Floppy ioctl range clamping fixes

Floppy ioctl range clamping fixes

[Linus Torvalds]

Hmm. I just realized when I saw Sasha's autoselect patches flying by
that the floppy ioctl fixes didn't get marked for stable, but they
probably should be.

There's four commits:

  da99466ac243 floppy: fix out-of-bounds read in copy_buffer
  9b04609b7840 floppy: fix invalid pointer dereference in drive_name
  5635f897ed83 floppy: fix out-of-bounds read in next_valid_format
  f3554aeb9912 floppy: fix div-by-zero in setup_format_params

that look like stable material - even if I sincerely hope that the
floppy driver isn't critical for anybody.

I leave it to the stable people to decide if they care. I don't think
the hardware matters any more, but I could imagine that people still
use it for some virtual images and have a floppy device inside a VM
for that reason.

                Linus

Re: Floppy ioctl range clamping fixes

[Greg KH]

On Thu, Jul 18, 2019 at 09:29:06PM -0700, Linus Torvalds wrote:
> Hmm. I just realized when I saw Sasha's autoselect patches flying by
> that the floppy ioctl fixes didn't get marked for stable, but they
> probably should be.
> 
> There's four commits:
> 
>   da99466ac243 floppy: fix out-of-bounds read in copy_buffer
>   9b04609b7840 floppy: fix invalid pointer dereference in drive_name
>   5635f897ed83 floppy: fix out-of-bounds read in next_valid_format
>   f3554aeb9912 floppy: fix div-by-zero in setup_format_params
> 
> that look like stable material - even if I sincerely hope that the
> floppy driver isn't critical for anybody.
> 
> I leave it to the stable people to decide if they care. I don't think
> the hardware matters any more, but I could imagine that people still
> use it for some virtual images and have a floppy device inside a VM
> for that reason.

Thanks for the reminder, I'll queue these up for the next round of
stable releases after the next ones go out in a day or so.

greg k-h

Re: Floppy ioctl range clamping fixes

[Sasha Levin]

On Fri, Jul 19, 2019 at 07:07:18PM +0200, Greg KH wrote:
>On Thu, Jul 18, 2019 at 09:29:06PM -0700, Linus Torvalds wrote:
>> Hmm. I just realized when I saw Sasha's autoselect patches flying by
>> that the floppy ioctl fixes didn't get marked for stable, but they
>> probably should be.
>>
>> There's four commits:
>>
>>   da99466ac243 floppy: fix out-of-bounds read in copy_buffer
>>   9b04609b7840 floppy: fix invalid pointer dereference in drive_name
>>   5635f897ed83 floppy: fix out-of-bounds read in next_valid_format
>>   f3554aeb9912 floppy: fix div-by-zero in setup_format_params
>>
>> that look like stable material - even if I sincerely hope that the
>> floppy driver isn't critical for anybody.
>>
>> I leave it to the stable people to decide if they care. I don't think
>> the hardware matters any more, but I could imagine that people still
>> use it for some virtual images and have a floppy device inside a VM
>> for that reason.
>
>Thanks for the reminder, I'll queue these up for the next round of
>stable releases after the next ones go out in a day or so.

I've queued them up, thanks!

--
Thanks,
Sasha