Thanks for the review . >While looking at the code I saw that we escape two critical arrays via >BufImg_GetRasInfo(). >I guess its okay if all caller code to GetRasInfo() in the JDK are properly >closed with the associated release function; still it made me a bit >apprehensive.
I hope too that the callers of GetRasInfo always call the release function 😊 , I briefly checked and it looked okay to me . Best regards, Matthias From: Thomas Stüfe <thomas.stu...@gmail.com> Sent: Dienstag, 10. September 2019 08:17 To: Baesken, Matthias <matthias.baes...@sap.com> Cc: 2d-dev@openjdk.java.net Subject: Re: [OpenJDK 2D-Dev] RFR [XS] 8230769: BufImg_SetupICM add ReleasePrimitiveArrayCritical call in early return Looks good Matthias. While looking at the code I saw that we escape two critical arrays via BufImg_GetRasInfo(). I guess its okay if all caller code to GetRasInfo() in the JDK are properly closed with the associated release function; still it made me a bit apprehensive. Cheers, Thomas On Mon, Sep 9, 2019 at 4:52 PM Baesken, Matthias <matthias.baes...@sap.com<mailto:matthias.baes...@sap.com>> wrote: Hello, please review the following webrev . I noticed that in function BufImg_SetupICM, in case of initCubemap returning NULL we currently miss a ReleasePrimitiveArrayCritical call; this should be added. Bug/webrev : https://bugs.openjdk.java.net/browse/JDK-8230769 http://cr.openjdk.java.net/~mbaesken/webrevs/8230769.0/ Thanks, Matthias
