RFR: 8166727: javac crashed: [jimage.dll+0x1942] ImageStrings::find+0x28
Jim Laskey
james.laskey at oracle.com
Sun Apr 4 01:20:15 UTC 2021
I’ll remove the code and elucidate in the commentary.
> On Apr 3, 2021, at 8:08 PM, David Holmes <David.Holmes at oracle.com> wrote:
>
> On 4/04/2021 12:02 am, Jim Laskey wrote:
>> Is it in bad form to declare a #define CLOSE_JIMAGE 0 ?
>
> Still dead code.
>
> This is a core-libs style call anyway.
>
> Cheers,
> David
>
>>
>>>> On Apr 3, 2021, at 9:49 AM, David Holmes <David.Holmes at oracle.com> wrote:
>>>
>>> On 2/04/2021 5:34 pm, Alan Bateman wrote:
>>>> On Thu, 1 Apr 2021 18:48:15 GMT, Jim Laskey <jlaskey at openjdk.org> wrote:
>>>>>> src/java.base/share/native/libjimage/imageFile.cpp line 219:
>>>>>>
>>>>>>> 217: // WARNING: Should never close the jimage file.
>>>>>>> 218: // Threads may still be running at shutdown.
>>>>>>> 219: #if 0
>>>>>>
>>>>>> Are you keeping the code in order to re-visit it again? Just wondering why it's not deleted.
>>>>>
>>>>> Leaving the code as an example of what is required to close in case the topic gets revisited in the future and I get hit by a bus or dementia.
>>>> Okay although I assume someone will spot this and be tempted to remove it.
>>>
>>> I didn't comment on this as I assumed it would contravene core-libs coding guidelines. I'm suprised to see dead code kept this way (there are existing cases elsewhere but it isn't considered good form). Previous code can always be retrieved from the repo history.
>>>
>>> Cheers,
>>> David
>>>
>>>> -------------
>>>> PR: https://git.openjdk.java.net/jdk/pull/3304
More information about the core-libs-dev
mailing list