RFC: linux-aarch64 and LSE support

[dean.long at oracle.com]

On 9/23/22 5:58 AM, Andrew Haley wrote:
> On 9/23/22 12:40, dean.long at oracle.com wrote:
>> I have to admit it is a bit surprising and counter-intuitive to me to
>> not treat the CASAL as atomic.  If the memory model allows the 2nd store
>> to be moved "inside" the CASAL, then that means it might be impossible
>> to provide a total order of the AArch64 instructions that reproduces the
>> observed memory effects, and that makes me wonder if it would break
>> tools like "rr".
> 
> Is that an "all the world's an x86" kind of intuition, though? We're
> all influenced more than we know by our history. Even Aleksey was
> surprised by
> https://twitter.com/shipilev/status/1400154160918581248
> in which AArch64 produces "Interesting  Whoa".
> 

For me it's more getting stuck thinking in terms of machine 
instructions.  If I can conceptually make the switch to "operations" and 
memory effects, then everything is fine.

dl

PS - I didn't mean to derail this thread!