[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: background_image fails to display image when timeout_style=hidden
From: |
John Z. Bohach |
Subject: |
Re: background_image fails to display image when timeout_style=hidden |
Date: |
Mon, 20 Feb 2017 09:16:49 -0800 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:43.0) Gecko/20100101 Firefox/43.0 SeaMonkey/2.40 |
>> Could someone please confirm if this behavior is what they also see:
>>
>> I have a background image set, and is displayed as expected, _AS LONG AS_
>>
>> 'set timeout_style=hidden'
>>
>> is NOT set. In other words, if I have
>>
>> 'set timeout_style=countdown'
>>
>> or nothing (meaning 'set timeout_style=menu') in my grub.cfg, background
>> works
>> as expected. If I change _only_ the value, to "hidden", so it becomes
>>
>> 'set timeout_style=hidden'
>>
>> the background image is not displayed. No other change is necessary to
>> demonstrate this behavior.
>>
>> This is on i386-pc installation.
>>
>> Is this an intentional design choice (a poor choice IMHO if it was), or is it
>> a bug, or something else?
>>
>
> Given that "hidden" timeout style counts down *before* displaying menu,
> I tend to think it is intentional. Do you see normal picture after
> pressing ESC?
Yes, it does display when 'ESC' is pressed. But displaying the 'menu' vs.
displaying the 'background' is not the same thing.
I rather hoped the two would be independent. Besides, what does "countdown"
have to do with whether or not the background is shown? This is a rather
inconsistent implementation. To wit, if there is a background image loaded,
it displays regardless of timeout_style=[countdown|menu]. Why should "hidden"
receive different treatment?
I mean, I know why, roughly. The screen rendering isn't called until
something is displayed, either menu, or countdown, and background by itself
isn't enough. But that is a bug. If I didn't want a background even in the
"hidden" case, I would not have set one. If I could find out where such
rendering takes place, I would patch it myself. But its a rather large chunk
of code to dig through, and for now, I don't have the time...
Thanks,
Z.