Re:Intel Itanium IA-64 Support Removed with the Linux 6.7 Ke...
by masklinn on Sunday November 05, 2023 @02:13AM
(#38149037)
Also “notoriously weak” memory model even as multithreading was taking off, and
> The processor may lose track of your LDx_L if you perform any memory access other than a matching STx_C, or if you perform a branch instruction, or if you trigger a trap (such as executing an emulated instruction), or if you execute more than 20 instructions after the LDx_L.
(LDxL and STxC are locked load and conditional store)
Re:Intel Itanium IA-64 Support Removed with the Linux 6.7 Ke...
by colejohnson66 on Sunday November 05, 2023 @08:07AM
(#38150812)
What do you mean “lose track of”?
Re:Intel Itanium IA-64 Support Removed with the Linux 6.7 Ke...
by masklinn on Sunday November 05, 2023 @08:19AM
(#38150883)
Discard / clear the lock flag, so the store never succeeds.
Re:Intel Itanium IA-64 Support Removed with the Linux 6.7 Ke...
by pm215 on Sunday November 05, 2023 @10:28AM
(#38151919)
That is pretty standard for LL/SC instructions -- you'll find similar constraints on the Arm equivalent insns. In practice this doesn't matter because you only use them in a limited set of handcoded sequences to implement the atomic primitives you actually wanted, so you can always be sure that your LL/SC loop obeys the rules and makes forward progress.
Re:Intel Itanium IA-64 Support Removed with the Linux 6.7 Ke...
by cmrdporcupine on Sunday November 05, 2023 @08:24PM
(#38157713)
Was going to say. x86 I think has spoiled people in terms of what they expect in terms of memory model / consistency.
The answer is to be careful as a dev, and hopefully let the compiler and your libraries take care of the finger details.
* no byte operations
* no flags, thus no reasonable way to check for overflow (similar to one of the problems RISC-V is having, though at least they pretend to care).