Removed aarch64 from start.sh scripts

[adamperkowski]

BEFORE MERGING THIS RECONSIDER ADDING ARM64 SUPPORT BACK #445 (comment)

Type of Change

• Bug fix

Description

Linutil no longer supports ARM64. It's misleading to have start.sh try to download it and just not work.
Not clearly showing "Unsupported architecture" leads to confusion, confusion leads to research, possibly opening more duplicate issues, and overall waste of time.

Issue related to PR

• Resolves Does not work for aarch64 ARM right now #444
• Resolves Does not work for aarch64 ARM #450

Checklist

• I have performed a self-review of my own code.
• My changes generate no errors/warnings/merge conflicts.

[nnyyxxxx]

I agree with this change, but there is another approach: Adding arrch64 ARM BACK in to linutil. I think this PR should be held off for now.

[adamperkowski]

but there is another approach: Adding arrch64 ARM BACK in to linutil. I think this PR should be held off for now.

I agree ARM support could be added back. Maintaining it is not so hard after all. I think @ChrisTitusTech sould decide.

[lj3954]

aarch64 was removed from the build workflow after it started failing in CI after #153 was merged. Why it failed? Nobody knows. I successfully built it on my system, and Chris built it on his; theoretically, there shouldn't be any difference in CI since cross uses a docker image.

Removing aarch64 from the start scripts is a pointless endeavour. Users of aarch64 systems will still try to launch the script, and be unable to due to the x86_64 binary downloaded being incompatible. Not really any better than pointing to an unresolvable URL. I think this PR should be closed.

[adamperkowski]

aarch64 was removed from the build workflow after it started failing in CI after #153 was merged. Why it failed? Nobody knows. I successfully built it on my system, and Chris built it on his; theoretically, there shouldn't be any difference in CI since cross uses a docker image.

Should be added back then.

Removing aarch64 from the start scripts is a pointless endeavour. Users of aarch64 systems will still try to launch the script, and be unable to due to the x86_64 binary downloaded being incompatible. Not really any better than pointing to an unresolvable URL. I think this PR should be closed.

I disagree. Not clearly showing "Unsupported architecture" leads to confusion, confusion leads to research, possibly opening more duplicate issues, and overall waste of time.