diff --git a/source/lib/rocprofiler-sdk/counters/tests/metrics_test.h b/source/lib/rocprofiler-sdk/counters/tests/metrics_test.h
index 9cea346493..58626b60c5 100644
--- a/source/lib/rocprofiler-sdk/counters/tests/metrics_test.h
+++ b/source/lib/rocprofiler-sdk/counters/tests/metrics_test.h
@@ -184,7 +184,1287 @@ static const std::unordered_map<std::string, std::vector<std::vector<std::string
        "TCP",
        "6",
        "<None>",
-       "TCP stalls TA data interface. Now Windowed."}}}};
+       "TCP stalls TA data interface. Now Windowed."},
+      {"SQ_WAVES_LT_32",
+       "SQ",
+       "9",
+       "<None>",
+       "Count number of waves sent <32 active threads sent to SQs. "
+       "This value represents the number of waves that an each individual SIMD has enqueued during "
+       "the collection timeframe (for dispatch profiling this is the timeframe of kernel execution, "
+       "for agent profiling it is the timeframe between start_context and read counter data) with "
+       "less than 32 threads. A sum of all SQ_WAVES_LT_32 values will give the total number of "
+       "waves with 32 threads enqueued during the collection timeframe by the application. "
+       "Returns one value per-SE (aggregates of SIMD values). "
+       "Useful for checking for wavefront occupancy."},
+      {"TCC_ALL_TC_OP_WB_WRITEBACK",
+       "TCC",
+       "73",
+       "<None>",
+       "Number of writebacks due to all TC_OP writeback requests."},
+      {"GRBM_UTCL2_BUSY",
+       "GRBM",
+       "34",
+       "<None>",
+       "The Unified Translation Cache Level-2 (UTCL2) block is busy."},
+      {"SPI_RA_SGPR_SIMD_FULL_CSN",
+       "SPI",
+       "115",
+       "<None>",
+       "Sum of SIMD where SGPR can't take csn wave when !fits. Source is RA0"},
+      {"TCP_TCC_NC_ATOMIC_REQ",
+       "TCP",
+       "77",
+       "<None>",
+       "Total atomic requests with NC mtype from this TCP to all TCCs"},
+      {"SQC_DCACHE_HITS",
+       "SQ",
+       "291",
+       "<None>",
+       "Number of cache hits. (per-SQ, per-Bank, nondeterministic)"},
+      {"CPC_CPC_TCIU_IDLE",
+       "CPC",
+       "29",
+       "<None>",
+       "CPC TCIU interface Idle."},
+      {"SPI_CSN_WAVE",
+       "SPI",
+       "52",
+       "<None>",
+       "Number of waves. Requires SPI_DEBUG_CNTL.DEBUG_PIPE_SEL to select source, "
+       "DEBUG_PIPE_SEL = 1, source is CS1; DEBUG_PIPE_SEL = 2, source is CS2; "
+       "DEBUG_PIPE_SEL = 3, source is CS3; default, source is CS0;"},
+      {"SQ_INST_CYCLES_VMEM_WR",
+       "SQ",
+       "77",
+       "<None>",
+       "The number of cycles needed to send addr and cmd data for VMEM write instructions. "
+       "This value is returned on a per-SE (aggregate of values in SIMDs in the SE) basis "
+       "with units in quad-cycles(4 cycles)."},
+      {"TCC_EA0_WR_UNCACHED_32B",
+       "TCC",
+       "29",
+       "<None>",
+       "Number of 32-byte write/atomic going over the TC_EA_wrreq interface due to uncached "
+       "traffic. Note that CC mtypes can produce uncached requests, and those are included "
+       "in this. A 64-byte request will be counted as 2"},
+      {"TA_ADDR_STALLED_BY_TD_CYCLES",
+       "TA",
+       "55",
+       "<None>",
+       "Number of cycles addr path stalled by TD. Perf_Windowing not supported for this "
+       "counter."},
+      {"TCP_TOTAL_WRITEBACK_INVALIDATES",
+       "TCP",
+       "45",
+       "<None>",
+       "Total number of cache invalidates. Equals TCP_PERF_SEL_TOTAL_WBINVL1+ "
+       "TCP_PERF_SEL_TOTAL_WBINVL1_VOL+ TCP_PERF_SEL_CP_TCP_INVALIDATE+ "
+       "TCP_PERF_SEL_SQ_TCP_INVALIDATE_VOL. Not Windowed."},
+      {"TCC_PROBE_ALL",
+       "TCC",
+       "10",
+       "<None>",
+       "Number of external probe requests with with EA_TCC_preq_all== 1. Not windowable."},
+      {"TCC_CC_REQ",
+       "TCC",
+       "7",
+       "<None>",
+       "The number of coherently cached requests. This is measured at the tag block."},
+      {"SPI_RA_REQ_NO_ALLOC_CSN",
+       "SPI",
+       "85",
+       "<None>",
+       "Arb cycles with CSn req and no CSn alloc. Source is RA0"},
+      {"CPC_ME1_DC0_SPI_BUSY",
+       "CPC",
+       "33",
+       "<None>",
+       "CPC Me1 Processor Busy."},
+      {"SQ_WAVES_RESTORED",
+       "SQ",
+       "159",
+       "<None>",
+       "Count number of context-restored waves sent to SQs. This value represents the number "
+       "of waves whos current register state has been restored from a register bank during "
+       "the collection timeframe (for dispatch profiling this is the timeframe of kernel "
+       "execution, for agent profiling it is the timeframe between start_context and read "
+       "counter data). Context saving/restoring is a slow operation and should be limited. "
+       "High values can also indicate that stalling may be taking place (waiting for free "
+       "register space). Returns one value per-SE (aggregates of SIMD values)."},
+      {"CPF_CPF_TCIU_IDLE",
+       "CPF",
+       "27",
+       "<None>",
+       "CPF TCIU interface Idle."},
+      {"TCP_TCC_ATOMIC_WITH_RET_REQ",
+       "TCP",
+       "71",
+       "<None>",
+       "Total atomic with return requests from TCP to all TCCs"},
+      {"SQC_DCACHE_REQ_READ_8",
+       "SQ",
+       "326",
+       "<None>",
+       "Number of constant cache 8 dw read requests. (per-SQ)"},
+      {"TCC_STREAMING_REQ",
+       "TCC",
+       "4",
+       "<None>",
+       "Number of streaming requests. This is measured at the tag block."},
+      {"SQ_INSTS_SMEM_NORM",
+       "SQ",
+       "161",
+       "<None>",
+       "Number of SMEM instructions issued normalized to match the level of memory accessed "
+       "(i.e. scratch, global, etc). This normalized value is designed to give a hint of "
+       "high cost memory actions being used. The formula used to calculate this value is "
+       "the following (INST_COUNT *2 for load/store; INST_COUNT*2 atomic; INST_COUNT*2 "
+       "memtime; INST_COUNT*4 wb/inv). This value is returned per-SE (aggregate of values "
+       "in SIMDs in the SE)."},
+      {"SQC_ICACHE_MISSES",
+       "SQ",
+       "272",
+       "<None>",
+       "Number of cache misses, includes uncached requests. (per-SQ, per-Bank, "
+       "nondeterministic)"},
+      {"SQ_WAVES_LT_64",
+       "SQ",
+       "7",
+       "<None>",
+       "Count number of waves with <64 active threads sent to SQs. This value represents "
+       "the number of waves that an each individual SIMD has enqueued during the collection "
+       "timeframe (for dispatch profiling this is the timeframe of kernel execution, for "
+       "agent profiling it is the timeframe between start_context and read counter data) "
+       "with less than 64 threads. A sum of all SQ_WAVES_LT_64 values will give the total "
+       "number of waves with 64 threads enqueued during the collection timeframe by the "
+       "application. Returns one value per-SE (aggregates of SIMD values). Useful for "
+       "checking for wavefront occupancy."},
+      {"TCP_TCC_NC_WRITE_REQ",
+       "TCP",
+       "76",
+       "<None>",
+       "Total write requests with NC mtype from this TCP to all TCCs"},
+      {"SQ_LDS_ATOMIC_RETURN",
+       "SQ",
+       "98",
+       "<None>",
+       "The number of atomic return cycles in LDS (local data store). This value is returned "
+       "on a per-SE (aggregate of values in SIMDs in the SE) basis."},
+      {"CPF_CMP_UTCL1_STALL_ON_TRANSLATION",
+       "CPF",
+       "20",
+       "<None>",
+       "One of the Compute UTCL1s is stalled waiting on translation, XNACK or PENDING "
+       "response."},
+      {"TCC_CYCLE",
+       "TCC",
+       "1",
+       "<None>",
+       "Number of cycles. Not windowable."},
+      {"TCP_GATE_EN2",
+       "TCP",
+       "1",
+       "<None>",
+       "TCP core clocks are turned on. Not Windowed."},
+      {"TCC_WRITEBACK",
+       "TCC",
+       "22",
+       "<None>",
+       "Number of lines written back to main memory. This includes writebacks of dirty lines "
+       "and uncached write/atomic requests."},
+      {"SPI_CSN_WINDOW_VALID",
+       "SPI",
+       "47",
+       "<None>",
+       "Clock count enabled by perfcounter_start event. Requires "
+       "SPI_DEBUG_CNTL.DEBUG_PIPE_SEL to select source, DEBUG_PIPE_SEL = 1, source is CS1; "
+       "DEBUG_PIPE_SEL = 2, source is CS2; DEBUG_PIPE_SEL = 3, source is CS3; default, "
+       "source is CS0;"},
+      {"SQ_ACCUM_PREV_HIRES",
+       "SQ",
+       "158",
+       "<None>",
+       "This is a hardware register that can be used for accumulating values for other "
+       "counters. This is useful in expressions where you want to integrate over time. "
+       "This counter is primarily for use with derived counters supplied by rocprof."},
+      {"TCP_TCC_UC_WRITE_REQ",
+       "TCP",
+       "79",
+       "<None>",
+       "Total write requests with UC mtype from this TCP to all TCCs"},
+      {"TCP_UTCL1_TRANSLATION_MISS",
+       "TCP",
+       "48",
+       "<None>",
+       "Total utcl1 translation misses"},
+      {"GRBM_TA_BUSY",
+       "GRBM",
+       "13",
+       "<None>",
+       "Any of the Texture Pipes (TA) are busy in the shader engine(s)."},
+      {"TCC_EA0_ATOMIC_LEVEL",
+       "TCC",
+       "37",
+       "<None>",
+       "The sum of the number of EA atomics in flight. This is primarily meant for measure "
+       "average EA atomic latency. Average atomic latency = "
+       "TCC_PERF_SEL_EA_WRREQ_ATOMIC_LEVEL/TCC_PERF_SEL_EA_WRREQ_ATOMIC."},
+      {"SQ_IFETCH_LEVEL",
+       "SQ",
+       "89",
+       "<None>",
+       "Number of inflight instruction fetch requests from the cache. This is a value "
+       "returned per-sharder engine. Best used with accumlate() functions as part of a "
+       "derived counter."},
+      {"TCC_RW_REQ",
+       "TCC",
+       "8",
+       "<None>",
+       "The number of RW requests. This is measured at the tag block."},
+      {"TCP_TOTAL_WRITE",
+       "TCP",
+       "32",
+       "<None>",
+       "Total number of local write pixels/buffers from TA. Equals "
+       "TCP_PERF_SEL_TOTAL_MISS_LRU_WRITE+ TCP_PERF_SEL_TOTAL_MISS_EVICT_WRITE"},
+      {"TCP_TOTAL_ACCESSES",
+       "TCP",
+       "29",
+       "<None>",
+       "Total number of pixels/buffers from TA. Equals "
+       "TCP_PERF_SEL_TOTAL_READ+TCP_PERF_SEL_TOTAL_NONREAD"},
+      {"SQC_DCACHE_REQ_READ_16",
+       "SQ",
+       "327",
+       "<None>",
+       "Number of constant cache 16 dw read requests. (per-SQ)"},
+      {"SQ_WAIT_ANY",
+       "SQ",
+       "58",
+       "<None>",
+       "Number of wave-cycles spent waiting for anything (per-simd, nondeterministic). "
+       "Units in quad-cycles(4 cycles)"},
+      {"SQ_CYCLES",
+       "SQ",
+       "2",
+       "<None>",
+       "Clock cycles. Value is returned per-SIMD."},
+      {"GRBM_SPI_BUSY",
+       "GRBM",
+       "11",
+       "<None>",
+       "Any of the Shader Pipe Interpolators (SPI) are busy in the shader engine(s)."},
+      {"SQ_INSTS_MFMA",
+       "SQ",
+       "27",
+       "<None>",
+       "Total number of MFMA (Matrix-Fused-Multiply-Add) instructions issued. This value is "
+       "returned per-SE (aggregate of values in SIMDs in the SE). See AMD ISAs for more "
+       "information on MFMA instructions."},
+      {"GRBM_CP_BUSY",
+       "GRBM",
+       "3",
+       "<None>",
+       "Any of the Command Processor (CPG/CPC/CPF) blocks are busy."},
+      {"SQ_ACCUM_PREV",
+       "SQ",
+       "1",
+       "<None>",
+       "This is a hardware register that can be used for accumulating values for other "
+       "counters. This is useful in expressions where you want to integrate over time. "
+       "Only accumulates once every 4 cycles. This counter is primarily for use with "
+       "derived counters supplied by rocprof."},
+      {"TCP_TCC_CC_WRITE_REQ",
+       "TCP",
+       "82",
+       "<None>",
+       "Total write requests with CC mtype from this TCP to all TCCs"},
+      {"SPI_CSN_NUM_THREADGROUPS",
+       "SPI",
+       "49",
+       "<None>",
+       "Number of threadgroups launched. Requires SPI_DEBUG_CNTL.DEBUG_PIPE_SEL to select "
+       "source, DEBUG_PIPE_SEL = 1, source is CS1; DEBUG_PIPE_SEL = 2, source is CS2; "
+       "DEBUG_PIPE_SEL = 3, source is CS3; default, source is CS0;"},
+      {"TCP_TCC_NC_READ_REQ",
+       "TCP",
+       "75",
+       "<None>",
+       "Total read requests with NC mtype from this TCP to all TCCs"},
+      {"TCP_TD_TCP_STALL_CYCLES",
+       "TCP",
+       "7",
+       "<None>",
+       "TD stalls TCP"},
+      {"SQ_INSTS_SENDMSG",
+       "SQ",
+       "40",
+       "<None>",
+       "Total number of Sendmsg (typically an interrupt to the CPU host) instructions "
+       "issued. This value is returned per-SE (aggregate of values in SIMDs in the SE). "
+       "See AMD ISAs for more information on Sendmsg instructions."},
+      {"TA_ADDR_STALLED_BY_TC_CYCLES",
+       "TA",
+       "54",
+       "<None>",
+       "Number of cycles addr path stalled by TC. Perf_Windowing not supported for this "
+       "counter."},
+      {"TA_BUFFER_WRITE_WAVEFRONTS",
+       "TA",
+       "46",
+       "<None>",
+       "Number of buffer write wavefronts processed by TA."},
+      {"SQ_ACTIVE_INST_EXP_GDS",
+       "SQ",
+       "74",
+       "<None>",
+       "Number of cycles each wave spends working on EXPORT or GDS instructions. This value "
+       "represents the number of cycles each wave spends executing instructions "
+       "synchronizing workgroups across the device (global data sync). High values "
+       "indicates large amounts of time spent waiting on communication between CUs. This "
+       "value is returned on a per-SE (aggregate of values in SIMDs in the SE) basis with "
+       "units in quad-cycles(4 cycles). See AMD ISAs for more information on GDS "
+       "instructions."},
+      {"TCC_EA0_WRREQ_DRAM_CREDIT_STALL",
+       "TCC",
+       "33",
+       "<None>",
+       "Number of cycles a EA write request was stalled because the interface was out of "
+       "DRAM credits."},
+      {"TCC_WRITE",
+       "TCC",
+       "13",
+       "<None>",
+       "Number of write requests."},
+      {"SPI_RA_VGPR_SIMD_FULL_CSN",
+       "SPI",
+       "109",
+       "<None>",
+       "Sum of SIMD where VGPR can't take csn wave when !fits. Source is RA0"},
+      {"TCP_TCC_UC_READ_REQ",
+       "TCP",
+       "78",
+       "<None>",
+       "Total read requests with UC mtype from this TCP to all TCCs"},
+      {"TCC_EA0_RDREQ_DRAM",
+       "TCC",
+       "43",
+       "<None>",
+       "Number of TCC/EA read requests (either 32-byte or 64-byte) destined for DRAM (MC)."},
+      {"TCC_EA0_WRREQ_IO_CREDIT_STALL",
+       "TCC",
+       "31",
+       "<None>",
+       "Number of cycles a EA write request was stalled because the interface was out of IO "
+       "credits."},
+      {"TCC_TOO_MANY_EA_WRREQS_STALL",
+       "TCC",
+       "34",
+       "<None>",
+       "Number of cycles the TCC could not send a EA write request because it already "
+       "reached its maximum number of pending EA write requests."},
+      {"TCP_TOTAL_READ",
+       "TCP",
+       "30",
+       "<None>",
+       "Total number of read pixels/buffers from TA. Equals "
+       "TCP_PERF_SEL_TOTAL_HIT_LRU_READ + TCP_PERF_SEL_TOTAL_MISS_LRU_READ + "
+       "TCP_PERF_SEL_TOTAL_MISS_EVICT_READ"},
+      {"SQ_INSTS_VMEM",
+       "SQ",
+       "30",
+       "<None>",
+       "The number of VMEM (GPU Memory) instructions issued. The value is returned per-SE "
+       "(aggregate of values in SIMDs in the SE)."},
+      {"SPI_RA_WAVE_SIMD_FULL_CSN",
+       "SPI",
+       "103",
+       "<None>",
+       "Sum of SIMD where WAVE can't take csn wave when !fits. Source is RA0"},
+      {"SQ_INSTS_VSKIPPED",
+       "SQ",
+       "41",
+       "<None>",
+       "The number of vector instructions skipped. This can occur when the S_SETVSKIP bit "
+       "is enabled on certain instructions. Often this is used as an alturnative to "
+       "branching (a compiler may replace a branch with setting this bit to skip the "
+       "operation, typically as a performance optimization). The value is returned per-SE "
+       "(aggregate of values in SIMDs in the SE)."},
+      {"SQ_ITEMS",
+       "SQ",
+       "14",
+       "<None>",
+       "Number of valid items per wave. This value is returned on a per-SE (aggregate of "
+       "values in SIMDs in the SE) basis."},
+      {"SQ_LEVEL_WAVES",
+       "SQ",
+       "5",
+       "<None>",
+       "Track the number of waves. Set ACCUM_PREV for the next counter to use this. This "
+       "value is returned on a per-SIMD basis."},
+      {"SQC_TC_DATA_WRITE_REQ",
+       "SQ",
+       "265",
+       "<None>",
+       "Number of data write requests to the TC (No-Masking, nondeterministic)"},
+      {"SQ_INST_LEVEL_LDS",
+       "SQ",
+       "44",
+       "<None>",
+       "Number of in-flight LDS instructions. This value represents the number of "
+       "instructions each wave spends executing instructions accessing the local data store "
+       "(data shared between SIMDs on the same CU). Set next counter to ACCUM_PREV and "
+       "divide by INSTS_LDS for average latency. Includes FLAT instructions. This value is "
+       "returned on a per-SE (aggregate of values in SIMDs in the SE) basis."},
+      {"TCP_TOTAL_CACHE_ACCESSES",
+       "TCP",
+       "60",
+       "<None>",
+       "Count of total cache line (tag) accesses (includes hits and misses)."},
+      {"TA_BUFFER_WAVEFRONTS",
+       "TA",
+       "44",
+       "<None>",
+       "Number of buffer wavefronts processed by TA."},
+      {"SQ_WAVES_EQ_64",
+       "SQ",
+       "6",
+       "<None>",
+       "Count number of waves with exactly 64 active threads sent to SQs. This value "
+       "represents the number of waves that an each individual SIMD has enqueued during "
+       "the collection timeframe (for dispatch profiling this is the timeframe of kernel "
+       "execution, for agent profiling it is the timeframe between start_context and read "
+       "counter data) with exactly 64 threads. A sum of all SQ_WAVES_EQ_64 values will "
+       "give the total number of waves with 64 threads enqueued during the collection "
+       "timeframe by the application. Returns one value per-SE (aggregates of SIMD values). "
+       "Useful for checking for wavefront occupancy."},
+      {"TCP_WRITE_TAGCONFLICT_STALL_CYCLES",
+       "TCP",
+       "12",
+       "<None>",
+       "Tagram conflict stall on a write"},
+      {"SQC_TC_INST_REQ",
+       "SQ",
+       "263",
+       "<None>",
+       "Number of insruction requests to the TC (No-Masking, nondeterministic)"},
+      {"SQC_TC_DATA_ATOMIC_REQ",
+       "SQ",
+       "266",
+       "<None>",
+       "Number of data atomic requests to the TC (No-Masking, nondeterministic)"},
+      {"TCC_EA0_RDREQ_IO_CREDIT_STALL",
+       "TCC",
+       "41",
+       "<None>",
+       "Number of cycles there was a stall because the read request interface was out of IO "
+       "credits. Stalls occur regardless of whether a read needed to be performed or not."},
+      {"TCP_TCC_CC_ATOMIC_REQ",
+       "TCP",
+       "83",
+       "<None>",
+       "Total atomic requests with CC mtype from this TCP to all TCCs"},
+      {"TCP_TCC_UC_ATOMIC_REQ",
+       "TCP",
+       "80",
+       "<None>",
+       "Total atomic requests with UC mtype from this TCP to all TCCs"},
+      {"SQC_ICACHE_MISSES_DUPLICATE",
+       "SQ",
+       "273",
+       "<None>",
+       "Number of misses that were duplicates (access to a non-resident, miss pending CL). "
+       "(per-SQ, per-Bank, nondeterministic)"},
+      {"TCC_EA0_RDREQ_LEVEL",
+       "TCC",
+       "44",
+       "<None>",
+       "The sum of the number of TCC/EA read requests in flight. This is primarily meant "
+       "for measure average EA read latency. Average read latency = "
+       "TCC_PERF_SEL_EA_RDREQ_LEVEL/TCC_PERF_SEL_EA_RDREQ."},
+      {"TA_BUFFER_TOTAL_CYCLES",
+       "TA",
+       "49",
+       "<None>",
+       "Number of buffer cycles issued to TC."},
+      {"SQ_WAIT_INST_ANY",
+       "SQ",
+       "61",
+       "<None>",
+       "Number of wave-cycles spent waiting for any instruction issue. Units in "
+       "quad-cycles(4 cycles)."},
+      {"SQ_WAVE_CYCLES",
+       "SQ",
+       "47",
+       "<None>",
+       "The cycles spent executing waves in the CUs. This value is reported per-SE "
+       "(aggregates of SIMD values) and is nondeterministic. Units are in quad-cycles (4 "
+       "cycles). Useful for determining how much time is spent executing wave code vs "
+       "overhead/waiting. Low cycle count relative to actual number of cycles processed by "
+       "the CU can indicate that the CU is stalling or is overloaded."},
+      {"TCC_EA0_WRREQ_DRAM",
+       "TCC",
+       "103",
+       "<None>",
+       "Number of TCC/EA write requests (either 32-byte of 64-byte) destined for DRAM (MC)."},
+      {"TCP_TCR_TCP_STALL_CYCLES",
+       "TCP",
+       "8",
+       "<None>",
+       "TCR stalls TCP_TCR_req interface"},
+      {"TCP_TCC_RW_READ_REQ",
+       "TCP",
+       "85",
+       "<None>",
+       "Total write requests with RW mtype from this TCP to all TCCs"},
+      {"SQ_INST_LEVEL_VMEM",
+       "SQ",
+       "42",
+       "<None>",
+       "Number of in-flight VMEM instructions. Set next counter to ACCUM_PREV and divide by "
+       "INSTS_VMEM for average latency. Includes FLAT instructions. This value is returned "
+       "on a per-SE (aggregate of values in SIMDs in the SE) basis."},
+      {"TCP_TCC_CC_READ_REQ",
+       "TCP",
+       "81",
+       "<None>",
+       "Total write requests with CC mtype from this TCP to all TCCs"},
+      {"SQ_ACTIVE_INST_VMEM",
+       "SQ",
+       "70",
+       "<None>",
+       "Number of cycles each wave spends working on a VMEM instructions. This value "
+       "represents the number of cycles each wave spends executing vector memory "
+       "instructions. High values indicates a large amount of time spent executing vector "
+       "memory operations. This value is returned on a per-SE (aggregate of values in "
+       "SIMDs in the SE) basis with units in quad-cycles(4 cycles)."},
+      {"SQ_IFETCH",
+       "SQ",
+       "88",
+       "<None>",
+       "Number of instruction fetch requests from L1I (instruction) cache. This is a value "
+       "returned per-SIMD."},
+      {"TCP_TCC_READ_REQ",
+       "TCP",
+       "69",
+       "<None>",
+       "Total read requests from TCP to all TCCs"},
+      {"SQC_DCACHE_REQ",
+       "SQ",
+       "290",
+       "<None>",
+       "Number of requests (post-bank-serialization). (per-SQ, per-Bank)"},
+      {"CPC_CPC_STAT_STALL",
+       "CPC",
+       "27",
+       "<None>",
+       "CPC Stalled."},
+      {"TCP_GATE_EN1",
+       "TCP",
+       "0",
+       "<None>",
+       "TCP interface clocks are turned on. Not Windowed."},
+      {"TCP_PENDING_STALL_CYCLES",
+       "TCP",
+       "22",
+       "<None>",
+       "Stall due to data pending from L2"},
+      {"SQC_DCACHE_MISSES_DUPLICATE",
+       "SQ",
+       "293",
+       "<None>",
+       "Number of misses that were duplicates (access to a non-resident, miss pending CL). "
+       "(per-SQ, per-Bank, nondeterministic)"},
+      {"CPF_CPF_STAT_IDLE",
+       "CPF",
+       "24",
+       "<None>",
+       "CPF Idle."},
+      {"TCP_VOLATILE",
+       "TCP",
+       "28",
+       "<None>",
+       "Total number of L1 volatile pixels/buffers from TA"},
+      {"CPC_CPC_TCIU_BUSY",
+       "CPC",
+       "28",
+       "<None>",
+       "CPC TCIU interface Busy."},
+      {"SQC_DCACHE_REQ_READ_2",
+       "SQ",
+       "324",
+       "<None>",
+       "Number of constant cache 2 dw read requests. (per-SQ)"},
+      {"CPC_CPC_STAT_BUSY",
+       "CPC",
+       "25",
+       "<None>",
+       "CPC Busy."},
+      {"TCP_TCP_LATENCY",
+       "TCP",
+       "65",
+       "<None>",
+       "Total TCP wave latency (from first clock of wave entering to first clock of wave "
+       "leaving), divide by TA_TCP_STATE_READ to avg wave latency"},
+      {"TCP_UTCL1_TRANSLATION_HIT",
+       "TCP",
+       "49",
+       "<None>",
+       "Total utcl1 translation hits"},
+      {"SQ_INST_LEVEL_SMEM",
+       "SQ",
+       "43",
+       "<None>",
+       "Number of in-flight SMEM instructions (*2 load/store; *2 atomic; *2 memtime; *4 "
+       "wb/inv). Set next counter to ACCUM_PREV and divide by INSTS_SMEM for average "
+       "latency per smem request. Falls slightly short of total request latency because "
+       "some fetches are divided into two requests that may finish at different times and "
+       "this counter collects the average latency of the two. This value is returned on a "
+       "per-SE (aggregate of values in SIMDs in the SE) basis."},
+      {"SPI_SWC_CSC_WR",
+       "SPI",
+       "189",
+       "<None>",
+       "Number of clocks to write CSC waves to SGPRs (need to multiply this value by 4) "
+       "Requires SPI_DEBUG_CNTL.DEBUG_PIPE_SEL to select source, DEBUG_PIPE_SEL = 1, "
+       "source is CS1; DEBUG_PIPE_SEL = 2, source is CS2; DEBUG_PIPE_SEL = 3, source is "
+       "CS3; default, source is CS0;"},
+      {"TCC_READ",
+       "TCC",
+       "12",
+       "<None>",
+       "Number of read requests. Compressed reads are included in this, but metadata reads "
+       "are not included."},
+      {"TD_LOAD_WAVEFRONT",
+       "TD",
+       "25",
+       "<None>",
+       "Count the wavefronts with opcode = load, include atomics and store."},
+      {"GRBM_EA_BUSY",
+       "GRBM",
+       "35",
+       "<None>",
+       "The Efficiency Arbiter (EA) block is busy."},
+      {"SPI_RA_WVLIM_STALL_CSN",
+       "SPI",
+       "133",
+       "<None>",
+       "Number of clocks csn is stalled due to WAVE LIMIT."},
+      {"SPI_RA_BAR_CU_FULL_CSN",
+       "SPI",
+       "123",
+       "<None>",
+       "Sum of CU where BARRIER can't take csn wave when !fits. Source is RA0"},
+      {"TD_TD_BUSY",
+       "TD",
+       "1",
+       "<None>",
+       "TD is processing or waiting for data. Perf_Windowing not supported for this "
+       "counter."},
+      {"SQC_ICACHE_REQ",
+       "SQ",
+       "270",
+       "<None>",
+       "Number of requests. (per-SQ, per-Bank)"},
+      {"TCC_ATOMIC",
+       "TCC",
+       "14",
+       "<None>",
+       "Number of atomic requests of all types."},
+      {"TCP_ATOMIC_TAGCONFLICT_STALL_CYCLES",
+       "TCP",
+       "13",
+       "<None>",
+       "Tagram conflict stall on an atomic"},
+      {"CPF_CPF_STAT_BUSY",
+       "CPF",
+       "23",
+       "<None>",
+       "CPF Busy."},
+      {"TCC_EA0_WRREQ_LEVEL",
+       "TCC",
+       "35",
+       "<None>",
+       "The sum of the number of EA write requests in flight. This is primarily meant for "
+       "measure average EA write latency. Average write latency = "
+       "TCC_PERF_SEL_EA_WRREQ_LEVEL/TCC_PERF_SEL_EA_WRREQ."},
+      {"SPI_RA_RES_STALL_CSN",
+       "SPI",
+       "91",
+       "<None>",
+       "Arb cycles with CSn req and no CSn fits. Source is RA0"},
+      {"TCC_BUSY",
+       "TCC",
+       "2",
+       "<None>",
+       "Number of cycles we have a request pending. Not windowable."},
+      {"SQC_DCACHE_INPUT_VALID_READYB",
+       "SQ",
+       "260",
+       "<None>",
+       "Input stalled by SQC (per-SQ, nondeterministic, unwindowed)"},
+      {"SQ_WAVES_SAVED",
+       "SQ",
+       "160",
+       "<None>",
+       "Count number of context-saved waves sent to SQs. This value represents the number "
+       "of waves whos current register state has been saved to a register bank during the "
+       "collection timeframe (for dispatch profiling this is the timeframe of kernel "
+       "execution, for agent profiling it is the timeframe between start_context and read "
+       "counter data) . Context saving/restoring is a slow operation and should be limited. "
+       "High values can also indicate that stalling may be taking place (waiting for free "
+       "register space). Returns one value per-SE (aggregates of SIMD values)."},
+      {"SQ_ACTIVE_INST_LDS",
+       "SQ",
+       "71",
+       "<None>",
+       "Number of cycles each wave spends working on LDS instructions. This value represents "
+       "the number of cycles each wave spends executing instructions accessing the local "
+       "data store (data shared between SIMDs on the same CU). High values indicates a "
+       "large amount of reading/writing to this shared memory space. This value is returned "
+       "on a per-SE (aggregate of values in SIMDs in the SE) basis with units in "
+       "quad-cycles(4 cycles). See AMD ISAs for more information on LDS instructions."},
+      {"TCC_EA0_RD_UNCACHED_32B",
+       "TCC",
+       "40",
+       "<None>",
+       "Number of 32-byte TCC/EA read due to uncached traffic. A 64-byte request will be "
+       "counted as 2"},
+      {"TCP_TCC_RW_WRITE_REQ",
+       "TCP",
+       "86",
+       "<None>",
+       "Total write requests with RW mtype from this TCP to all TCCs"},
+      {"TCC_EA0_RDREQ_DRAM_CREDIT_STALL",
+       "TCC",
+       "43",
+       "<None>",
+       "Number of cycles there was a stall because the read request interface was out of "
+       "DRAM credits. Stalls occur regardless of whether a read needed to be performed or "
+       "not."},
+      {"SPI_VWC_CSC_WR",
+       "SPI",
+       "195",
+       "<None>",
+       "Number of clocks to write CSC waves to VGPRs (need to multiply this value by 4) "
+       "Requires SPI_DEBUG_CNTL.DEBUG_PIPE_SEL to select source, DEBUG_PIPE_SEL = 1, "
+       "source is CS1; DEBUG_PIPE_SEL = 2, source is CS2; DEBUG_PIPE_SEL = 3, source is "
+       "CS3; default, source is CS0;"},
+      {"SQC_DCACHE_REQ_READ_4",
+       "SQ",
+       "325",
+       "<None>",
+       "Number of constant cache 4 dw read requests. (per-SQ)"},
+      {"SQ_WAVES_LT_16",
+       "SQ",
+       "10",
+       "<None>",
+       "Count number of waves sent <16 active threads sent to SQs. (per-simd, emulated, "
+       "global). This value represents the number of waves that an each individual SIMD "
+       "has enqueued during the collection timeframe (for dispatch profiling this is the "
+       "timeframe of kernel execution, for agent profiling it is the timeframe between "
+       "start_context and read counter data) with less than 16 threads. A sum of all "
+       "SQ_WAVES_LT_16 values will give the total number of waves with 16 threads enqueued "
+       "during the collection timeframe by the application. Returns one value per-SE "
+       "(aggregates of SIMD values). Useful for checking for wavefront occupancy."},
+      {"SQC_DCACHE_ATOMIC",
+       "SQ",
+       "298",
+       "<None>",
+       "Number of atomic requests. (per-SQ, per-Bank)"},
+      {"TCC_EA0_RDREQ_GMI_CREDIT_STALL",
+       "TCC",
+       "42",
+       "<None>",
+       "Number of cycles there was a stall because the read request interface was out of "
+       "GMI credits. Stalls occur regardless of whether a read needed to be performed or "
+       "not."},
+      {"SPI_RA_REQ_NO_ALLOC",
+       "SPI",
+       "79",
+       "<None>",
+       "Arb cycles with requests but no allocation. Source is RA0"},
+      {"SQC_DCACHE_MISSES",
+       "SQ",
+       "292",
+       "<None>",
+       "Number of cache misses, includes uncached requests. (per-SQ, per-Bank, "
+       "nondeterministic)"},
+      {"TCC_NC_REQ",
+       "TCC",
+       "5",
+       "<None>",
+       "The number of noncoherently cached requests. This is measured at the tag block."},
+      {"SQ_ACTIVE_INST_FLAT",
+       "SQ",
+       "76",
+       "<None>",
+       "Number of cycles each wave spends working on FLAT instructions. This value "
+       "represents the number of cycles each wave spends executing instructions accessing "
+       "flat scratch memory locations. High values indicates a large amount of "
+       "reading/writing to scratch memory on the device. This value is returned on a "
+       "per-SE (aggregate of values in SIMDs in the SE) basis with units in quad-cycles(4 "
+       "cycles). See AMD ISAs for more information on FLAT instructions."},
+      {"SQC_TC_DATA_READ_REQ",
+       "SQ",
+       "264",
+       "<None>",
+       "Number of data read requests to the TC (No-Masking, nondeterministic)"},
+      {"TCP_TCC_READ_REQ_LATENCY",
+       "TCP",
+       "66",
+       "<None>",
+       "Total TCP->TCC request latency for reads and atomics with return. Not Windowed."},
+      {"CPC_UTCL1_STALL_ON_TRANSLATION",
+       "CPC",
+       "24",
+       "<None>",
+       "One of the UTCL1s is stalled waiting on translation, XNACK or PENDING response."},
+      {"TCC_PROBE",
+       "TCC",
+       "9",
+       "<None>",
+       "Number of probe requests. Not windowable."},
+      {"TA_BUFFER_ATOMIC_WAVEFRONTS",
+       "TA",
+       "47",
+       "<None>",
+       "Number of buffer atomic wavefronts processed by TA."},
+      {"TA_DATA_STALLED_BY_TC_CYCLES",
+       "TA",
+       "56",
+       "<None>",
+       "Number of cycles data path stalled by TC. Perf_Windowing not supported for this "
+       "counter."},
+      {"TA_FLAT_WAVEFRONTS",
+       "TA",
+       "100",
+       "<None>",
+       "Number of flat opcode wavfronts processed by the TA."},
+      {"TA_TOTAL_WAVEFRONTS",
+       "TA",
+       "32",
+       "<None>",
+       "Total number of wavefronts processed by TA."},
+      {"CPC_CPC_STAT_IDLE",
+       "CPC",
+       "26",
+       "<None>",
+       "CPC Idle."},
+      {"CPC_CPC_UTCL2IU_STALL",
+       "CPC",
+       "32",
+       "<None>",
+       "CPC UTCL2 interface Stalled waiting on Free, Tags or Translation."},
+      {"TCC_NORMAL_WRITEBACK",
+       "TCC",
+       "68",
+       "<None>",
+       "Number of writebacks due to requests that are not writeback requests."},
+      {"SQ_INST_CYCLES_SMEM",
+       "SQ",
+       "84",
+       "<None>",
+       "The number of cycles needed to execute scalar memory reads (SMEM). This value is "
+       "returned on a per-SE (aggregate of values in SIMDs in the SE) basis with units in "
+       "quad-cycles(4 cycles)."},
+      {"SQ_BUSY_CYCLES",
+       "SQ",
+       "3",
+       "<None>",
+       "Number of clock cycles there are active waves in a shader engine (as reported by "
+       "the distributed sequencer). This value does not denote the number of active waves, "
+       "only the clock cycle in which any wave is present in a SE. This value is returned "
+       "on a per-shader engine basis in clock cycles."},
+      {"SQ_LDS_UNALIGNED_STALL",
+       "SQ",
+       "96",
+       "<None>",
+       "Number of cycles LDS (local data store) is stalled processing flat unaligned "
+       "load/store ops. This value is returned on a per-SE (aggregate of values in SIMDs "
+       "in the SE) basis."},
+      {"TA_BUFFER_READ_WAVEFRONTS",
+       "TA",
+       "45",
+       "<None>",
+       "Number of buffer read wavefronts processed by TA."},
+      {"TA_FLAT_ATOMIC_WAVEFRONTS",
+       "TA",
+       "103",
+       "<None>",
+       "Number of flat opcode atomics processed by the TA."},
+      {"GRBM_TC_BUSY",
+       "GRBM",
+       "28",
+       "<None>",
+       "Any of the Texture Cache Blocks (TCP/TCI/TCA/TCC) are busy."},
+      {"GRBM_CPC_BUSY",
+       "GRBM",
+       "30",
+       "<None>",
+       "The Command Processor Compute (CPC) is busy."},
+      {"TCP_UTCL1_PERMISSION_MISS",
+       "TCP",
+       "50",
+       "<None>",
+       "Total utcl1 permission misses"},
+      {"SPI_RA_BULKY_CU_FULL_CSN",
+       "SPI",
+       "125",
+       "<None>",
+       "Sum of CU where BULKY can't take csn wave when !fits. Source is RA0"},
+      {"TCP_TA_TCP_STATE_READ",
+       "TCP",
+       "27",
+       "<None>",
+       "Number of state reads"},
+      {"TCP_TCC_WRITE_REQ",
+       "TCP",
+       "70",
+       "<None>",
+       "Total write requests from TCP to all TCCs"},
+      {"TCP_TCC_RW_ATOMIC_REQ",
+       "TCP",
+       "87",
+       "<None>",
+       "Total atomic requests with RW mtype from this TCP to all TCCs"},
+      {"SQ_ACTIVE_INST_MISC",
+       "SQ",
+       "75",
+       "<None>",
+       "Number of cycles each wave spends working on a BRANCH or SENDMSG instructions. This "
+       "value represents the number of cycles each wave spends executing instructions "
+       "performing control flow branching and message sending. This value is returned on a "
+       "per-SE (aggregate of values in SIMDs in the SE) basis with units in quad-cycles(4 "
+       "cycles). See AMD ISAs for more information on BRANCH and SENDMSG instructions."},
+      {"SQ_ACTIVE_INST_SCA",
+       "SQ",
+       "73",
+       "<None>",
+       "Number of cycles each wave spends working on a SALU or SMEM instructions. This "
+       "value represents the number of cycles each wave spends executing scalar ALU or "
+       "scalar memory instructions. On MI200/300 platforms, there is a single ALU per CU. "
+       "High values indicates a large amount of time spent executing scalar instructions. "
+       "This value is returned on a per-SE (aggregate of values in SIMDs in the SE) basis "
+       "with units in quad-cycles(4 cycles). See AMD ISAs for more information on SALU "
+       "and SMEM instructions."},
+      {"TD_COALESCABLE_WAVEFRONT",
+       "TD",
+       "32",
+       "<None>",
+       "Count wavefronts that TA finds coalescable."},
+      {"SPI_RA_TMP_STALL_CSN",
+       "SPI",
+       "97",
+       "<None>",
+       "Cycles where csn wants to req but does not fit in temp space."},
+      {"SQ_BUSY_CU_CYCLES",
+       "SQ",
+       "13",
+       "<None>",
+       "Number of quad-cycles each CU is busy. Can be used to calculate the percentage of "
+       "time each CU is busy. This value is returned on a per-SE (aggregate of values in "
+       "SIMDs in the SE) basis with units in quad-cycles(4 cycles)."},
+      {"TCA_BUSY",
+       "TCA",
+       "2",
+       "<None>",
+       "Number of cycles we have a request pending. Not windowable."},
+      {"TCP_TOTAL_ATOMIC_WITH_RET",
+       "TCP",
+       "38",
+       "<None>",
+       "Total number of atomic with return pixels/buffers from TA"},
+      {"SQC_ICACHE_HITS",
+       "SQ",
+       "271",
+       "<None>",
+       "Number of cache hits. (per-SQ, per-Bank, nondeterministic)"},
+      {"TCA_CYCLE",
+       "TCA",
+       "1",
+       "<None>",
+       "Number of cycles. Not windowable."},
+      {"SPI_CSN_BUSY",
+       "SPI",
+       "48",
+       "<None>",
+       "Number of clocks with outstanding waves (SPI or SH). Requires "
+       "SPI_DEBUG_CNTL.DEBUG_PIPE_SEL to select source, DEBUG_PIPE_SEL = 1, source is "
+       "CS1; DEBUG_PIPE_SEL = 2, source is CS2; DEBUG_PIPE_SEL = 3, source is CS3; "
+       "default, source is CS0;"},
+      {"TCC_UC_REQ",
+       "TCC",
+       "6",
+       "<None>",
+       "The number of uncached requests. This is measured at the tag block."},
+      {"SQC_DCACHE_REQ_READ_1",
+       "SQ",
+       "323",
+       "<None>",
+       "Number of constant cache 1 dw read requests. (per-SQ)"},
+      {"SQ_ACTIVE_INST_ANY",
+       "SQ",
+       "69",
+       "<None>",
+       "Number of cycles each wave spends working on any type of instruction. Useful in "
+       "determining percentage of time spend executing wave workloads (see WaveExec). This "
+       "value is returned on a per-SE (aggregate of values in SIMDs in the SE) basis with "
+       "units in quad-cycles(4 cycles)."},
+      {"TCP_TCC_ATOMIC_WITHOUT_RET_REQ",
+       "TCP",
+       "72",
+       "<None>",
+       "Total atomic without return requests from TCP to all TCCs"},
+      {"TCP_TCC_WRITE_REQ_LATENCY",
+       "TCP",
+       "67",
+       "<None>",
+       "Total TCP->TCC request latency for writes and atomics without return. Not Windowed."},
+      {"TCP_READ_TAGCONFLICT_STALL_CYCLES",
+       "TCP",
+       "11",
+       "<None>",
+       "Tagram conflict stall on a read"},
+      {"TCC_EA0_WRREQ_STALL",
+       "TCC",
+       "30",
+       "<None>",
+       "Number of cycles a write request was stalled."},
+      {"SQC_TC_REQ",
+       "SQ",
+       "262",
+       "<None>",
+       "Total number of TC requests that were issued by instruction and constant caches. "
+       "(No-Masking, nondeterministic)"},
+      {"CPF_CPF_STAT_STALL",
+       "CPF",
+       "25",
+       "<None>",
+       "CPF Stalled."},
+      {"TCC_ALL_TC_OP_INV_EVICT",
+       "TCC",
+       "80",
+       "<None>",
+       "Number of evictions due to all TC_OP invalidate requests."},
+      {"SQ_LDS_ADDR_CONFLICT",
+       "SQ",
+       "95",
+       "<None>",
+       "Number of cycles LDS (local data store) is stalled by address conflicts. This value "
+       "is returned on a per-SE (aggregate of values in SIMDs in the SE) basis."},
+      {"SQ_LDS_IDX_ACTIVE",
+       "SQ",
+       "99",
+       "<None>",
+       "Number of cycles LDS (local data store) is used for indexed (non-direct,"
+       "non-interpolation) operations. This value is returned on a per-SE (aggregate of "
+       "values in SIMDs in the SE) basis."},
+      {"SQ_INST_CYCLES_VMEM_RD",
+       "SQ",
+       "78",
+       "<None>",
+       "The number of cycles needed to send addr and cmd data for VMEM read instructions. "
+       "This value is returned on a per-SE (aggregate of values in SIMDs in the SE) basis "
+       "with units in quad-cycles(4 cycles)."},
+      {"CPC_CPC_UTCL2IU_BUSY",
+       "CPC",
+       "30",
+       "<None>",
+       "CPC UTCL2 interface Busy."},
+      {"TCP_UTCL1_REQUEST",
+       "TCP",
+       "47",
+       "<None>",
+       "Total CLIENT_UTCL1 NORMAL requests"},
+      {"CPF_CPF_TCIU_STALL",
+       "CPF",
+       "28",
+       "<None>",
+       "CPF TCIU interface Stalled waiting on Free, Tags."},
+      {"SQ_INSTS_BRANCH",
+       "SQ",
+       "39",
+       "<None>",
+       "Total number of BRANCH instructions issued. This value is returned per-SE "
+       "(aggregate of values in SIMDs in the SE). This value SHOULD NOT be used in "
+       "combination with SQ_ACTIVE_INST_MISC to calculate latency. SQ_ACTIVE_INST_MISC "
+       "includes both BRANCH and SENDMSG instructions while this is only BRANCH."},
+      {"SPI_RA_LDS_CU_FULL_CSN",
+       "SPI",
+       "120",
+       "<None>",
+       "Sum of CU where LDS can't take csn wave when !fits. Source is RA0"},
+      {"TD_ATOMIC_WAVEFRONT",
+       "TD",
+       "26",
+       "<None>",
+       "Count the wavefronts with opcode = atomic."},
+      {"SQ_INSTS_EXP_GDS",
+       "SQ",
+       "38",
+       "<None>",
+       "Total number of EXPORT or GDS (global wave state) instructions issued. When used in "
+       "combination with SQ_ACTIVE_INST_EXP_GDS (cycle count for executing instructions) "
+       "the average latency of EXPORT/GDS instruction execution can be calculated "
+       "(SQ_ACTIVE_INST_EXP_GDS / SQ_INSTS_EXP_GDS). This value is returned per-SE "
+       "(aggregate of values in SIMDs in the SE)."},
+      {"SQC_TC_STALL",
+       "SQ",
+       "267",
+       "<None>",
+       "Valid request stalled TC request interface (no-credits). (No-Masking, "
+       "nondeterministic, unwindowed)"},
+      {"CPF_CPF_TCIU_BUSY",
+       "CPF",
+       "26",
+       "<None>",
+       "CPF TCIU interface Busy."},
+      {"TCC_EA0_WRREQ_GMI_CREDIT_STALL",
+       "TCC",
+       "32",
+       "<None>",
+       "Number of cycles a EA write request was stalled because the interface was out of "
+       "GMI credits."},
+      {"GRBM_CPF_BUSY",
+       "GRBM",
+       "31",
+       "<None>",
+       "The Command Processor Fetchers (CPF) is busy."},
+      {"SQ_WAVES_LT_48",
+       "SQ",
+       "8",
+       "<None>",
+       "Count number of waves with <48 active threads sent to SQs. This value represents "
+       "the number of waves that an each individual SIMD has enqueued during the collection "
+       "timeframe (for dispatch profiling this is the timeframe of kernel execution, for "
+       "agent profiling it is the timeframe between start_context and read counter data) "
+       "with less than 48 threads. A sum of all SQ_WAVES_LT_48 values will give the total "
+       "number of waves with 48 threads enqueued during the collection timeframe by the "
+       "application. Returns one value per-SE (aggregates of SIMD values). Useful for "
+       "checking for wavefront occupancy."},
+      {"TCC_EA0_ATOMIC",
+       "TCC",
+       "36",
+       "<None>",
+       "Number of transactions going over the TC_EA_wrreq interface that are actually "
+       "atomic requests."},
+      {"TD_TC_STALL",
+       "TD",
+       "15",
+       "<None>",
+       "TD is stalled waiting for TC data."},
+      {"SPI_RA_TGLIM_CU_FULL_CSN",
+       "SPI",
+       "127",
+       "<None>",
+       "Cycles where csn wants to req but all CU are at tg_limit"},
+      {"TA_BUFFER_COALESCED_WRITE_CYCLES",
+       "TA",
+       "53",
+       "<None>",
+       "Number of buffer coalesced write cycles issued to TC."},
+      {"TCP_TOTAL_ATOMIC_WITHOUT_RET",
+       "TCP",
+       "39",
+       "<None>",
+       "Total number of atomic without return pixels/buffers from TA"},
+      {"CPC_ME1_BUSY_FOR_PACKET_DECODE",
+       "CPC",
+       "13",
+       "<None>",
+       "Me1 busy for packet decode."},
+      {"SQ_INSTS",
+       "SQ",
+       "25",
+       "<None>",
+       "Total number of instructions issued. When used in combination with "
+       "SQ_ACTIVE_INST_ANY (cycle count for executing instructions) the average latency of "
+       "instruction execution can be calculated (SQ_ACTIVE_INST_ANY / SQ_INSTS). This "
+       "value is returned per-SE (aggregate of values in SIMDs in the SE)."},
+      {"TCC_NORMAL_EVICT",
+       "TCC",
+       "74",
+       "<None>",
+       "Number of evictions due to requests that are not invalidate or probe requests."},
+      {"CPC_CPC_UTCL2IU_IDLE",
+       "CPC",
+       "31",
+       "<None>",
+       "CPC UTCL2 interface Idle."},
+      {"TCC_REQ",
+       "TCC",
+       "3",
+       "<None>",
+       "Number of requests of all types. This is measured at the tag block. This may be "
+       "more than the number of requests arriving at the TCC, but it is a good indication "
+       "of the total amount of work that needs to be performed."},
+      {"TCC_TAG_STALL",
+       "TCC",
+       "45",
+       "<None>",
+       "Number of cycles the normal request pipeline in the tag was stalled for any reason. "
+       "Normally, stalls of this nature are measured exactly from one point the pipeline, "
+       "but that is not the case for this counter. Probes can stall the pipeline at a "
+       "variety of places, and there is no single point that can reasonably measure the "
+       "total stalls accurately."},
+      {"TD_STORE_WAVEFRONT",
+       "TD",
+       "27",
+       "<None>",
+       "Count the wavefronts with opcode = store."},
+      {"TA_BUFFER_COALESCED_READ_CYCLES",
+       "TA",
+       "52",
+       "<None>",
+       "Number of buffer coalesced read cycles issued to TC."},
+      {"SQ_LDS_MEM_VIOLATIONS",
+       "SQ",
+       "97",
+       "<None>",
+       "Number of threads that have a memory violation in the LDS (local data store). This "
+       "value is returned on a per-SE (aggregate of values in SIMDs in the SE) basis."},
+      {"TCC_CLIENT184_REQ",
+       "TCC",
+       "312",
+       "<None>",
+       "Number of cycles client184 sent a request to this TCC."},
+      {"TCC_CLIENT185_REQ",
+       "TCC",
+       "313",
+       "<None>",
+       "Number of cycles client185 sent a request to this TCC."},
+      {"TCC_CLIENT186_REQ",
+       "TCC",
+       "314",
+       "<None>",
+       "Number of cycles client186 sent a request to this TCC."},
+      {"TCC_CLIENT187_REQ",
+       "TCC",
+       "315",
+       "<None>",
+       "Number of cycles client187 sent a request to this TCC."},
+      {"TCC_CLIENT188_REQ",
+       "TCC",
+       "316",
+       "<None>",
+       "Number of cycles client188 sent a request to this TCC."},
+      {"TCC_CLIENT189_REQ",
+       "TCC",
+       "317",
+       "<None>",
+       "Number of cycles client189 sent a request to this TCC."},
+      {"TCC_CLIENT190_REQ",
+       "TCC",
+       "318",
+       "<None>",
+       "Number of cycles client190 sent a request to this TCC."},
+      {"TCC_CLIENT191_REQ",
+       "TCC",
+       "319",
+       "<None>",
+       "Number of cycles client191 sent a request to this TCC."}}}};
 
 static const std::unordered_map<std::string, std::vector<std::vector<std::string>>> derived_gfx908 =
     {{"gfx908",
@@ -431,4 +1711,533 @@ static const std::unordered_map<std::string, std::vector<std::vector<std::string
         "",
         "100*reduce(SQ_LDS_BANK_CONFLICT,sum)/reduce(GRBM_GUI_ACTIVE,max)/CU_NUM",
         "The percentage of GPUTime LDS is stalled by bank conflicts. Value range: 0\% (optimal) "
-        "to 100\% (bad)."}}}};
+        "to 100\% (bad)."},
+       {"TCC_WRITEBACK_sum",
+        "",
+        "",
+        "reduce(TCC_WRITEBACK,sum)",
+        "Number of lines written back to main memory. This includes writebacks of dirty lines and "
+        "uncached write/atomic requests. Sum over TCC instances."},
+       {"TCC_CC_REQ_sum",
+        "",
+        "",
+        "reduce(TCC_CC_REQ,sum)",
+        "The number of coherently cached requests. This is measured at the tag block. Sum over TCC "
+        "instances."},
+       {"TA_DATA_STALLED_BY_TC_CYCLES_sum",
+        "",
+        "",
+        "reduce(TA_DATA_STALLED_BY_TC_CYCLES,sum)",
+        "Number of cycles data path stalled by TC. Perf_Windowing not supported for this counter. "
+        "Sum over TA instances."},
+       {"TA_TA_BUSY_sum",
+        "",
+        "",
+        "reduce(TA_TA_BUSY,sum)",
+        "TA block is busy. Perf_Windowing not supported for this counter. Sum over TA instances."},
+       {"TCP_TOTAL_WRITE_sum",
+        "",
+        "",
+        "reduce(TCP_TOTAL_WRITE,sum)",
+        "Total number of local write pixels/buffers from TA. Equals "
+        "TCP_PERF_SEL_TOTAL_MISS_LRU_WRITE+ TCP_PERF_SEL_TOTAL_MISS_EVICT_WRITE. Sum over TCP "
+        "instances."},
+       {"TCP_TCC_CC_WRITE_REQ_sum",
+        "",
+        "",
+        "reduce(TCP_TCC_CC_WRITE_REQ,sum)",
+        "Total write requests with CC mtype from this TCP to all TCCs Sum over TCP instances."},
+       {"TA_BUFFER_COALESCED_READ_CYCLES_sum",
+        "",
+        "",
+        "reduce(TA_BUFFER_COALESCED_READ_CYCLES,sum)",
+        "Number of buffer coalesced read cycles issued to TC. Sum over TA instances."},
+       {"TD_TD_BUSY_sum",
+        "",
+        "",
+        "reduce(TD_TD_BUSY,sum)",
+        "TD is processing or waiting for data. Perf_Windowing not supported for this counter. "
+        "Sum over TD instances."},
+       {"TA_BUFFER_WRITE_WAVEFRONTS_sum",
+        "",
+        "",
+        "reduce(TA_BUFFER_WRITE_WAVEFRONTS,sum)",
+        "Number of buffer write wavefronts processed by TA. Sum over TA instances."},
+       {"TCP_TCC_UC_READ_REQ_sum",
+        "",
+        "",
+        "reduce(TCP_TCC_UC_READ_REQ,sum)",
+        "Total read requests with UC mtype from this TCP to all TCCs Sum over TCP instances."},
+       {"TCP_TCC_UC_WRITE_REQ_sum",
+        "",
+        "",
+        "reduce(TCP_TCC_UC_WRITE_REQ,sum)",
+        "Total write requests with UC mtype from this TCP to all TCCs Sum over TCP instances."},
+       {"TCP_TCC_CC_READ_REQ_sum",
+        "",
+        "",
+        "reduce(TCP_TCC_CC_READ_REQ,sum)",
+        "Total write requests with CC mtype from this TCP to all TCCs Sum over TCP instances."},
+       {"TCP_VOLATILE_sum",
+        "",
+        "",
+        "reduce(TCP_VOLATILE,sum)",
+        "Total number of L1 volatile pixels/buffers from TA. Sum over TCP instances."},
+       {"TCP_TOTAL_CACHE_ACCESSES_sum",
+        "",
+        "",
+        "reduce(TCP_TOTAL_CACHE_ACCESSES,sum)",
+        "Count of total cache line (tag) accesses (includes hits and misses). Sum over TCP "
+        "instances."},
+       {"TA_BUFFER_COALESCED_WRITE_CYCLES_sum",
+        "",
+        "",
+        "reduce(TA_BUFFER_COALESCED_WRITE_CYCLES,sum)",
+        "Number of buffer coalesced write cycles issued to TC. Sum over TA instances."},
+       {"TCP_TCC_CC_ATOMIC_REQ_sum",
+        "",
+        "",
+        "reduce(TCP_TCC_CC_ATOMIC_REQ,sum)",
+        "Total atomic requests with CC mtype from this TCP to all TCCs Sum over TCP instances."},
+       {"TCC_EA0_WR_UNCACHED_32B_sum",
+        "",
+        "",
+        "reduce(TCC_EA0_WR_UNCACHED_32B,sum)",
+        "Number of 32-byte write/atomic going over the TC_EA_wrreq interface due to uncached "
+        "traffic. Note that CC mtypes can produce uncached requests, and those are included in "
+        "this. A 64-byte request will be counted as 2. Sum over TCC instances."},
+       {"TA_FLAT_WAVEFRONTS_sum",
+        "",
+        "",
+        "reduce(TA_FLAT_WAVEFRONTS,sum)",
+        "Number of flat opcode wavfronts processed by the TA. Sum over TA instances."},
+       {"TCP_TOTAL_ACCESSES_sum",
+        "",
+        "",
+        "reduce(TCP_TOTAL_ACCESSES,sum)",
+        "Total number of pixels/buffers from TA. Equals "
+        "TCP_PERF_SEL_TOTAL_READ+TCP_PERF_SEL_TOTAL_NONREAD. Sum over TCP instances."},
+       {"TA_FLAT_ATOMIC_WAVEFRONTS_sum",
+        "",
+        "",
+        "reduce(TA_FLAT_ATOMIC_WAVEFRONTS,sum)",
+        "Number of flat opcode atomics processed by the TA. Sum over TA instances."},
+       {"TCP_GATE_EN2_sum",
+        "",
+        "",
+        "reduce(TCP_GATE_EN2,sum)",
+        "TCP core clocks are turned on. Not Windowed. Sum over TCP instances."},
+       {"TCP_TCC_ATOMIC_WITH_RET_REQ_sum",
+        "",
+        "",
+        "reduce(TCP_TCC_ATOMIC_WITH_RET_REQ,sum)",
+        "Total atomic with return requests from TCP to all TCCs Sum over TCP instances."},
+       {"TCC_EA0_WRREQ_LEVEL_sum",
+        "",
+        "",
+        "reduce(TCC_EA0_WRREQ_LEVEL,sum)",
+        "The sum of the number of EA write requests in flight. This is primarily meant for "
+        "measure average EA write latency. Average write latency = "
+        "TCC_PERF_SEL_EA_WRREQ_LEVEL/TCC_PERF_SEL_EA_WRREQ. Sum over TCC instances."},
+       {"TA_BUFFER_ATOMIC_WAVEFRONTS_sum",
+        "",
+        "",
+        "reduce(TA_BUFFER_ATOMIC_WAVEFRONTS,sum)",
+        "Number of buffer atomic wavefronts processed by TA. Sum over TA instances."},
+       {"TCP_PENDING_STALL_CYCLES_sum",
+        "",
+        "",
+        "reduce(TCP_PENDING_STALL_CYCLES,sum)",
+        "Stall due to data pending from L2. Sum over TCP instances."},
+       {"TCC_EA0_ATOMIC_LEVEL_sum",
+        "",
+        "",
+        "reduce(TCC_EA0_ATOMIC_LEVEL,sum)",
+        "The sum of the number of EA atomics in flight. This is primarily meant for measure "
+        "average EA atomic latency. Average atomic latency = "
+        "TCC_PERF_SEL_EA_WRREQ_ATOMIC_LEVEL/TCC_PERF_SEL_EA_WRREQ_ATOMIC. Sum over TCC "
+        "instances."},
+       {"TCC_NC_REQ_sum",
+        "",
+        "",
+        "reduce(TCC_NC_REQ,sum)",
+        "The number of noncoherently cached requests. This is measured at the tag block. Sum over "
+        "TCC instances."},
+       {"TCP_TCC_RW_WRITE_REQ_sum",
+        "",
+        "",
+        "reduce(TCP_TCC_RW_WRITE_REQ,sum)",
+        "Total write requests with RW mtype from this TCP to all TCCs. Sum over TCP instances."},
+       {"TCC_NORMAL_EVICT_sum",
+        "",
+        "",
+        "reduce(TCC_NORMAL_EVICT,sum)",
+        "Number of evictions due to requests that are not invalidate or probe requests. Sum over "
+        "TCC instances."},
+       {"TCP_TCC_WRITE_REQ_sum",
+        "",
+        "",
+        "reduce(TCP_TCC_WRITE_REQ,sum)",
+        "Total write requests from TCP to all TCCs Sum over TCP instances."},
+       {"TD_LOAD_WAVEFRONT_sum",
+        "",
+        "",
+        "reduce(TD_LOAD_WAVEFRONT,sum)",
+        "Count the wavefronts with opcode = load, include atomics and store. Sum over TD "
+        "instances."},
+       {"TD_STORE_WAVEFRONT_sum",
+        "",
+        "",
+        "reduce(TD_STORE_WAVEFRONT,sum)",
+        "Count the wavefronts with opcode = store. Sum over TD instances."},
+       {"TCP_TCC_NC_READ_REQ_sum",
+        "",
+        "",
+        "reduce(TCP_TCC_NC_READ_REQ,sum)",
+        "Total read requests with NC mtype from this TCP to all TCCs Sum over TCP instances."},
+       {"TCP_TOTAL_ATOMIC_WITH_RET_sum",
+        "",
+        "",
+        "reduce(TCP_TOTAL_ATOMIC_WITH_RET,sum)",
+        "Total number of atomic with return pixels/buffers from TA. Sum over TCP instances."},
+       {"TCP_TCP_LATENCY_sum",
+        "",
+        "",
+        "reduce(TCP_TCP_LATENCY,sum)",
+        "Total TCP wave latency (from first clock of wave entering to first clock of wave "
+        "leaving), divide by TA_TCP_STATE_READ to avg wave latency Sum over TCP instances."},
+       {"TD_COALESCABLE_WAVEFRONT_sum",
+        "",
+        "",
+        "reduce(TD_COALESCABLE_WAVEFRONT,sum)",
+        "Count wavefronts that TA finds coalescable. Sum over TD instances."},
+       {"TA_BUFFER_WAVEFRONTS_sum",
+        "",
+        "",
+        "reduce(TA_BUFFER_WAVEFRONTS,sum)",
+        "Number of buffer wavefronts processed by TA. Sum over TA instances."},
+       {"TCP_TCC_READ_REQ_LATENCY_sum",
+        "",
+        "",
+        "reduce(TCP_TCC_READ_REQ_LATENCY,sum)",
+        "Total TCP->TCC request latency for reads and atomics with return. Not Windowed. Sum over "
+        "TCP instances."},
+       {"TCC_EA0_RDREQ_GMI_CREDIT_STALL_sum",
+        "",
+        "",
+        "reduce(TCC_EA0_RDREQ_GMI_CREDIT_STALL,sum)",
+        "Number of cycles there was a stall because the read request interface was out of GMI "
+        "credits. Stalls occur regardless of whether a read needed to be performed or not. Sum "
+        "over TCC instances."},
+       {"TCC_TAG_STALL_sum",
+        "",
+        "",
+        "reduce(TCC_TAG_STALL,sum)",
+        "Total number of cycles the normal request pipeline in the tag is stalled for any reason."},
+       {"TCP_TA_TCP_STATE_READ_sum",
+        "",
+        "",
+        "reduce(TCP_TA_TCP_STATE_READ,sum)",
+        "Number of state reads Sum over TCP instances."},
+       {"TCC_ATOMIC_sum",
+        "",
+        "",
+        "reduce(TCC_ATOMIC,sum)",
+        "Number of atomic requests of all types. Sum over TCC instances."},
+       {"TCP_TCC_READ_REQ_sum",
+        "",
+        "",
+        "reduce(TCP_TCC_READ_REQ,sum)",
+        "Total read requests from TCP to all TCCs Sum over TCP instances."},
+       {"TCP_TCC_NC_ATOMIC_REQ_sum",
+        "",
+        "",
+        "reduce(TCP_TCC_NC_ATOMIC_REQ,sum)",
+        "Total atomic requests with NC mtype from this TCP to all TCCs Sum over TCP instances."},
+       {"TCC_READ_sum",
+        "",
+        "",
+        "reduce(TCC_READ,sum)",
+        "Number of read requests. Compressed reads are included in this, but metadata reads are "
+        "not included. Sum over TCC instances."},
+       {"TA_BUFFER_TOTAL_CYCLES_sum",
+        "",
+        "",
+        "reduce(TA_BUFFER_TOTAL_CYCLES,sum)",
+        "Number of buffer cycles issued to TC. Sum over TA instances."},
+       {"TCP_TOTAL_WRITEBACK_INVALIDATES_sum",
+        "",
+        "",
+        "reduce(TCP_TOTAL_WRITEBACK_INVALIDATES,sum)",
+        "Total number of cache invalidates. Equals TCP_PERF_SEL_TOTAL_WBINVL1+ "
+        "TCP_PERF_SEL_TOTAL_WBINVL1_VOL+ TCP_PERF_SEL_CP_TCP_INVALIDATE+ "
+        "TCP_PERF_SEL_SQ_TCP_INVALIDATE_VOL. Not Windowed. Sum over TCP instances."},
+       {"TCC_EA0_ATOMIC_sum",
+        "",
+        "",
+        "reduce(TCC_EA0_ATOMIC,sum)",
+        "Number of transactions going over the TC_EA_wrreq interface that are actually atomic "
+        "requests. Sum over TCC instances."},
+       {"TCC_STREAMING_REQ_sum",
+        "",
+        "",
+        "reduce(TCC_STREAMING_REQ,sum)",
+        "Number of streaming requests. This is measured at the tag block. Sum over TCC instances."},
+       {"TA_ADDR_STALLED_BY_TC_CYCLES_sum",
+        "",
+        "",
+        "reduce(TA_ADDR_STALLED_BY_TC_CYCLES,sum)",
+        "Number of cycles addr path stalled by TC. Perf_Windowing not supported for this counter. "
+        "Sum over TA instances."},
+       {"TCC_EA0_WRREQ_DRAM_sum",
+        "",
+        "",
+        "reduce(TCC_EA0_WRREQ_DRAM,sum)",
+        "Number of TCC/EA write requests (either 32-byte of 64-byte) destined for DRAM (MC). Sum "
+        "over TCC instances."},
+       {"TCC_EA0_RDREQ_DRAM_sum",
+        "",
+        "",
+        "reduce(TCC_EA0_RDREQ_DRAM,sum)",
+        "Number of TCC/EA read requests (either 32-byte or 64-byte) destined for DRAM (MC). Sum "
+        "over TCC instances."},
+       {"TD_ATOMIC_WAVEFRONT_sum",
+        "",
+        "",
+        "reduce(TD_ATOMIC_WAVEFRONT,sum)",
+        "Count the wavefronts with opcode = atomic. Sum over TD instances."},
+       {"TCC_EA0_RDREQ_DRAM_CREDIT_STALL_sum",
+        "",
+        "",
+        "reduce(TCC_EA0_RDREQ_DRAM_CREDIT_STALL,sum)",
+        "Number of cycles there was a stall because the read request interface was out of DRAM "
+        "credits. Stalls occur regardless of whether a read needed to be performed or not. Sum "
+        "over TCC instances."},
+       {"TCC_ALL_TC_OP_INV_EVICT_sum",
+        "",
+        "",
+        "reduce(TCC_ALL_TC_OP_INV_EVICT,sum)",
+        "Number of evictions due to all TC_OP invalidate requests. Sum over TCC instances."},
+       {"TCP_TD_TCP_STALL_CYCLES_sum",
+        "",
+        "",
+        "reduce(TCP_TD_TCP_STALL_CYCLES,sum)",
+        "TD stalls TCP. Sum over TCP instances."},
+       {"TCP_WRITE_TAGCONFLICT_STALL_CYCLES_sum",
+        "",
+        "",
+        "reduce(TCP_WRITE_TAGCONFLICT_STALL_CYCLES,sum)",
+        "Tagram conflict stall on a write. Sum over TCP instances."},
+       {"TCC_EA0_WRREQ_STALL_sum",
+        "",
+        "",
+        "reduce(TCC_EA0_WRREQ_STALL,sum)",
+        "Number of cycles a write request was stalled. Sum over TCC instances."},
+       {"TCP_TCC_UC_ATOMIC_REQ_sum",
+        "",
+        "",
+        "reduce(TCP_TCC_UC_ATOMIC_REQ,sum)",
+        "Total atomic requests with UC mtype from this TCP to all TCCs Sum over TCP instances."},
+       {"TCP_ATOMIC_TAGCONFLICT_STALL_CYCLES_sum",
+        "",
+        "",
+        "reduce(TCP_ATOMIC_TAGCONFLICT_STALL_CYCLES,sum)",
+        "Tagram conflict stall on an atomic. Sum over TCP instances."},
+       {"TCP_GATE_EN1_sum",
+        "",
+        "",
+        "reduce(TCP_GATE_EN1,sum)",
+        "TCP interface clocks are turned on. Not Windowed. Sum over TCP instances."},
+       {"TD_TC_STALL_sum",
+        "",
+        "",
+        "reduce(TD_TC_STALL,sum)",
+        "TD is stalled waiting for TC data. Sum over TD instances."},
+       {"TCC_ALL_TC_OP_WB_WRITEBACK_sum",
+        "",
+        "",
+        "reduce(TCC_ALL_TC_OP_WB_WRITEBACK,sum)",
+        "Number of writebacks due to all TC_OP writeback requests. Sum over TCC instances."},
+       {"TCC_UC_REQ_sum",
+        "",
+        "",
+        "reduce(TCC_UC_REQ,sum)",
+        "The number of uncached requests. This is measured at the tag block. Sum over TCC "
+        "instances."},
+       {"TCC_EA0_RD_UNCACHED_32B_sum",
+        "",
+        "",
+        "reduce(TCC_EA0_RD_UNCACHED_32B,sum)",
+        "Number of 32-byte TCC/EA read due to uncached traffic. A 64-byte request will be counted "
+        "as 2 Sum over TCC instances."},
+       {"TCP_UTCL1_PERMISSION_MISS_sum",
+        "",
+        "",
+        "reduce(TCP_UTCL1_PERMISSION_MISS,sum)",
+        "Total utcl1 permission misses Sum over TCP instances."},
+       {"TCC_TOO_MANY_EA_WRREQS_STALL_sum",
+        "",
+        "",
+        "reduce(TCC_TOO_MANY_EA_WRREQS_STALL,sum)",
+        "Number of cycles the TCC could not send a EA write request because it already reached its "
+        "maximum number of pending EA write requests. Sum over TCC instances."},
+       {"TCP_TCC_WRITE_REQ_LATENCY_sum",
+        "",
+        "",
+        "reduce(TCP_TCC_WRITE_REQ_LATENCY,sum)",
+        "Total TCP->TCC request latency for writes and atomics without return. Not Windowed. Sum "
+        "over TCP instances."},
+       {"TCC_REQ_sum",
+        "",
+        "",
+        "reduce(TCC_REQ,sum)",
+        "Number of requests of all types. This is measured at the tag block. This may be more "
+        "than the number of requests arriving at the TCC, but it is a good indication of the "
+        "total amount of work that needs to be performed. Sum over TCC instances."},
+       {"TCP_TCR_TCP_STALL_CYCLES_sum",
+        "",
+        "",
+        "reduce(TCP_TCR_TCP_STALL_CYCLES,sum)",
+        "TCR stalls TCP_TCR_req interface. Sum over TCP instances."},
+       {"TCP_UTCL1_REQUEST_sum",
+        "",
+        "",
+        "reduce(TCP_UTCL1_REQUEST,sum)",
+        "Total CLIENT_UTCL1 NORMAL requests Sum over TCP instances."},
+       {"TCP_TCC_ATOMIC_WITHOUT_RET_REQ_sum",
+        "",
+        "",
+        "reduce(TCP_TCC_ATOMIC_WITHOUT_RET_REQ,sum)",
+        "Total atomic without return requests from TCP to all TCCs Sum over TCP instances."},
+       {"TA_ADDR_STALLED_BY_TD_CYCLES_sum",
+        "",
+        "",
+        "reduce(TA_ADDR_STALLED_BY_TD_CYCLES,sum)",
+        "Number of cycles addr path stalled by TD. Perf_Windowing not supported for this counter. "
+        "Sum over TA instances."},
+       {"TCP_TCC_NC_WRITE_REQ_sum",
+        "",
+        "",
+        "reduce(TCP_TCC_NC_WRITE_REQ,sum)",
+        "Total write requests with NC mtype from this TCP to all TCCs Sum over TCP instances."},
+       {"TCC_EA0_RDREQ_LEVEL_sum",
+        "",
+        "",
+        "reduce(TCC_EA0_RDREQ_LEVEL,sum)",
+        "The sum of the number of TCC/EA read requests in flight. This is primarily meant for "
+        "measure average EA read latency. Average read latency = "
+        "TCC_PERF_SEL_EA_RDREQ_LEVEL/TCC_PERF_SEL_EA_RDREQ. Sum over TCC instances."},
+       {"TCC_RW_REQ_sum",
+        "",
+        "",
+        "reduce(TCC_RW_REQ,sum)",
+        "The number of RW requests. This is measured at the tag block. Sum over TCC instances."},
+       {"TCC_EA0_WRREQ_IO_CREDIT_STALL_sum",
+        "",
+        "",
+        "reduce(TCC_EA0_WRREQ_IO_CREDIT_STALL,sum)",
+        "Number of cycles a EA write request was stalled because the interface was out of IO "
+        "credits. Sum over TCC instances."},
+       {"TCP_READ_TAGCONFLICT_STALL_CYCLES_sum",
+        "",
+        "",
+        "reduce(TCP_READ_TAGCONFLICT_STALL_CYCLES,sum)",
+        "Tagram conflict stall on a read. Sum over TCP instances."},
+       {"TCC_BUSY_sum",
+        "",
+        "",
+        "reduce(TCC_BUSY,sum)",
+        "Number of cycles we have a request pending. Not windowable. Sum over TCC instances."},
+       {"TCC_BUSY_avr",
+        "",
+        "",
+        "reduce(TCC_BUSY,avr)",
+        "TCC_BUSY avr over all memory channels."},
+       {"TCC_PROBE_sum",
+        "",
+        "",
+        "reduce(TCC_PROBE,sum)",
+        "Number of probe requests. Not windowable. Sum over TCC instances."},
+       {"TCC_PROBE_ALL_sum",
+        "",
+        "",
+        "reduce(TCC_PROBE_ALL,sum)",
+        "Number of external probe requests with with EA_TCC_preq_all== 1. Not windowable. Sum "
+        "over TCC instances."},
+       {"TA_BUFFER_READ_WAVEFRONTS_sum",
+        "",
+        "",
+        "reduce(TA_BUFFER_READ_WAVEFRONTS,sum)",
+        "Number of buffer read wavefronts processed by TA. Sum over TA instances."},
+       {"TCP_TOTAL_READ_sum",
+        "",
+        "",
+        "reduce(TCP_TOTAL_READ,sum)",
+        "Total number of read pixels/buffers from TA. Equals TCP_PERF_SEL_TOTAL_HIT_LRU_READ + "
+        "TCP_PERF_SEL_TOTAL_MISS_LRU_READ + TCP_PERF_SEL_TOTAL_MISS_EVICT_READ. Sum over TCP "
+        "instances."},
+       {"TCC_WRITE_sum",
+        "",
+        "",
+        "reduce(TCC_WRITE,sum)",
+        "Number of write requests. Sum over TCC instances."},
+       {"TCC_EA0_RDREQ_IO_CREDIT_STALL_sum",
+        "",
+        "",
+        "reduce(TCC_EA0_RDREQ_IO_CREDIT_STALL,sum)",
+        "Number of cycles there was a stall because the read request interface was out of IO "
+        "credits. Stalls occur regardless of whether a read needed to be performed or not. Sum "
+        "over TCC instances."},
+       {"TCC_EA0_WRREQ_GMI_CREDIT_STALL_sum",
+        "",
+        "",
+        "reduce(TCC_EA0_WRREQ_GMI_CREDIT_STALL,sum)",
+        "Number of cycles a EA write request was stalled because the interface was out of GMI "
+        "credits. Sum over TCC instances."},
+       {"TCP_UTCL1_TRANSLATION_HIT_sum",
+        "",
+        "",
+        "reduce(TCP_UTCL1_TRANSLATION_HIT,sum)",
+        "Total utcl1 translation hits Sum over TCP instances."},
+       {"TCC_CYCLE_sum",
+        "",
+        "",
+        "reduce(TCC_CYCLE,sum)",
+        "Number of cycles. Not windowable. Sum over TCC instances."},
+       {"TCC_EA0_WRREQ_DRAM_CREDIT_STALL_sum",
+        "",
+        "",
+        "reduce(TCC_EA0_WRREQ_DRAM_CREDIT_STALL,sum)",
+        "Number of cycles a EA write request was stalled because the interface was out of DRAM "
+        "credits. Sum over TCC instances."},
+       {"TCP_TCC_RW_ATOMIC_REQ_sum",
+        "",
+        "",
+        "reduce(TCP_TCC_RW_ATOMIC_REQ,sum)",
+        "Total atomic requests with RW mtype from this TCP to all TCCs. Sum over TCP instances."},
+       {"TCP_TCC_RW_READ_REQ_sum",
+        "",
+        "",
+        "reduce(TCP_TCC_RW_READ_REQ,sum)",
+        "Total write requests with RW mtype from this TCP to all TCCs. Sum over TCP instances."},
+       {"TCP_UTCL1_TRANSLATION_MISS_sum",
+        "",
+        "",
+        "reduce(TCP_UTCL1_TRANSLATION_MISS,sum)",
+        "Total utcl1 translation misses Sum over TCP instances."},
+       {"TA_TOTAL_WAVEFRONTS_sum",
+        "",
+        "",
+        "reduce(TA_TOTAL_WAVEFRONTS,sum)",
+        "Total number of wavefronts processed by TA. Sum over TA instances."},
+       {"TCP_TOTAL_ATOMIC_WITHOUT_RET_sum",
+        "",
+        "",
+        "reduce(TCP_TOTAL_ATOMIC_WITHOUT_RET,sum)",
+        "Total number of atomic without return pixels/buffers from TA Sum over TCP instances."},
+       {"TCC_NORMAL_WRITEBACK_sum",
+        "",
+        "",
+        "reduce(TCC_NORMAL_WRITEBACK,sum)",
+        "Number of writebacks due to requests that are not writeback requests. Sum over TCC "
+        "instances."}}}};
diff --git a/source/share/rocprofiler-sdk/counter_defs.yaml b/source/share/rocprofiler-sdk/counter_defs.yaml
index 82243e87c6..4072c46437 100644
--- a/source/share/rocprofiler-sdk/counter_defs.yaml
+++ b/source/share/rocprofiler-sdk/counter_defs.yaml
@@ -42,6 +42,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -54,6 +55,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -66,6 +68,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -78,6 +81,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -90,6 +94,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -102,6 +107,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -114,6 +120,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -126,6 +133,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -138,6 +146,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -150,6 +159,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -162,6 +172,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -310,6 +321,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -322,6 +334,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -334,6 +347,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -346,6 +360,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -358,6 +373,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -370,6 +386,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -382,6 +399,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1171,6 +1189,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1183,6 +1202,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1200,6 +1220,7 @@ rocprofiler-sdk:
       - gfx1031
       - gfx1032
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1217,6 +1238,7 @@ rocprofiler-sdk:
       - gfx1031
       - gfx1032
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1286,6 +1308,7 @@ rocprofiler-sdk:
       - gfx1031
       - gfx1032
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1303,6 +1326,7 @@ rocprofiler-sdk:
       - gfx1031
       - gfx1032
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1315,6 +1339,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1327,6 +1352,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1769,6 +1795,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1782,6 +1809,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1795,6 +1823,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1809,6 +1838,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1821,6 +1851,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1833,6 +1864,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1845,6 +1877,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1857,6 +1890,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1869,6 +1903,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1881,6 +1916,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1893,6 +1929,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1905,6 +1942,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1917,6 +1955,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1929,6 +1968,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1941,6 +1981,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1953,6 +1994,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1967,6 +2009,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -1993,6 +2036,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2509,6 +2553,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2532,6 +2577,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2544,6 +2590,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2556,6 +2603,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2568,6 +2616,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2580,6 +2629,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2592,6 +2642,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2604,6 +2655,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2616,6 +2668,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2628,6 +2681,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2640,6 +2694,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2663,6 +2718,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2686,6 +2742,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2698,6 +2755,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2710,6 +2768,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2773,6 +2832,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2785,6 +2845,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2797,6 +2858,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2809,6 +2871,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2821,6 +2884,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2833,6 +2897,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2859,6 +2924,7 @@ rocprofiler-sdk:
       - gfx1200
       - gfx1201
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -2875,6 +2941,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 185
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 158
     - architectures:
       - gfx940
       - gfx941
@@ -2895,6 +2965,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 96
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 69
     - architectures:
       - gfx940
       - gfx941
@@ -2917,6 +2991,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 101
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 74
     - architectures:
       - gfx940
       - gfx941
@@ -2938,6 +3016,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 103
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 76
     - architectures:
       - gfx940
       - gfx941
@@ -2959,6 +3041,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 98
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 71
     - architectures:
       - gfx940
       - gfx941
@@ -2980,6 +3066,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 102
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 75
     - architectures:
       - gfx940
       - gfx941
@@ -3002,6 +3092,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 100
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 73
     - architectures:
       - gfx940
       - gfx941
@@ -3054,6 +3148,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 97
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 70
     - architectures:
       - gfx940
       - gfx941
@@ -3071,6 +3169,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -3097,6 +3196,7 @@ rocprofiler-sdk:
       - gfx1200
       - gfx1201
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -3109,6 +3209,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -3123,6 +3224,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 115
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 88
     - architectures:
       - gfx940
       - gfx941
@@ -3142,6 +3247,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 116
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 89
     - architectures:
       - gfx940
       - gfx941
@@ -3160,6 +3269,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -3176,6 +3286,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 64
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 39
     - architectures:
       - gfx940
       - gfx941
@@ -3196,6 +3310,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 63
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 38
     - architectures:
       - gfx940
       - gfx941
@@ -3379,6 +3497,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 52
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 27
     - architectures:
       - gfx940
       - gfx941
@@ -3445,6 +3567,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 65
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 40
     - architectures:
       - gfx940
       - gfx941
@@ -3513,6 +3639,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 188
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 161
     - architectures:
       - gfx940
       - gfx941
@@ -4039,6 +4169,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 55
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 30
     - architectures:
       - gfx940
       - gfx941
@@ -4118,6 +4252,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 66
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 41
     - architectures:
       - gfx940
       - gfx941
@@ -4243,6 +4381,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 111
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 84
     - architectures:
       - gfx940
       - gfx941
@@ -4289,6 +4431,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 105
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 78
     - architectures:
       - gfx940
       - gfx941
@@ -4308,6 +4454,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 104
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 77
     - architectures:
       - gfx940
       - gfx941
@@ -4358,6 +4508,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 69
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 44
     - architectures:
       - gfx940
       - gfx941
@@ -4392,6 +4546,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 68
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 43
     - architectures:
       - gfx940
       - gfx941
@@ -4411,6 +4569,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 67
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 42
     - architectures:
       - gfx940
       - gfx941
@@ -4428,6 +4590,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -4443,6 +4606,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 122
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 95
     - architectures:
       - gfx940
       - gfx941
@@ -4462,6 +4629,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 125
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 98
     - architectures:
       - gfx940
       - gfx941
@@ -4510,6 +4681,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 126
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 99
     - architectures:
       - gfx940
       - gfx941
@@ -4529,6 +4704,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 124
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 97
     - architectures:
       - gfx940
       - gfx941
@@ -4548,6 +4727,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 123
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 96
     - architectures:
       - gfx940
       - gfx941
@@ -4573,6 +4756,7 @@ rocprofiler-sdk:
       event: 7
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -4643,6 +4827,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 85
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 58
     - architectures:
       - gfx940
       - gfx941
@@ -4682,6 +4870,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 88
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 61
     - architectures:
       - gfx940
       - gfx941
@@ -4838,6 +5030,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -4855,6 +5048,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -4872,6 +5066,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -4889,6 +5084,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -4906,6 +5102,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -4924,6 +5121,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 186
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 159
     - architectures:
       - gfx940
       - gfx941
@@ -4946,6 +5147,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 187
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 160
     - architectures:
       - gfx940
       - gfx941
@@ -5004,6 +5209,10 @@ rocprofiler-sdk:
       - gfx90a
       block: SQ
       event: 74
+    - architectures:
+      - gfx908
+      block: SQ
+      event: 47
     - architectures:
       - gfx940
       - gfx941
@@ -5256,6 +5465,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TA
       event: 54
     - architectures:
@@ -5271,6 +5481,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -5282,6 +5493,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TA
       event: 55
     - architectures:
@@ -5297,6 +5509,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -5308,6 +5521,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TA
       event: 47
     - architectures:
@@ -5323,6 +5537,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -5334,6 +5549,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TA
       event: 52
     - architectures:
@@ -5349,6 +5565,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -5360,6 +5577,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TA
       event: 53
     - architectures:
@@ -5375,6 +5593,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -5413,6 +5632,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TA
       event: 45
     - architectures:
@@ -5428,6 +5648,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -5466,6 +5687,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TA
       event: 49
     - architectures:
@@ -5481,6 +5703,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -5492,6 +5715,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TA
       event: 44
     - architectures:
@@ -5507,6 +5731,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -5518,6 +5743,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TA
       event: 46
     - architectures:
@@ -5533,6 +5759,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -5625,6 +5852,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TA
       event: 56
     - architectures:
@@ -5640,6 +5868,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -5651,6 +5880,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TA
       event: 103
     - architectures:
@@ -5666,6 +5896,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -5757,6 +5988,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TA
       event: 100
     - architectures:
@@ -5772,6 +6004,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -5848,6 +6081,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -5859,6 +6093,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TA
       event: 32
     - architectures:
@@ -5874,6 +6109,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -5956,6 +6192,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -5979,6 +6216,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6002,6 +6240,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6017,6 +6256,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6028,6 +6268,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6043,6 +6284,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6054,6 +6296,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6069,6 +6312,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6080,6 +6324,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6092,6 +6337,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6103,6 +6349,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6114,6 +6361,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6129,6 +6377,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6140,6 +6389,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6152,6 +6402,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6162,6 +6413,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6177,6 +6429,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6192,6 +6445,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6203,6 +6457,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6250,6 +6505,10 @@ rocprofiler-sdk:
     description: Number of TCC/EA read requests (either 32-byte or 64-byte) destined for DRAM (MC).
     properties: []
     definitions:
+    - architectures:
+      - gfx908
+      block: TCC
+      event: 43
     - architectures:
       - gfx940
       - gfx941
@@ -6266,6 +6525,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6281,6 +6541,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6291,6 +6552,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6302,6 +6564,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6317,6 +6580,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6328,6 +6592,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6343,6 +6608,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6354,6 +6620,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6369,6 +6636,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6389,6 +6657,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6404,6 +6673,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6454,6 +6724,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6468,6 +6739,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6483,6 +6755,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6493,6 +6766,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6503,6 +6777,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6518,6 +6793,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6528,6 +6804,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6543,6 +6820,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6554,6 +6832,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6569,6 +6848,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6593,6 +6873,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6607,6 +6888,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6630,6 +6912,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -6646,6 +6929,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7172,6 +7456,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx90a
       - gfx940
       - gfx941
@@ -7188,6 +7473,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7199,6 +7485,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7214,6 +7501,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7224,6 +7512,7 @@ rocprofiler-sdk:
     properties: []
     definitions:
     - architectures:
+      - gfx908
       - gfx90a
       - gfx940
       - gfx941
@@ -7240,6 +7529,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7251,6 +7541,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7266,6 +7557,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7281,6 +7573,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7306,6 +7599,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7317,6 +7611,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7333,6 +7628,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7345,6 +7641,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7362,6 +7659,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7373,6 +7671,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7388,6 +7687,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7399,6 +7699,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7414,6 +7715,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7427,6 +7729,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7442,6 +7745,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7454,6 +7758,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7470,6 +7775,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7481,6 +7787,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7496,6 +7803,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7507,6 +7815,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7523,6 +7832,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7539,6 +7849,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7550,6 +7861,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7783,6 +8095,70 @@ rocprofiler-sdk:
       - gfx950
       block: TCC
       event: 116
+  - name: TCC_CLIENT184_REQ
+    description: 'Number of cycles client184 sent a request to this TCC.'
+    properties: []
+    definitions:
+    - architectures:
+      - gfx908
+      block: TCC
+      event: 312
+  - name: TCC_CLIENT185_REQ
+    description: 'Number of cycles client185 sent a request to this TCC.'
+    properties: []
+    definitions:
+    - architectures:
+      - gfx908
+      block: TCC
+      event: 313
+  - name: TCC_CLIENT186_REQ
+    description: 'Number of cycles client186 sent a request to this TCC.'
+    properties: []
+    definitions:
+    - architectures:
+      - gfx908
+      block: TCC
+      event: 314
+  - name: TCC_CLIENT187_REQ
+    description: 'Number of cycles client187 sent a request to this TCC.'
+    properties: []
+    definitions:
+    - architectures:
+      - gfx908
+      block: TCC
+      event: 315
+  - name: TCC_CLIENT188_REQ
+    description: 'Number of cycles client188 sent a request to this TCC.'
+    properties: []
+    definitions:
+    - architectures:
+      - gfx908
+      block: TCC
+      event: 316
+  - name: TCC_CLIENT189_REQ
+    description: 'Number of cycles client189 sent a request to this TCC.'
+    properties: []
+    definitions:
+    - architectures:
+      - gfx908
+      block: TCC
+      event: 317
+  - name: TCC_CLIENT190_REQ
+    description: 'Number of cycles client190 sent a request to this TCC.'
+    properties: []
+    definitions:
+    - architectures:
+      - gfx908
+      block: TCC
+      event: 318
+  - name: TCC_CLIENT191_REQ
+    description: 'Number of cycles client191 sent a request to this TCC.'
+    properties: []
+    definitions:
+    - architectures:
+      - gfx908
+      block: TCC
+      event: 319
   - name: TCC_EA0_RDREQ_64B_sum
     description: Number of 64-byte TCC/EA read requests. Sum over TCP instances.
     properties: []
@@ -7954,6 +8330,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 13
     - architectures:
@@ -7969,6 +8346,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7980,6 +8358,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -7992,6 +8371,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8003,6 +8383,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8015,6 +8396,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8026,6 +8408,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 22
     - architectures:
@@ -8041,6 +8424,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8052,6 +8436,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 11
     - architectures:
@@ -8067,6 +8452,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8078,6 +8464,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 27
     - architectures:
@@ -8093,6 +8480,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8104,6 +8492,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 72
     - architectures:
@@ -8122,6 +8511,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8133,6 +8523,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 71
     - architectures:
@@ -8151,6 +8542,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8162,6 +8554,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 83
     - architectures:
@@ -8180,6 +8573,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8191,6 +8585,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 81
     - architectures:
@@ -8209,6 +8604,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8220,6 +8616,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 82
     - architectures:
@@ -8238,6 +8635,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8249,6 +8647,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 77
     - architectures:
@@ -8267,6 +8666,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8278,6 +8678,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 75
     - architectures:
@@ -8296,6 +8697,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8307,6 +8709,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 76
     - architectures:
@@ -8325,6 +8728,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8336,6 +8740,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 69
     - architectures:
@@ -8354,6 +8759,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 66
     - architectures:
@@ -8366,6 +8772,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx950
       expression: reduce(TCP_TCC_READ_REQ_LATENCY,sum)
   - name: TCP_TCC_READ_REQ_sum
@@ -8374,6 +8781,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8385,6 +8793,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 87
     - architectures:
@@ -8403,6 +8812,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8414,6 +8824,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 85
     - architectures:
@@ -8432,6 +8843,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8443,6 +8855,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 86
     - architectures:
@@ -8461,6 +8874,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8472,6 +8886,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 80
     - architectures:
@@ -8490,6 +8905,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8501,6 +8917,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 78
     - architectures:
@@ -8519,6 +8936,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8530,6 +8948,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 79
     - architectures:
@@ -8548,6 +8967,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8559,6 +8979,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 70
     - architectures:
@@ -8577,6 +8998,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 67
     - architectures:
@@ -8589,6 +9011,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx950
       expression: reduce(TCP_TCC_WRITE_REQ_LATENCY,sum)
   - name: TCP_TCC_WRITE_REQ_sum
@@ -8597,6 +9020,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8609,6 +9033,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 65
     - architectures:
@@ -8622,6 +9047,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx950
       expression: reduce(TCP_TCP_LATENCY,sum)
   - name: TCP_TCP_TA_DATA_STALL_CYCLES
@@ -8676,6 +9102,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8688,6 +9115,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8699,6 +9127,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8711,6 +9140,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8722,6 +9152,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 29
     - architectures:
@@ -8738,6 +9169,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8749,6 +9181,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 39
     - architectures:
@@ -8764,6 +9197,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8775,6 +9209,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 38
     - architectures:
@@ -8790,6 +9225,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8801,6 +9237,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8816,6 +9253,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8828,6 +9266,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 30
     - architectures:
@@ -8844,6 +9283,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8855,6 +9295,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 32
     - architectures:
@@ -8871,6 +9312,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 45
     - architectures:
@@ -8890,6 +9332,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8902,6 +9345,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8913,6 +9357,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 50
     - architectures:
@@ -8931,6 +9376,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8942,6 +9388,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 47
     - architectures:
@@ -8960,6 +9407,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -8971,6 +9419,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 49
     - architectures:
@@ -8989,6 +9438,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -9000,6 +9450,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 48
     - architectures:
@@ -9018,6 +9469,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -9029,6 +9481,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 28
     - architectures:
@@ -9044,6 +9497,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -9055,6 +9509,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TCP
       event: 12
     - architectures:
@@ -9070,6 +9525,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -9396,6 +9852,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TD
       event: 26
     - architectures:
@@ -9411,6 +9868,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -9422,6 +9880,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TD
       event: 32
     - architectures:
@@ -9437,6 +9896,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -9448,6 +9908,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TD
       event: 25
     - architectures:
@@ -9463,6 +9924,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -9500,6 +9962,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TD
       event: 27
     - architectures:
@@ -9515,6 +9978,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -9526,6 +9990,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       block: TD
       event: 15
     - architectures:
@@ -9541,6 +10006,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -9552,6 +10018,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942
@@ -9564,6 +10031,7 @@ rocprofiler-sdk:
     definitions:
     - architectures:
       - gfx90a
+      - gfx908
       - gfx940
       - gfx941
       - gfx942