51x_chip *)(us->extra);
    int retval;
    u8 buf[16];

    retval = rts51x_read_status(us, lun, buf, 16, &(chip->status_len));
    if (retval != STATUS_SUCCESS)
        return -EIO;

    usb_stor_dbg(us, "chip->status_len = %d\n", chip->status_len);

    chip->status[lun].vid = ((u16) buf[0] << 8) | buf[1]; //
ZERO_SIZE_PTR DEREFERENCE :(
    chip->status[lun].pid = ((u16) buf[2] << 8) | buf[3];
    chip->status[lun].cur_lun = buf[4];
    chip->status[lun].card_type = buf[5];
    chip->status[lun].total_lun = buf[6];
    chip->status[lun].fw_ver = ((u16) buf[7] << 8) | buf[8];
    chip->status[lun].phy_exist = buf[9];
    chip->status[lun].multi_flag = buf[10];
    chip->status[lun].multi_card = buf[11];
    chip->status[lun].log_exist = buf[12];
    if (chip->status_len == 16) {
        chip->status[lun].detailed_type.detailed_type1 = buf[13];
        chip->status[lun].function[0] = buf[14];
        chip->status[lun].function[1] = buf[15];
    }
    return 0;
}

Need to change a declaration of max_lun. (u8)

             reply	other threads:[~2019-10-16  9:22 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-10-16  9:21 zzoru [this message]
2019-10-16 12:59 ` WARNING in __alloc_pages_nodemask Greg KH
  -- strict thread matches above, loose matches on Subject: below --
2019-09-16 18:49 syzbot
2019-09-16 20:00 ` David Rientjes
2019-09-16 20:57   ` Matthew Wilcox
2019-09-19 18:29     ` Andrey Konovalov

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=CALRZ7UuEBvMtL6Q3TOKjXdh9B4X5bcHUdbO7i20TmPShZHmKZg@mail.gmail.com \
    --to=zzoru007@gmail.com \
    --cc=dokyungs@uci.edu \
    --cc=gregkh@linuxfoundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-usb@vger.kernel.org \
    --cc=stern@rowland.harvard.edu \
    --cc=syzkaller@googlegroups.com \
    --cc=usb-storage@lists.one-eyed-alien.net \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).