[clang] [llvm] [CHERI][RISCV][AArch64][MC] Add shared support for aligning globals to an "alignment granule", which is applied to both the alignment and the allocated size. (PR #203872)

[Owen Anderson via cfe-commits]

resistor wrote:

@fmayer Have you have a chance to take a look at the issue?

Barring that, would it make sense to land this without the MTE swap-over, and then someone better able to test MTE could land that part later?

https://github.com/llvm/llvm-project/pull/203872