From: "Gautham R. Shenoy" <e...@linux.vnet.ibm.com> Hi,
This is the seventh iteration of the patchset to add support for big-core on POWER9. This patch also optimizes the task placement on such big-core systems. The previous versions can be found here: v6: https://lkml.org/lkml/2018/8/9/119 v5: https://lkml.org/lkml/2018/8/6/587 v4: https://lkml.org/lkml/2018/7/24/79 v3: https://lkml.org/lkml/2018/7/6/255 v2: https://lkml.org/lkml/2018/7/3/401 v1: https://lkml.org/lkml/2018/5/11/245 Changes : v6 --> v7: - Addressed the review comments from Srikar in Patch 1. - For building the SMT level sched-domain with small_core_sibling_mask, parse the "ibm,thread-groups" property of the CPU node only once, i.e when the CPU is made online for the first time. Description: ~~~~~~~~~~~~~~~~~~~~ A pair of IBM POWER9 SMT4 cores can be fused together to form a big-core with 8 SMT threads. This can be discovered via the "ibm,thread-groups" CPU property in the device tree which will indicate which group of threads that share the L1 cache, translation cache and instruction data flow. If there are multiple such group of threads, then the core is a big-core. Furthermore, on POWER9 the thread-ids of such a big-core is obtained by interleaving the thread-ids of the component SMT4 cores. Eg: Threads in the pair of component SMT4 cores of an interleaved big-core are numbered {0,2,4,6} and {1,3,5,7} respectively. ------------------------- | L1 Cache | ---------------------------------- |L2| | | | | | | 0 | 2 | 4 | 6 |Small Core0 |C | | | | | Big |a -------------------------- Core |c | | | | | |h | 1 | 3 | 5 | 7 | Small Core1 |e | | | | | ----------------------------- | L1 Cache | -------------------------- On such a big-core system, when multiple tasks are scheduled to run on the big-core, we get the best performance when the tasks are spread across the pair of SMT4 cores. Eg: Suppose there 4 tasks {p1, p2, p3, p4} are run on a big core, then An Example of Optimal Task placement: -------------------------- | | | | | | 0 | 2 | 4 | 6 | Small Core0 | (p1)| (p2)| | | Big Core -------------------------- | | | | | | 1 | 3 | 5 | 7 | Small Core1 | | (p3)| | (p4) | -------------------------- An example of Suboptimal Task placement: -------------------------- | | | | | | 0 | 2 | 4 | 6 | Small Core0 | (p1)| (p2)| | (p4)| Big Core -------------------------- | | | | | | 1 | 3 | 5 | 7 | Small Core1 | | (p3)| | | -------------------------- In order to achieve optimal task placement, on big-core systems, we define the SMT level sched-domain to consist of the threads belonging to the small cores. The CACHE level sched domain will consist of all the threads belonging to the big-core. With this, the Linux Kernel load-balancer will ensure that the tasks are spread across all the component small cores in the system, thereby yielding optimum performance. Furthermore, this solution works correctly across all SMT modes (8,4,2), as the interleaved thread-ids ensures that when we go to lower SMT modes (4,2) the threads are offlined in a descending order, thereby leaving equal number of threads from the component small cores online as illustrated below. With Patches: (ppc64_cpu --smt=on) : SMT domain ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ CPU0 attaching sched-domain(s): domain-0: span=0,2,4,6 level=SMT groups: 0:{ span=0 cap=294 }, 2:{ span=2 cap=294 }, 4:{ span=4 cap=294 }, 6:{ span=6 cap=294 } CPU1 attaching sched-domain(s): domain-0: span=1,3,5,7 level=SMT groups: 1:{ span=1 cap=294 }, 3:{ span=3 cap=294 }, 5:{ span=5 cap=294 }, 7:{ span=7 cap=294 } Optimal Task placement (SMT 8) -------------------------- | | | | | | 0 | 2 | 4 | 6 | Small Core0 | (p1)| (p2)| | | Big Core -------------------------- | | | | | | 1 | 3 | 5 | 7 | Small Core1 | | (p3)| | (p4) | -------------------------- With Patches : (ppc64_cpu --smt=4) : SMT domain ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ CPU0 attaching sched-domain(s): domain-0: span=0,2 level=SMT groups: 0:{ span=0 cap=589 }, 2:{ span=2 cap=589 } CPU1 attaching sched-domain(s): domain-0: span=1,3 level=SMT groups: 1:{ span=1 cap=589 }, 3:{ span=3 cap=589 } Optimal Task placement (SMT 4) -------------------------- | | | | | | 0 | 2 | 4 | 6 | Small Core0 | (p1)| (p2)| Off | Off | Big Core -------------------------- | | | | | | 1 | 3 | 5 | 7 | Small Core1 | (p4)| (p3)| Off | Off | -------------------------- With Patches : (ppc64_cpu --smt=2) : SMT domain ceases to exist. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Optimal Task placement (SMT 2) -------------------------- | (p2)| | | | | 0 | 2 | 4 | 6 | Small Core0 | (p1)| Off | Off | Off | Big Core -------------------------- | (p3)| | | | | 1 | 3 | 5 | 7 | Small Core1 | (p4)| Off | Off | Off | -------------------------- Thus, as an added advantage in SMT=2 mode, we will only have 3 levels in the sched-domain topology (CACHE, DIE and NUMA). The SMT levels, without the patches are as follows. Without Patches: (ppc64_cpu --smt=on) : SMT domain ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ CPU0 attaching sched-domain(s): domain-0: span=0-7 level=SMT groups: 0:{ span=0 cap=147 }, 1:{ span=1 cap=147 }, 2:{ span=2 cap=147 }, 3:{ span=3 cap=147 }, 4:{ span=4 cap=147 }, 5:{ span=5 cap=147 }, 6:{ span=6 cap=147 }, 7:{ span=7 cap=147 } CPU1 attaching sched-domain(s): domain-0: span=0-7 level=SMT groups: 1:{ span=1 cap=147 }, 2:{ span=2 cap=147 }, 3:{ span=3 cap=147 }, 4:{ span=4 cap=147 }, 5:{ span=5 cap=147 }, 6:{ span=6 cap=147 }, 7:{ span=7 cap=147 }, 0:{ span=0 cap=147 } Without Patches: (ppc64_cpu --smt=4) : SMT domain ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ CPU0 attaching sched-domain(s): domain-0: span=0-3 level=SMT groups: 0:{ span=0 cap=294 }, 1:{ span=1 cap=294 }, 2:{ span=2 cap=294 }, 3:{ span=3 cap=294 }, CPU1 attaching sched-domain(s): domain-0: span=0-3 level=SMT groups: 1:{ span=1 cap=294 }, 2:{ span=2 cap=294 }, 3:{ span=3 cap=294 }, 0:{ span=0 cap=294 } Without Patches: (ppc64_cpu --smt=2) : SMT domain ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ CPU0 attaching sched-domain(s): domain-0: span=0-1 level=SMT groups: 0:{ span=0 cap=589 }, 1:{ span=1 cap=589 }, CPU1 attaching sched-domain(s): domain-0: span=0-1 level=SMT groups: 1:{ span=1 cap=589 }, 0:{ span=0 cap=589 }, This patchset contains two patches which on detecting the presence of big-cores, defines the SMT level sched domain to correspond to the threads of the small cores. Patch 1: adds support to detect the presence of big-cores and reports the small-core siblings of each CPU X via the sysfs file "/sys/devices/system/cpu/cpuX/small_core_siblings". Patch 2: Defines the SMT level sched domain to correspond to the threads of the small cores. Results: ~~~~~~~~~~~~~~~~~ 1) 2 thread ebizzy ~~~~~~~~~~~~~~~~~~~~~~ Experimental results for ebizzy with 2 threads, bound to a single big-core show a marked improvement with this patchset over the 4.18.0 vanilla kernel. The result of 100 such runs for 4.18-rc7 kernel and the 4.18 + big-core-smt-patches are as follows 4.18.0 vanilla ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ records/s : # samples : Histogram ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ [0 - 1000000] : 0 : # [1000000 - 2000000] : 11 : ### [2000000 - 3000000] : 9 : ## [3000000 - 4000000] : 9 : ## [4000000 - 5000000] : 0 : # [5000000 - 6000000] : 71 : ############### 4.18.0 + big-core-smt-patches ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ records/s : # samples : Histogram ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ [0 - 1000000] : 0 : # [1000000 - 2000000] : 0 : # [2000000 - 3000000] : 16 : #### [3000000 - 4000000] : 0 : # [4000000 - 5000000] : 1 : # [5000000 - 6000000] : 83 : ################# 2) Hackbench (perf bench sched pipe) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 500 iterations of the hackbench run both on 4.18.0 vanilla kernel and v4.18.0 + big-core-smt-patches. More samples in the lower numbered buckets is better. We can observe that for nearly 60% of the samples are in the 4-5 seconds range when hackbench is run with this patchset as opposed to < 20% of the time when the hackbench is run on the vanilla kernel. Similarly, nearly 80% of the samples are within the 4-6 seconds range when hackbench is run with the patchset when compared with 50% of the samples in the same range when hackbench is run on the vanilla kernel. Though as a downside, we do see ~10% of the samples in the 7-9 seconds range with the patchset as compared to 2% of the samples in the same range without the patchset. 4.18.0 vanilla ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 4 - 5 seconds : 74 samples 5 - 6 seconds : 169 samples 6 - 7 seconds : 248 samples 7 - 8 seconds : 6 samples 8 - 9 seconds : 3 samples 4.18.0 + big-core-smt-patches ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 4 - 5 seconds : 289 samples 5 - 6 seconds : 99 samples 6 - 7 seconds : 58 samples 7 - 8 seconds : 28 samples 8 - 9 seconds : 26 samples Gautham R. Shenoy (2): powerpc: Detect the presence of big-cores via "ibm,thread-groups" powerpc: Use cpu_smallcore_sibling_mask at SMT level on bigcores Documentation/ABI/testing/sysfs-devices-system-cpu | 8 ++ arch/powerpc/include/asm/cputhreads.h | 25 ++++ arch/powerpc/kernel/setup-common.c | 151 +++++++++++++++++++++ arch/powerpc/kernel/smp.c | 136 ++++++++++++++++++- arch/powerpc/kernel/sysfs.c | 38 ++++++ 5 files changed, 356 insertions(+), 2 deletions(-) -- 1.9.4