Re: RFR: 8185862: AWT Assertion Failure in ::GetDIBits(hBMDC, hBM, 0, 1, 0, gpBitmapInfo, 0) 'awt_Win32GraphicsDevice.cpp', at line 185 [v4]

[Alexey Ivanov]

On Fri, 16 Feb 2024 12:38:02 GMT, Christoph Langer <clan...@openjdk.org> wrote:

>> src/java.desktop/windows/native/libawt/windows/awt_Win32GraphicsDevice.cpp 
>> line 200:
>> 
>>> 198:     }
>>> 199: 
>>> 200:     if (colorData->bitsperpixel > 8) {
>> 
>> So at this time, the code below will not be run if an error occurs. This may 
>> lead to leaving the fields of `gpBitmapInfo` uninitialised.
>> 
>> I'm unsure how critical it is… even though we're (presumably) on a headless 
>> system.
>
> OK, yes, I'll turn this back to previous behavior. Rather don't touch 
> something at all if we have no idea... 😉

I prefer *backing out*… — it makes the code cleaner. At the same time, I am 
unsure about the consequences of doing so. I could be safer to run the code 
below, especially if you take into account that there's a fallback already:

https://github.com/openjdk/jdk/blob/fb90af881badf143163c7d0b9961152c2a12cd84/src/java.desktop/windows/native/libawt/windows/awt_Win32GraphicsDevice.cpp#L192-L199

-------------

PR Review Comment: https://git.openjdk.org/jdk/pull/17614#discussion_r1492417059