Fix alignment value for Aligned512M.

[tobz]

Simple update to the alignment parameter for the generated Aligned512M struct.

[oliver-giersch]

Thanks for your contribution, I am a bit conflicting about what I want to do with the align utility, though. Apparently, there are some issues with using #[repr(align(X))] with values greater than the OS page size (when used with static values, I believe) and there was some talk about restricting values to 4K. I am not sure what has become of that, but I had been planning to remove all implementations above 4K for version 0.3.0, which I have had lying around for some time now.

Out of interest, do you have a particular use case for needing an alignment of 512M?

[tobz]

No use for 512M at all, just happened to see that it was the wrong value. Fine with you merging or closing, to be honest. :)

…

On Tue, Oct 27, 2020, 3:56 AM oliver-giersch ***@***.***> wrote: Thanks for your contribution, I am a bit conflicting about what I want to do with the align utility, though. Apparently, there are some issues <rust-lang/rust#70193> with using #[repr(align(X))] with values greater than the OS page size (when used with static values, I believe) and there was some talk about restricting values to 4K. I am not sure what has become of that, but I had been planning to remove all implementations above 4K for version 0.3.0, which I have had lying around for some time now. Out of interest, do you have a particular use case for needing an alignment of 512M? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#1 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AABWLFZQIYC4PZ4WNM23MALSMZ4KNANCNFSM4S6LWV5Q> .

[oliver-giersch]

From what I've read, the plans for alignment restrictions are canned for now, so I think I'm going to keep huge alignments for the 0.3.0 release and maybe mention the issues involving statics. Thanks for the fix!