RFR: 8310939: [c1] The visibility of write-volatile requires membar_volatile instead of membar

[Andrew Haley]

On Tue, 27 Jun 2023 12:46:14 GMT, SUN Guoyun <duke at openjdk.org> wrote:

> For c1 now, a volatile write case:
> 
> membar_release // LoadStore | StoreStore
> write volatile
> membar
> 
> Just like c2, here `membar` should be defined `membar_volatile` clearly, then for risc-v, ppc and loongarch can use StoreLoad for `membar_volatile` for better performance.
> 
> Testing:
> GHA testing
> jtreg tier1-3 for loongarch64

For a little more clarity: let's imagine that you wanted to have a sequentially-consistent set of stores and loads, using plain memory accesses and fences.

This would work:

str r1, [x1] ; fullFence; ldr r1, [x1]; fullFence; str r1, [x2]; fullFence; ldr r1, [x2]

but this would not:

str r1, [x1] ; storeLoad; ldr r1, [x1]; storeLoad; str r1, [x2]; storeLoad; ldr r1, [x2]

This is why I wouldn't call storeLoad "membar_volatile".

-------------

PR Comment: https://git.openjdk.org/jdk/pull/14677#issuecomment-1647903628