News for hackers. Stuff that matters.
Re:Nvidia GB10's Memory Subsystem, from the CPU Side by Neywiny on Wednesday December 31, 2025 @11:48AM (#46445791)
I don't understand on one of the later graphs the core to core latency for strix halo goes out to 32 cores but he says only has 16 cores?
Re:Nvidia GB10's Memory Subsystem, from the CPU Side by wtallis on Wednesday December 31, 2025 @11:56AM (#46445864)
AMD's cores have SMT, allowing them to run two threads at a time and appear to the OS and its scheduler as two logical cores despite being implemented as a single physical core.
Re:Nvidia GB10's Memory Subsystem, from the CPU Side by Neywiny on Wednesday December 31, 2025 @12:23PM (#46446157)
What pattern in the data shows that's what's being measured? I would expect to see basically 0 latency between adjacent "cores" then since L1 is shared per thread?
Re:Nvidia GB10's Memory Subsystem, from the CPU Side by monocasa on Wednesday December 31, 2025 @01:30PM (#46446844)
Co resident threads might not get any speed up here since coherency instructions are functionally operations on the L2 cache.