cachepc-linux

Fork of AMDESE/linux with modifications for CachePC side-channel attack
git clone https://git.sinitax.com/sinitax/cachepc-linux
Log | Files | Refs | README | LICENSE | sfeed.txt

hisi-pmu.rst (5031B)


      1======================================================
      2HiSilicon SoC uncore Performance Monitoring Unit (PMU)
      3======================================================
      4
      5The HiSilicon SoC chip includes various independent system device PMUs
      6such as L3 cache (L3C), Hydra Home Agent (HHA) and DDRC. These PMUs are
      7independent and have hardware logic to gather statistics and performance
      8information.
      9
     10The HiSilicon SoC encapsulates multiple CPU and IO dies. Each CPU cluster
     11(CCL) is made up of 4 cpu cores sharing one L3 cache; each CPU die is
     12called Super CPU cluster (SCCL) and is made up of 6 CCLs. Each SCCL has
     13two HHAs (0 - 1) and four DDRCs (0 - 3), respectively.
     14
     15HiSilicon SoC uncore PMU driver
     16-------------------------------
     17
     18Each device PMU has separate registers for event counting, control and
     19interrupt, and the PMU driver shall register perf PMU drivers like L3C,
     20HHA and DDRC etc. The available events and configuration options shall
     21be described in the sysfs, see:
     22
     23/sys/devices/hisi_sccl{X}_<l3c{Y}/hha{Y}/ddrc{Y}>/, or
     24/sys/bus/event_source/devices/hisi_sccl{X}_<l3c{Y}/hha{Y}/ddrc{Y}>.
     25The "perf list" command shall list the available events from sysfs.
     26
     27Each L3C, HHA and DDRC is registered as a separate PMU with perf. The PMU
     28name will appear in event listing as hisi_sccl<sccl-id>_module<index-id>.
     29where "sccl-id" is the identifier of the SCCL and "index-id" is the index of
     30module.
     31
     32e.g. hisi_sccl3_l3c0/rd_hit_cpipe is READ_HIT_CPIPE event of L3C index #0 in
     33SCCL ID #3.
     34
     35e.g. hisi_sccl1_hha0/rx_operations is RX_OPERATIONS event of HHA index #0 in
     36SCCL ID #1.
     37
     38The driver also provides a "cpumask" sysfs attribute, which shows the CPU core
     39ID used to count the uncore PMU event.
     40
     41Example usage of perf::
     42
     43  $# perf list
     44  hisi_sccl3_l3c0/rd_hit_cpipe/ [kernel PMU event]
     45  ------------------------------------------
     46  hisi_sccl3_l3c0/wr_hit_cpipe/ [kernel PMU event]
     47  ------------------------------------------
     48  hisi_sccl1_l3c0/rd_hit_cpipe/ [kernel PMU event]
     49  ------------------------------------------
     50  hisi_sccl1_l3c0/wr_hit_cpipe/ [kernel PMU event]
     51  ------------------------------------------
     52
     53  $# perf stat -a -e hisi_sccl3_l3c0/rd_hit_cpipe/ sleep 5
     54  $# perf stat -a -e hisi_sccl3_l3c0/config=0x02/ sleep 5
     55
     56For HiSilicon uncore PMU v2 whose identifier is 0x30, the topology is the same
     57as PMU v1, but some new functions are added to the hardware.
     58
     59(a) L3C PMU supports filtering by core/thread within the cluster which can be
     60specified as a bitmap::
     61
     62  $# perf stat -a -e hisi_sccl3_l3c0/config=0x02,tt_core=0x3/ sleep 5
     63
     64This will only count the operations from core/thread 0 and 1 in this cluster.
     65
     66(b) Tracetag allow the user to chose to count only read, write or atomic
     67operations via the tt_req parameeter in perf. The default value counts all
     68operations. tt_req is 3bits, 3'b100 represents read operations, 3'b101
     69represents write operations, 3'b110 represents atomic store operations and
     703'b111 represents atomic non-store operations, other values are reserved::
     71
     72  $# perf stat -a -e hisi_sccl3_l3c0/config=0x02,tt_req=0x4/ sleep 5
     73
     74This will only count the read operations in this cluster.
     75
     76(c) Datasrc allows the user to check where the data comes from. It is 5 bits.
     77Some important codes are as follows:
     785'b00001: comes from L3C in this die;
     795'b01000: comes from L3C in the cross-die;
     805'b01001: comes from L3C which is in another socket;
     815'b01110: comes from the local DDR;
     825'b01111: comes from the cross-die DDR;
     835'b10000: comes from cross-socket DDR;
     84etc, it is mainly helpful to find that the data source is nearest from the CPU
     85cores. If datasrc_cfg is used in the multi-chips, the datasrc_skt shall be
     86configured in perf command::
     87
     88  $# perf stat -a -e hisi_sccl3_l3c0/config=0xb9,datasrc_cfg=0xE/,
     89  hisi_sccl3_l3c0/config=0xb9,datasrc_cfg=0xF/ sleep 5
     90
     91(d)Some HiSilicon SoCs encapsulate multiple CPU and IO dies. Each CPU die
     92contains several Compute Clusters (CCLs). The I/O dies are called Super I/O
     93clusters (SICL) containing multiple I/O clusters (ICLs). Each CCL/ICL in the
     94SoC has a unique ID. Each ID is 11bits, include a 6-bit SCCL-ID and 5-bit
     95CCL/ICL-ID. For I/O die, the ICL-ID is followed by:
     965'b00000: I/O_MGMT_ICL;
     975'b00001: Network_ICL;
     985'b00011: HAC_ICL;
     995'b10000: PCIe_ICL;
    100
    101Users could configure IDs to count data come from specific CCL/ICL, by setting
    102srcid_cmd & srcid_msk, and data desitined for specific CCL/ICL by setting
    103tgtid_cmd & tgtid_msk. A set bit in srcid_msk/tgtid_msk means the PMU will not
    104check the bit when matching against the srcid_cmd/tgtid_cmd.
    105
    106If all of these options are disabled, it can works by the default value that
    107doesn't distinguish the filter condition and ID information and will return
    108the total counter values in the PMU counters.
    109
    110The current driver does not support sampling. So "perf record" is unsupported.
    111Also attach to a task is unsupported as the events are all uncore.
    112
    113Note: Please contact the maintainer for a complete list of events supported for
    114the PMU devices in the SoC and its information if needed.