Re: Memory ordering

On 11/17/2024 6:03 AM, jseigh wrote:

On 11/16/24 16:21, Chris M. Thomasson wrote:
 

Fwiw, in C++ std::memory_order_consume is useful for traversing a node based stack of something in RCU. In most systems it only acts like a compiler barrier. On the Alpha, it must emit a membar instruction. Iirc, mb for alpha? Cannot remember that one right now.

 That got deprecated.  Too hard for compilers to deal with.  It's now
same as memory_order_acquire.
 Which brings up an interesting point.  Even if the hardware memory
memory model is strongly ordered, compilers can reorder stuff,
so you still have to program as if a weak memory model was in
effect.  

Or maybe disable reordering or optimization altogether
for those target architectures.

^^^^^^^^^^^^
Yeah. No shit. Some people say the std::memory_order_* is too complex. Others says its okay. Shit happens.
I remember way before C++11 I would feel nervous about keeping link time optimizations on because I thought they mess around with my custom assembly language code for my sensitive thread sync algorithms. I thought that a compiler would say okay this is calling into externally assembled/compiled code. No optimizations.... Well, link time optimization scared me.