> Yeah, "word tearing" is an overloaded term. You can also consider splitting > a word across cachelines as potentially causing a tear as a store/load > involves two cachelines
If a word was split across cache lines, it is by definition not aligned, so the guarantees of an atomic write don't apply, cache or no cache. On Wed, Jan 18, 2017 at 8:02 AM, Vitaly Davidovich <vita...@gmail.com> wrote: > > > On Tue, Jan 17, 2017 at 3:39 PM, Aleksey Shipilev > <aleksey.shipi...@gmail.com> wrote: >> >> On 01/17/2017 12:55 PM, Vitaly Davidovich wrote: >> > Atomicity of values isn't something I'd assume happens automatically. >> > Word >> > tearing isn't observable from single threaded code. >> >> On 01/17/2017 09:17 PM, Michael Barker wrote: >> > That was my understanding too. Normal load/stores on 32 bit JVMs would >> > tear 64 >> > bit values. Although, I think object references are guaranteed to be >> > written >> > atomically. >> >> (Triggered by my pet peeve) >> >> Please don't confuse "access atomicity" and "word tearing". Access >> atomicity >> means reading/writing the value in full. Word tearing (at least per JLS >> 17.6) >> means that fields are considered distinct, and the read/write of one field >> cannot disturb neighbors. > > Yeah, "word tearing" is an overloaded term. You can also consider splitting > a word across cachelines as potentially causing a tear as a store/load > involves two cachelines. But my point was really that word tearing and > access atomicity (or lack thereof) aren't observable in single threaded > code. >> >> >> Speaking of access atomicity, the upcoming value types would probably be >> non-access-atomic by default too, because enforcing access atomicity for >> widths >> larger than a machine one is painful. volatile longs/doubles had it easy >> on >> 32-bit VMs with 64-bit FPUs, compared to what variable-len values would >> have to >> experience. > > I don't see how they could be access atomic either. >> >> >> Thanks, >> -Aleksey >> >> >> -- >> You received this message because you are subscribed to the Google Groups >> "mechanical-sympathy" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to mechanical-sympathy+unsubscr...@googlegroups.com. >> For more options, visit https://groups.google.com/d/optout. > > > -- > You received this message because you are subscribed to the Google Groups > "mechanical-sympathy" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to mechanical-sympathy+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups "mechanical-sympathy" group. To unsubscribe from this group and stop receiving emails from it, send an email to mechanical-sympathy+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
