<metricname="SQ_WAVES"block=SQevent=4descr="Count number of waves sent to SQs. (per-simd, emulated, global)"></metric>
<metricname="SQ_INSTS_VALU"block=SQevent=26descr="Number of VALU instructions issued. (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_VMEM_WR"block=SQevent=27descr="Number of VMEM write instructions issued (including FLAT). (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_VMEM_RD"block=SQevent=28descr="Number of VMEM read instructions issued (including FLAT). (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_SALU"block=SQevent=30descr="Number of SALU instructions issued. (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_SMEM"block=SQevent=31descr="Number of SMEM instructions issued. (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_FLAT"block=SQevent=32descr="Number of FLAT instructions issued. (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_FLAT_LDS_ONLY"block=SQevent=33descr="Number of FLAT instructions issued that read/wrote only from/to LDS (only works if EARLY_TA_DONE is enabled). (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_LDS"block=SQevent=34descr="Number of LDS instructions issued (including FLAT). (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_GDS"block=SQevent=35descr="Number of GDS instructions issued. (per-simd, emulated)"></metric>
<metricname="SQ_WAIT_INST_LDS"block=SQevent=61descr="Number of wave-cycles spent waiting for LDS instruction issue. In units of 4 cycles. (per-simd, nondeterministic)"></metric>
<metricname="SQ_ACTIVE_INST_VALU"block=SQevent=69descr="Number of cycles the SQ instruction arbiter is working on a VALU instruction. (per-simd, nondeterministic)"></metric>
<metricname="SQ_INST_CYCLES_SALU"block=SQevent=86descr="Number of cycles needed to execute non-memory read scalar operations. (per-simd, emulated)"></metric>
<metricname="SQ_THREAD_CYCLES_VALU"block=SQevent=89descr="Number of thread-cycles used to execute VALU operations (similar to INST_CYCLES_VALU but multiplied by # of active threads). (per-simd)"></metric>
<metricname="SQ_LDS_BANK_CONFLICT"block=SQevent=97descr="Number of cycles LDS is stalled by bank conflicts. (emulated)"></metric>
<metricname="TCC_HIT"block=TCCevent=18descr="Number of cache hits."></metric>
<metricname="TCC_MISS"block=TCCevent=19descr="Number of cache misses. UC reads count as misses."></metric>
<metricname="TCC_MC_RDREQ"block=TCCevent=35descr="Number of 32-byte reads. The hardware actually does 64-byte reads but the number is adjusted to provide uniformity."></metric>
<metricname="TCC_MC_WRREQ"block=TCCevent=26descr="Number of 32-byte transactions going over the TC_MC_wrreq interface. Atomics may travel over the same interface and are generally classified as write requests."></metric>
<metricname="TCC_MC_WRREQ_STALL"block=TCCevent=28descr="Number of cycles a write request was stalled."></metric>
<metricname="SQ_WAVES"block=SQevent=4descr="Count number of waves sent to SQs. (per-simd, emulated, global)"></metric>
<metricname="SQ_INSTS_VALU"block=SQevent=26descr="Number of VALU instructions issued. (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_VMEM_WR"block=SQevent=27descr="Number of VMEM write instructions issued (including FLAT). (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_VMEM_RD"block=SQevent=28descr="Number of VMEM read instructions issued (including FLAT). (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_SALU"block=SQevent=30descr="Number of SALU instructions issued. (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_SMEM"block=SQevent=31descr="Number of SMEM instructions issued. (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_FLAT"block=SQevent=32descr="Number of FLAT instructions issued. (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_FLAT_LDS_ONLY"block=SQevent=33descr="Number of FLAT instructions issued that read/wrote only from/to LDS (only works if EARLY_TA_DONE is enabled). (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_LDS"block=SQevent=34descr="Number of LDS instructions issued (including FLAT). (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_GDS"block=SQevent=35descr="Number of GDS instructions issued. (per-simd, emulated)"></metric>
<metricname="SQ_WAIT_INST_LDS"block=SQevent=63descr="Number of wave-cycles spent waiting for LDS instruction issue. In units of 4 cycles. (per-simd, nondeterministic)"></metric>
<metricname="SQ_ACTIVE_INST_VALU"block=SQevent=71descr="regspec 71? Number of cycles the SQ instruction arbiter is working on a VALU instruction. (per-simd, nondeterministic)"></metric>
<metricname="SQ_INST_CYCLES_SALU"block=SQevent=84descr="Number of cycles needed to execute non-memory read scalar operations. (per-simd, emulated)"></metric>
<metricname="SQ_THREAD_CYCLES_VALU"block=SQevent=85descr="Number of thread-cycles used to execute VALU operations (similar to INST_CYCLES_VALU but multiplied by # of active threads). (per-simd)"></metric>
<metricname="SQ_LDS_BANK_CONFLICT"block=SQevent=93descr="Number of cycles LDS is stalled by bank conflicts. (emulated)"></metric>
<metricname="TCC_HIT"block=TCCevent=20descr="Number of cache hits."></metric>
<metricname="TCC_MISS"block=TCCevent=22descr="Number of cache misses. UC reads count as misses."></metric>
<metricname="TCC_EA_WRREQ"block=TCCevent=29descr="Number of transactions (either 32-byte or 64-byte) going over the TC_EA_wrreq interface. Atomics may travel over the same interface and are generally classified as write requests. This does not include probe commands."></metric>
<metricname="TCC_EA_WRREQ_64B"block=TCCevent=30descr="Number of 64-byte transactions going (64-byte write or CMPSWAP) over the TC_EA_wrreq interface."></metric>
<metricname="TCC_EA_WRREQ_STALL"block=TCCevent=33descr="Number of cycles a write request was stalled."></metric>
<metricname="TCC_EA_RDREQ"block=TCCevent=41descr="Number of TCC/EA read requests (either 32-byte or 64-byte)"></metric>
<metricname="TCC_EA_RDREQ_32B"block=TCCevent=42descr="Number of 32-byte TCC/EA read requests"></metric>
<metricname="TCC_EA1_WRREQ"block=TCCevent=256descr="Number of transactions (either 32-byte or 64-byte) going over the TC_EA_wrreq interface. Atomics may travel over the same interface and are generally classified as write requests. This does not include probe commands."></metric>
<metricname="TCC_EA1_WRREQ_64B"block=TCCevent=257descr="Number of 64-byte transactions going (64-byte write or CMPSWAP) over the TC_EA_wrreq interface."></metric>
<metricname="TCC_EA1_WRREQ_STALL"block=TCCevent=260descr="Number of cycles a write request was stalled."></metric>
<metricname="TCC_EA1_RDREQ"block=TCCevent=267descr="Number of TCC/EA read requests (either 32-byte or 64-byte)"></metric>
<metricname="TCC_EA1_RDREQ_32B"block=TCCevent=268descr="Number of 32-byte TCC/EA read requests"></metric>
<metricname="SQ_INSTS_VMEM_WR"block=SQevent=28descr="Number of VMEM write instructions issued (including FLAT). (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_VMEM_RD"block=SQevent=29descr="Number of VMEM read instructions issued (including FLAT). (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_SALU"block=SQevent=31descr="Number of SALU instructions issued. (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_SMEM"block=SQevent=32descr="Number of SMEM instructions issued. (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_FLAT"block=SQevent=33descr="Number of FLAT instructions issued. (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_FLAT_LDS_ONLY"block=SQevent=34descr="Number of FLAT instructions issued that read/wrote only from/to LDS (only works if EARLY_TA_DONE is enabled). (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_LDS"block=SQevent=35descr="Number of LDS instructions issued (including FLAT). (per-simd, emulated)"></metric>
<metricname="SQ_INSTS_GDS"block=SQevent=36descr="Number of GDS instructions issued. (per-simd, emulated)"></metric>
<metricname="SQ_WAIT_INST_LDS"block=SQevent=64descr="Number of wave-cycles spent waiting for LDS instruction issue. In units of 4 cycles. (per-simd, nondeterministic)"></metric>
<metricname="SQ_ACTIVE_INST_VALU"block=SQevent=72descr="regspec 71? Number of cycles the SQ instruction arbiter is working on a VALU instruction. (per-simd, nondeterministic)"></metric>
<metricname="SQ_INST_CYCLES_SALU"block=SQevent=85descr="Number of cycles needed to execute non-memory read scalar operations. (per-simd, emulated)"></metric>
<metricname="SQ_THREAD_CYCLES_VALU"block=SQevent=86descr="Number of thread-cycles used to execute VALU operations (similar to INST_CYCLES_VALU but multiplied by # of active threads). (per-simd)"></metric>
<metricname="SQ_LDS_BANK_CONFLICT"block=SQevent=94descr="Number of cycles LDS is stalled by bank conflicts. (emulated)"></metric>
<metricname="TCC_HIT"block=TCCevent=17descr="Number of cache hits."></metric>
<metricname="TCC_MISS"block=TCCevent=19descr="Number of cache misses. UC reads count as misses."></metric>
<metricname="TCC_EA_WRREQ"block=TCCevent=26descr="Number of transactions (either 32-byte or 64-byte) going over the TC_EA_wrreq interface. Atomics may travel over the same interface and are generally classified as write requests. This does not include probe commands."></metric>
<metricname="TCC_EA_WRREQ_64B"block=TCCevent=27descr="Number of 64-byte transactions going (64-byte write or CMPSWAP) over the TC_EA_wrreq interface."></metric>
<metricname="TCC_EA_WRREQ_STALL"block=TCCevent=30descr="Number of cycles a write request was stalled."></metric>
<metricname="TCC_EA_RDREQ"block=TCCevent=38descr="Number of TCC/EA read requests (either 32-byte or 64-byte)"></metric>
<metricname="TCC_EA_RDREQ_32B"block=TCCevent=39descr="Number of 32-byte TCC/EA read requests"></metric>