[PATCH] comedi: dt2815: add hardware detection to prevent crash

[PATCH] comedi: dt2815: add hardware detection to prevent crash

[Deepanshu Kartikey]

The dt2815 driver crashes when attached to I/O ports without actual
hardware present. This occurs because syzkaller or users can attach
the driver to arbitrary I/O addresses via COMEDI_DEVCONFIG ioctl.

When no hardware exists at the specified port, inb() operations return
0xff (floating bus), but outb() operations can trigger page faults due
to undefined behavior, especially under race conditions:

  BUG: unable to handle page fault for address: 000000007fffff90
  #PF: supervisor write access in kernel mode
  #PF: error_code(0x0002) - not-present page
  RIP: 0010:dt2815_attach+0x6e0/0x1110

Add hardware detection by reading the status register before attempting
any write operations. If the read returns 0xff, assume no hardware is
present and fail the attach with -ENODEV. This prevents crashes from
outb() operations on non-existent hardware.

Reported-by: syzbot+72f94b474d6e50b71ffc@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=72f94b474d6e50b71ffc
Tested-by: syzbot+72f94b474d6e50b71ffc@syzkaller.appspotmail.com
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Deepanshu Kartikey <kartikey406@gmail.com>
---
 drivers/comedi/drivers/dt2815.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/comedi/drivers/dt2815.c b/drivers/comedi/drivers/dt2815.c
index 03ba2fd18a21..7c642860f127 100644
--- a/drivers/comedi/drivers/dt2815.c
+++ b/drivers/comedi/drivers/dt2815.c
@@ -175,6 +175,18 @@ static int dt2815_attach(struct comedi_device *dev, struct comedi_devconfig *it)
 		    ? current_range_type : voltage_range_type;
 	}
 
+	/*
+	 * Check if hardware is present before attempting any I/O operations.
+	 * Reading 0xff from status register typically indicates no hardware
+	 * on the bus (floating bus reads as all 1s).
+	 */
+	if (inb(dev->iobase + DT2815_STATUS) == 0xff) {
+		dev_err(dev->class_dev,
+			"No hardware detected at I/O base 0x%lx\n",
+			dev->iobase);
+		return -ENODEV;
+	}
+
 	/* Init the 2815 */
 	outb(0x00, dev->iobase + DT2815_STATUS);
 	for (i = 0; i < 100; i++) {
-- 
2.43.0

Re: [PATCH] comedi: dt2815: add hardware detection to prevent crash

[Deepanshu Kartikey]

On Mon, Mar 9, 2026 at 4:19 PM Deepanshu Kartikey <kartikey406@gmail.com> wrote:
>
> The dt2815 driver crashes when attached to I/O ports without actual
> hardware present. This occurs because syzkaller or users can attach
> the driver to arbitrary I/O addresses via COMEDI_DEVCONFIG ioctl.
>
> When no hardware exists at the specified port, inb() operations return
> 0xff (floating bus), but outb() operations can trigger page faults due
> to undefined behavior, especially under race conditions:
>
>   BUG: unable to handle page fault for address: 000000007fffff90
>   #PF: supervisor write access in kernel mode
>   #PF: error_code(0x0002) - not-present page
>   RIP: 0010:dt2815_attach+0x6e0/0x1110
>
> Add hardware detection by reading the status register before attempting
> any write operations. If the read returns 0xff, assume no hardware is
> present and fail the attach with -ENODEV. This prevents crashes from
> outb() operations on non-existent hardware.
>
> Reported-by: syzbot+72f94b474d6e50b71ffc@syzkaller.appspotmail.com
> Closes: https://syzkaller.appspot.com/bug?extid=72f94b474d6e50b71ffc
> Tested-by: syzbot+72f94b474d6e50b71ffc@syzkaller.appspotmail.com
> Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
> Signed-off-by: Deepanshu Kartikey <kartikey406@gmail.com>
> ---
>  drivers/comedi/drivers/dt2815.c | 12 ++++++++++++
>  1 file changed, 12 insertions(+)
>
> diff --git a/drivers/comedi/drivers/dt2815.c b/drivers/comedi/drivers/dt2815.c
> index 03ba2fd18a21..7c642860f127 100644
> --- a/drivers/comedi/drivers/dt2815.c
> +++ b/drivers/comedi/drivers/dt2815.c
> @@ -175,6 +175,18 @@ static int dt2815_attach(struct comedi_device *dev, struct comedi_devconfig *it)
>                     ? current_range_type : voltage_range_type;
>         }
>
> +       /*
> +        * Check if hardware is present before attempting any I/O operations.
> +        * Reading 0xff from status register typically indicates no hardware
> +        * on the bus (floating bus reads as all 1s).
> +        */
> +       if (inb(dev->iobase + DT2815_STATUS) == 0xff) {
> +               dev_err(dev->class_dev,
> +                       "No hardware detected at I/O base 0x%lx\n",
> +                       dev->iobase);
> +               return -ENODEV;
> +       }
> +
>         /* Init the 2815 */
>         outb(0x00, dev->iobase + DT2815_STATUS);
>         for (i = 0; i < 100; i++) {
> --
> 2.43.0
>


Hi Greg,

Gentle ping on this patch. This patch was submitted on 26th jan 2026.
It. It is  Reviewed-by: Ian Abbott <abbotti@mev.co.uk>

Subject: [PATCH] comedi: dt2815: add hardware
detection to prevent crash
Link: [https://lore.kernel.org/all/20260126070458.10974-1-kartikey406@gmail.com/T/]

Let me know if any changes required.

Thanks

Re: [PATCH] comedi: dt2815: add hardware detection to prevent crash

[Deepanshu Kartikey]

On Sun, Mar 15, 2026 at 7:48 AM Deepanshu Kartikey
<kartikey406@gmail.com> wrote:
>
> On Mon, Mar 9, 2026 at 4:19 PM Deepanshu Kartikey <kartikey406@gmail.com> wrote:
> >
> > The dt2815 driver crashes when attached to I/O ports without actual
> > hardware present. This occurs because syzkaller or users can attach
> > the driver to arbitrary I/O addresses via COMEDI_DEVCONFIG ioctl.
> >
> > When no hardware exists at the specified port, inb() operations return
> > 0xff (floating bus), but outb() operations can trigger page faults due
> > to undefined behavior, especially under race conditions:
> >
> >   BUG: unable to handle page fault for address: 000000007fffff90
> >   #PF: supervisor write access in kernel mode
> >   #PF: error_code(0x0002) - not-present page
> >   RIP: 0010:dt2815_attach+0x6e0/0x1110
> >
> > Add hardware detection by reading the status register before attempting
> > any write operations. If the read returns 0xff, assume no hardware is
> > present and fail the attach with -ENODEV. This prevents crashes from
> > outb() operations on non-existent hardware.
> >
> > Reported-by: syzbot+72f94b474d6e50b71ffc@syzkaller.appspotmail.com
> > Closes: https://syzkaller.appspot.com/bug?extid=72f94b474d6e50b71ffc
> > Tested-by: syzbot+72f94b474d6e50b71ffc@syzkaller.appspotmail.com
> > Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
> > Signed-off-by: Deepanshu Kartikey <kartikey406@gmail.com>
> > ---
> >  drivers/comedi/drivers/dt2815.c | 12 ++++++++++++
> >  1 file changed, 12 insertions(+)
> >
> > diff --git a/drivers/comedi/drivers/dt2815.c b/drivers/comedi/drivers/dt2815.c
> > index 03ba2fd18a21..7c642860f127 100644
> > --- a/drivers/comedi/drivers/dt2815.c
> > +++ b/drivers/comedi/drivers/dt2815.c
> > @@ -175,6 +175,18 @@ static int dt2815_attach(struct comedi_device *dev, struct comedi_devconfig *it)
> >                     ? current_range_type : voltage_range_type;
> >         }
> >
> > +       /*
> > +        * Check if hardware is present before attempting any I/O operations.
> > +        * Reading 0xff from status register typically indicates no hardware
> > +        * on the bus (floating bus reads as all 1s).
> > +        */
> > +       if (inb(dev->iobase + DT2815_STATUS) == 0xff) {
> > +               dev_err(dev->class_dev,
> > +                       "No hardware detected at I/O base 0x%lx\n",
> > +                       dev->iobase);
> > +               return -ENODEV;
> > +       }
> > +
> >         /* Init the 2815 */
> >         outb(0x00, dev->iobase + DT2815_STATUS);
> >         for (i = 0; i < 100; i++) {
> > --
> > 2.43.0
> >
>
>
Hi Greg,
Gentle ping on this patch  . I submitted this patch on 26th jan 2026.
It is reviewed by : Reviewed-by: Ian Abbott <abbotti@mev.co.uk>

Please let me know if anything else required to get it merged

Thanks

Deepanshu

[PATCH] comedi: dt2815: add hardware detection to prevent crash

[Deepanshu Kartikey]

The dt2815 driver crashes when attached to I/O ports without actual
hardware present. This occurs because syzkaller or users can attach
the driver to arbitrary I/O addresses via COMEDI_DEVCONFIG ioctl.

When no hardware exists at the specified port, inb() operations return
0xff (floating bus), but outb() operations can trigger page faults due
to undefined behavior, especially under race conditions:

  BUG: unable to handle page fault for address: 000000007fffff90
  #PF: supervisor write access in kernel mode
  #PF: error_code(0x0002) - not-present page
  RIP: 0010:dt2815_attach+0x6e0/0x1110

Add hardware detection by reading the status register before attempting
any write operations. If the read returns 0xff, assume no hardware is
present and fail the attach with -ENODEV. This prevents crashes from
outb() operations on non-existent hardware.

Reported-by: syzbot+72f94b474d6e50b71ffc@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=72f94b474d6e50b71ffc
Tested-by: syzbot+72f94b474d6e50b71ffc@syzkaller.appspotmail.com
Signed-off-by: Deepanshu Kartikey <kartikey406@gmail.com>
---
 drivers/comedi/drivers/dt2815.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/comedi/drivers/dt2815.c b/drivers/comedi/drivers/dt2815.c
index 03ba2fd18a21..7c642860f127 100644
--- a/drivers/comedi/drivers/dt2815.c
+++ b/drivers/comedi/drivers/dt2815.c
@@ -175,6 +175,18 @@ static int dt2815_attach(struct comedi_device *dev, struct comedi_devconfig *it)
 		    ? current_range_type : voltage_range_type;
 	}
 
+	/*
+	 * Check if hardware is present before attempting any I/O operations.
+	 * Reading 0xff from status register typically indicates no hardware
+	 * on the bus (floating bus reads as all 1s).
+	 */
+	if (inb(dev->iobase + DT2815_STATUS) == 0xff) {
+		dev_err(dev->class_dev,
+			"No hardware detected at I/O base 0x%lx\n",
+			dev->iobase);
+		return -ENODEV;
+	}
+
 	/* Init the 2815 */
 	outb(0x00, dev->iobase + DT2815_STATUS);
 	for (i = 0; i < 100; i++) {
-- 
2.43.0

Re: [PATCH] comedi: dt2815: add hardware detection to prevent crash

[Ian Abbott]

On 26/01/2026 07:04, Deepanshu Kartikey wrote:
> The dt2815 driver crashes when attached to I/O ports without actual
> hardware present. This occurs because syzkaller or users can attach
> the driver to arbitrary I/O addresses via COMEDI_DEVCONFIG ioctl.
> 
> When no hardware exists at the specified port, inb() operations return
> 0xff (floating bus), but outb() operations can trigger page faults due
> to undefined behavior, especially under race conditions:
> 
>    BUG: unable to handle page fault for address: 000000007fffff90
>    #PF: supervisor write access in kernel mode
>    #PF: error_code(0x0002) - not-present page
>    RIP: 0010:dt2815_attach+0x6e0/0x1110
> 
> Add hardware detection by reading the status register before attempting
> any write operations. If the read returns 0xff, assume no hardware is
> present and fail the attach with -ENODEV. This prevents crashes from
> outb() operations on non-existent hardware.
> 
> Reported-by: syzbot+72f94b474d6e50b71ffc@syzkaller.appspotmail.com
> Closes: https://syzkaller.appspot.com/bug?extid=72f94b474d6e50b71ffc
> Tested-by: syzbot+72f94b474d6e50b71ffc@syzkaller.appspotmail.com
> Signed-off-by: Deepanshu Kartikey <kartikey406@gmail.com>
> ---
>   drivers/comedi/drivers/dt2815.c | 12 ++++++++++++
>   1 file changed, 12 insertions(+)
> 
> diff --git a/drivers/comedi/drivers/dt2815.c b/drivers/comedi/drivers/dt2815.c
> index 03ba2fd18a21..7c642860f127 100644
> --- a/drivers/comedi/drivers/dt2815.c
> +++ b/drivers/comedi/drivers/dt2815.c
> @@ -175,6 +175,18 @@ static int dt2815_attach(struct comedi_device *dev, struct comedi_devconfig *it)
>   		    ? current_range_type : voltage_range_type;
>   	}
>   
> +	/*
> +	 * Check if hardware is present before attempting any I/O operations.
> +	 * Reading 0xff from status register typically indicates no hardware
> +	 * on the bus (floating bus reads as all 1s).
> +	 */
> +	if (inb(dev->iobase + DT2815_STATUS) == 0xff) {
> +		dev_err(dev->class_dev,
> +			"No hardware detected at I/O base 0x%lx\n",
> +			dev->iobase);
> +		return -ENODEV;
> +	}
> +
>   	/* Init the 2815 */
>   	outb(0x00, dev->iobase + DT2815_STATUS);
>   	for (i = 0; i < 100; i++) {

Thanks.  Looks good with some reservations.

It's probably fine for this device (assuming the status register cannot 
legitimately read back 0xff before the board has been initialized), but 
the same technique might not work for other devices (for example 8255, 
which has 4 registers, 3 of which can legitimately read 0xff, and the 
other being a write-only register).

In general, we assume the device is being configured correctly as 
configuration requires CAP_SYSADMIN privileges.

I think I will start adding some base address sanity checks to the 
drivers.  For example, The DT2815 hardware jumpers to configure the base 
address base addresses in the range 0x200 to 0x3FE aligned on 2-byte 
boundaries.

Reviewed-by: Ian Abbott <abbotti@mev.co.uk>

-- 
-=( Ian Abbott <abbotti@mev.co.uk> || MEV Ltd. is a company  )=-
-=( registered in England & Wales.  Regd. number: 02862268.  )=-
-=( Regd. addr.: S11 & 12 Building 67, Europa Business Park, )=-
-=( Bird Hall Lane, STOCKPORT, SK3 0XA, UK. || www.mev.co.uk )=-

Re: [PATCH] comedi: dt2815: add hardware detection to prevent crash

[Deepanshu Kartikey]

On Mon, Jan 26, 2026 at 6:34 PM Ian Abbott <abbotti@mev.co.uk> wrote:
>
> On 26/01/2026 07:04, Deepanshu Kartikey wrote:
> > The dt2815 driver crashes when attached to I/O ports without actual
> > hardware present. This occurs because syzkaller or users can attach
> > the driver to arbitrary I/O addresses via COMEDI_DEVCONFIG ioctl.
> >
> > When no hardware exists at the specified port, inb() operations return
> > 0xff (floating bus), but outb() operations can trigger page faults due
> > to undefined behavior, especially under race conditions:
> >
> >    BUG: unable to handle page fault for address: 000000007fffff90
> >    #PF: supervisor write access in kernel mode
> >    #PF: error_code(0x0002) - not-present page
> >    RIP: 0010:dt2815_attach+0x6e0/0x1110
> >
> > Add hardware detection by reading the status register before attempting
> > any write operations. If the read returns 0xff, assume no hardware is
> > present and fail the attach with -ENODEV. This prevents crashes from
> > outb() operations on non-existent hardware.
> >
> > Reported-by: syzbot+72f94b474d6e50b71ffc@syzkaller.appspotmail.com
> > Closes: https://syzkaller.appspot.com/bug?extid=72f94b474d6e50b71ffc
> > Tested-by: syzbot+72f94b474d6e50b71ffc@syzkaller.appspotmail.com
> > Signed-off-by: Deepanshu Kartikey <kartikey406@gmail.com>
> > ---
> >   drivers/comedi/drivers/dt2815.c | 12 ++++++++++++
> >   1 file changed, 12 insertions(+)
> >
> > diff --git a/drivers/comedi/drivers/dt2815.c b/drivers/comedi/drivers/dt2815.c
> > index 03ba2fd18a21..7c642860f127 100644
> > --- a/drivers/comedi/drivers/dt2815.c
> > +++ b/drivers/comedi/drivers/dt2815.c
> > @@ -175,6 +175,18 @@ static int dt2815_attach(struct comedi_device *dev, struct comedi_devconfig *it)
> >                   ? current_range_type : voltage_range_type;
> >       }
> >
> > +     /*
> > +      * Check if hardware is present before attempting any I/O operations.
> > +      * Reading 0xff from status register typically indicates no hardware
> > +      * on the bus (floating bus reads as all 1s).
> > +      */
> > +     if (inb(dev->iobase + DT2815_STATUS) == 0xff) {
> > +             dev_err(dev->class_dev,
> > +                     "No hardware detected at I/O base 0x%lx\n",
> > +                     dev->iobase);
> > +             return -ENODEV;
> > +     }
> > +
> >       /* Init the 2815 */
> >       outb(0x00, dev->iobase + DT2815_STATUS);
> >       for (i = 0; i < 100; i++) {
>
> Thanks.  Looks good with some reservations.
>
> It's probably fine for this device (assuming the status register cannot
> legitimately read back 0xff before the board has been initialized), but
> the same technique might not work for other devices (for example 8255,
> which has 4 registers, 3 of which can legitimately read 0xff, and the
> other being a write-only register).
>
> In general, we assume the device is being configured correctly as
> configuration requires CAP_SYSADMIN privileges.
>
> I think I will start adding some base address sanity checks to the
> drivers.  For example, The DT2815 hardware jumpers to configure the base
> address base addresses in the range 0x200 to 0x3FE aligned on 2-byte
> boundaries.
>
> Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
>

Hi Ian,
Gentle ping on the dt2815 hardware detection patch that you approved on

[26th jan 2026] Subject: [PATCH] comedi: dt2815: add hardware
detection to prevent crash
Link: [https://lore.kernel.org/all/20260126070458.10974-1-kartikey406@gmail.com/T/

It's been more than 30 days since approval.
Could you please let me know the status or if any further action is
needed from my side?

Thank you,
Deepanshu

Re: [PATCH] comedi: dt2815: add hardware detection to prevent crash

[Ian Abbott]

On 07/03/2026 00:59, Deepanshu Kartikey wrote:
> On Mon, Jan 26, 2026 at 6:34 PM Ian Abbott <abbotti@mev.co.uk> wrote:
>>
>> On 26/01/2026 07:04, Deepanshu Kartikey wrote:
>>> The dt2815 driver crashes when attached to I/O ports without actual
>>> hardware present. This occurs because syzkaller or users can attach
>>> the driver to arbitrary I/O addresses via COMEDI_DEVCONFIG ioctl.
>>>
>>> When no hardware exists at the specified port, inb() operations return
>>> 0xff (floating bus), but outb() operations can trigger page faults due
>>> to undefined behavior, especially under race conditions:
>>>
>>>     BUG: unable to handle page fault for address: 000000007fffff90
>>>     #PF: supervisor write access in kernel mode
>>>     #PF: error_code(0x0002) - not-present page
>>>     RIP: 0010:dt2815_attach+0x6e0/0x1110
>>>
>>> Add hardware detection by reading the status register before attempting
>>> any write operations. If the read returns 0xff, assume no hardware is
>>> present and fail the attach with -ENODEV. This prevents crashes from
>>> outb() operations on non-existent hardware.
>>>
>>> Reported-by: syzbot+72f94b474d6e50b71ffc@syzkaller.appspotmail.com
>>> Closes: https://syzkaller.appspot.com/bug?extid=72f94b474d6e50b71ffc
>>> Tested-by: syzbot+72f94b474d6e50b71ffc@syzkaller.appspotmail.com
>>> Signed-off-by: Deepanshu Kartikey <kartikey406@gmail.com>
>>> ---
>>>    drivers/comedi/drivers/dt2815.c | 12 ++++++++++++
>>>    1 file changed, 12 insertions(+)
>>>
>>> diff --git a/drivers/comedi/drivers/dt2815.c b/drivers/comedi/drivers/dt2815.c
>>> index 03ba2fd18a21..7c642860f127 100644
>>> --- a/drivers/comedi/drivers/dt2815.c
>>> +++ b/drivers/comedi/drivers/dt2815.c
>>> @@ -175,6 +175,18 @@ static int dt2815_attach(struct comedi_device *dev, struct comedi_devconfig *it)
>>>                    ? current_range_type : voltage_range_type;
>>>        }
>>>
>>> +     /*
>>> +      * Check if hardware is present before attempting any I/O operations.
>>> +      * Reading 0xff from status register typically indicates no hardware
>>> +      * on the bus (floating bus reads as all 1s).
>>> +      */
>>> +     if (inb(dev->iobase + DT2815_STATUS) == 0xff) {
>>> +             dev_err(dev->class_dev,
>>> +                     "No hardware detected at I/O base 0x%lx\n",
>>> +                     dev->iobase);
>>> +             return -ENODEV;
>>> +     }
>>> +
>>>        /* Init the 2815 */
>>>        outb(0x00, dev->iobase + DT2815_STATUS);
>>>        for (i = 0; i < 100; i++) {
>>
>> Thanks.  Looks good with some reservations.
>>
>> It's probably fine for this device (assuming the status register cannot
>> legitimately read back 0xff before the board has been initialized), but
>> the same technique might not work for other devices (for example 8255,
>> which has 4 registers, 3 of which can legitimately read 0xff, and the
>> other being a write-only register).
>>
>> In general, we assume the device is being configured correctly as
>> configuration requires CAP_SYSADMIN privileges.
>>
>> I think I will start adding some base address sanity checks to the
>> drivers.  For example, The DT2815 hardware jumpers to configure the base
>> address base addresses in the range 0x200 to 0x3FE aligned on 2-byte
>> boundaries.
>>
>> Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
>>
> 
> Hi Ian,
> Gentle ping on the dt2815 hardware detection patch that you approved on
> 
> [26th jan 2026] Subject: [PATCH] comedi: dt2815: add hardware
> detection to prevent crash
> Link: [https://lore.kernel.org/all/20260126070458.10974-1-kartikey406@gmail.com/T/
> 
> It's been more than 30 days since approval.
> Could you please let me know the status or if any further action is
> needed from my side?
> 
> Thank you,
> Deepanshu

Hi Deepanshu,

Please could you resend the patch and add Greg Kroah-Hartman to the Cc: 
list?  Greg generally commits Comedi patches to his "char-misc" 
repository on my behalf.  You can add my Reviewed-by: line at the same time.

Thanks,
Ian

-- 
-=( Ian Abbott <abbotti@mev.co.uk> || MEV Ltd. is a company  )=-
-=( registered in England & Wales.  Regd. number: 02862268.  )=-
-=( Regd. addr.: S11 & 12 Building 67, Europa Business Park, )=-
-=( Bird Hall Lane, STOCKPORT, SK3 0XA, UK. || www.mev.co.uk )=-