On 01/17/13 16:43, Daniel P. Berrange wrote:
I beg to differ:
"This technology is informally called CMT (Clustered Multi-Thread) and formally called "module" by the AMD's marketing service, but whose real name is "Clustered Integer" Core Technology. In terms of hardware complexity and functionality, this "module" is midway between a true dual-core processor and its integer power (each thread having a fully independent integer core) and a single core processor that has the SMT ability, which can create a dual threads processor but with the power of one (each thread shares the resources of the core with the other thread)."
(source: http://en.wikipedia.org/wiki/Bulldozer_(microarchitecture)#BULLDOZER_Core_.2... )
oVirt/vdsm considers the "Module" as a core with two threads, where others consider it more as separate cores. The performance depends on the type of task that is being run on the module.
So what information does VDSM use to identify this hardware topology.
VDSM uses the <topology> element (subelement of cpu) in the capabilities and multiplies the numbers by the number of numa nodes. As the data there is taken from nodeinfo, this breaks on some systems.
In some form or another we need to be providing the more detailed and accurate topology infomation in the capabilities NUMA description.
What does 'hwloc --no-io --no-caches' shown on a Bulldozer machine ?
# hwloc-info --no-io --no-caches depth 0: 1 Machine (type #1) depth 1: 2 Socket (type #3) depth 2: 4 NUMANode (type #2) depth 3: 32 Core (type #5) depth 4: 32 PU (type #6) hwloc recognizes it as cores ... # lscpu Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Byte Order: Little Endian CPU(s): 32 On-line CPU(s) list: 0-31 Thread(s) per core: 2 Core(s) per socket: 8 Socket(s): 2 ... lscpu as as threads ... NUMA node(s): 4 Vendor ID: AuthenticAMD CPU family: 21 Model: 1 Stepping: 2 CPU MHz: 1400.000 BogoMIPS: 4389.80 Virtualization: AMD-V L1d cache: 16K L1i cache: 64K L2 cache: 2048K L3 cache: 6144K NUMA node0 CPU(s): 0-7 NUMA node1 CPU(s): 8-15 NUMA node2 CPU(s): 16-23 NUMA node3 CPU(s): 24-31
Daniel