AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[SWDEV-496693] GPU metrics 1.7

Browse Source 
Changes:
    - Added new GPU metrics:
      1) XGMI link status - Up/Down; 1 = up; 0 = down
      2) Graphics clocks below host limit (per XCP)
         accumulators -> used to help calculate a violation status
      3) VRAM max bandwidth at max memory clock
    - Updated rocm-smi --showmetrics to include new metrics.
    Units/values reflect as indicated by driver, may differ
    from AMD SMI or other ROCm SMI interfaces which
    use these fields.
    - N/A fields means the device does not support providing this
    data.

Change-Id: I17b313345f15070a76b3a30dd8d5645d212d601b
Signed-off-by: Charis Poag <Charis.Poag@amd.com>


[ROCm/rocm_smi_lib commit: 4de2168866]
This commit is contained in:
Charis Poag
2024-12-12 17:56:09 -06:00
parent 5d7555e586
commit 7b3c814501
8 changed files with 969 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files
projects/rocm-smi-lib/include/rocm_smi/rocm_smi.h
+20 -1
View File
@@ -40,6 +40,7 @@
* DEALINGS WITH THE SOFTWARE.
*
*/
#ifndef ROCM_SMI_ROCM_SMI_H_
#define ROCM_SMI_ROCM_SMI_H_
@@ -987,6 +988,9 @@ typedef struct metrics_table_header_t metrics_table_header_t;
* @brief The following structures hold the gpu statistics for a device.
*/
struct amdgpu_xcp_metrics_t {
/*
* v1.6 additions
*/
/* Utilization Instantaneous (%) */
uint32_t gfx_busy_inst[RSMI_MAX_NUM_XCC];
uint16_t jpeg_busy[RSMI_MAX_NUM_JPEG_ENGS];
@@ -994,6 +998,12 @@ struct amdgpu_xcp_metrics_t {
/* Utilization Accumulated (%) */
uint64_t gfx_busy_acc[RSMI_MAX_NUM_XCC];
/*
* v1.7 additions
*/
/* Total App Clock Counter Accumulated */
uint64_t gfx_below_host_limit_acc[RSMI_MAX_NUM_XCC];
};
typedef struct {
@@ -1173,7 +1183,7 @@ typedef struct {
/**
* Accumulated throttler residencies
*
* Socket (thermal) -
* Socket (thermal) -
* Socket thermal violation % (greater than 0% is a violation);
* aka TVIOL
*
@@ -1197,6 +1207,15 @@ typedef struct {
/* PCIE other end recovery counter */
uint32_t pcie_lc_perf_other_end_recovery;
/*
* v1.7 additions
*/
/* VRAM max bandwidth at max memory clock (GB/s) */
uint64_t vram_max_bandwidth;
/* XGMI link status(up/down) */
uint16_t xgmi_link_status[RSMI_MAX_NUM_XGMI_LINKS];
/// \endcond
} rsmi_gpu_metrics_t;
projects/rocm-smi-lib/include/rocm_smi/rocm_smi_gpu_metrics.h
+149 -2
View File
@@ -109,6 +109,19 @@ struct AMDGpuMetricsHeader_v1_t {
uint8_t m_content_revision;
};
struct amdgpu_xcp_metrics_v1_1 {
/* Utilization Instantaneous (%) */
uint32_t gfx_busy_inst[kRSMI_MAX_NUM_XCC];
uint16_t jpeg_busy[kRSMI_MAX_JPEG_ENGINES];
uint16_t vcn_busy[kRSMI_MAX_NUM_VCNS];
/* Utilization Accumulated (%) */
uint64_t gfx_busy_acc[kRSMI_MAX_NUM_XCC];
/* Total App Clock Counter Accumulated */
uint64_t gfx_below_host_limit_acc[kRSMI_MAX_NUM_XCC];
};
struct amdgpu_xcp_metrics {
/* Utilization Instantaneous (%) */
uint32_t gfx_busy_inst[kRSMI_MAX_NUM_XCC];
@@ -572,7 +585,107 @@ struct AMDGpuMetrics_v16_t {
/* PCIE other end recovery counter */
uint32_t m_pcie_lc_perf_other_end_recovery;
};
using AMGpuMetricsLatest_t = AMDGpuMetrics_v16_t;
struct AMDGpuMetrics_v17_t {
~AMDGpuMetrics_v17_t() = default;
struct AMDGpuMetricsHeader_v1_t m_common_header;
/* Temperature (Celsius) */
uint16_t m_temperature_hotspot;
uint16_t m_temperature_mem;
uint16_t m_temperature_vrsoc;
/* Power (Watts) */
uint16_t m_current_socket_power;
/* Utilization (%) */
uint16_t m_average_gfx_activity;
uint16_t m_average_umc_activity; // memory controller
/* VRAM max bandwidth at max memory clock */
uint64_t m_vram_max_bandwidth; // new for 1.7
/* Energy (15.259uJ (2^-16) units) */
uint64_t m_energy_accumulator;
/* Driver attached timestamp (in ns) */
uint64_t m_system_clock_counter;
/* Accumulation cycle counter */
uint32_t m_accumulation_counter;
/* Accumulated throttler residencies */
uint32_t m_prochot_residency_acc;
uint32_t m_ppt_residency_acc;
uint32_t m_socket_thm_residency_acc;
uint32_t m_vr_thm_residency_acc;
uint32_t m_hbm_thm_residency_acc;
/* Clock Lock Status. Each bit corresponds to clock instance */
uint32_t m_gfxclk_lock_status;
/* Link width (number of lanes) and speed (in 0.1 GT/s) */
uint16_t m_pcie_link_width;
uint16_t m_pcie_link_speed;
/* XGMI bus width and bitrate (in Gbps) */
uint16_t m_xgmi_link_width;
uint16_t m_xgmi_link_speed;
/* Utilization Accumulated (%) */
uint32_t m_gfx_activity_acc;
uint32_t m_mem_activity_acc;
/*PCIE accumulated bandwidth (GB/sec) */
uint64_t m_pcie_bandwidth_acc;
/*PCIE instantaneous bandwidth (GB/sec) */
uint64_t m_pcie_bandwidth_inst;
/* PCIE L0 to recovery state transition accumulated count */
uint64_t m_pcie_l0_to_recov_count_acc;
/* PCIE replay accumulated count */
uint64_t m_pcie_replay_count_acc;
/* PCIE replay rollover accumulated count */
uint64_t m_pcie_replay_rover_count_acc;
/* PCIE NAK sent accumulated count */
uint32_t m_pcie_nak_sent_count_acc;
/* PCIE NAK received accumulated count */
uint32_t m_pcie_nak_rcvd_count_acc;
/* XGMI accumulated data transfer size(KiloBytes) */
uint64_t m_xgmi_read_data_acc[kRSMI_MAX_NUM_XGMI_LINKS];
uint64_t m_xgmi_write_data_acc[kRSMI_MAX_NUM_XGMI_LINKS];
/* XGMI link status(up/down) */
uint16_t m_xgmi_link_status[kRSMI_MAX_NUM_XGMI_LINKS]; // new for 1.7
uint16_t m_padding;
/* PMFW attached timestamp (10ns resolution) */
uint64_t m_firmware_timestamp;
/* Current clocks (Mhz) */
uint16_t m_current_gfxclk[kRSMI_MAX_NUM_GFX_CLKS];
uint16_t m_current_socclk[kRSMI_MAX_NUM_CLKS];
uint16_t m_current_vclk0[kRSMI_MAX_NUM_CLKS];
uint16_t m_current_dclk0[kRSMI_MAX_NUM_CLKS];
uint16_t m_current_uclk;
/* Number of current partition */
uint16_t m_num_partition;
/* XCP metrics stats */
struct amdgpu_xcp_metrics_v1_1 m_xcp_stats[kRSMI_MAX_NUM_XCP];
/* PCIE other end recovery counter */
uint32_t m_pcie_lc_perf_other_end_recovery;
};
using AMGpuMetricsLatest_t = AMDGpuMetrics_v17_t;
/**
* This is GPU Metrics version that gets to public access.
@@ -787,8 +900,11 @@ enum class AMDGpuMetricsUnitType_t : AMDGpuMetricTypeId_t
kMetricJpegBusy, // v1.6
kMetricVcnBusy, // v1.6
kMetricGfxBusyAcc, // v1.6
kMetricPcieLCPerfOtherEndRecov, // v1.6
kMetricVramMaxBandwidth, // v1.7
kMetricXgmiLinkStatus, // v1.7
kMetricGfxBelowHostLimitAccumulator, // v1.7
};
using AMDGpuMetricsUnitTypeTranslationTbl_t = std::map<AMDGpuMetricsUnitType_t, std::string>;
@@ -826,6 +942,7 @@ enum class AMDGpuMetricVersionFlags_t : AMDGpuMetricVersionFlagId_t
kGpuMetricV14 = (0x1 << 4),
kGpuMetricV15 = (0x1 << 5),
kGpuMetricV16 = (0x1 << 6),
kGpuMetricV17 = (0x1 << 7),
};
using AMDGpuMetricVersionTranslationTbl_t = std::map<uint16_t, AMDGpuMetricVersionFlags_t>;
using GpuMetricTypePtr_t = std::shared_ptr<void>;
@@ -1044,6 +1161,36 @@ class GpuMetricsBase_v16_t final : public GpuMetricsBase_t {
std::shared_ptr<AMDGpuMetrics_v16_t> m_gpu_metric_ptr;
};
class GpuMetricsBase_v17_t final : public GpuMetricsBase_t {
public:
~GpuMetricsBase_v17_t() = default;
size_t sizeof_metric_table() override {
return sizeof(AMDGpuMetrics_v17_t);
}
GpuMetricTypePtr_t get_metrics_table() override {
if (!m_gpu_metric_ptr) {
m_gpu_metric_ptr.reset(&m_gpu_metrics_tbl, [](AMDGpuMetrics_v17_t*){});
}
assert(m_gpu_metric_ptr != nullptr);
return m_gpu_metric_ptr;
}
void dump_internal_metrics_table() override;
AMDGpuMetricVersionFlags_t get_gpu_metrics_version_used() override {
return AMDGpuMetricVersionFlags_t::kGpuMetricV17;
}
rsmi_status_t populate_metrics_dynamic_tbl() override;
AMGpuMetricsPublicLatestTupl_t copy_internal_to_external_metrics() override;
private:
AMDGpuMetrics_v17_t m_gpu_metrics_tbl;
std::shared_ptr<AMDGpuMetrics_v17_t> m_gpu_metric_ptr;
};
template<typename T>
rsmi_status_t rsmi_dev_gpu_metrics_info_query(uint32_t dv_ind,
AMDGpuMetricsUnitType_t metric_counter, T& metric_value);