Re: [PATCH v3 3/4] fbmem: Check screen resolution change against font size

[Helge Deller <deller@gmx.de>]

On 6/26/22 10:49, Daniel Vetter wrote:
> On Sun, Jun 26, 2022 at 01:56:18AM +0200, Helge Deller wrote:
>> On 6/26/22 01:12, Helge Deller wrote:
>>> On 6/26/22 00:37, Daniel Vetter wrote:
>>>> On Sun, Jun 26, 2022 at 12:33:53AM +0200, Daniel Vetter wrote:
>>>>> On Sun, Jun 26, 2022 at 12:06:29AM +0200, Helge Deller wrote:
>>>>>> Enhance the check in the FBIOPUT_VSCREENINFO ioctl handler to verify if the
>>>>>> user-provided new screen size is bigger than the current font size.
>>>>>>
>>>>>> Signed-off-by: Helge Deller <deller@gmx.de>
>>>>>> Cc: stable@vger.kernel.org # v5.4+
>>>>>
>>>>> Please squash with previous patch. You might also want to add a note there
>>>>> that on older kernels backporters need to open-code
>>>>> fbcon_info_from_console instead, since it only exists since
>>>>> 409d6c95f9c6 ("fbcon: Introduce wrapper for console->fb_info lookup")
>>>>
>>>> Maybe easier would be to include that patch in the backports instead of
>>>> open coding.
>>>
>>> I was afraid that WARN_CONSOLE_UNLOCKED() hadn't been backported.
>>> But it seems it's in v4.19+ (from patch 56e6c104e4f15), so that's ok.
>>>
>>> So, yes, it seems pushing 409d6c95f9c6 backwards is probably best.
>>
>> It would be the best solution, but sadly 409d6c95f9c6 can't easily be backported.
>> So, probably my other approach (fix up afterwards with extra patch) is
>> the way to go.
>
> Ah right there's some conflicts with the restoration/removal of scroll
> accel.
>
>> What's your thought on this ?
>
> I guess just open code in a separate backport is simplest.

I think my just-sent series is somewhat smarter... use old open-coding
in patch which goes backwards, and then just fix up in v5.19 (where
commit 409d6c95f9c6 was added).

Helge



> -Daniel
>
>>
>> Helge
>>
>>
>>
>>> Will try that approach now.
>>>
>>> Helge
>>>
>>>  I think that's what Greg generally prefers at least, less
>>>> divergence between stable kernels.
>>>> -Daniel
>>>>
>>>>>
>>>>> With these two nits: Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
>>>>>
>>>>>> ---
>>>>>>  drivers/video/fbdev/core/fbmem.c | 4 +++-
>>>>>>  1 file changed, 3 insertions(+), 1 deletion(-)
>>>>>>
>>>>>> diff --git a/drivers/video/fbdev/core/fbmem.c b/drivers/video/fbdev/core/fbmem.c
>>>>>> index afa2863670f3..160389365a36 100644
>>>>>> --- a/drivers/video/fbdev/core/fbmem.c
>>>>>> +++ b/drivers/video/fbdev/core/fbmem.c
>>>>>> @@ -1106,7 +1106,9 @@ static long do_fb_ioctl(struct fb_info *info, unsigned int cmd,
>>>>>>  			return -EFAULT;
>>>>>>  		console_lock();
>>>>>>  		lock_fb_info(info);
>>>>>> -		ret = fb_set_var(info, &var);
>>>>>> +		ret = fbcon_modechange_possible(info, &var);
>>>>>> +		if (!ret)
>>>>>> +			ret = fb_set_var(info, &var);
>>>>>>  		if (!ret)
>>>>>>  			fbcon_update_vcs(info, var.activate & FB_ACTIVATE_ALL);
>>>>>>  		unlock_fb_info(info);
>>>>>> --
>>>>>> 2.35.3
>>>>>>
>>>>>
>>>>> --
>>>>> Daniel Vetter
>>>>> Software Engineer, Intel Corporation
>>>>> http://blog.ffwll.ch
>>>>
>>>
>>
>