felipecrv commented on PR #35565: URL: https://github.com/apache/arrow/pull/35565#issuecomment-1552991769
> A. Demonstrated potential performance penalties (which were often CPU-specific) associated with the unaligned loads "A" makes compute kernel performance more noisy and unpredictable. IMO, that alone could be a reason to do our best to feed aligned data to compute kernels even with the potential alloc and copy before the kernel runs. > B. Were specific to 32-bit ARM processors (which crash on unaligned loads) "B" is a sign that relying on unaligned memory access makes us much more exposed to CPU bugs. In trusting modern CPUs unaligned-memory access machinery, we expose ourselves to risk. That machinery is even less trustworthy on SIMD instruction sets. CPU vendors started supporting unaligned memory access as a response to bad software, not because it makes for elegant and reliable hardware design. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
