FWIW, I’m working a lot with Macs, both Intel-based and Apple Silicon-based.
In my experience, it’s been never useful to build “fat” binaries. They brought nothing but problems.
In our place we build x86_64 binaries for Intel, and AARCH64 for Silicon - and is been perfectly satisfactory here. I recommend considering the same (proven) approach.
We considered this approach. For context, the software I'm working on is a commercial closed-source application library, containing only .so/.dylib/.dll files, delivered in a SDK+Runtime package. Our customers integrate our software into their application and that gets deployed to either the cloud or on-prem. Originally we planned to have 2 packages for Mac: 1 for Intel, 1 for Apple Silicon. But as we thought about the use case of deploying to Mac endpoints, it seemed like Universal binaries would be easier (ie. "One size fits all").
Your experience is interesting to me. Can you elaborate on the problems you've faced with Universal ("fat") binaries?
Regards,
Nick
You received this message because you are subscribed to the Google Groups "openssl-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openssl-users+unsubscribe@xxxxxxxxxxx.
To view this discussion visit https://groups.google.com/a/openssl.org/d/msgid/openssl-users/a775e722-60a0-2179-5553-6bfeb8497d66%40codesniffer.com.
