On 06/05/2022 01:22, Jason Hatton wrote: > I believe it would be best to only change the behavior of files that are > multiples of 2^32 exactly. Changing the behavior of all files larger than > 4GBi may not be good. I like the idea of using 0x80000000 instead of 1. I think, having had a further think, that the 0x80000000 would be a better balance as it would, as you say, only affect exact multiples of 4GiB, and fix the 'racy' confusion. > I have a patch file, but I'm not sure how to actually submit it. I'm going to > attempt using outlook. > > Jason You maybe able to use the `git format-patch` to generate an easily applied patch, and `git send-email` to post it to the list without damage from outlook/gmail/whatever. Se the bottom part of https://public-inbox.org/git/CY4PR16MB165558026798A6E51E6A005FAFC59@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx/ which has "Reply instructions:" or just start with a simple .patch or .txt attachment (or in-line) of the diff to get a good visual. The full discussion about submitting patches is at https://github.com/git/git/blob/master/Documentation/SubmittingPatches (and other hosts), but don't let perfect be the enemy of useful; -- Philip
