[R-SIG-Mac] is there an M1 virtual machine image available for linux?

[Prof Brian Ripley]

On 15/12/2020 03:26, Simon Urbanek wrote:
> To my best knowledge the only way to test anything on M1 is to buy one. As far as I know there is no virtual solution yet. If you see one, let me know.
> 
> As for CRAN there is no official arm64 build yet - the hardware just arrived and I need to setup the nightly builds first before moving to packages. However, Brian Ripley has run a first pass over all packages to give authors chance to look at what breaks on the new architecture - it is listed on the check page in the additional checks section.

(Listed as of last night European time and still being tweaked.)

>>From what I can see some of your packages broke on the regular check systems and have not been fixed after several requests which why your packages were under threat to be removed, it has nothing to fo with M1 as far as I can tell.

Indeed.  I have only reported as 'additional issues' whose root causes 
are obvious from the logs and in the vast majority of cases could be 
fixed without access to hardware (e.g. by changing a tolerance or using 
a suggested package conditionally).

> 
> Cheers,
> Simon
> 
> 
> 
>> On 15/12/2020, at 4:02 PM, Joshua N Pritikin <jpritikin using pobox.com> wrote:
>>
>> On Tue, Dec 15, 2020 at 03:54:12PM +1300, Simon Urbanek wrote:
>>> I don't think that is feasible (assuming you mean macOS 11 arm64
>>> system running in a VM). Note that M1 is not just the CPU, there is
>>> much tighter integration of all the components so you'd have to
>>> emulate a lot more than that and also all the Apple security
>>> components etc. in order to be able to boot macOS on an arm emulation.
>>
>> OK, so how do I test my package on M1?
>>
>> I've been threatened with removal from CRAN if I don't get it working.
>>
>> Thank you.
>>
> 
> _______________________________________________
> R-SIG-Mac mailing list
> R-SIG-Mac using r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-mac
> 


-- 
Brian D. Ripley,                  ripley using stats.ox.ac.uk
Emeritus Professor of Applied Statistics, University of Oxford