sanchayanmaity/gem5
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
gem5/tests/long/fs/10.linux-boot/ref/alpha/linux/tsunami-switcheroo-full/stats.txt

1742 lines
198 KiB
Text
Raw Normal View History

tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
---------- Begin Simulation Statistics ----------
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
sim_seconds 1.842705 # Number of seconds simulated
sim_ticks 1842705252000 # Number of ticks simulated
final_tick 1842705252000 # Number of ticks from beginning of simulation (restored from checkpoints and never reset)
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
sim_freq 1000000000000 # Frequency of simulated ticks
stats: Cumulative stats update This patch updates the stats to reflect the: 1) addition of the internal queue in SimpleMemory, 2) moving of the memory class outside FSConfig, 3) fixing up of the 2D vector printing format, 4) specifying burst size and interface width for the DRAM instead of relying on cache-line size, 5) performing merging in the DRAM controller write buffer, and 6) fixing how idle cycles are counted in the atomic and timing CPU models. The main reason for bundling them up is to minimise the changeset size.
2013-08-19 09:52:36 +02:00
host_inst_rate 308319 # Simulator instruction rate (inst/s)
host_op_rate 308319 # Simulator op (including micro ops) rate (op/s)
host_tick_rate 7821132493 # Simulator tick rate (ticks/s)
host_mem_usage 307864 # Number of bytes of host memory used
host_seconds 235.61 # Real time elapsed on the host
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
sim_insts 72641883 # Number of instructions simulated
sim_ops 72641883 # Number of ops (including micro ops) simulated
system.physmem.bytes_read::cpu0.inst 488448 # Number of bytes read from this memory
system.physmem.bytes_read::cpu0.data 20049216 # Number of bytes read from this memory
regressions: update due to cache latency fix
2013-03-28 00:36:21 +01:00
system.physmem.bytes_read::tsunami.ide 2652352 # Number of bytes read from this memory
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.bytes_read::cpu1.inst 147328 # Number of bytes read from this memory
system.physmem.bytes_read::cpu1.data 2290432 # Number of bytes read from this memory
system.physmem.bytes_read::cpu2.inst 282112 # Number of bytes read from this memory
system.physmem.bytes_read::cpu2.data 2525760 # Number of bytes read from this memory
system.physmem.bytes_read::total 28435648 # Number of bytes read from this memory
system.physmem.bytes_inst_read::cpu0.inst 488448 # Number of instructions bytes read from this memory
system.physmem.bytes_inst_read::cpu1.inst 147328 # Number of instructions bytes read from this memory
system.physmem.bytes_inst_read::cpu2.inst 282112 # Number of instructions bytes read from this memory
system.physmem.bytes_inst_read::total 917888 # Number of instructions bytes read from this memory
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.physmem.bytes_written::writebacks 7459584 # Number of bytes written to this memory
system.physmem.bytes_written::total 7459584 # Number of bytes written to this memory
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.num_reads::cpu0.inst 7632 # Number of read requests responded to by this memory
system.physmem.num_reads::cpu0.data 313269 # Number of read requests responded to by this memory
regressions: update due to cache latency fix
2013-03-28 00:36:21 +01:00
system.physmem.num_reads::tsunami.ide 41443 # Number of read requests responded to by this memory
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.num_reads::cpu1.inst 2302 # Number of read requests responded to by this memory
system.physmem.num_reads::cpu1.data 35788 # Number of read requests responded to by this memory
system.physmem.num_reads::cpu2.inst 4408 # Number of read requests responded to by this memory
system.physmem.num_reads::cpu2.data 39465 # Number of read requests responded to by this memory
system.physmem.num_reads::total 444307 # Number of read requests responded to by this memory
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.physmem.num_writes::writebacks 116556 # Number of write requests responded to by this memory
system.physmem.num_writes::total 116556 # Number of write requests responded to by this memory
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.bw_read::cpu0.inst 265071 # Total read bandwidth from this memory (bytes/s)
system.physmem.bw_read::cpu0.data 10880316 # Total read bandwidth from this memory (bytes/s)
system.physmem.bw_read::tsunami.ide 1439379 # Total read bandwidth from this memory (bytes/s)
system.physmem.bw_read::cpu1.inst 79952 # Total read bandwidth from this memory (bytes/s)
system.physmem.bw_read::cpu1.data 1242973 # Total read bandwidth from this memory (bytes/s)
system.physmem.bw_read::cpu2.inst 153097 # Total read bandwidth from this memory (bytes/s)
system.physmem.bw_read::cpu2.data 1370680 # Total read bandwidth from this memory (bytes/s)
system.physmem.bw_read::total 15431468 # Total read bandwidth from this memory (bytes/s)
system.physmem.bw_inst_read::cpu0.inst 265071 # Instruction read bandwidth from this memory (bytes/s)
system.physmem.bw_inst_read::cpu1.inst 79952 # Instruction read bandwidth from this memory (bytes/s)
system.physmem.bw_inst_read::cpu2.inst 153097 # Instruction read bandwidth from this memory (bytes/s)
system.physmem.bw_inst_read::total 498120 # Instruction read bandwidth from this memory (bytes/s)
system.physmem.bw_write::writebacks 4048170 # Write bandwidth from this memory (bytes/s)
system.physmem.bw_write::total 4048170 # Write bandwidth from this memory (bytes/s)
system.physmem.bw_total::writebacks 4048170 # Total bandwidth to/from this memory (bytes/s)
system.physmem.bw_total::cpu0.inst 265071 # Total bandwidth to/from this memory (bytes/s)
system.physmem.bw_total::cpu0.data 10880316 # Total bandwidth to/from this memory (bytes/s)
system.physmem.bw_total::tsunami.ide 1439379 # Total bandwidth to/from this memory (bytes/s)
system.physmem.bw_total::cpu1.inst 79952 # Total bandwidth to/from this memory (bytes/s)
system.physmem.bw_total::cpu1.data 1242973 # Total bandwidth to/from this memory (bytes/s)
system.physmem.bw_total::cpu2.inst 153097 # Total bandwidth to/from this memory (bytes/s)
system.physmem.bw_total::cpu2.data 1370680 # Total bandwidth to/from this memory (bytes/s)
system.physmem.bw_total::total 19479638 # Total bandwidth to/from this memory (bytes/s)
stats: Cumulative stats update This patch updates the stats to reflect the: 1) addition of the internal queue in SimpleMemory, 2) moving of the memory class outside FSConfig, 3) fixing up of the 2D vector printing format, 4) specifying burst size and interface width for the DRAM instead of relying on cache-line size, 5) performing merging in the DRAM controller write buffer, and 6) fixing how idle cycles are counted in the atomic and timing CPU models. The main reason for bundling them up is to minimise the changeset size.
2013-08-19 09:52:36 +02:00
system.physmem.readReqs 99238 # Total number of read requests accepted by DRAM controller
system.physmem.writeReqs 44800 # Total number of write requests accepted by DRAM controller
system.physmem.readBursts 99238 # Total number of DRAM read bursts. Each DRAM read request translates to either one or multiple DRAM read bursts
system.physmem.writeBursts 44800 # Total number of DRAM write bursts. Each DRAM write request translates to either one or multiple DRAM write bursts
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.bytesRead 6351232 # Total number of bytes read from memory
system.physmem.bytesWritten 2867200 # Total number of bytes written to memory
system.physmem.bytesConsumedRd 6351232 # bytesRead derated as per pkt->getSize()
system.physmem.bytesConsumedWr 2867200 # bytesWritten derated as per pkt->getSize()
stats: Cumulative stats update This patch updates the stats to reflect the: 1) addition of the internal queue in SimpleMemory, 2) moving of the memory class outside FSConfig, 3) fixing up of the 2D vector printing format, 4) specifying burst size and interface width for the DRAM instead of relying on cache-line size, 5) performing merging in the DRAM controller write buffer, and 6) fixing how idle cycles are counted in the atomic and timing CPU models. The main reason for bundling them up is to minimise the changeset size.
2013-08-19 09:52:36 +02:00
system.physmem.servicedByWrQ 11 # Number of DRAM read bursts serviced by write Q
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.physmem.neitherReadNorWrite 44 # Reqs where no action is needed
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.perBankRdReqs::0 6232 # Track reads on a per bank basis
system.physmem.perBankRdReqs::1 6043 # Track reads on a per bank basis
system.physmem.perBankRdReqs::2 6220 # Track reads on a per bank basis
system.physmem.perBankRdReqs::3 6348 # Track reads on a per bank basis
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.physmem.perBankRdReqs::4 5767 # Track reads on a per bank basis
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.perBankRdReqs::5 6398 # Track reads on a per bank basis
system.physmem.perBankRdReqs::6 6152 # Track reads on a per bank basis
system.physmem.perBankRdReqs::7 6059 # Track reads on a per bank basis
system.physmem.perBankRdReqs::8 6519 # Track reads on a per bank basis
system.physmem.perBankRdReqs::9 6372 # Track reads on a per bank basis
system.physmem.perBankRdReqs::10 6626 # Track reads on a per bank basis
system.physmem.perBankRdReqs::11 6008 # Track reads on a per bank basis
system.physmem.perBankRdReqs::12 5967 # Track reads on a per bank basis
system.physmem.perBankRdReqs::13 6231 # Track reads on a per bank basis
system.physmem.perBankRdReqs::14 6240 # Track reads on a per bank basis
system.physmem.perBankRdReqs::15 6045 # Track reads on a per bank basis
system.physmem.perBankWrReqs::0 2861 # Track writes on a per bank basis
system.physmem.perBankWrReqs::1 2670 # Track writes on a per bank basis
system.physmem.perBankWrReqs::2 2847 # Track writes on a per bank basis
system.physmem.perBankWrReqs::3 2964 # Track writes on a per bank basis
system.physmem.perBankWrReqs::4 2622 # Track writes on a per bank basis
system.physmem.perBankWrReqs::5 3000 # Track writes on a per bank basis
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.physmem.perBankWrReqs::6 2942 # Track writes on a per bank basis
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.perBankWrReqs::7 2703 # Track writes on a per bank basis
system.physmem.perBankWrReqs::8 3213 # Track writes on a per bank basis
system.physmem.perBankWrReqs::9 2742 # Track writes on a per bank basis
system.physmem.perBankWrReqs::10 3001 # Track writes on a per bank basis
system.physmem.perBankWrReqs::11 2449 # Track writes on a per bank basis
system.physmem.perBankWrReqs::12 2468 # Track writes on a per bank basis
system.physmem.perBankWrReqs::13 2705 # Track writes on a per bank basis
system.physmem.perBankWrReqs::14 2852 # Track writes on a per bank basis
system.physmem.perBankWrReqs::15 2761 # Track writes on a per bank basis
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.physmem.numRdRetry 0 # Number of times rd buffer was full causing retry
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.physmem.numWrRetry 0 # Number of times wr buffer was full causing retry
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.totGap 1841692926500 # Total gap between requests
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.physmem.readPktSize::0 0 # Categorize read packet sizes
system.physmem.readPktSize::1 0 # Categorize read packet sizes
system.physmem.readPktSize::2 0 # Categorize read packet sizes
system.physmem.readPktSize::3 0 # Categorize read packet sizes
system.physmem.readPktSize::4 0 # Categorize read packet sizes
system.physmem.readPktSize::5 0 # Categorize read packet sizes
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.readPktSize::6 99238 # Categorize read packet sizes
stats: Update stats to reflect SimpleDRAM changes This patch bumps the stats to reflect the slight change in how the retry is handled, and also the pruning of some redundant stats.
2013-03-01 19:20:30 +01:00
system.physmem.writePktSize::0 0 # Categorize write packet sizes
system.physmem.writePktSize::1 0 # Categorize write packet sizes
system.physmem.writePktSize::2 0 # Categorize write packet sizes
system.physmem.writePktSize::3 0 # Categorize write packet sizes
system.physmem.writePktSize::4 0 # Categorize write packet sizes
system.physmem.writePktSize::5 0 # Categorize write packet sizes
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.writePktSize::6 44800 # Categorize write packet sizes
system.physmem.rdQLenPdf::0 67489 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::1 12659 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::2 6294 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::3 2227 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::4 1387 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::5 1264 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::6 650 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::7 635 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::8 621 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::9 612 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::10 600 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::11 599 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::12 588 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::13 864 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::14 994 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::15 932 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::16 505 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::17 183 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::18 84 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::19 39 # What read queue length does an incoming req see
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.physmem.rdQLenPdf::20 1 # What read queue length does an incoming req see
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.physmem.rdQLenPdf::21 0 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::22 0 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::23 0 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::24 0 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::25 0 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::26 0 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::27 0 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::28 0 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::29 0 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::30 0 # What read queue length does an incoming req see
system.physmem.rdQLenPdf::31 0 # What read queue length does an incoming req see
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.wrQLenPdf::0 1381 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::1 1406 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::2 1848 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::3 1963 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::4 1959 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::5 1957 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::6 1956 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::7 1952 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::8 1949 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::9 1949 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::10 1946 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::11 1943 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::12 1942 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::13 1941 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::14 1939 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::15 1937 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::16 1937 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::17 1934 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::18 1933 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::19 1931 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::20 1929 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::21 1928 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::22 1926 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::23 626 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::24 568 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::25 120 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::26 0 # What write queue length does an incoming req see
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.physmem.wrQLenPdf::27 0 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::28 0 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::29 0 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::30 0 # What write queue length does an incoming req see
system.physmem.wrQLenPdf::31 0 # What write queue length does an incoming req see
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.bytesPerActivate::samples 15760 # Bytes accessed per row activation
system.physmem.bytesPerActivate::mean 584.832487 # Bytes accessed per row activation
system.physmem.bytesPerActivate::gmean 171.909397 # Bytes accessed per row activation
system.physmem.bytesPerActivate::stdev 1926.760563 # Bytes accessed per row activation
system.physmem.bytesPerActivate::64-67 6603 41.90% 41.90% # Bytes accessed per row activation
system.physmem.bytesPerActivate::128-131 2572 16.32% 58.22% # Bytes accessed per row activation
system.physmem.bytesPerActivate::192-195 1454 9.23% 67.44% # Bytes accessed per row activation
system.physmem.bytesPerActivate::256-259 899 5.70% 73.15% # Bytes accessed per row activation
system.physmem.bytesPerActivate::320-323 642 4.07% 77.22% # Bytes accessed per row activation
system.physmem.bytesPerActivate::384-387 535 3.39% 80.62% # Bytes accessed per row activation
system.physmem.bytesPerActivate::448-451 370 2.35% 82.96% # Bytes accessed per row activation
system.physmem.bytesPerActivate::512-515 312 1.98% 84.94% # Bytes accessed per row activation
system.physmem.bytesPerActivate::576-579 250 1.59% 86.53% # Bytes accessed per row activation
system.physmem.bytesPerActivate::640-643 195 1.24% 87.77% # Bytes accessed per row activation
system.physmem.bytesPerActivate::704-707 235 1.49% 89.26% # Bytes accessed per row activation
system.physmem.bytesPerActivate::768-771 190 1.21% 90.46% # Bytes accessed per row activation
system.physmem.bytesPerActivate::832-835 101 0.64% 91.10% # Bytes accessed per row activation
system.physmem.bytesPerActivate::896-899 71 0.45% 91.55% # Bytes accessed per row activation
system.physmem.bytesPerActivate::960-963 63 0.40% 91.95% # Bytes accessed per row activation
system.physmem.bytesPerActivate::1024-1027 80 0.51% 92.46% # Bytes accessed per row activation
system.physmem.bytesPerActivate::1088-1091 51 0.32% 92.79% # Bytes accessed per row activation
system.physmem.bytesPerActivate::1152-1155 28 0.18% 92.96% # Bytes accessed per row activation
system.physmem.bytesPerActivate::1216-1219 32 0.20% 93.17% # Bytes accessed per row activation
system.physmem.bytesPerActivate::1280-1283 74 0.47% 93.64% # Bytes accessed per row activation
system.physmem.bytesPerActivate::1344-1347 51 0.32% 93.96% # Bytes accessed per row activation
system.physmem.bytesPerActivate::1408-1411 34 0.22% 94.18% # Bytes accessed per row activation
system.physmem.bytesPerActivate::1472-1475 173 1.10% 95.27% # Bytes accessed per row activation
system.physmem.bytesPerActivate::1536-1539 86 0.55% 95.82% # Bytes accessed per row activation
system.physmem.bytesPerActivate::1600-1603 27 0.17% 95.99% # Bytes accessed per row activation
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.physmem.bytesPerActivate::1664-1667 14 0.09% 96.08% # Bytes accessed per row activation
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.bytesPerActivate::1728-1731 12 0.08% 96.15% # Bytes accessed per row activation
system.physmem.bytesPerActivate::1792-1795 22 0.14% 96.29% # Bytes accessed per row activation
system.physmem.bytesPerActivate::1856-1859 10 0.06% 96.36% # Bytes accessed per row activation
system.physmem.bytesPerActivate::1920-1923 4 0.03% 96.38% # Bytes accessed per row activation
system.physmem.bytesPerActivate::1984-1987 2 0.01% 96.40% # Bytes accessed per row activation
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.physmem.bytesPerActivate::2048-2051 6 0.04% 96.43% # Bytes accessed per row activation
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.bytesPerActivate::2112-2115 7 0.04% 96.48% # Bytes accessed per row activation
system.physmem.bytesPerActivate::2176-2179 1 0.01% 96.48% # Bytes accessed per row activation
system.physmem.bytesPerActivate::2304-2307 3 0.02% 96.50% # Bytes accessed per row activation
system.physmem.bytesPerActivate::2368-2371 1 0.01% 96.51% # Bytes accessed per row activation
system.physmem.bytesPerActivate::2432-2435 2 0.01% 96.52% # Bytes accessed per row activation
system.physmem.bytesPerActivate::2496-2499 1 0.01% 96.53% # Bytes accessed per row activation
system.physmem.bytesPerActivate::2560-2563 2 0.01% 96.54% # Bytes accessed per row activation
system.physmem.bytesPerActivate::2624-2627 1 0.01% 96.55% # Bytes accessed per row activation
system.physmem.bytesPerActivate::2688-2691 1 0.01% 96.55% # Bytes accessed per row activation
system.physmem.bytesPerActivate::2816-2819 2 0.01% 96.57% # Bytes accessed per row activation
system.physmem.bytesPerActivate::2880-2883 3 0.02% 96.59% # Bytes accessed per row activation
system.physmem.bytesPerActivate::3008-3011 3 0.02% 96.61% # Bytes accessed per row activation
system.physmem.bytesPerActivate::3072-3075 1 0.01% 96.61% # Bytes accessed per row activation
system.physmem.bytesPerActivate::3520-3523 1 0.01% 96.62% # Bytes accessed per row activation
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.physmem.bytesPerActivate::3584-3587 1 0.01% 96.62% # Bytes accessed per row activation
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.bytesPerActivate::3648-3651 2 0.01% 96.64% # Bytes accessed per row activation
system.physmem.bytesPerActivate::3712-3715 1 0.01% 96.64% # Bytes accessed per row activation
system.physmem.bytesPerActivate::3904-3907 2 0.01% 96.66% # Bytes accessed per row activation
system.physmem.bytesPerActivate::4224-4227 1 0.01% 96.66% # Bytes accessed per row activation
system.physmem.bytesPerActivate::4672-4675 1 0.01% 96.67% # Bytes accessed per row activation
system.physmem.bytesPerActivate::4928-4931 1 0.01% 96.68% # Bytes accessed per row activation
system.physmem.bytesPerActivate::5056-5059 2 0.01% 96.69% # Bytes accessed per row activation
system.physmem.bytesPerActivate::5312-5315 2 0.01% 96.70% # Bytes accessed per row activation
system.physmem.bytesPerActivate::5696-5699 1 0.01% 96.71% # Bytes accessed per row activation
system.physmem.bytesPerActivate::6464-6467 1 0.01% 96.71% # Bytes accessed per row activation
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.physmem.bytesPerActivate::7808-7811 1 0.01% 96.72% # Bytes accessed per row activation
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.bytesPerActivate::8000-8003 1 0.01% 96.73% # Bytes accessed per row activation
system.physmem.bytesPerActivate::8192-8195 383 2.43% 99.16% # Bytes accessed per row activation
system.physmem.bytesPerActivate::11456-11459 1 0.01% 99.16% # Bytes accessed per row activation
system.physmem.bytesPerActivate::13824-13827 1 0.01% 99.17% # Bytes accessed per row activation
system.physmem.bytesPerActivate::14016-14019 1 0.01% 99.18% # Bytes accessed per row activation
system.physmem.bytesPerActivate::14400-14403 1 0.01% 99.18% # Bytes accessed per row activation
system.physmem.bytesPerActivate::15040-15043 2 0.01% 99.19% # Bytes accessed per row activation
system.physmem.bytesPerActivate::15360-15363 6 0.04% 99.23% # Bytes accessed per row activation
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.physmem.bytesPerActivate::15552-15555 1 0.01% 99.24% # Bytes accessed per row activation
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.bytesPerActivate::15744-15747 1 0.01% 99.24% # Bytes accessed per row activation
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.physmem.bytesPerActivate::16384-16387 111 0.70% 99.95% # Bytes accessed per row activation
system.physmem.bytesPerActivate::16448-16451 1 0.01% 99.96% # Bytes accessed per row activation
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.bytesPerActivate::16512-16515 2 0.01% 99.97% # Bytes accessed per row activation
system.physmem.bytesPerActivate::16640-16643 3 0.02% 99.99% # Bytes accessed per row activation
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.physmem.bytesPerActivate::16832-16835 1 0.01% 99.99% # Bytes accessed per row activation
system.physmem.bytesPerActivate::17088-17091 1 0.01% 100.00% # Bytes accessed per row activation
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.bytesPerActivate::total 15760 # Bytes accessed per row activation
system.physmem.totQLat 1910826000 # Total cycles spent in queuing delays
system.physmem.totMemAccLat 3572864750 # Sum of mem lat for all requests
system.physmem.totBusLat 496135000 # Total cycles spent in databus access
system.physmem.totBankLat 1165903750 # Total cycles spent in bank access
system.physmem.avgQLat 19257.12 # Average queueing delay per request
system.physmem.avgBankLat 11749.86 # Average bank access latency per request
stats: Update stats for regressions using SimpleDDR3 This patch updates the regression stats to reflect that they are using the SimpleDDR3 controller by default.
2013-01-31 13:49:16 +01:00
system.physmem.avgBusLat 5000.00 # Average bus latency per request
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.avgMemAccLat 36006.98 # Average memory access latency
system.physmem.avgRdBW 3.45 # Average achieved read bandwidth in MB/s
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.physmem.avgWrBW 1.56 # Average achieved write bandwidth in MB/s
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.avgConsumedRdBW 3.45 # Average consumed read bandwidth in MB/s
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.physmem.avgConsumedWrBW 1.56 # Average consumed write bandwidth in MB/s
stats: Update stats for regressions using SimpleDDR3 This patch updates the regression stats to reflect that they are using the SimpleDDR3 controller by default.
2013-01-31 13:49:16 +01:00
system.physmem.peakBW 12800.00 # Theoretical peak bandwidth in MB/s
system.physmem.busUtil 0.04 # Data bus utilization in percentage
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.physmem.avgRdQLen 0.00 # Average read queue length over time
regressions: update due to cache latency fix
2013-03-28 00:36:21 +01:00
system.physmem.avgWrQLen 0.17 # Average write queue length over time
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.physmem.readRowHits 92920 # Number of row buffer hits during reads
system.physmem.writeRowHits 35346 # Number of row buffer hits during writes
system.physmem.readRowHitRate 93.64 # Row buffer hit rate for reads
system.physmem.writeRowHitRate 78.90 # Row buffer hit rate for writes
system.physmem.avgGap 12786160.09 # Average gap between requests
system.membus.throughput 19523578 # Throughput (bytes/s)
system.membus.trans_dist::ReadReq 45592 # Transaction distribution
system.membus.trans_dist::ReadResp 45560 # Transaction distribution
system.membus.trans_dist::WriteReq 3756 # Transaction distribution
system.membus.trans_dist::WriteResp 3756 # Transaction distribution
system.membus.trans_dist::Writeback 44800 # Transaction distribution
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.membus.trans_dist::UpgradeReq 46 # Transaction distribution
system.membus.trans_dist::SCUpgradeReq 1 # Transaction distribution
system.membus.trans_dist::UpgradeResp 47 # Transaction distribution
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.membus.trans_dist::ReadExReq 56741 # Transaction distribution
system.membus.trans_dist::ReadExResp 56741 # Transaction distribution
system.membus.trans_dist::BadAddressError 32 # Transaction distribution
system.membus.pkt_count_system.l2c.mem_side::system.bridge.slave 13322 # Packet count per connected master and slave (bytes)
system.membus.pkt_count_system.l2c.mem_side::system.physmem.port 191660 # Packet count per connected master and slave (bytes)
system.membus.pkt_count_system.l2c.mem_side::system.membus.badaddr_responder.pio 64 # Packet count per connected master and slave (bytes)
system.membus.pkt_count_system.l2c.mem_side::total 205046 # Packet count per connected master and slave (bytes)
system.membus.pkt_count_system.iocache.mem_side::system.physmem.port 51865 # Packet count per connected master and slave (bytes)
system.membus.pkt_count_system.iocache.mem_side::total 51865 # Packet count per connected master and slave (bytes)
system.membus.pkt_count::total 256911 # Packet count per connected master and slave (bytes)
system.membus.tot_pkt_size_system.l2c.mem_side::system.bridge.slave 15754 # Cumulative packet size per connected master and slave (bytes)
system.membus.tot_pkt_size_system.l2c.mem_side::system.physmem.port 7009472 # Cumulative packet size per connected master and slave (bytes)
system.membus.tot_pkt_size_system.l2c.mem_side::total 7025226 # Cumulative packet size per connected master and slave (bytes)
system.membus.tot_pkt_size_system.iocache.mem_side::system.physmem.port 2208960 # Cumulative packet size per connected master and slave (bytes)
system.membus.tot_pkt_size_system.iocache.mem_side::total 2208960 # Cumulative packet size per connected master and slave (bytes)
system.membus.tot_pkt_size::total 9234186 # Cumulative packet size per connected master and slave (bytes)
system.membus.data_through_bus 35966088 # Total data (bytes)
system.membus.snoop_data_through_bus 10112 # Total snoop data (bytes)
system.membus.reqLayer0.occupancy 12465000 # Layer occupancy (ticks)
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.membus.reqLayer0.utilization 0.0 # Layer utilization (%)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.membus.reqLayer1.occupancy 516080000 # Layer occupancy (ticks)
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.membus.reqLayer1.utilization 0.0 # Layer utilization (%)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.membus.reqLayer2.occupancy 39000 # Layer occupancy (ticks)
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.membus.reqLayer2.utilization 0.0 # Layer utilization (%)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.membus.respLayer1.occupancy 771793954 # Layer occupancy (ticks)
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.membus.respLayer1.utilization 0.0 # Layer utilization (%)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.membus.respLayer2.occupancy 156435750 # Layer occupancy (ticks)
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.membus.respLayer2.utilization 0.0 # Layer utilization (%)
stats: Cumulative stats update This patch updates the stats to reflect the: 1) addition of the internal queue in SimpleMemory, 2) moving of the memory class outside FSConfig, 3) fixing up of the 2D vector printing format, 4) specifying burst size and interface width for the DRAM instead of relying on cache-line size, 5) performing merging in the DRAM controller write buffer, and 6) fixing how idle cycles are counted in the atomic and timing CPU models. The main reason for bundling them up is to minimise the changeset size.
2013-08-19 09:52:36 +02:00
system.l2c.tags.replacements 337384 # number of replacements
system.l2c.tags.tagsinuse 65423.390976 # Cycle average of tags in use
system.l2c.tags.total_refs 2471195 # Total number of references to valid blocks.
system.l2c.tags.sampled_refs 402547 # Sample count of references to valid blocks.
system.l2c.tags.avg_refs 6.138898 # Average number of references to valid blocks.
system.l2c.tags.warmup_cycle 614754000 # Cycle when the warmup percentage was hit.
system.l2c.tags.occ_blocks::writebacks 54840.022307 # Average occupied blocks per requestor
system.l2c.tags.occ_blocks::cpu0.inst 2455.785986 # Average occupied blocks per requestor
system.l2c.tags.occ_blocks::cpu0.data 2733.317890 # Average occupied blocks per requestor
system.l2c.tags.occ_blocks::cpu1.inst 573.564095 # Average occupied blocks per requestor
system.l2c.tags.occ_blocks::cpu1.data 593.217562 # Average occupied blocks per requestor
system.l2c.tags.occ_blocks::cpu2.inst 2104.783507 # Average occupied blocks per requestor
system.l2c.tags.occ_blocks::cpu2.data 2122.699630 # Average occupied blocks per requestor
system.l2c.tags.occ_percent::writebacks 0.836792 # Average percentage of cache occupancy
system.l2c.tags.occ_percent::cpu0.inst 0.037472 # Average percentage of cache occupancy
system.l2c.tags.occ_percent::cpu0.data 0.041707 # Average percentage of cache occupancy
system.l2c.tags.occ_percent::cpu1.inst 0.008752 # Average percentage of cache occupancy
system.l2c.tags.occ_percent::cpu1.data 0.009052 # Average percentage of cache occupancy
system.l2c.tags.occ_percent::cpu2.inst 0.032116 # Average percentage of cache occupancy
system.l2c.tags.occ_percent::cpu2.data 0.032390 # Average percentage of cache occupancy
system.l2c.tags.occ_percent::total 0.998282 # Average percentage of cache occupancy
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.l2c.ReadReq_hits::cpu0.inst 518817 # number of ReadReq hits
system.l2c.ReadReq_hits::cpu0.data 493229 # number of ReadReq hits
system.l2c.ReadReq_hits::cpu1.inst 124693 # number of ReadReq hits
system.l2c.ReadReq_hits::cpu1.data 83730 # number of ReadReq hits
system.l2c.ReadReq_hits::cpu2.inst 293247 # number of ReadReq hits
system.l2c.ReadReq_hits::cpu2.data 239325 # number of ReadReq hits
system.l2c.ReadReq_hits::total 1753041 # number of ReadReq hits
system.l2c.Writeback_hits::writebacks 835257 # number of Writeback hits
system.l2c.Writeback_hits::total 835257 # number of Writeback hits
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.l2c.UpgradeReq_hits::cpu0.data 3 # number of UpgradeReq hits
system.l2c.UpgradeReq_hits::cpu1.data 1 # number of UpgradeReq hits
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.l2c.UpgradeReq_hits::cpu2.data 3 # number of UpgradeReq hits
system.l2c.UpgradeReq_hits::total 7 # number of UpgradeReq hits
system.l2c.SCUpgradeReq_hits::cpu2.data 1 # number of SCUpgradeReq hits
system.l2c.SCUpgradeReq_hits::total 1 # number of SCUpgradeReq hits
system.l2c.ReadExReq_hits::cpu0.data 92556 # number of ReadExReq hits
system.l2c.ReadExReq_hits::cpu1.data 26552 # number of ReadExReq hits
system.l2c.ReadExReq_hits::cpu2.data 67733 # number of ReadExReq hits
system.l2c.ReadExReq_hits::total 186841 # number of ReadExReq hits
system.l2c.demand_hits::cpu0.inst 518817 # number of demand (read+write) hits
system.l2c.demand_hits::cpu0.data 585785 # number of demand (read+write) hits
system.l2c.demand_hits::cpu1.inst 124693 # number of demand (read+write) hits
system.l2c.demand_hits::cpu1.data 110282 # number of demand (read+write) hits
system.l2c.demand_hits::cpu2.inst 293247 # number of demand (read+write) hits
system.l2c.demand_hits::cpu2.data 307058 # number of demand (read+write) hits
system.l2c.demand_hits::total 1939882 # number of demand (read+write) hits
system.l2c.overall_hits::cpu0.inst 518817 # number of overall hits
system.l2c.overall_hits::cpu0.data 585785 # number of overall hits
system.l2c.overall_hits::cpu1.inst 124693 # number of overall hits
system.l2c.overall_hits::cpu1.data 110282 # number of overall hits
system.l2c.overall_hits::cpu2.inst 293247 # number of overall hits
system.l2c.overall_hits::cpu2.data 307058 # number of overall hits
system.l2c.overall_hits::total 1939882 # number of overall hits
system.l2c.ReadReq_misses::cpu0.inst 7632 # number of ReadReq misses
system.l2c.ReadReq_misses::cpu0.data 237318 # number of ReadReq misses
system.l2c.ReadReq_misses::cpu1.inst 2302 # number of ReadReq misses
system.l2c.ReadReq_misses::cpu1.data 17365 # number of ReadReq misses
system.l2c.ReadReq_misses::cpu2.inst 4408 # number of ReadReq misses
system.l2c.ReadReq_misses::cpu2.data 18542 # number of ReadReq misses
system.l2c.ReadReq_misses::total 287567 # number of ReadReq misses
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.l2c.UpgradeReq_misses::cpu0.data 8 # number of UpgradeReq misses
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.l2c.UpgradeReq_misses::cpu2.data 11 # number of UpgradeReq misses
system.l2c.UpgradeReq_misses::total 19 # number of UpgradeReq misses
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.l2c.SCUpgradeReq_misses::cpu2.data 1 # number of SCUpgradeReq misses
system.l2c.SCUpgradeReq_misses::total 1 # number of SCUpgradeReq misses
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.l2c.ReadExReq_misses::cpu0.data 76225 # number of ReadExReq misses
system.l2c.ReadExReq_misses::cpu1.data 18470 # number of ReadExReq misses
system.l2c.ReadExReq_misses::cpu2.data 21026 # number of ReadExReq misses
system.l2c.ReadExReq_misses::total 115721 # number of ReadExReq misses
system.l2c.demand_misses::cpu0.inst 7632 # number of demand (read+write) misses
system.l2c.demand_misses::cpu0.data 313543 # number of demand (read+write) misses
system.l2c.demand_misses::cpu1.inst 2302 # number of demand (read+write) misses
system.l2c.demand_misses::cpu1.data 35835 # number of demand (read+write) misses
system.l2c.demand_misses::cpu2.inst 4408 # number of demand (read+write) misses
system.l2c.demand_misses::cpu2.data 39568 # number of demand (read+write) misses
system.l2c.demand_misses::total 403288 # number of demand (read+write) misses
system.l2c.overall_misses::cpu0.inst 7632 # number of overall misses
system.l2c.overall_misses::cpu0.data 313543 # number of overall misses
system.l2c.overall_misses::cpu1.inst 2302 # number of overall misses
system.l2c.overall_misses::cpu1.data 35835 # number of overall misses
system.l2c.overall_misses::cpu2.inst 4408 # number of overall misses
system.l2c.overall_misses::cpu2.data 39568 # number of overall misses
system.l2c.overall_misses::total 403288 # number of overall misses
system.l2c.ReadReq_miss_latency::cpu1.inst 186892499 # number of ReadReq miss cycles
system.l2c.ReadReq_miss_latency::cpu1.data 1138263250 # number of ReadReq miss cycles
system.l2c.ReadReq_miss_latency::cpu2.inst 380152250 # number of ReadReq miss cycles
system.l2c.ReadReq_miss_latency::cpu2.data 1198688249 # number of ReadReq miss cycles
system.l2c.ReadReq_miss_latency::total 2903996248 # number of ReadReq miss cycles
system.l2c.UpgradeReq_miss_latency::cpu2.data 263498 # number of UpgradeReq miss cycles
system.l2c.UpgradeReq_miss_latency::total 263498 # number of UpgradeReq miss cycles
system.l2c.ReadExReq_miss_latency::cpu1.data 1209958491 # number of ReadExReq miss cycles
system.l2c.ReadExReq_miss_latency::cpu2.data 1715474224 # number of ReadExReq miss cycles
system.l2c.ReadExReq_miss_latency::total 2925432715 # number of ReadExReq miss cycles
system.l2c.demand_miss_latency::cpu1.inst 186892499 # number of demand (read+write) miss cycles
system.l2c.demand_miss_latency::cpu1.data 2348221741 # number of demand (read+write) miss cycles
system.l2c.demand_miss_latency::cpu2.inst 380152250 # number of demand (read+write) miss cycles
system.l2c.demand_miss_latency::cpu2.data 2914162473 # number of demand (read+write) miss cycles
system.l2c.demand_miss_latency::total 5829428963 # number of demand (read+write) miss cycles
system.l2c.overall_miss_latency::cpu1.inst 186892499 # number of overall miss cycles
system.l2c.overall_miss_latency::cpu1.data 2348221741 # number of overall miss cycles
system.l2c.overall_miss_latency::cpu2.inst 380152250 # number of overall miss cycles
system.l2c.overall_miss_latency::cpu2.data 2914162473 # number of overall miss cycles
system.l2c.overall_miss_latency::total 5829428963 # number of overall miss cycles
system.l2c.ReadReq_accesses::cpu0.inst 526449 # number of ReadReq accesses(hits+misses)
system.l2c.ReadReq_accesses::cpu0.data 730547 # number of ReadReq accesses(hits+misses)
system.l2c.ReadReq_accesses::cpu1.inst 126995 # number of ReadReq accesses(hits+misses)
system.l2c.ReadReq_accesses::cpu1.data 101095 # number of ReadReq accesses(hits+misses)
system.l2c.ReadReq_accesses::cpu2.inst 297655 # number of ReadReq accesses(hits+misses)
system.l2c.ReadReq_accesses::cpu2.data 257867 # number of ReadReq accesses(hits+misses)
system.l2c.ReadReq_accesses::total 2040608 # number of ReadReq accesses(hits+misses)
system.l2c.Writeback_accesses::writebacks 835257 # number of Writeback accesses(hits+misses)
system.l2c.Writeback_accesses::total 835257 # number of Writeback accesses(hits+misses)
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.l2c.UpgradeReq_accesses::cpu0.data 11 # number of UpgradeReq accesses(hits+misses)
system.l2c.UpgradeReq_accesses::cpu1.data 1 # number of UpgradeReq accesses(hits+misses)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.l2c.UpgradeReq_accesses::cpu2.data 14 # number of UpgradeReq accesses(hits+misses)
system.l2c.UpgradeReq_accesses::total 26 # number of UpgradeReq accesses(hits+misses)
system.l2c.SCUpgradeReq_accesses::cpu2.data 2 # number of SCUpgradeReq accesses(hits+misses)
system.l2c.SCUpgradeReq_accesses::total 2 # number of SCUpgradeReq accesses(hits+misses)
system.l2c.ReadExReq_accesses::cpu0.data 168781 # number of ReadExReq accesses(hits+misses)
system.l2c.ReadExReq_accesses::cpu1.data 45022 # number of ReadExReq accesses(hits+misses)
system.l2c.ReadExReq_accesses::cpu2.data 88759 # number of ReadExReq accesses(hits+misses)
system.l2c.ReadExReq_accesses::total 302562 # number of ReadExReq accesses(hits+misses)
system.l2c.demand_accesses::cpu0.inst 526449 # number of demand (read+write) accesses
system.l2c.demand_accesses::cpu0.data 899328 # number of demand (read+write) accesses
system.l2c.demand_accesses::cpu1.inst 126995 # number of demand (read+write) accesses
system.l2c.demand_accesses::cpu1.data 146117 # number of demand (read+write) accesses
system.l2c.demand_accesses::cpu2.inst 297655 # number of demand (read+write) accesses
system.l2c.demand_accesses::cpu2.data 346626 # number of demand (read+write) accesses
system.l2c.demand_accesses::total 2343170 # number of demand (read+write) accesses
system.l2c.overall_accesses::cpu0.inst 526449 # number of overall (read+write) accesses
system.l2c.overall_accesses::cpu0.data 899328 # number of overall (read+write) accesses
system.l2c.overall_accesses::cpu1.inst 126995 # number of overall (read+write) accesses
system.l2c.overall_accesses::cpu1.data 146117 # number of overall (read+write) accesses
system.l2c.overall_accesses::cpu2.inst 297655 # number of overall (read+write) accesses
system.l2c.overall_accesses::cpu2.data 346626 # number of overall (read+write) accesses
system.l2c.overall_accesses::total 2343170 # number of overall (read+write) accesses
system.l2c.ReadReq_miss_rate::cpu0.inst 0.014497 # miss rate for ReadReq accesses
system.l2c.ReadReq_miss_rate::cpu0.data 0.324850 # miss rate for ReadReq accesses
system.l2c.ReadReq_miss_rate::cpu1.inst 0.018127 # miss rate for ReadReq accesses
system.l2c.ReadReq_miss_rate::cpu1.data 0.171769 # miss rate for ReadReq accesses
system.l2c.ReadReq_miss_rate::cpu2.inst 0.014809 # miss rate for ReadReq accesses
system.l2c.ReadReq_miss_rate::cpu2.data 0.071905 # miss rate for ReadReq accesses
system.l2c.ReadReq_miss_rate::total 0.140922 # miss rate for ReadReq accesses
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.l2c.UpgradeReq_miss_rate::cpu0.data 0.727273 # miss rate for UpgradeReq accesses
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.l2c.UpgradeReq_miss_rate::cpu2.data 0.785714 # miss rate for UpgradeReq accesses
system.l2c.UpgradeReq_miss_rate::total 0.730769 # miss rate for UpgradeReq accesses
system.l2c.SCUpgradeReq_miss_rate::cpu2.data 0.500000 # miss rate for SCUpgradeReq accesses
system.l2c.SCUpgradeReq_miss_rate::total 0.500000 # miss rate for SCUpgradeReq accesses
system.l2c.ReadExReq_miss_rate::cpu0.data 0.451621 # miss rate for ReadExReq accesses
system.l2c.ReadExReq_miss_rate::cpu1.data 0.410244 # miss rate for ReadExReq accesses
system.l2c.ReadExReq_miss_rate::cpu2.data 0.236889 # miss rate for ReadExReq accesses
system.l2c.ReadExReq_miss_rate::total 0.382470 # miss rate for ReadExReq accesses
system.l2c.demand_miss_rate::cpu0.inst 0.014497 # miss rate for demand accesses
system.l2c.demand_miss_rate::cpu0.data 0.348641 # miss rate for demand accesses
system.l2c.demand_miss_rate::cpu1.inst 0.018127 # miss rate for demand accesses
system.l2c.demand_miss_rate::cpu1.data 0.245249 # miss rate for demand accesses
system.l2c.demand_miss_rate::cpu2.inst 0.014809 # miss rate for demand accesses
system.l2c.demand_miss_rate::cpu2.data 0.114152 # miss rate for demand accesses
system.l2c.demand_miss_rate::total 0.172112 # miss rate for demand accesses
system.l2c.overall_miss_rate::cpu0.inst 0.014497 # miss rate for overall accesses
system.l2c.overall_miss_rate::cpu0.data 0.348641 # miss rate for overall accesses
system.l2c.overall_miss_rate::cpu1.inst 0.018127 # miss rate for overall accesses
system.l2c.overall_miss_rate::cpu1.data 0.245249 # miss rate for overall accesses
system.l2c.overall_miss_rate::cpu2.inst 0.014809 # miss rate for overall accesses
system.l2c.overall_miss_rate::cpu2.data 0.114152 # miss rate for overall accesses
system.l2c.overall_miss_rate::total 0.172112 # miss rate for overall accesses
system.l2c.ReadReq_avg_miss_latency::cpu1.inst 81187.010860 # average ReadReq miss latency
system.l2c.ReadReq_avg_miss_latency::cpu1.data 65549.280161 # average ReadReq miss latency
system.l2c.ReadReq_avg_miss_latency::cpu2.inst 86241.436025 # average ReadReq miss latency
system.l2c.ReadReq_avg_miss_latency::cpu2.data 64647.192806 # average ReadReq miss latency
system.l2c.ReadReq_avg_miss_latency::total 10098.503124 # average ReadReq miss latency
system.l2c.UpgradeReq_avg_miss_latency::cpu2.data 23954.363636 # average UpgradeReq miss latency
system.l2c.UpgradeReq_avg_miss_latency::total 13868.315789 # average UpgradeReq miss latency
system.l2c.ReadExReq_avg_miss_latency::cpu1.data 65509.393124 # average ReadExReq miss latency
system.l2c.ReadExReq_avg_miss_latency::cpu2.data 81588.234757 # average ReadExReq miss latency
system.l2c.ReadExReq_avg_miss_latency::total 25280.050423 # average ReadExReq miss latency
system.l2c.demand_avg_miss_latency::cpu1.inst 81187.010860 # average overall miss latency
system.l2c.demand_avg_miss_latency::cpu1.data 65528.721669 # average overall miss latency
system.l2c.demand_avg_miss_latency::cpu2.inst 86241.436025 # average overall miss latency
system.l2c.demand_avg_miss_latency::cpu2.data 73649.476168 # average overall miss latency
system.l2c.demand_avg_miss_latency::total 14454.754327 # average overall miss latency
system.l2c.overall_avg_miss_latency::cpu1.inst 81187.010860 # average overall miss latency
system.l2c.overall_avg_miss_latency::cpu1.data 65528.721669 # average overall miss latency
system.l2c.overall_avg_miss_latency::cpu2.inst 86241.436025 # average overall miss latency
system.l2c.overall_avg_miss_latency::cpu2.data 73649.476168 # average overall miss latency
system.l2c.overall_avg_miss_latency::total 14454.754327 # average overall miss latency
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.l2c.blocked_cycles::no_mshrs 0 # number of cycles access was blocked
system.l2c.blocked_cycles::no_targets 0 # number of cycles access was blocked
system.l2c.blocked::no_mshrs 0 # number of cycles access was blocked
system.l2c.blocked::no_targets 0 # number of cycles access was blocked
system.l2c.avg_blocked_cycles::no_mshrs nan # average number of cycles each access was blocked
system.l2c.avg_blocked_cycles::no_targets nan # average number of cycles each access was blocked
system.l2c.fast_writes 0 # number of fast writes performed
system.l2c.cache_copies 0 # number of cache copies performed
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.l2c.writebacks::writebacks 75044 # number of writebacks
system.l2c.writebacks::total 75044 # number of writebacks
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.l2c.ReadReq_mshr_misses::cpu1.inst 2302 # number of ReadReq MSHR misses
system.l2c.ReadReq_mshr_misses::cpu1.data 17365 # number of ReadReq MSHR misses
system.l2c.ReadReq_mshr_misses::cpu2.inst 4408 # number of ReadReq MSHR misses
system.l2c.ReadReq_mshr_misses::cpu2.data 18542 # number of ReadReq MSHR misses
system.l2c.ReadReq_mshr_misses::total 42617 # number of ReadReq MSHR misses
system.l2c.UpgradeReq_mshr_misses::cpu2.data 11 # number of UpgradeReq MSHR misses
system.l2c.UpgradeReq_mshr_misses::total 11 # number of UpgradeReq MSHR misses
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.l2c.SCUpgradeReq_mshr_misses::cpu2.data 1 # number of SCUpgradeReq MSHR misses
system.l2c.SCUpgradeReq_mshr_misses::total 1 # number of SCUpgradeReq MSHR misses
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.l2c.ReadExReq_mshr_misses::cpu1.data 18470 # number of ReadExReq MSHR misses
system.l2c.ReadExReq_mshr_misses::cpu2.data 21026 # number of ReadExReq MSHR misses
system.l2c.ReadExReq_mshr_misses::total 39496 # number of ReadExReq MSHR misses
system.l2c.demand_mshr_misses::cpu1.inst 2302 # number of demand (read+write) MSHR misses
system.l2c.demand_mshr_misses::cpu1.data 35835 # number of demand (read+write) MSHR misses
system.l2c.demand_mshr_misses::cpu2.inst 4408 # number of demand (read+write) MSHR misses
system.l2c.demand_mshr_misses::cpu2.data 39568 # number of demand (read+write) MSHR misses
system.l2c.demand_mshr_misses::total 82113 # number of demand (read+write) MSHR misses
system.l2c.overall_mshr_misses::cpu1.inst 2302 # number of overall MSHR misses
system.l2c.overall_mshr_misses::cpu1.data 35835 # number of overall MSHR misses
system.l2c.overall_mshr_misses::cpu2.inst 4408 # number of overall MSHR misses
system.l2c.overall_mshr_misses::cpu2.data 39568 # number of overall MSHR misses
system.l2c.overall_mshr_misses::total 82113 # number of overall MSHR misses
system.l2c.ReadReq_mshr_miss_latency::cpu1.inst 157803001 # number of ReadReq MSHR miss cycles
system.l2c.ReadReq_mshr_miss_latency::cpu1.data 920718250 # number of ReadReq MSHR miss cycles
system.l2c.ReadReq_mshr_miss_latency::cpu2.inst 324243750 # number of ReadReq MSHR miss cycles
system.l2c.ReadReq_mshr_miss_latency::cpu2.data 970102251 # number of ReadReq MSHR miss cycles
system.l2c.ReadReq_mshr_miss_latency::total 2372867252 # number of ReadReq MSHR miss cycles
system.l2c.UpgradeReq_mshr_miss_latency::cpu2.data 261008 # number of UpgradeReq MSHR miss cycles
system.l2c.UpgradeReq_mshr_miss_latency::total 261008 # number of UpgradeReq MSHR miss cycles
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.l2c.SCUpgradeReq_mshr_miss_latency::cpu2.data 10001 # number of SCUpgradeReq MSHR miss cycles
system.l2c.SCUpgradeReq_mshr_miss_latency::total 10001 # number of SCUpgradeReq MSHR miss cycles
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.l2c.ReadExReq_mshr_miss_latency::cpu1.data 978413509 # number of ReadExReq MSHR miss cycles
system.l2c.ReadExReq_mshr_miss_latency::cpu2.data 1456000776 # number of ReadExReq MSHR miss cycles
system.l2c.ReadExReq_mshr_miss_latency::total 2434414285 # number of ReadExReq MSHR miss cycles
system.l2c.demand_mshr_miss_latency::cpu1.inst 157803001 # number of demand (read+write) MSHR miss cycles
system.l2c.demand_mshr_miss_latency::cpu1.data 1899131759 # number of demand (read+write) MSHR miss cycles
system.l2c.demand_mshr_miss_latency::cpu2.inst 324243750 # number of demand (read+write) MSHR miss cycles
system.l2c.demand_mshr_miss_latency::cpu2.data 2426103027 # number of demand (read+write) MSHR miss cycles
system.l2c.demand_mshr_miss_latency::total 4807281537 # number of demand (read+write) MSHR miss cycles
system.l2c.overall_mshr_miss_latency::cpu1.inst 157803001 # number of overall MSHR miss cycles
system.l2c.overall_mshr_miss_latency::cpu1.data 1899131759 # number of overall MSHR miss cycles
system.l2c.overall_mshr_miss_latency::cpu2.inst 324243750 # number of overall MSHR miss cycles
system.l2c.overall_mshr_miss_latency::cpu2.data 2426103027 # number of overall MSHR miss cycles
system.l2c.overall_mshr_miss_latency::total 4807281537 # number of overall MSHR miss cycles
system.l2c.ReadReq_mshr_uncacheable_latency::cpu1.data 275693000 # number of ReadReq MSHR uncacheable cycles
system.l2c.ReadReq_mshr_uncacheable_latency::cpu2.data 292819000 # number of ReadReq MSHR uncacheable cycles
system.l2c.ReadReq_mshr_uncacheable_latency::total 568512000 # number of ReadReq MSHR uncacheable cycles
system.l2c.WriteReq_mshr_uncacheable_latency::cpu1.data 341668500 # number of WriteReq MSHR uncacheable cycles
system.l2c.WriteReq_mshr_uncacheable_latency::cpu2.data 404628500 # number of WriteReq MSHR uncacheable cycles
system.l2c.WriteReq_mshr_uncacheable_latency::total 746297000 # number of WriteReq MSHR uncacheable cycles
system.l2c.overall_mshr_uncacheable_latency::cpu1.data 617361500 # number of overall MSHR uncacheable cycles
system.l2c.overall_mshr_uncacheable_latency::cpu2.data 697447500 # number of overall MSHR uncacheable cycles
system.l2c.overall_mshr_uncacheable_latency::total 1314809000 # number of overall MSHR uncacheable cycles
system.l2c.ReadReq_mshr_miss_rate::cpu1.inst 0.018127 # mshr miss rate for ReadReq accesses
system.l2c.ReadReq_mshr_miss_rate::cpu1.data 0.171769 # mshr miss rate for ReadReq accesses
system.l2c.ReadReq_mshr_miss_rate::cpu2.inst 0.014809 # mshr miss rate for ReadReq accesses
system.l2c.ReadReq_mshr_miss_rate::cpu2.data 0.071905 # mshr miss rate for ReadReq accesses
system.l2c.ReadReq_mshr_miss_rate::total 0.020884 # mshr miss rate for ReadReq accesses
system.l2c.UpgradeReq_mshr_miss_rate::cpu2.data 0.785714 # mshr miss rate for UpgradeReq accesses
system.l2c.UpgradeReq_mshr_miss_rate::total 0.423077 # mshr miss rate for UpgradeReq accesses
system.l2c.SCUpgradeReq_mshr_miss_rate::cpu2.data 0.500000 # mshr miss rate for SCUpgradeReq accesses
system.l2c.SCUpgradeReq_mshr_miss_rate::total 0.500000 # mshr miss rate for SCUpgradeReq accesses
system.l2c.ReadExReq_mshr_miss_rate::cpu1.data 0.410244 # mshr miss rate for ReadExReq accesses
system.l2c.ReadExReq_mshr_miss_rate::cpu2.data 0.236889 # mshr miss rate for ReadExReq accesses
system.l2c.ReadExReq_mshr_miss_rate::total 0.130539 # mshr miss rate for ReadExReq accesses
system.l2c.demand_mshr_miss_rate::cpu1.inst 0.018127 # mshr miss rate for demand accesses
system.l2c.demand_mshr_miss_rate::cpu1.data 0.245249 # mshr miss rate for demand accesses
system.l2c.demand_mshr_miss_rate::cpu2.inst 0.014809 # mshr miss rate for demand accesses
system.l2c.demand_mshr_miss_rate::cpu2.data 0.114152 # mshr miss rate for demand accesses
system.l2c.demand_mshr_miss_rate::total 0.035044 # mshr miss rate for demand accesses
system.l2c.overall_mshr_miss_rate::cpu1.inst 0.018127 # mshr miss rate for overall accesses
system.l2c.overall_mshr_miss_rate::cpu1.data 0.245249 # mshr miss rate for overall accesses
system.l2c.overall_mshr_miss_rate::cpu2.inst 0.014809 # mshr miss rate for overall accesses
system.l2c.overall_mshr_miss_rate::cpu2.data 0.114152 # mshr miss rate for overall accesses
system.l2c.overall_mshr_miss_rate::total 0.035044 # mshr miss rate for overall accesses
system.l2c.ReadReq_avg_mshr_miss_latency::cpu1.inst 68550.391399 # average ReadReq mshr miss latency
system.l2c.ReadReq_avg_mshr_miss_latency::cpu1.data 53021.494385 # average ReadReq mshr miss latency
system.l2c.ReadReq_avg_mshr_miss_latency::cpu2.inst 73558.019510 # average ReadReq mshr miss latency
system.l2c.ReadReq_avg_mshr_miss_latency::cpu2.data 52319.180833 # average ReadReq mshr miss latency
system.l2c.ReadReq_avg_mshr_miss_latency::total 55678.889927 # average ReadReq mshr miss latency
system.l2c.UpgradeReq_avg_mshr_miss_latency::cpu2.data 23728 # average UpgradeReq mshr miss latency
system.l2c.UpgradeReq_avg_mshr_miss_latency::total 23728 # average UpgradeReq mshr miss latency
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.l2c.SCUpgradeReq_avg_mshr_miss_latency::cpu2.data 10001 # average SCUpgradeReq mshr miss latency
system.l2c.SCUpgradeReq_avg_mshr_miss_latency::total 10001 # average SCUpgradeReq mshr miss latency
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.l2c.ReadExReq_avg_mshr_miss_latency::cpu1.data 52973.119058 # average ReadExReq mshr miss latency
system.l2c.ReadExReq_avg_mshr_miss_latency::cpu2.data 69247.635118 # average ReadExReq mshr miss latency
system.l2c.ReadExReq_avg_mshr_miss_latency::total 61636.983112 # average ReadExReq mshr miss latency
system.l2c.demand_avg_mshr_miss_latency::cpu1.inst 68550.391399 # average overall mshr miss latency
system.l2c.demand_avg_mshr_miss_latency::cpu1.data 52996.560876 # average overall mshr miss latency
system.l2c.demand_avg_mshr_miss_latency::cpu2.inst 73558.019510 # average overall mshr miss latency
system.l2c.demand_avg_mshr_miss_latency::cpu2.data 61314.775248 # average overall mshr miss latency
system.l2c.demand_avg_mshr_miss_latency::total 58544.707135 # average overall mshr miss latency
system.l2c.overall_avg_mshr_miss_latency::cpu1.inst 68550.391399 # average overall mshr miss latency
system.l2c.overall_avg_mshr_miss_latency::cpu1.data 52996.560876 # average overall mshr miss latency
system.l2c.overall_avg_mshr_miss_latency::cpu2.inst 73558.019510 # average overall mshr miss latency
system.l2c.overall_avg_mshr_miss_latency::cpu2.data 61314.775248 # average overall mshr miss latency
system.l2c.overall_avg_mshr_miss_latency::total 58544.707135 # average overall mshr miss latency
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.l2c.ReadReq_avg_mshr_uncacheable_latency::cpu1.data inf # average ReadReq mshr uncacheable latency
system.l2c.ReadReq_avg_mshr_uncacheable_latency::cpu2.data inf # average ReadReq mshr uncacheable latency
system.l2c.ReadReq_avg_mshr_uncacheable_latency::total inf # average ReadReq mshr uncacheable latency
system.l2c.WriteReq_avg_mshr_uncacheable_latency::cpu1.data inf # average WriteReq mshr uncacheable latency
system.l2c.WriteReq_avg_mshr_uncacheable_latency::cpu2.data inf # average WriteReq mshr uncacheable latency
system.l2c.WriteReq_avg_mshr_uncacheable_latency::total inf # average WriteReq mshr uncacheable latency
system.l2c.overall_avg_mshr_uncacheable_latency::cpu1.data inf # average overall mshr uncacheable latency
system.l2c.overall_avg_mshr_uncacheable_latency::cpu2.data inf # average overall mshr uncacheable latency
system.l2c.overall_avg_mshr_uncacheable_latency::total inf # average overall mshr uncacheable latency
system.l2c.no_allocate_misses 0 # Number of misses that were no-allocate
stats: Cumulative stats update This patch updates the stats to reflect the: 1) addition of the internal queue in SimpleMemory, 2) moving of the memory class outside FSConfig, 3) fixing up of the 2D vector printing format, 4) specifying burst size and interface width for the DRAM instead of relying on cache-line size, 5) performing merging in the DRAM controller write buffer, and 6) fixing how idle cycles are counted in the atomic and timing CPU models. The main reason for bundling them up is to minimise the changeset size.
2013-08-19 09:52:36 +02:00
system.iocache.tags.replacements 41685 # number of replacements
system.iocache.tags.tagsinuse 1.254957 # Cycle average of tags in use
system.iocache.tags.total_refs 0 # Total number of references to valid blocks.
system.iocache.tags.sampled_refs 41701 # Sample count of references to valid blocks.
system.iocache.tags.avg_refs 0 # Average number of references to valid blocks.
system.iocache.tags.warmup_cycle 1694872745000 # Cycle when the warmup percentage was hit.
system.iocache.tags.occ_blocks::tsunami.ide 1.254957 # Average occupied blocks per requestor
system.iocache.tags.occ_percent::tsunami.ide 0.078435 # Average percentage of cache occupancy
system.iocache.tags.occ_percent::total 0.078435 # Average percentage of cache occupancy
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.iocache.ReadReq_misses::tsunami.ide 173 # number of ReadReq misses
system.iocache.ReadReq_misses::total 173 # number of ReadReq misses
system.iocache.WriteReq_misses::tsunami.ide 41552 # number of WriteReq misses
system.iocache.WriteReq_misses::total 41552 # number of WriteReq misses
system.iocache.demand_misses::tsunami.ide 41725 # number of demand (read+write) misses
system.iocache.demand_misses::total 41725 # number of demand (read+write) misses
system.iocache.overall_misses::tsunami.ide 41725 # number of overall misses
system.iocache.overall_misses::total 41725 # number of overall misses
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.iocache.ReadReq_miss_latency::tsunami.ide 9629212 # number of ReadReq miss cycles
system.iocache.ReadReq_miss_latency::total 9629212 # number of ReadReq miss cycles
system.iocache.WriteReq_miss_latency::tsunami.ide 4353407559 # number of WriteReq miss cycles
system.iocache.WriteReq_miss_latency::total 4353407559 # number of WriteReq miss cycles
system.iocache.demand_miss_latency::tsunami.ide 4363036771 # number of demand (read+write) miss cycles
system.iocache.demand_miss_latency::total 4363036771 # number of demand (read+write) miss cycles
system.iocache.overall_miss_latency::tsunami.ide 4363036771 # number of overall miss cycles
system.iocache.overall_miss_latency::total 4363036771 # number of overall miss cycles
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.iocache.ReadReq_accesses::tsunami.ide 173 # number of ReadReq accesses(hits+misses)
system.iocache.ReadReq_accesses::total 173 # number of ReadReq accesses(hits+misses)
system.iocache.WriteReq_accesses::tsunami.ide 41552 # number of WriteReq accesses(hits+misses)
system.iocache.WriteReq_accesses::total 41552 # number of WriteReq accesses(hits+misses)
system.iocache.demand_accesses::tsunami.ide 41725 # number of demand (read+write) accesses
system.iocache.demand_accesses::total 41725 # number of demand (read+write) accesses
system.iocache.overall_accesses::tsunami.ide 41725 # number of overall (read+write) accesses
system.iocache.overall_accesses::total 41725 # number of overall (read+write) accesses
system.iocache.ReadReq_miss_rate::tsunami.ide 1 # miss rate for ReadReq accesses
system.iocache.ReadReq_miss_rate::total 1 # miss rate for ReadReq accesses
system.iocache.WriteReq_miss_rate::tsunami.ide 1 # miss rate for WriteReq accesses
system.iocache.WriteReq_miss_rate::total 1 # miss rate for WriteReq accesses
system.iocache.demand_miss_rate::tsunami.ide 1 # miss rate for demand accesses
system.iocache.demand_miss_rate::total 1 # miss rate for demand accesses
system.iocache.overall_miss_rate::tsunami.ide 1 # miss rate for overall accesses
system.iocache.overall_miss_rate::total 1 # miss rate for overall accesses
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.iocache.ReadReq_avg_miss_latency::tsunami.ide 55660.184971 # average ReadReq miss latency
system.iocache.ReadReq_avg_miss_latency::total 55660.184971 # average ReadReq miss latency
system.iocache.WriteReq_avg_miss_latency::tsunami.ide 104770.108755 # average WriteReq miss latency
system.iocache.WriteReq_avg_miss_latency::total 104770.108755 # average WriteReq miss latency
system.iocache.demand_avg_miss_latency::tsunami.ide 104566.489419 # average overall miss latency
system.iocache.demand_avg_miss_latency::total 104566.489419 # average overall miss latency
system.iocache.overall_avg_miss_latency::tsunami.ide 104566.489419 # average overall miss latency
system.iocache.overall_avg_miss_latency::total 104566.489419 # average overall miss latency
system.iocache.blocked_cycles::no_mshrs 114649 # number of cycles access was blocked
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.iocache.blocked_cycles::no_targets 0 # number of cycles access was blocked
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.iocache.blocked::no_mshrs 11461 # number of cycles access was blocked
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.iocache.blocked::no_targets 0 # number of cycles access was blocked
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.iocache.avg_blocked_cycles::no_mshrs 10.003403 # average number of cycles each access was blocked
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.iocache.avg_blocked_cycles::no_targets nan # average number of cycles each access was blocked
system.iocache.fast_writes 0 # number of fast writes performed
system.iocache.cache_copies 0 # number of cache copies performed
system.iocache.writebacks::writebacks 41512 # number of writebacks
system.iocache.writebacks::total 41512 # number of writebacks
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.iocache.ReadReq_mshr_misses::tsunami.ide 70 # number of ReadReq MSHR misses
system.iocache.ReadReq_mshr_misses::total 70 # number of ReadReq MSHR misses
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.iocache.WriteReq_mshr_misses::tsunami.ide 17280 # number of WriteReq MSHR misses
system.iocache.WriteReq_mshr_misses::total 17280 # number of WriteReq MSHR misses
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.iocache.demand_mshr_misses::tsunami.ide 17350 # number of demand (read+write) MSHR misses
system.iocache.demand_mshr_misses::total 17350 # number of demand (read+write) MSHR misses
system.iocache.overall_mshr_misses::tsunami.ide 17350 # number of overall MSHR misses
system.iocache.overall_mshr_misses::total 17350 # number of overall MSHR misses
system.iocache.ReadReq_mshr_miss_latency::tsunami.ide 5987712 # number of ReadReq MSHR miss cycles
system.iocache.ReadReq_mshr_miss_latency::total 5987712 # number of ReadReq MSHR miss cycles
system.iocache.WriteReq_mshr_miss_latency::tsunami.ide 3454277559 # number of WriteReq MSHR miss cycles
system.iocache.WriteReq_mshr_miss_latency::total 3454277559 # number of WriteReq MSHR miss cycles
system.iocache.demand_mshr_miss_latency::tsunami.ide 3460265271 # number of demand (read+write) MSHR miss cycles
system.iocache.demand_mshr_miss_latency::total 3460265271 # number of demand (read+write) MSHR miss cycles
system.iocache.overall_mshr_miss_latency::tsunami.ide 3460265271 # number of overall MSHR miss cycles
system.iocache.overall_mshr_miss_latency::total 3460265271 # number of overall MSHR miss cycles
system.iocache.ReadReq_mshr_miss_rate::tsunami.ide 0.404624 # mshr miss rate for ReadReq accesses
system.iocache.ReadReq_mshr_miss_rate::total 0.404624 # mshr miss rate for ReadReq accesses
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.iocache.WriteReq_mshr_miss_rate::tsunami.ide 0.415864 # mshr miss rate for WriteReq accesses
system.iocache.WriteReq_mshr_miss_rate::total 0.415864 # mshr miss rate for WriteReq accesses
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.iocache.demand_mshr_miss_rate::tsunami.ide 0.415818 # mshr miss rate for demand accesses
system.iocache.demand_mshr_miss_rate::total 0.415818 # mshr miss rate for demand accesses
system.iocache.overall_mshr_miss_rate::tsunami.ide 0.415818 # mshr miss rate for overall accesses
system.iocache.overall_mshr_miss_rate::total 0.415818 # mshr miss rate for overall accesses
system.iocache.ReadReq_avg_mshr_miss_latency::tsunami.ide 85538.742857 # average ReadReq mshr miss latency
system.iocache.ReadReq_avg_mshr_miss_latency::total 85538.742857 # average ReadReq mshr miss latency
system.iocache.WriteReq_avg_mshr_miss_latency::tsunami.ide 199900.321701 # average WriteReq mshr miss latency
system.iocache.WriteReq_avg_mshr_miss_latency::total 199900.321701 # average WriteReq mshr miss latency
system.iocache.demand_avg_mshr_miss_latency::tsunami.ide 199438.920519 # average overall mshr miss latency
system.iocache.demand_avg_mshr_miss_latency::total 199438.920519 # average overall mshr miss latency
system.iocache.overall_avg_mshr_miss_latency::tsunami.ide 199438.920519 # average overall mshr miss latency
system.iocache.overall_avg_mshr_miss_latency::total 199438.920519 # average overall mshr miss latency
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.iocache.no_allocate_misses 0 # Number of misses that were no-allocate
system.disk0.dma_read_full_pages 0 # Number of full page size DMA reads (not PRD).
system.disk0.dma_read_bytes 1024 # Number of bytes transfered via DMA reads (not PRD).
system.disk0.dma_read_txs 1 # Number of DMA read transactions (not PRD).
system.disk0.dma_write_full_pages 298 # Number of full page size DMA writes.
system.disk0.dma_write_bytes 2651136 # Number of bytes transfered via DMA writes.
system.disk0.dma_write_txs 395 # Number of DMA write transactions.
system.disk2.dma_read_full_pages 0 # Number of full page size DMA reads (not PRD).
system.disk2.dma_read_bytes 0 # Number of bytes transfered via DMA reads (not PRD).
system.disk2.dma_read_txs 0 # Number of DMA read transactions (not PRD).
system.disk2.dma_write_full_pages 1 # Number of full page size DMA writes.
system.disk2.dma_write_bytes 8192 # Number of bytes transfered via DMA writes.
system.disk2.dma_write_txs 1 # Number of DMA write transactions.
system.cpu0.dtb.fetch_hits 0 # ITB hits
system.cpu0.dtb.fetch_misses 0 # ITB misses
system.cpu0.dtb.fetch_acv 0 # ITB acv
system.cpu0.dtb.fetch_accesses 0 # ITB accesses
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.dtb.read_hits 4909978 # DTB read hits
system.cpu0.dtb.read_misses 6100 # DTB read misses
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.cpu0.dtb.read_acv 126 # DTB read access violations
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.dtb.read_accesses 428319 # DTB read accesses
system.cpu0.dtb.write_hits 3504299 # DTB write hits
system.cpu0.dtb.write_misses 671 # DTB write misses
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.cpu0.dtb.write_acv 84 # DTB write access violations
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.dtb.write_accesses 163761 # DTB write accesses
system.cpu0.dtb.data_hits 8414277 # DTB hits
system.cpu0.dtb.data_misses 6771 # DTB misses
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.cpu0.dtb.data_acv 210 # DTB access violations
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.dtb.data_accesses 592080 # DTB accesses
system.cpu0.itb.fetch_hits 2758234 # ITB hits
system.cpu0.itb.fetch_misses 3034 # ITB misses
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.cpu0.itb.fetch_acv 104 # ITB acv
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.itb.fetch_accesses 2761268 # ITB accesses
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu0.itb.read_hits 0 # DTB read hits
system.cpu0.itb.read_misses 0 # DTB read misses
system.cpu0.itb.read_acv 0 # DTB read access violations
system.cpu0.itb.read_accesses 0 # DTB read accesses
system.cpu0.itb.write_hits 0 # DTB write hits
system.cpu0.itb.write_misses 0 # DTB write misses
system.cpu0.itb.write_acv 0 # DTB write access violations
system.cpu0.itb.write_accesses 0 # DTB write accesses
system.cpu0.itb.data_hits 0 # DTB hits
system.cpu0.itb.data_misses 0 # DTB misses
system.cpu0.itb.data_acv 0 # DTB access violations
system.cpu0.itb.data_accesses 0 # DTB accesses
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.numCycles 928316891 # number of cpu cycles simulated
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu0.numWorkItemsStarted 0 # number of work items this cpu started
system.cpu0.numWorkItemsCompleted 0 # number of work items this cpu completed
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.committedInsts 33736461 # Number of instructions committed
system.cpu0.committedOps 33736461 # Number of ops (including micro ops) committed
system.cpu0.num_int_alu_accesses 31599588 # Number of integer alu accesses
system.cpu0.num_fp_alu_accesses 169686 # Number of float alu accesses
system.cpu0.num_func_calls 810809 # number of times a function call or return occured
system.cpu0.num_conditional_control_insts 4665593 # number of instructions that are conditional controls
system.cpu0.num_int_insts 31599588 # number of integer instructions
system.cpu0.num_fp_insts 169686 # number of float instructions
system.cpu0.num_int_register_reads 44374544 # number of times the integer registers were read
system.cpu0.num_int_register_writes 23060255 # number of times the integer registers were written
system.cpu0.num_fp_register_reads 87629 # number of times the floating registers were read
system.cpu0.num_fp_register_writes 89168 # number of times the floating registers were written
system.cpu0.num_mem_refs 8444409 # number of memory refs
system.cpu0.num_load_insts 4931349 # Number of load instructions
system.cpu0.num_store_insts 3513060 # Number of store instructions
stats: Cumulative stats update This patch updates the stats to reflect the: 1) addition of the internal queue in SimpleMemory, 2) moving of the memory class outside FSConfig, 3) fixing up of the 2D vector printing format, 4) specifying burst size and interface width for the DRAM instead of relying on cache-line size, 5) performing merging in the DRAM controller write buffer, and 6) fixing how idle cycles are counted in the atomic and timing CPU models. The main reason for bundling them up is to minimise the changeset size.
2013-08-19 09:52:36 +02:00
system.cpu0.num_idle_cycles 903633014.989213 # Number of idle cycles
system.cpu0.num_busy_cycles 24683876.010787 # Number of busy cycles
system.cpu0.not_idle_fraction 0.026590 # Percentage of non-idle cycles
system.cpu0.idle_fraction 0.973410 # Percentage of idle cycles
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu0.kern.inst.arm 0 # number of arm instructions executed
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.kern.inst.quiesce 6419 # number of quiesce instructions executed
system.cpu0.kern.inst.hwrei 211396 # number of hwrei instructions executed
system.cpu0.kern.ipl_count::0 74806 40.97% 40.97% # number of times we switched to this ipl
stats: Update stats for regressions using SimpleDDR3 This patch updates the regression stats to reflect that they are using the SimpleDDR3 controller by default.
2013-01-31 13:49:16 +01:00
system.cpu0.kern.ipl_count::21 203 0.11% 41.08% # number of times we switched to this ipl
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.cpu0.kern.ipl_count::22 1879 1.03% 42.11% # number of times we switched to this ipl
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.kern.ipl_count::31 105698 57.89% 100.00% # number of times we switched to this ipl
system.cpu0.kern.ipl_count::total 182586 # number of times we switched to this ipl
system.cpu0.kern.ipl_good::0 73439 49.30% 49.30% # number of times we switched to this ipl from a different ipl
stats: Update stats for regressions using SimpleDDR3 This patch updates the regression stats to reflect that they are using the SimpleDDR3 controller by default.
2013-01-31 13:49:16 +01:00
system.cpu0.kern.ipl_good::21 203 0.14% 49.44% # number of times we switched to this ipl from a different ipl
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.cpu0.kern.ipl_good::22 1879 1.26% 50.70% # number of times we switched to this ipl from a different ipl
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.kern.ipl_good::31 73439 49.30% 100.00% # number of times we switched to this ipl from a different ipl
system.cpu0.kern.ipl_good::total 148960 # number of times we switched to this ipl from a different ipl
system.cpu0.kern.ipl_ticks::0 1819515680500 98.74% 98.74% # number of cycles we spent at this ipl
system.cpu0.kern.ipl_ticks::21 39349500 0.00% 98.74% # number of cycles we spent at this ipl
system.cpu0.kern.ipl_ticks::22 365678500 0.02% 98.76% # number of cycles we spent at this ipl
system.cpu0.kern.ipl_ticks::31 22783774000 1.24% 100.00% # number of cycles we spent at this ipl
system.cpu0.kern.ipl_ticks::total 1842704482500 # number of cycles we spent at this ipl
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.cpu0.kern.ipl_used::0 0.981726 # fraction of swpipl calls that actually changed the ipl
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu0.kern.ipl_used::21 1 # fraction of swpipl calls that actually changed the ipl
system.cpu0.kern.ipl_used::22 1 # fraction of swpipl calls that actually changed the ipl
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.kern.ipl_used::31 0.694800 # fraction of swpipl calls that actually changed the ipl
system.cpu0.kern.ipl_used::total 0.815835 # fraction of swpipl calls that actually changed the ipl
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu0.kern.syscall::2 8 2.45% 2.45% # number of syscalls executed
system.cpu0.kern.syscall::3 30 9.20% 11.66% # number of syscalls executed
system.cpu0.kern.syscall::4 4 1.23% 12.88% # number of syscalls executed
system.cpu0.kern.syscall::6 42 12.88% 25.77% # number of syscalls executed
system.cpu0.kern.syscall::12 1 0.31% 26.07% # number of syscalls executed
system.cpu0.kern.syscall::15 1 0.31% 26.38% # number of syscalls executed
system.cpu0.kern.syscall::17 15 4.60% 30.98% # number of syscalls executed
system.cpu0.kern.syscall::19 10 3.07% 34.05% # number of syscalls executed
system.cpu0.kern.syscall::20 6 1.84% 35.89% # number of syscalls executed
system.cpu0.kern.syscall::23 4 1.23% 37.12% # number of syscalls executed
system.cpu0.kern.syscall::24 6 1.84% 38.96% # number of syscalls executed
system.cpu0.kern.syscall::33 11 3.37% 42.33% # number of syscalls executed
system.cpu0.kern.syscall::41 2 0.61% 42.94% # number of syscalls executed
system.cpu0.kern.syscall::45 54 16.56% 59.51% # number of syscalls executed
system.cpu0.kern.syscall::47 6 1.84% 61.35% # number of syscalls executed
system.cpu0.kern.syscall::48 10 3.07% 64.42% # number of syscalls executed
system.cpu0.kern.syscall::54 10 3.07% 67.48% # number of syscalls executed
system.cpu0.kern.syscall::58 1 0.31% 67.79% # number of syscalls executed
system.cpu0.kern.syscall::59 7 2.15% 69.94% # number of syscalls executed
system.cpu0.kern.syscall::71 54 16.56% 86.50% # number of syscalls executed
system.cpu0.kern.syscall::73 3 0.92% 87.42% # number of syscalls executed
system.cpu0.kern.syscall::74 16 4.91% 92.33% # number of syscalls executed
system.cpu0.kern.syscall::87 1 0.31% 92.64% # number of syscalls executed
system.cpu0.kern.syscall::90 3 0.92% 93.56% # number of syscalls executed
system.cpu0.kern.syscall::92 9 2.76% 96.32% # number of syscalls executed
system.cpu0.kern.syscall::97 2 0.61% 96.93% # number of syscalls executed
system.cpu0.kern.syscall::98 2 0.61% 97.55% # number of syscalls executed
system.cpu0.kern.syscall::132 4 1.23% 98.77% # number of syscalls executed
system.cpu0.kern.syscall::144 2 0.61% 99.39% # number of syscalls executed
system.cpu0.kern.syscall::147 2 0.61% 100.00% # number of syscalls executed
system.cpu0.kern.syscall::total 326 # number of syscalls executed
system.cpu0.kern.callpal::cserve 1 0.00% 0.00% # number of callpals executed
system.cpu0.kern.callpal::wrmces 1 0.00% 0.00% # number of callpals executed
system.cpu0.kern.callpal::wrfen 1 0.00% 0.00% # number of callpals executed
system.cpu0.kern.callpal::wrvptptr 1 0.00% 0.00% # number of callpals executed
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.kern.callpal::swpctx 4176 2.17% 2.17% # number of callpals executed
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu0.kern.callpal::tbi 54 0.03% 2.20% # number of callpals executed
system.cpu0.kern.callpal::wrent 7 0.00% 2.21% # number of callpals executed
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.kern.callpal::swpipl 175327 91.20% 93.41% # number of callpals executed
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.cpu0.kern.callpal::rdps 6783 3.53% 96.94% # number of callpals executed
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu0.kern.callpal::wrkgp 1 0.00% 96.94% # number of callpals executed
system.cpu0.kern.callpal::wrusp 7 0.00% 96.94% # number of callpals executed
system.cpu0.kern.callpal::rdusp 9 0.00% 96.94% # number of callpals executed
system.cpu0.kern.callpal::whami 2 0.00% 96.95% # number of callpals executed
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.cpu0.kern.callpal::rti 5176 2.69% 99.64% # number of callpals executed
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu0.kern.callpal::callsys 515 0.27% 99.91% # number of callpals executed
system.cpu0.kern.callpal::imb 181 0.09% 100.00% # number of callpals executed
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.kern.callpal::total 192242 # number of callpals executed
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.cpu0.kern.mode_switch::kernel 5923 # number of protection mode switches
stats: Update stats for O3 switching fix.
2013-04-22 19:20:33 +02:00
system.cpu0.kern.mode_switch::user 1738 # number of protection mode switches
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.kern.mode_switch::idle 2095 # number of protection mode switches
system.cpu0.kern.mode_good::kernel 1908
stats: Update stats for O3 switching fix.
2013-04-22 19:20:33 +02:00
system.cpu0.kern.mode_good::user 1738
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.kern.mode_good::idle 170
system.cpu0.kern.mode_switch_good::kernel 0.322134 # fraction of useful protection mode switches
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu0.kern.mode_switch_good::user 1 # fraction of useful protection mode switches
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.kern.mode_switch_good::idle 0.081146 # fraction of useful protection mode switches
system.cpu0.kern.mode_switch_good::total 0.391144 # fraction of useful protection mode switches
system.cpu0.kern.mode_ticks::kernel 29786026000 1.62% 1.62% # number of ticks spent at the given mode
system.cpu0.kern.mode_ticks::user 2614250500 0.14% 1.76% # number of ticks spent at the given mode
system.cpu0.kern.mode_ticks::idle 1810304201500 98.24% 100.00% # number of ticks spent at the given mode
system.cpu0.kern.swap_context 4177 # number of times the context was actually changed
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.tsunami.ethernet.descDMAReads 0 # Number of descriptors the device read w/ DMA
system.tsunami.ethernet.descDMAWrites 0 # Number of descriptors the device wrote w/ DMA
system.tsunami.ethernet.descDmaReadBytes 0 # number of descriptor bytes read w/ DMA
system.tsunami.ethernet.descDmaWriteBytes 0 # number of descriptor bytes write w/ DMA
system.tsunami.ethernet.postedSwi 0 # number of software interrupts posted to CPU
system.tsunami.ethernet.coalescedSwi nan # average number of Swi's coalesced into each post
system.tsunami.ethernet.totalSwi 0 # total number of Swi written to ISR
system.tsunami.ethernet.postedRxIdle 0 # number of rxIdle interrupts posted to CPU
system.tsunami.ethernet.coalescedRxIdle nan # average number of RxIdle's coalesced into each post
system.tsunami.ethernet.totalRxIdle 0 # total number of RxIdle written to ISR
system.tsunami.ethernet.postedRxOk 0 # number of RxOk interrupts posted to CPU
system.tsunami.ethernet.coalescedRxOk nan # average number of RxOk's coalesced into each post
system.tsunami.ethernet.totalRxOk 0 # total number of RxOk written to ISR
system.tsunami.ethernet.postedRxDesc 0 # number of RxDesc interrupts posted to CPU
system.tsunami.ethernet.coalescedRxDesc nan # average number of RxDesc's coalesced into each post
system.tsunami.ethernet.totalRxDesc 0 # total number of RxDesc written to ISR
system.tsunami.ethernet.postedTxOk 0 # number of TxOk interrupts posted to CPU
system.tsunami.ethernet.coalescedTxOk nan # average number of TxOk's coalesced into each post
system.tsunami.ethernet.totalTxOk 0 # total number of TxOk written to ISR
system.tsunami.ethernet.postedTxIdle 0 # number of TxIdle interrupts posted to CPU
system.tsunami.ethernet.coalescedTxIdle nan # average number of TxIdle's coalesced into each post
system.tsunami.ethernet.totalTxIdle 0 # total number of TxIdle written to ISR
system.tsunami.ethernet.postedTxDesc 0 # number of TxDesc interrupts posted to CPU
system.tsunami.ethernet.coalescedTxDesc nan # average number of TxDesc's coalesced into each post
system.tsunami.ethernet.totalTxDesc 0 # total number of TxDesc written to ISR
system.tsunami.ethernet.postedRxOrn 0 # number of RxOrn posted to CPU
system.tsunami.ethernet.coalescedRxOrn nan # average number of RxOrn's coalesced into each post
system.tsunami.ethernet.totalRxOrn 0 # total number of RxOrn written to ISR
system.tsunami.ethernet.coalescedTotal nan # average number of interrupts coalesced into each post
system.tsunami.ethernet.postedInterrupts 0 # number of posts to CPU
system.tsunami.ethernet.droppedPackets 0 # number of packets dropped
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.toL2Bus.throughput 110422039 # Throughput (bytes/s)
system.toL2Bus.trans_dist::ReadReq 786602 # Transaction distribution
system.toL2Bus.trans_dist::ReadResp 786555 # Transaction distribution
system.toL2Bus.trans_dist::WriteReq 3756 # Transaction distribution
system.toL2Bus.trans_dist::WriteResp 3756 # Transaction distribution
system.toL2Bus.trans_dist::Writeback 371447 # Transaction distribution
system.toL2Bus.trans_dist::UpgradeReq 15 # Transaction distribution
system.toL2Bus.trans_dist::SCUpgradeReq 2 # Transaction distribution
system.toL2Bus.trans_dist::UpgradeResp 17 # Transaction distribution
system.toL2Bus.trans_dist::ReadExReq 151061 # Transaction distribution
system.toL2Bus.trans_dist::ReadExResp 133781 # Transaction distribution
system.toL2Bus.trans_dist::BadAddressError 32 # Transaction distribution
stats: Cumulative stats update This patch updates the stats to reflect the: 1) addition of the internal queue in SimpleMemory, 2) moving of the memory class outside FSConfig, 3) fixing up of the 2D vector printing format, 4) specifying burst size and interface width for the DRAM instead of relying on cache-line size, 5) performing merging in the DRAM controller write buffer, and 6) fixing how idle cycles are counted in the atomic and timing CPU models. The main reason for bundling them up is to minimise the changeset size.
2013-08-19 09:52:36 +02:00
system.toL2Bus.pkt_count_system.cpu0.icache.mem_side::system.l2c.cpu_side 849315 # Packet count per connected master and slave (bytes)
system.toL2Bus.pkt_count_system.cpu0.dcache.mem_side::system.l2c.cpu_side 1370344 # Packet count per connected master and slave (bytes)
system.toL2Bus.pkt_count::total 2219659 # Packet count per connected master and slave (bytes)
system.toL2Bus.tot_pkt_size_system.cpu0.icache.mem_side::system.l2c.cpu_side 27177600 # Cumulative packet size per connected master and slave (bytes)
system.toL2Bus.tot_pkt_size_system.cpu0.dcache.mem_side::system.l2c.cpu_side 55325386 # Cumulative packet size per connected master and slave (bytes)
system.toL2Bus.tot_pkt_size::total 82502986 # Cumulative packet size per connected master and slave (bytes)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.toL2Bus.data_through_bus 203464200 # Total data (bytes)
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.toL2Bus.snoop_data_through_bus 11072 # Total snoop data (bytes)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.toL2Bus.reqLayer0.occupancy 2135432500 # Layer occupancy (ticks)
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.toL2Bus.reqLayer0.utilization 0.1 # Layer utilization (%)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.toL2Bus.snoopLayer0.occupancy 247500 # Layer occupancy (ticks)
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.toL2Bus.snoopLayer0.utilization 0.0 # Layer utilization (%)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.toL2Bus.respLayer0.occupancy 1913139810 # Layer occupancy (ticks)
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.toL2Bus.respLayer0.utilization 0.1 # Layer utilization (%)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.toL2Bus.respLayer1.occupancy 2237602233 # Layer occupancy (ticks)
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.toL2Bus.respLayer1.utilization 0.1 # Layer utilization (%)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.iobus.throughput 1469136 # Throughput (bytes/s)
system.iobus.trans_dist::ReadReq 2975 # Transaction distribution
system.iobus.trans_dist::ReadResp 2975 # Transaction distribution
system.iobus.trans_dist::WriteReq 21036 # Transaction distribution
system.iobus.trans_dist::WriteResp 21036 # Transaction distribution
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.iobus.pkt_count_system.bridge.master::system.tsunami.cchip.pio 2342 # Packet count per connected master and slave (bytes)
system.iobus.pkt_count_system.bridge.master::system.tsunami.pchip.pio 140 # Packet count per connected master and slave (bytes)
system.iobus.pkt_count_system.bridge.master::system.tsunami.io.pio 66 # Packet count per connected master and slave (bytes)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.iobus.pkt_count_system.bridge.master::system.tsunami.uart.pio 8320 # Packet count per connected master and slave (bytes)
system.iobus.pkt_count_system.bridge.master::system.tsunami.ide.pio 2420 # Packet count per connected master and slave (bytes)
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.iobus.pkt_count_system.bridge.master::system.tsunami.ide-pciconf 34 # Packet count per connected master and slave (bytes)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.iobus.pkt_count_system.bridge.master::total 13322 # Packet count per connected master and slave (bytes)
system.iobus.pkt_count_system.tsunami.ide.dma::system.iocache.cpu_side 34700 # Packet count per connected master and slave (bytes)
system.iobus.pkt_count_system.tsunami.ide.dma::total 34700 # Packet count per connected master and slave (bytes)
system.iobus.pkt_count::total 48022 # Packet count per connected master and slave (bytes)
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.iobus.tot_pkt_size_system.bridge.master::system.tsunami.cchip.pio 9368 # Cumulative packet size per connected master and slave (bytes)
system.iobus.tot_pkt_size_system.bridge.master::system.tsunami.pchip.pio 560 # Cumulative packet size per connected master and slave (bytes)
system.iobus.tot_pkt_size_system.bridge.master::system.tsunami.io.pio 61 # Cumulative packet size per connected master and slave (bytes)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.iobus.tot_pkt_size_system.bridge.master::system.tsunami.uart.pio 4160 # Cumulative packet size per connected master and slave (bytes)
system.iobus.tot_pkt_size_system.bridge.master::system.tsunami.ide.pio 1574 # Cumulative packet size per connected master and slave (bytes)
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.iobus.tot_pkt_size_system.bridge.master::system.tsunami.ide-pciconf 31 # Cumulative packet size per connected master and slave (bytes)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.iobus.tot_pkt_size_system.bridge.master::total 15754 # Cumulative packet size per connected master and slave (bytes)
system.iobus.tot_pkt_size_system.tsunami.ide.dma::system.iocache.cpu_side 1107376 # Cumulative packet size per connected master and slave (bytes)
system.iobus.tot_pkt_size_system.tsunami.ide.dma::total 1107376 # Cumulative packet size per connected master and slave (bytes)
system.iobus.tot_pkt_size::total 1123130 # Cumulative packet size per connected master and slave (bytes)
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.iobus.data_through_bus 2707184 # Total data (bytes)
system.iobus.reqLayer0.occupancy 2208000 # Layer occupancy (ticks)
system.iobus.reqLayer0.utilization 0.0 # Layer utilization (%)
system.iobus.reqLayer1.occupancy 105000 # Layer occupancy (ticks)
system.iobus.reqLayer1.utilization 0.0 # Layer utilization (%)
system.iobus.reqLayer22.occupancy 57000 # Layer occupancy (ticks)
system.iobus.reqLayer22.utilization 0.0 # Layer utilization (%)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.iobus.reqLayer23.occupancy 6200000 # Layer occupancy (ticks)
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.iobus.reqLayer23.utilization 0.0 # Layer utilization (%)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.iobus.reqLayer25.occupancy 1827000 # Layer occupancy (ticks)
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.iobus.reqLayer25.utilization 0.0 # Layer utilization (%)
system.iobus.reqLayer26.occupancy 20000 # Layer occupancy (ticks)
system.iobus.reqLayer26.utilization 0.0 # Layer utilization (%)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.iobus.reqLayer29.occupancy 157303021 # Layer occupancy (ticks)
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.iobus.reqLayer29.utilization 0.0 # Layer utilization (%)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.iobus.respLayer0.occupancy 9566000 # Layer occupancy (ticks)
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.iobus.respLayer0.utilization 0.0 # Layer utilization (%)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.iobus.respLayer1.occupancy 17990250 # Layer occupancy (ticks)
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.iobus.respLayer1.utilization 0.0 # Layer utilization (%)
stats: Cumulative stats update This patch updates the stats to reflect the: 1) addition of the internal queue in SimpleMemory, 2) moving of the memory class outside FSConfig, 3) fixing up of the 2D vector printing format, 4) specifying burst size and interface width for the DRAM instead of relying on cache-line size, 5) performing merging in the DRAM controller write buffer, and 6) fixing how idle cycles are counted in the atomic and timing CPU models. The main reason for bundling them up is to minimise the changeset size.
2013-08-19 09:52:36 +02:00
system.cpu0.icache.tags.replacements 950451 # number of replacements
system.cpu0.icache.tags.tagsinuse 511.192015 # Cycle average of tags in use
system.cpu0.icache.tags.total_refs 43221003 # Total number of references to valid blocks.
system.cpu0.icache.tags.sampled_refs 950962 # Sample count of references to valid blocks.
system.cpu0.icache.tags.avg_refs 45.449769 # Average number of references to valid blocks.
system.cpu0.icache.tags.warmup_cycle 10381115250 # Cycle when the warmup percentage was hit.
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.icache.tags.occ_blocks::cpu0.inst 246.999230 # Average occupied blocks per requestor
system.cpu0.icache.tags.occ_blocks::cpu1.inst 99.674980 # Average occupied blocks per requestor
system.cpu0.icache.tags.occ_blocks::cpu2.inst 164.517805 # Average occupied blocks per requestor
system.cpu0.icache.tags.occ_percent::cpu0.inst 0.482420 # Average percentage of cache occupancy
system.cpu0.icache.tags.occ_percent::cpu1.inst 0.194678 # Average percentage of cache occupancy
system.cpu0.icache.tags.occ_percent::cpu2.inst 0.321324 # Average percentage of cache occupancy
stats: Cumulative stats update This patch updates the stats to reflect the: 1) addition of the internal queue in SimpleMemory, 2) moving of the memory class outside FSConfig, 3) fixing up of the 2D vector printing format, 4) specifying burst size and interface width for the DRAM instead of relying on cache-line size, 5) performing merging in the DRAM controller write buffer, and 6) fixing how idle cycles are counted in the atomic and timing CPU models. The main reason for bundling them up is to minimise the changeset size.
2013-08-19 09:52:36 +02:00
system.cpu0.icache.tags.occ_percent::total 0.998422 # Average percentage of cache occupancy
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.icache.ReadReq_hits::cpu0.inst 33216972 # number of ReadReq hits
system.cpu0.icache.ReadReq_hits::cpu1.inst 7763860 # number of ReadReq hits
system.cpu0.icache.ReadReq_hits::cpu2.inst 2240171 # number of ReadReq hits
system.cpu0.icache.ReadReq_hits::total 43221003 # number of ReadReq hits
system.cpu0.icache.demand_hits::cpu0.inst 33216972 # number of demand (read+write) hits
system.cpu0.icache.demand_hits::cpu1.inst 7763860 # number of demand (read+write) hits
system.cpu0.icache.demand_hits::cpu2.inst 2240171 # number of demand (read+write) hits
system.cpu0.icache.demand_hits::total 43221003 # number of demand (read+write) hits
system.cpu0.icache.overall_hits::cpu0.inst 33216972 # number of overall hits
system.cpu0.icache.overall_hits::cpu1.inst 7763860 # number of overall hits
system.cpu0.icache.overall_hits::cpu2.inst 2240171 # number of overall hits
system.cpu0.icache.overall_hits::total 43221003 # number of overall hits
system.cpu0.icache.ReadReq_misses::cpu0.inst 526470 # number of ReadReq misses
system.cpu0.icache.ReadReq_misses::cpu1.inst 126995 # number of ReadReq misses
system.cpu0.icache.ReadReq_misses::cpu2.inst 313704 # number of ReadReq misses
system.cpu0.icache.ReadReq_misses::total 967169 # number of ReadReq misses
system.cpu0.icache.demand_misses::cpu0.inst 526470 # number of demand (read+write) misses
system.cpu0.icache.demand_misses::cpu1.inst 126995 # number of demand (read+write) misses
system.cpu0.icache.demand_misses::cpu2.inst 313704 # number of demand (read+write) misses
system.cpu0.icache.demand_misses::total 967169 # number of demand (read+write) misses
system.cpu0.icache.overall_misses::cpu0.inst 526470 # number of overall misses
system.cpu0.icache.overall_misses::cpu1.inst 126995 # number of overall misses
system.cpu0.icache.overall_misses::cpu2.inst 313704 # number of overall misses
system.cpu0.icache.overall_misses::total 967169 # number of overall misses
system.cpu0.icache.ReadReq_miss_latency::cpu1.inst 1823123501 # number of ReadReq miss cycles
system.cpu0.icache.ReadReq_miss_latency::cpu2.inst 4463889198 # number of ReadReq miss cycles
system.cpu0.icache.ReadReq_miss_latency::total 6287012699 # number of ReadReq miss cycles
system.cpu0.icache.demand_miss_latency::cpu1.inst 1823123501 # number of demand (read+write) miss cycles
system.cpu0.icache.demand_miss_latency::cpu2.inst 4463889198 # number of demand (read+write) miss cycles
system.cpu0.icache.demand_miss_latency::total 6287012699 # number of demand (read+write) miss cycles
system.cpu0.icache.overall_miss_latency::cpu1.inst 1823123501 # number of overall miss cycles
system.cpu0.icache.overall_miss_latency::cpu2.inst 4463889198 # number of overall miss cycles
system.cpu0.icache.overall_miss_latency::total 6287012699 # number of overall miss cycles
system.cpu0.icache.ReadReq_accesses::cpu0.inst 33743442 # number of ReadReq accesses(hits+misses)
system.cpu0.icache.ReadReq_accesses::cpu1.inst 7890855 # number of ReadReq accesses(hits+misses)
system.cpu0.icache.ReadReq_accesses::cpu2.inst 2553875 # number of ReadReq accesses(hits+misses)
system.cpu0.icache.ReadReq_accesses::total 44188172 # number of ReadReq accesses(hits+misses)
system.cpu0.icache.demand_accesses::cpu0.inst 33743442 # number of demand (read+write) accesses
system.cpu0.icache.demand_accesses::cpu1.inst 7890855 # number of demand (read+write) accesses
system.cpu0.icache.demand_accesses::cpu2.inst 2553875 # number of demand (read+write) accesses
system.cpu0.icache.demand_accesses::total 44188172 # number of demand (read+write) accesses
system.cpu0.icache.overall_accesses::cpu0.inst 33743442 # number of overall (read+write) accesses
system.cpu0.icache.overall_accesses::cpu1.inst 7890855 # number of overall (read+write) accesses
system.cpu0.icache.overall_accesses::cpu2.inst 2553875 # number of overall (read+write) accesses
system.cpu0.icache.overall_accesses::total 44188172 # number of overall (read+write) accesses
system.cpu0.icache.ReadReq_miss_rate::cpu0.inst 0.015602 # miss rate for ReadReq accesses
system.cpu0.icache.ReadReq_miss_rate::cpu1.inst 0.016094 # miss rate for ReadReq accesses
system.cpu0.icache.ReadReq_miss_rate::cpu2.inst 0.122835 # miss rate for ReadReq accesses
system.cpu0.icache.ReadReq_miss_rate::total 0.021888 # miss rate for ReadReq accesses
system.cpu0.icache.demand_miss_rate::cpu0.inst 0.015602 # miss rate for demand accesses
system.cpu0.icache.demand_miss_rate::cpu1.inst 0.016094 # miss rate for demand accesses
system.cpu0.icache.demand_miss_rate::cpu2.inst 0.122835 # miss rate for demand accesses
system.cpu0.icache.demand_miss_rate::total 0.021888 # miss rate for demand accesses
system.cpu0.icache.overall_miss_rate::cpu0.inst 0.015602 # miss rate for overall accesses
system.cpu0.icache.overall_miss_rate::cpu1.inst 0.016094 # miss rate for overall accesses
system.cpu0.icache.overall_miss_rate::cpu2.inst 0.122835 # miss rate for overall accesses
system.cpu0.icache.overall_miss_rate::total 0.021888 # miss rate for overall accesses
system.cpu0.icache.ReadReq_avg_miss_latency::cpu1.inst 14355.868349 # average ReadReq miss latency
system.cpu0.icache.ReadReq_avg_miss_latency::cpu2.inst 14229.621548 # average ReadReq miss latency
system.cpu0.icache.ReadReq_avg_miss_latency::total 6500.428259 # average ReadReq miss latency
system.cpu0.icache.demand_avg_miss_latency::cpu1.inst 14355.868349 # average overall miss latency
system.cpu0.icache.demand_avg_miss_latency::cpu2.inst 14229.621548 # average overall miss latency
system.cpu0.icache.demand_avg_miss_latency::total 6500.428259 # average overall miss latency
system.cpu0.icache.overall_avg_miss_latency::cpu1.inst 14355.868349 # average overall miss latency
system.cpu0.icache.overall_avg_miss_latency::cpu2.inst 14229.621548 # average overall miss latency
system.cpu0.icache.overall_avg_miss_latency::total 6500.428259 # average overall miss latency
system.cpu0.icache.blocked_cycles::no_mshrs 4785 # number of cycles access was blocked
system.cpu0.icache.blocked_cycles::no_targets 0 # number of cycles access was blocked
system.cpu0.icache.blocked::no_mshrs 202 # number of cycles access was blocked
system.cpu0.icache.blocked::no_targets 0 # number of cycles access was blocked
system.cpu0.icache.avg_blocked_cycles::no_mshrs 23.688119 # average number of cycles each access was blocked
system.cpu0.icache.avg_blocked_cycles::no_targets nan # average number of cycles each access was blocked
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu0.icache.fast_writes 0 # number of fast writes performed
system.cpu0.icache.cache_copies 0 # number of cache copies performed
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.icache.ReadReq_mshr_hits::cpu2.inst 16034 # number of ReadReq MSHR hits
system.cpu0.icache.ReadReq_mshr_hits::total 16034 # number of ReadReq MSHR hits
system.cpu0.icache.demand_mshr_hits::cpu2.inst 16034 # number of demand (read+write) MSHR hits
system.cpu0.icache.demand_mshr_hits::total 16034 # number of demand (read+write) MSHR hits
system.cpu0.icache.overall_mshr_hits::cpu2.inst 16034 # number of overall MSHR hits
system.cpu0.icache.overall_mshr_hits::total 16034 # number of overall MSHR hits
system.cpu0.icache.ReadReq_mshr_misses::cpu1.inst 126995 # number of ReadReq MSHR misses
system.cpu0.icache.ReadReq_mshr_misses::cpu2.inst 297670 # number of ReadReq MSHR misses
system.cpu0.icache.ReadReq_mshr_misses::total 424665 # number of ReadReq MSHR misses
system.cpu0.icache.demand_mshr_misses::cpu1.inst 126995 # number of demand (read+write) MSHR misses
system.cpu0.icache.demand_mshr_misses::cpu2.inst 297670 # number of demand (read+write) MSHR misses
system.cpu0.icache.demand_mshr_misses::total 424665 # number of demand (read+write) MSHR misses
system.cpu0.icache.overall_mshr_misses::cpu1.inst 126995 # number of overall MSHR misses
system.cpu0.icache.overall_mshr_misses::cpu2.inst 297670 # number of overall MSHR misses
system.cpu0.icache.overall_mshr_misses::total 424665 # number of overall MSHR misses
system.cpu0.icache.ReadReq_mshr_miss_latency::cpu1.inst 1568013499 # number of ReadReq MSHR miss cycles
system.cpu0.icache.ReadReq_mshr_miss_latency::cpu2.inst 3675865433 # number of ReadReq MSHR miss cycles
system.cpu0.icache.ReadReq_mshr_miss_latency::total 5243878932 # number of ReadReq MSHR miss cycles
system.cpu0.icache.demand_mshr_miss_latency::cpu1.inst 1568013499 # number of demand (read+write) MSHR miss cycles
system.cpu0.icache.demand_mshr_miss_latency::cpu2.inst 3675865433 # number of demand (read+write) MSHR miss cycles
system.cpu0.icache.demand_mshr_miss_latency::total 5243878932 # number of demand (read+write) MSHR miss cycles
system.cpu0.icache.overall_mshr_miss_latency::cpu1.inst 1568013499 # number of overall MSHR miss cycles
system.cpu0.icache.overall_mshr_miss_latency::cpu2.inst 3675865433 # number of overall MSHR miss cycles
system.cpu0.icache.overall_mshr_miss_latency::total 5243878932 # number of overall MSHR miss cycles
system.cpu0.icache.ReadReq_mshr_miss_rate::cpu1.inst 0.016094 # mshr miss rate for ReadReq accesses
system.cpu0.icache.ReadReq_mshr_miss_rate::cpu2.inst 0.116556 # mshr miss rate for ReadReq accesses
system.cpu0.icache.ReadReq_mshr_miss_rate::total 0.009610 # mshr miss rate for ReadReq accesses
system.cpu0.icache.demand_mshr_miss_rate::cpu1.inst 0.016094 # mshr miss rate for demand accesses
system.cpu0.icache.demand_mshr_miss_rate::cpu2.inst 0.116556 # mshr miss rate for demand accesses
system.cpu0.icache.demand_mshr_miss_rate::total 0.009610 # mshr miss rate for demand accesses
system.cpu0.icache.overall_mshr_miss_rate::cpu1.inst 0.016094 # mshr miss rate for overall accesses
system.cpu0.icache.overall_mshr_miss_rate::cpu2.inst 0.116556 # mshr miss rate for overall accesses
system.cpu0.icache.overall_mshr_miss_rate::total 0.009610 # mshr miss rate for overall accesses
system.cpu0.icache.ReadReq_avg_mshr_miss_latency::cpu1.inst 12347.049089 # average ReadReq mshr miss latency
system.cpu0.icache.ReadReq_avg_mshr_miss_latency::cpu2.inst 12348.793741 # average ReadReq mshr miss latency
system.cpu0.icache.ReadReq_avg_mshr_miss_latency::total 12348.272007 # average ReadReq mshr miss latency
system.cpu0.icache.demand_avg_mshr_miss_latency::cpu1.inst 12347.049089 # average overall mshr miss latency
system.cpu0.icache.demand_avg_mshr_miss_latency::cpu2.inst 12348.793741 # average overall mshr miss latency
system.cpu0.icache.demand_avg_mshr_miss_latency::total 12348.272007 # average overall mshr miss latency
system.cpu0.icache.overall_avg_mshr_miss_latency::cpu1.inst 12347.049089 # average overall mshr miss latency
system.cpu0.icache.overall_avg_mshr_miss_latency::cpu2.inst 12348.793741 # average overall mshr miss latency
system.cpu0.icache.overall_avg_mshr_miss_latency::total 12348.272007 # average overall mshr miss latency
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu0.icache.no_allocate_misses 0 # Number of misses that were no-allocate
stats: Cumulative stats update This patch updates the stats to reflect the: 1) addition of the internal queue in SimpleMemory, 2) moving of the memory class outside FSConfig, 3) fixing up of the 2D vector printing format, 4) specifying burst size and interface width for the DRAM instead of relying on cache-line size, 5) performing merging in the DRAM controller write buffer, and 6) fixing how idle cycles are counted in the atomic and timing CPU models. The main reason for bundling them up is to minimise the changeset size.
2013-08-19 09:52:36 +02:00
system.cpu0.dcache.tags.replacements 1391525 # number of replacements
system.cpu0.dcache.tags.tagsinuse 511.997811 # Cycle average of tags in use
system.cpu0.dcache.tags.total_refs 13285085 # Total number of references to valid blocks.
system.cpu0.dcache.tags.sampled_refs 1392037 # Sample count of references to valid blocks.
system.cpu0.dcache.tags.avg_refs 9.543629 # Average number of references to valid blocks.
system.cpu0.dcache.tags.warmup_cycle 10840000 # Cycle when the warmup percentage was hit.
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.dcache.tags.occ_blocks::cpu0.data 250.196143 # Average occupied blocks per requestor
system.cpu0.dcache.tags.occ_blocks::cpu1.data 130.348399 # Average occupied blocks per requestor
system.cpu0.dcache.tags.occ_blocks::cpu2.data 131.453270 # Average occupied blocks per requestor
system.cpu0.dcache.tags.occ_percent::cpu0.data 0.488664 # Average percentage of cache occupancy
system.cpu0.dcache.tags.occ_percent::cpu1.data 0.254587 # Average percentage of cache occupancy
system.cpu0.dcache.tags.occ_percent::cpu2.data 0.256745 # Average percentage of cache occupancy
stats: Cumulative stats update This patch updates the stats to reflect the: 1) addition of the internal queue in SimpleMemory, 2) moving of the memory class outside FSConfig, 3) fixing up of the 2D vector printing format, 4) specifying burst size and interface width for the DRAM instead of relying on cache-line size, 5) performing merging in the DRAM controller write buffer, and 6) fixing how idle cycles are counted in the atomic and timing CPU models. The main reason for bundling them up is to minimise the changeset size.
2013-08-19 09:52:36 +02:00
system.cpu0.dcache.tags.occ_percent::total 0.999996 # Average percentage of cache occupancy
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.dcache.ReadReq_hits::cpu0.data 4073389 # number of ReadReq hits
system.cpu0.dcache.ReadReq_hits::cpu1.data 1086662 # number of ReadReq hits
system.cpu0.dcache.ReadReq_hits::cpu2.data 2399601 # number of ReadReq hits
system.cpu0.dcache.ReadReq_hits::total 7559652 # number of ReadReq hits
system.cpu0.dcache.WriteReq_hits::cpu0.data 3208453 # number of WriteReq hits
system.cpu0.dcache.WriteReq_hits::cpu1.data 837545 # number of WriteReq hits
system.cpu0.dcache.WriteReq_hits::cpu2.data 1295843 # number of WriteReq hits
system.cpu0.dcache.WriteReq_hits::total 5341841 # number of WriteReq hits
system.cpu0.dcache.LoadLockedReq_hits::cpu0.data 116948 # number of LoadLockedReq hits
system.cpu0.dcache.LoadLockedReq_hits::cpu1.data 19346 # number of LoadLockedReq hits
system.cpu0.dcache.LoadLockedReq_hits::cpu2.data 47955 # number of LoadLockedReq hits
system.cpu0.dcache.LoadLockedReq_hits::total 184249 # number of LoadLockedReq hits
system.cpu0.dcache.StoreCondReq_hits::cpu0.data 126113 # number of StoreCondReq hits
system.cpu0.dcache.StoreCondReq_hits::cpu1.data 21379 # number of StoreCondReq hits
system.cpu0.dcache.StoreCondReq_hits::cpu2.data 51797 # number of StoreCondReq hits
system.cpu0.dcache.StoreCondReq_hits::total 199289 # number of StoreCondReq hits
system.cpu0.dcache.demand_hits::cpu0.data 7281842 # number of demand (read+write) hits
system.cpu0.dcache.demand_hits::cpu1.data 1924207 # number of demand (read+write) hits
system.cpu0.dcache.demand_hits::cpu2.data 3695444 # number of demand (read+write) hits
system.cpu0.dcache.demand_hits::total 12901493 # number of demand (read+write) hits
system.cpu0.dcache.overall_hits::cpu0.data 7281842 # number of overall hits
system.cpu0.dcache.overall_hits::cpu1.data 1924207 # number of overall hits
system.cpu0.dcache.overall_hits::cpu2.data 3695444 # number of overall hits
system.cpu0.dcache.overall_hits::total 12901493 # number of overall hits
system.cpu0.dcache.ReadReq_misses::cpu0.data 720818 # number of ReadReq misses
system.cpu0.dcache.ReadReq_misses::cpu1.data 98930 # number of ReadReq misses
system.cpu0.dcache.ReadReq_misses::cpu2.data 533963 # number of ReadReq misses
system.cpu0.dcache.ReadReq_misses::total 1353711 # number of ReadReq misses
system.cpu0.dcache.WriteReq_misses::cpu0.data 168792 # number of WriteReq misses
system.cpu0.dcache.WriteReq_misses::cpu1.data 45023 # number of WriteReq misses
system.cpu0.dcache.WriteReq_misses::cpu2.data 591984 # number of WriteReq misses
system.cpu0.dcache.WriteReq_misses::total 805799 # number of WriteReq misses
system.cpu0.dcache.LoadLockedReq_misses::cpu0.data 9729 # number of LoadLockedReq misses
system.cpu0.dcache.LoadLockedReq_misses::cpu1.data 2165 # number of LoadLockedReq misses
system.cpu0.dcache.LoadLockedReq_misses::cpu2.data 6797 # number of LoadLockedReq misses
system.cpu0.dcache.LoadLockedReq_misses::total 18691 # number of LoadLockedReq misses
system.cpu0.dcache.StoreCondReq_misses::cpu2.data 2 # number of StoreCondReq misses
system.cpu0.dcache.StoreCondReq_misses::total 2 # number of StoreCondReq misses
system.cpu0.dcache.demand_misses::cpu0.data 889610 # number of demand (read+write) misses
system.cpu0.dcache.demand_misses::cpu1.data 143953 # number of demand (read+write) misses
system.cpu0.dcache.demand_misses::cpu2.data 1125947 # number of demand (read+write) misses
system.cpu0.dcache.demand_misses::total 2159510 # number of demand (read+write) misses
system.cpu0.dcache.overall_misses::cpu0.data 889610 # number of overall misses
system.cpu0.dcache.overall_misses::cpu1.data 143953 # number of overall misses
system.cpu0.dcache.overall_misses::cpu2.data 1125947 # number of overall misses
system.cpu0.dcache.overall_misses::total 2159510 # number of overall misses
system.cpu0.dcache.ReadReq_miss_latency::cpu1.data 2262060250 # number of ReadReq miss cycles
system.cpu0.dcache.ReadReq_miss_latency::cpu2.data 9361678086 # number of ReadReq miss cycles
system.cpu0.dcache.ReadReq_miss_latency::total 11623738336 # number of ReadReq miss cycles
system.cpu0.dcache.WriteReq_miss_latency::cpu1.data 1620272009 # number of WriteReq miss cycles
system.cpu0.dcache.WriteReq_miss_latency::cpu2.data 17690504226 # number of WriteReq miss cycles
system.cpu0.dcache.WriteReq_miss_latency::total 19310776235 # number of WriteReq miss cycles
system.cpu0.dcache.LoadLockedReq_miss_latency::cpu1.data 28582500 # number of LoadLockedReq miss cycles
system.cpu0.dcache.LoadLockedReq_miss_latency::cpu2.data 102574499 # number of LoadLockedReq miss cycles
system.cpu0.dcache.LoadLockedReq_miss_latency::total 131156999 # number of LoadLockedReq miss cycles
system.cpu0.dcache.StoreCondReq_miss_latency::cpu2.data 38001 # number of StoreCondReq miss cycles
system.cpu0.dcache.StoreCondReq_miss_latency::total 38001 # number of StoreCondReq miss cycles
system.cpu0.dcache.demand_miss_latency::cpu1.data 3882332259 # number of demand (read+write) miss cycles
system.cpu0.dcache.demand_miss_latency::cpu2.data 27052182312 # number of demand (read+write) miss cycles
system.cpu0.dcache.demand_miss_latency::total 30934514571 # number of demand (read+write) miss cycles
system.cpu0.dcache.overall_miss_latency::cpu1.data 3882332259 # number of overall miss cycles
system.cpu0.dcache.overall_miss_latency::cpu2.data 27052182312 # number of overall miss cycles
system.cpu0.dcache.overall_miss_latency::total 30934514571 # number of overall miss cycles
system.cpu0.dcache.ReadReq_accesses::cpu0.data 4794207 # number of ReadReq accesses(hits+misses)
system.cpu0.dcache.ReadReq_accesses::cpu1.data 1185592 # number of ReadReq accesses(hits+misses)
system.cpu0.dcache.ReadReq_accesses::cpu2.data 2933564 # number of ReadReq accesses(hits+misses)
system.cpu0.dcache.ReadReq_accesses::total 8913363 # number of ReadReq accesses(hits+misses)
system.cpu0.dcache.WriteReq_accesses::cpu0.data 3377245 # number of WriteReq accesses(hits+misses)
system.cpu0.dcache.WriteReq_accesses::cpu1.data 882568 # number of WriteReq accesses(hits+misses)
system.cpu0.dcache.WriteReq_accesses::cpu2.data 1887827 # number of WriteReq accesses(hits+misses)
system.cpu0.dcache.WriteReq_accesses::total 6147640 # number of WriteReq accesses(hits+misses)
system.cpu0.dcache.LoadLockedReq_accesses::cpu0.data 126677 # number of LoadLockedReq accesses(hits+misses)
system.cpu0.dcache.LoadLockedReq_accesses::cpu1.data 21511 # number of LoadLockedReq accesses(hits+misses)
system.cpu0.dcache.LoadLockedReq_accesses::cpu2.data 54752 # number of LoadLockedReq accesses(hits+misses)
system.cpu0.dcache.LoadLockedReq_accesses::total 202940 # number of LoadLockedReq accesses(hits+misses)
system.cpu0.dcache.StoreCondReq_accesses::cpu0.data 126113 # number of StoreCondReq accesses(hits+misses)
system.cpu0.dcache.StoreCondReq_accesses::cpu1.data 21379 # number of StoreCondReq accesses(hits+misses)
system.cpu0.dcache.StoreCondReq_accesses::cpu2.data 51799 # number of StoreCondReq accesses(hits+misses)
system.cpu0.dcache.StoreCondReq_accesses::total 199291 # number of StoreCondReq accesses(hits+misses)
system.cpu0.dcache.demand_accesses::cpu0.data 8171452 # number of demand (read+write) accesses
system.cpu0.dcache.demand_accesses::cpu1.data 2068160 # number of demand (read+write) accesses
system.cpu0.dcache.demand_accesses::cpu2.data 4821391 # number of demand (read+write) accesses
system.cpu0.dcache.demand_accesses::total 15061003 # number of demand (read+write) accesses
system.cpu0.dcache.overall_accesses::cpu0.data 8171452 # number of overall (read+write) accesses
system.cpu0.dcache.overall_accesses::cpu1.data 2068160 # number of overall (read+write) accesses
system.cpu0.dcache.overall_accesses::cpu2.data 4821391 # number of overall (read+write) accesses
system.cpu0.dcache.overall_accesses::total 15061003 # number of overall (read+write) accesses
system.cpu0.dcache.ReadReq_miss_rate::cpu0.data 0.150352 # miss rate for ReadReq accesses
system.cpu0.dcache.ReadReq_miss_rate::cpu1.data 0.083444 # miss rate for ReadReq accesses
system.cpu0.dcache.ReadReq_miss_rate::cpu2.data 0.182019 # miss rate for ReadReq accesses
system.cpu0.dcache.ReadReq_miss_rate::total 0.151874 # miss rate for ReadReq accesses
system.cpu0.dcache.WriteReq_miss_rate::cpu0.data 0.049979 # miss rate for WriteReq accesses
system.cpu0.dcache.WriteReq_miss_rate::cpu1.data 0.051014 # miss rate for WriteReq accesses
system.cpu0.dcache.WriteReq_miss_rate::cpu2.data 0.313580 # miss rate for WriteReq accesses
system.cpu0.dcache.WriteReq_miss_rate::total 0.131075 # miss rate for WriteReq accesses
system.cpu0.dcache.LoadLockedReq_miss_rate::cpu0.data 0.076802 # miss rate for LoadLockedReq accesses
system.cpu0.dcache.LoadLockedReq_miss_rate::cpu1.data 0.100646 # miss rate for LoadLockedReq accesses
system.cpu0.dcache.LoadLockedReq_miss_rate::cpu2.data 0.124142 # miss rate for LoadLockedReq accesses
system.cpu0.dcache.LoadLockedReq_miss_rate::total 0.092101 # miss rate for LoadLockedReq accesses
system.cpu0.dcache.StoreCondReq_miss_rate::cpu2.data 0.000039 # miss rate for StoreCondReq accesses
system.cpu0.dcache.StoreCondReq_miss_rate::total 0.000010 # miss rate for StoreCondReq accesses
system.cpu0.dcache.demand_miss_rate::cpu0.data 0.108868 # miss rate for demand accesses
system.cpu0.dcache.demand_miss_rate::cpu1.data 0.069604 # miss rate for demand accesses
system.cpu0.dcache.demand_miss_rate::cpu2.data 0.233532 # miss rate for demand accesses
system.cpu0.dcache.demand_miss_rate::total 0.143384 # miss rate for demand accesses
system.cpu0.dcache.overall_miss_rate::cpu0.data 0.108868 # miss rate for overall accesses
system.cpu0.dcache.overall_miss_rate::cpu1.data 0.069604 # miss rate for overall accesses
system.cpu0.dcache.overall_miss_rate::cpu2.data 0.233532 # miss rate for overall accesses
system.cpu0.dcache.overall_miss_rate::total 0.143384 # miss rate for overall accesses
system.cpu0.dcache.ReadReq_avg_miss_latency::cpu1.data 22865.260790 # average ReadReq miss latency
system.cpu0.dcache.ReadReq_avg_miss_latency::cpu2.data 17532.447166 # average ReadReq miss latency
system.cpu0.dcache.ReadReq_avg_miss_latency::total 8586.573010 # average ReadReq miss latency
system.cpu0.dcache.WriteReq_avg_miss_latency::cpu1.data 35987.650956 # average WriteReq miss latency
system.cpu0.dcache.WriteReq_avg_miss_latency::cpu2.data 29883.416150 # average WriteReq miss latency
system.cpu0.dcache.WriteReq_avg_miss_latency::total 23964.755770 # average WriteReq miss latency
system.cpu0.dcache.LoadLockedReq_avg_miss_latency::cpu1.data 13202.078522 # average LoadLockedReq miss latency
system.cpu0.dcache.LoadLockedReq_avg_miss_latency::cpu2.data 15091.143004 # average LoadLockedReq miss latency
system.cpu0.dcache.LoadLockedReq_avg_miss_latency::total 7017.120486 # average LoadLockedReq miss latency
system.cpu0.dcache.StoreCondReq_avg_miss_latency::cpu2.data 19000.500000 # average StoreCondReq miss latency
system.cpu0.dcache.StoreCondReq_avg_miss_latency::total 19000.500000 # average StoreCondReq miss latency
system.cpu0.dcache.demand_avg_miss_latency::cpu1.data 26969.443214 # average overall miss latency
system.cpu0.dcache.demand_avg_miss_latency::cpu2.data 24026.159590 # average overall miss latency
system.cpu0.dcache.demand_avg_miss_latency::total 14324.784127 # average overall miss latency
system.cpu0.dcache.overall_avg_miss_latency::cpu1.data 26969.443214 # average overall miss latency
system.cpu0.dcache.overall_avg_miss_latency::cpu2.data 24026.159590 # average overall miss latency
system.cpu0.dcache.overall_avg_miss_latency::total 14324.784127 # average overall miss latency
system.cpu0.dcache.blocked_cycles::no_mshrs 557875 # number of cycles access was blocked
system.cpu0.dcache.blocked_cycles::no_targets 852 # number of cycles access was blocked
system.cpu0.dcache.blocked::no_mshrs 17918 # number of cycles access was blocked
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu0.dcache.blocked::no_targets 7 # number of cycles access was blocked
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.dcache.avg_blocked_cycles::no_mshrs 31.134892 # average number of cycles each access was blocked
system.cpu0.dcache.avg_blocked_cycles::no_targets 121.714286 # average number of cycles each access was blocked
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu0.dcache.fast_writes 0 # number of fast writes performed
system.cpu0.dcache.cache_copies 0 # number of cache copies performed
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu0.dcache.writebacks::writebacks 835257 # number of writebacks
system.cpu0.dcache.writebacks::total 835257 # number of writebacks
system.cpu0.dcache.ReadReq_mshr_hits::cpu2.data 281255 # number of ReadReq MSHR hits
system.cpu0.dcache.ReadReq_mshr_hits::total 281255 # number of ReadReq MSHR hits
system.cpu0.dcache.WriteReq_mshr_hits::cpu2.data 503456 # number of WriteReq MSHR hits
system.cpu0.dcache.WriteReq_mshr_hits::total 503456 # number of WriteReq MSHR hits
system.cpu0.dcache.LoadLockedReq_mshr_hits::cpu2.data 1394 # number of LoadLockedReq MSHR hits
system.cpu0.dcache.LoadLockedReq_mshr_hits::total 1394 # number of LoadLockedReq MSHR hits
system.cpu0.dcache.demand_mshr_hits::cpu2.data 784711 # number of demand (read+write) MSHR hits
system.cpu0.dcache.demand_mshr_hits::total 784711 # number of demand (read+write) MSHR hits
system.cpu0.dcache.overall_mshr_hits::cpu2.data 784711 # number of overall MSHR hits
system.cpu0.dcache.overall_mshr_hits::total 784711 # number of overall MSHR hits
system.cpu0.dcache.ReadReq_mshr_misses::cpu1.data 98930 # number of ReadReq MSHR misses
system.cpu0.dcache.ReadReq_mshr_misses::cpu2.data 252708 # number of ReadReq MSHR misses
system.cpu0.dcache.ReadReq_mshr_misses::total 351638 # number of ReadReq MSHR misses
system.cpu0.dcache.WriteReq_mshr_misses::cpu1.data 45023 # number of WriteReq MSHR misses
system.cpu0.dcache.WriteReq_mshr_misses::cpu2.data 88528 # number of WriteReq MSHR misses
system.cpu0.dcache.WriteReq_mshr_misses::total 133551 # number of WriteReq MSHR misses
system.cpu0.dcache.LoadLockedReq_mshr_misses::cpu1.data 2165 # number of LoadLockedReq MSHR misses
system.cpu0.dcache.LoadLockedReq_mshr_misses::cpu2.data 5403 # number of LoadLockedReq MSHR misses
system.cpu0.dcache.LoadLockedReq_mshr_misses::total 7568 # number of LoadLockedReq MSHR misses
system.cpu0.dcache.StoreCondReq_mshr_misses::cpu2.data 2 # number of StoreCondReq MSHR misses
system.cpu0.dcache.StoreCondReq_mshr_misses::total 2 # number of StoreCondReq MSHR misses
system.cpu0.dcache.demand_mshr_misses::cpu1.data 143953 # number of demand (read+write) MSHR misses
system.cpu0.dcache.demand_mshr_misses::cpu2.data 341236 # number of demand (read+write) MSHR misses
system.cpu0.dcache.demand_mshr_misses::total 485189 # number of demand (read+write) MSHR misses
system.cpu0.dcache.overall_mshr_misses::cpu1.data 143953 # number of overall MSHR misses
system.cpu0.dcache.overall_mshr_misses::cpu2.data 341236 # number of overall MSHR misses
system.cpu0.dcache.overall_mshr_misses::total 485189 # number of overall MSHR misses
system.cpu0.dcache.ReadReq_mshr_miss_latency::cpu1.data 2055649750 # number of ReadReq MSHR miss cycles
system.cpu0.dcache.ReadReq_mshr_miss_latency::cpu2.data 4240939408 # number of ReadReq MSHR miss cycles
system.cpu0.dcache.ReadReq_mshr_miss_latency::total 6296589158 # number of ReadReq MSHR miss cycles
system.cpu0.dcache.WriteReq_mshr_miss_latency::cpu1.data 1521155991 # number of WriteReq MSHR miss cycles
system.cpu0.dcache.WriteReq_mshr_miss_latency::cpu2.data 2563125747 # number of WriteReq MSHR miss cycles
system.cpu0.dcache.WriteReq_mshr_miss_latency::total 4084281738 # number of WriteReq MSHR miss cycles
system.cpu0.dcache.LoadLockedReq_mshr_miss_latency::cpu1.data 24249500 # number of LoadLockedReq MSHR miss cycles
system.cpu0.dcache.LoadLockedReq_mshr_miss_latency::cpu2.data 66698251 # number of LoadLockedReq MSHR miss cycles
system.cpu0.dcache.LoadLockedReq_mshr_miss_latency::total 90947751 # number of LoadLockedReq MSHR miss cycles
system.cpu0.dcache.StoreCondReq_mshr_miss_latency::cpu2.data 33999 # number of StoreCondReq MSHR miss cycles
system.cpu0.dcache.StoreCondReq_mshr_miss_latency::total 33999 # number of StoreCondReq MSHR miss cycles
system.cpu0.dcache.demand_mshr_miss_latency::cpu1.data 3576805741 # number of demand (read+write) MSHR miss cycles
system.cpu0.dcache.demand_mshr_miss_latency::cpu2.data 6804065155 # number of demand (read+write) MSHR miss cycles
system.cpu0.dcache.demand_mshr_miss_latency::total 10380870896 # number of demand (read+write) MSHR miss cycles
system.cpu0.dcache.overall_mshr_miss_latency::cpu1.data 3576805741 # number of overall MSHR miss cycles
system.cpu0.dcache.overall_mshr_miss_latency::cpu2.data 6804065155 # number of overall MSHR miss cycles
system.cpu0.dcache.overall_mshr_miss_latency::total 10380870896 # number of overall MSHR miss cycles
system.cpu0.dcache.ReadReq_mshr_uncacheable_latency::cpu1.data 294283500 # number of ReadReq MSHR uncacheable cycles
system.cpu0.dcache.ReadReq_mshr_uncacheable_latency::cpu2.data 312003000 # number of ReadReq MSHR uncacheable cycles
system.cpu0.dcache.ReadReq_mshr_uncacheable_latency::total 606286500 # number of ReadReq MSHR uncacheable cycles
system.cpu0.dcache.WriteReq_mshr_uncacheable_latency::cpu1.data 361937500 # number of WriteReq MSHR uncacheable cycles
system.cpu0.dcache.WriteReq_mshr_uncacheable_latency::cpu2.data 429433000 # number of WriteReq MSHR uncacheable cycles
system.cpu0.dcache.WriteReq_mshr_uncacheable_latency::total 791370500 # number of WriteReq MSHR uncacheable cycles
system.cpu0.dcache.overall_mshr_uncacheable_latency::cpu1.data 656221000 # number of overall MSHR uncacheable cycles
system.cpu0.dcache.overall_mshr_uncacheable_latency::cpu2.data 741436000 # number of overall MSHR uncacheable cycles
system.cpu0.dcache.overall_mshr_uncacheable_latency::total 1397657000 # number of overall MSHR uncacheable cycles
system.cpu0.dcache.ReadReq_mshr_miss_rate::cpu1.data 0.083444 # mshr miss rate for ReadReq accesses
system.cpu0.dcache.ReadReq_mshr_miss_rate::cpu2.data 0.086144 # mshr miss rate for ReadReq accesses
system.cpu0.dcache.ReadReq_mshr_miss_rate::total 0.039451 # mshr miss rate for ReadReq accesses
system.cpu0.dcache.WriteReq_mshr_miss_rate::cpu1.data 0.051014 # mshr miss rate for WriteReq accesses
system.cpu0.dcache.WriteReq_mshr_miss_rate::cpu2.data 0.046894 # mshr miss rate for WriteReq accesses
system.cpu0.dcache.WriteReq_mshr_miss_rate::total 0.021724 # mshr miss rate for WriteReq accesses
system.cpu0.dcache.LoadLockedReq_mshr_miss_rate::cpu1.data 0.100646 # mshr miss rate for LoadLockedReq accesses
system.cpu0.dcache.LoadLockedReq_mshr_miss_rate::cpu2.data 0.098681 # mshr miss rate for LoadLockedReq accesses
system.cpu0.dcache.LoadLockedReq_mshr_miss_rate::total 0.037292 # mshr miss rate for LoadLockedReq accesses
system.cpu0.dcache.StoreCondReq_mshr_miss_rate::cpu2.data 0.000039 # mshr miss rate for StoreCondReq accesses
system.cpu0.dcache.StoreCondReq_mshr_miss_rate::total 0.000010 # mshr miss rate for StoreCondReq accesses
system.cpu0.dcache.demand_mshr_miss_rate::cpu1.data 0.069604 # mshr miss rate for demand accesses
system.cpu0.dcache.demand_mshr_miss_rate::cpu2.data 0.070775 # mshr miss rate for demand accesses
system.cpu0.dcache.demand_mshr_miss_rate::total 0.032215 # mshr miss rate for demand accesses
system.cpu0.dcache.overall_mshr_miss_rate::cpu1.data 0.069604 # mshr miss rate for overall accesses
system.cpu0.dcache.overall_mshr_miss_rate::cpu2.data 0.070775 # mshr miss rate for overall accesses
system.cpu0.dcache.overall_mshr_miss_rate::total 0.032215 # mshr miss rate for overall accesses
system.cpu0.dcache.ReadReq_avg_mshr_miss_latency::cpu1.data 20778.830992 # average ReadReq mshr miss latency
system.cpu0.dcache.ReadReq_avg_mshr_miss_latency::cpu2.data 16781.975276 # average ReadReq mshr miss latency
system.cpu0.dcache.ReadReq_avg_mshr_miss_latency::total 17906.452539 # average ReadReq mshr miss latency
system.cpu0.dcache.WriteReq_avg_mshr_miss_latency::cpu1.data 33786.197965 # average WriteReq mshr miss latency
system.cpu0.dcache.WriteReq_avg_mshr_miss_latency::cpu2.data 28952.712667 # average WriteReq mshr miss latency
system.cpu0.dcache.WriteReq_avg_mshr_miss_latency::total 30582.187614 # average WriteReq mshr miss latency
system.cpu0.dcache.LoadLockedReq_avg_mshr_miss_latency::cpu1.data 11200.692841 # average LoadLockedReq mshr miss latency
system.cpu0.dcache.LoadLockedReq_avg_mshr_miss_latency::cpu2.data 12344.669813 # average LoadLockedReq mshr miss latency
system.cpu0.dcache.LoadLockedReq_avg_mshr_miss_latency::total 12017.408959 # average LoadLockedReq mshr miss latency
system.cpu0.dcache.StoreCondReq_avg_mshr_miss_latency::cpu2.data 16999.500000 # average StoreCondReq mshr miss latency
system.cpu0.dcache.StoreCondReq_avg_mshr_miss_latency::total 16999.500000 # average StoreCondReq mshr miss latency
system.cpu0.dcache.demand_avg_mshr_miss_latency::cpu1.data 24847.038554 # average overall mshr miss latency
system.cpu0.dcache.demand_avg_mshr_miss_latency::cpu2.data 19939.470498 # average overall mshr miss latency
system.cpu0.dcache.demand_avg_mshr_miss_latency::total 21395.519882 # average overall mshr miss latency
system.cpu0.dcache.overall_avg_mshr_miss_latency::cpu1.data 24847.038554 # average overall mshr miss latency
system.cpu0.dcache.overall_avg_mshr_miss_latency::cpu2.data 19939.470498 # average overall mshr miss latency
system.cpu0.dcache.overall_avg_mshr_miss_latency::total 21395.519882 # average overall mshr miss latency
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu0.dcache.ReadReq_avg_mshr_uncacheable_latency::cpu1.data inf # average ReadReq mshr uncacheable latency
system.cpu0.dcache.ReadReq_avg_mshr_uncacheable_latency::cpu2.data inf # average ReadReq mshr uncacheable latency
system.cpu0.dcache.ReadReq_avg_mshr_uncacheable_latency::total inf # average ReadReq mshr uncacheable latency
system.cpu0.dcache.WriteReq_avg_mshr_uncacheable_latency::cpu1.data inf # average WriteReq mshr uncacheable latency
system.cpu0.dcache.WriteReq_avg_mshr_uncacheable_latency::cpu2.data inf # average WriteReq mshr uncacheable latency
system.cpu0.dcache.WriteReq_avg_mshr_uncacheable_latency::total inf # average WriteReq mshr uncacheable latency
system.cpu0.dcache.overall_avg_mshr_uncacheable_latency::cpu1.data inf # average overall mshr uncacheable latency
system.cpu0.dcache.overall_avg_mshr_uncacheable_latency::cpu2.data inf # average overall mshr uncacheable latency
system.cpu0.dcache.overall_avg_mshr_uncacheable_latency::total inf # average overall mshr uncacheable latency
system.cpu0.dcache.no_allocate_misses 0 # Number of misses that were no-allocate
system.cpu1.dtb.fetch_hits 0 # ITB hits
system.cpu1.dtb.fetch_misses 0 # ITB misses
system.cpu1.dtb.fetch_acv 0 # ITB acv
system.cpu1.dtb.fetch_accesses 0 # ITB accesses
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu1.dtb.read_hits 1205047 # DTB read hits
system.cpu1.dtb.read_misses 1367 # DTB read misses
system.cpu1.dtb.read_acv 34 # DTB read access violations
system.cpu1.dtb.read_accesses 142944 # DTB read accesses
system.cpu1.dtb.write_hits 904403 # DTB write hits
system.cpu1.dtb.write_misses 185 # DTB write misses
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.cpu1.dtb.write_acv 23 # DTB write access violations
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu1.dtb.write_accesses 58533 # DTB write accesses
system.cpu1.dtb.data_hits 2109450 # DTB hits
system.cpu1.dtb.data_misses 1552 # DTB misses
system.cpu1.dtb.data_acv 57 # DTB access violations
system.cpu1.dtb.data_accesses 201477 # DTB accesses
system.cpu1.itb.fetch_hits 861634 # ITB hits
system.cpu1.itb.fetch_misses 693 # ITB misses
system.cpu1.itb.fetch_acv 30 # ITB acv
system.cpu1.itb.fetch_accesses 862327 # ITB accesses
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu1.itb.read_hits 0 # DTB read hits
system.cpu1.itb.read_misses 0 # DTB read misses
system.cpu1.itb.read_acv 0 # DTB read access violations
system.cpu1.itb.read_accesses 0 # DTB read accesses
system.cpu1.itb.write_hits 0 # DTB write hits
system.cpu1.itb.write_misses 0 # DTB write misses
system.cpu1.itb.write_acv 0 # DTB write access violations
system.cpu1.itb.write_accesses 0 # DTB write accesses
system.cpu1.itb.data_hits 0 # DTB hits
system.cpu1.itb.data_misses 0 # DTB misses
system.cpu1.itb.data_acv 0 # DTB access violations
system.cpu1.itb.data_accesses 0 # DTB accesses
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu1.numCycles 953630418 # number of cpu cycles simulated
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu1.numWorkItemsStarted 0 # number of work items this cpu started
system.cpu1.numWorkItemsCompleted 0 # number of work items this cpu completed
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu1.committedInsts 7889245 # Number of instructions committed
system.cpu1.committedOps 7889245 # Number of ops (including micro ops) committed
system.cpu1.num_int_alu_accesses 7344952 # Number of integer alu accesses
system.cpu1.num_fp_alu_accesses 44937 # Number of float alu accesses
system.cpu1.num_func_calls 213049 # number of times a function call or return occured
system.cpu1.num_conditional_control_insts 993802 # number of instructions that are conditional controls
system.cpu1.num_int_insts 7344952 # number of integer instructions
system.cpu1.num_fp_insts 44937 # number of float instructions
system.cpu1.num_int_register_reads 10269748 # number of times the integer registers were read
system.cpu1.num_int_register_writes 5343251 # number of times the integer registers were written
system.cpu1.num_fp_register_reads 24271 # number of times the floating registers were read
system.cpu1.num_fp_register_writes 24577 # number of times the floating registers were written
system.cpu1.num_mem_refs 2116682 # number of memory refs
system.cpu1.num_load_insts 1209934 # Number of load instructions
system.cpu1.num_store_insts 906748 # Number of store instructions
stats: Cumulative stats update This patch updates the stats to reflect the: 1) addition of the internal queue in SimpleMemory, 2) moving of the memory class outside FSConfig, 3) fixing up of the 2D vector printing format, 4) specifying burst size and interface width for the DRAM instead of relying on cache-line size, 5) performing merging in the DRAM controller write buffer, and 6) fixing how idle cycles are counted in the atomic and timing CPU models. The main reason for bundling them up is to minimise the changeset size.
2013-08-19 09:52:36 +02:00
system.cpu1.num_idle_cycles 923700977.463911 # Number of idle cycles
system.cpu1.num_busy_cycles 29929440.536089 # Number of busy cycles
system.cpu1.not_idle_fraction 0.031385 # Percentage of non-idle cycles
system.cpu1.idle_fraction 0.968615 # Percentage of idle cycles
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu1.kern.inst.arm 0 # number of arm instructions executed
system.cpu1.kern.inst.quiesce 0 # number of quiesce instructions executed
system.cpu1.kern.inst.hwrei 0 # number of hwrei instructions executed
system.cpu1.kern.mode_switch::kernel 0 # number of protection mode switches
system.cpu1.kern.mode_switch::user 0 # number of protection mode switches
system.cpu1.kern.mode_switch::idle 0 # number of protection mode switches
system.cpu1.kern.mode_good::kernel 0
system.cpu1.kern.mode_good::user 0
system.cpu1.kern.mode_good::idle 0
system.cpu1.kern.mode_switch_good::kernel nan # fraction of useful protection mode switches
system.cpu1.kern.mode_switch_good::user nan # fraction of useful protection mode switches
system.cpu1.kern.mode_switch_good::idle nan # fraction of useful protection mode switches
system.cpu1.kern.mode_switch_good::total nan # fraction of useful protection mode switches
system.cpu1.kern.mode_ticks::kernel 0 # number of ticks spent at the given mode
system.cpu1.kern.mode_ticks::user 0 # number of ticks spent at the given mode
system.cpu1.kern.mode_ticks::idle 0 # number of ticks spent at the given mode
system.cpu1.kern.swap_context 0 # number of times the context was actually changed
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.branchPred.lookups 9022316 # Number of BP lookups
system.cpu2.branchPred.condPredicted 8342315 # Number of conditional branches predicted
system.cpu2.branchPred.condIncorrect 122648 # Number of conditional branches incorrect
system.cpu2.branchPred.BTBLookups 7529449 # Number of BTB lookups
system.cpu2.branchPred.BTBHits 6410701 # Number of BTB hits
regressions: update stats due to branch predictor changes The actual statistical values are being updated for only two tests belonging to sparc architecture and inorder cpu: 00.hello and 02.insttest. For others the patch updates config.ini and name changes to statistical variables.
2013-01-24 19:29:00 +01:00
system.cpu2.branchPred.BTBCorrect 0 # Number of correct BTB predictions (this stat may not work properly.
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.branchPred.BTBHitPct 85.141702 # BTB Hit Percentage
system.cpu2.branchPred.usedRAS 283187 # Number of times the RAS was used to get a target.
system.cpu2.branchPred.RASInCorrect 12478 # Number of incorrect RAS predictions.
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.dtb.fetch_hits 0 # ITB hits
system.cpu2.dtb.fetch_misses 0 # ITB misses
system.cpu2.dtb.fetch_acv 0 # ITB acv
system.cpu2.dtb.fetch_accesses 0 # ITB accesses
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.dtb.read_hits 3192037 # DTB read hits
system.cpu2.dtb.read_misses 11608 # DTB read misses
system.cpu2.dtb.read_acv 121 # DTB read access violations
system.cpu2.dtb.read_accesses 216573 # DTB read accesses
system.cpu2.dtb.write_hits 2009173 # DTB write hits
system.cpu2.dtb.write_misses 2522 # DTB write misses
system.cpu2.dtb.write_acv 106 # DTB write access violations
system.cpu2.dtb.write_accesses 81978 # DTB write accesses
system.cpu2.dtb.data_hits 5201210 # DTB hits
system.cpu2.dtb.data_misses 14130 # DTB misses
stats: Update the stats to reflect bus and memory changes This patch updates the stats to reflect the addition of the bus stats, and changes to the bus layers. In addition it updates the stats to match the addition of the static pipeline latency of the memory conotroller and the addition of a stat tracking the bytes per activate.
2013-05-30 18:54:18 +02:00
system.cpu2.dtb.data_acv 227 # DTB access violations
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.dtb.data_accesses 298551 # DTB accesses
system.cpu2.itb.fetch_hits 369667 # ITB hits
system.cpu2.itb.fetch_misses 5681 # ITB misses
system.cpu2.itb.fetch_acv 262 # ITB acv
system.cpu2.itb.fetch_accesses 375348 # ITB accesses
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.itb.read_hits 0 # DTB read hits
system.cpu2.itb.read_misses 0 # DTB read misses
system.cpu2.itb.read_acv 0 # DTB read access violations
system.cpu2.itb.read_accesses 0 # DTB read accesses
system.cpu2.itb.write_hits 0 # DTB write hits
system.cpu2.itb.write_misses 0 # DTB write misses
system.cpu2.itb.write_acv 0 # DTB write access violations
system.cpu2.itb.write_accesses 0 # DTB write accesses
system.cpu2.itb.data_hits 0 # DTB hits
system.cpu2.itb.data_misses 0 # DTB misses
system.cpu2.itb.data_acv 0 # DTB access violations
system.cpu2.itb.data_accesses 0 # DTB accesses
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.numCycles 31245078 # number of cpu cycles simulated
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.numWorkItemsStarted 0 # number of work items this cpu started
system.cpu2.numWorkItemsCompleted 0 # number of work items this cpu completed
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.fetch.icacheStallCycles 8348883 # Number of cycles fetch is stalled on an Icache miss
system.cpu2.fetch.Insts 36663716 # Number of instructions fetch has processed
system.cpu2.fetch.Branches 9022316 # Number of branches that fetch encountered
system.cpu2.fetch.predictedBranches 6693888 # Number of branches that fetch has predicted taken
system.cpu2.fetch.Cycles 8736568 # Number of cycles fetch has run and was not squashing or blocked
system.cpu2.fetch.SquashCycles 602984 # Number of cycles fetch has spent squashing
system.cpu2.fetch.BlockedCycles 9694630 # Number of cycles fetch has spent blocked
system.cpu2.fetch.MiscStallCycles 11222 # Number of cycles fetch has spent waiting on interrupts, or bad addresses, or out of MSHRs
system.cpu2.fetch.PendingDrainCycles 1957 # Number of cycles fetch has spent waiting on pipes to drain
system.cpu2.fetch.PendingTrapStallCycles 63711 # Number of stall cycles due to pending traps
system.cpu2.fetch.PendingQuiesceStallCycles 86195 # Number of stall cycles due to pending quiesce instructions
system.cpu2.fetch.IcacheWaitRetryStallCycles 437 # Number of stall cycles due to full MSHR
system.cpu2.fetch.CacheLines 2553880 # Number of cache lines fetched
system.cpu2.fetch.IcacheSquashes 85053 # Number of outstanding Icache misses that were squashed
system.cpu2.fetch.rateDist::samples 27335965 # Number of instructions fetched each cycle (Total)
system.cpu2.fetch.rateDist::mean 1.341226 # Number of instructions fetched each cycle (Total)
system.cpu2.fetch.rateDist::stdev 2.294449 # Number of instructions fetched each cycle (Total)
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.fetch.rateDist::underflows 0 0.00% 0.00% # Number of instructions fetched each cycle (Total)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.fetch.rateDist::0 18599397 68.04% 68.04% # Number of instructions fetched each cycle (Total)
system.cpu2.fetch.rateDist::1 269863 0.99% 69.03% # Number of instructions fetched each cycle (Total)
system.cpu2.fetch.rateDist::2 429102 1.57% 70.60% # Number of instructions fetched each cycle (Total)
system.cpu2.fetch.rateDist::3 4885317 17.87% 88.47% # Number of instructions fetched each cycle (Total)
system.cpu2.fetch.rateDist::4 756803 2.77% 91.24% # Number of instructions fetched each cycle (Total)
system.cpu2.fetch.rateDist::5 166340 0.61% 91.85% # Number of instructions fetched each cycle (Total)
system.cpu2.fetch.rateDist::6 191609 0.70% 92.55% # Number of instructions fetched each cycle (Total)
system.cpu2.fetch.rateDist::7 429140 1.57% 94.12% # Number of instructions fetched each cycle (Total)
system.cpu2.fetch.rateDist::8 1608394 5.88% 100.00% # Number of instructions fetched each cycle (Total)
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.fetch.rateDist::overflows 0 0.00% 100.00% # Number of instructions fetched each cycle (Total)
system.cpu2.fetch.rateDist::min_value 0 # Number of instructions fetched each cycle (Total)
system.cpu2.fetch.rateDist::max_value 8 # Number of instructions fetched each cycle (Total)
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.fetch.rateDist::total 27335965 # Number of instructions fetched each cycle (Total)
system.cpu2.fetch.branchRate 0.288760 # Number of branch fetches per cycle
system.cpu2.fetch.rate 1.173424 # Number of inst fetches per cycle
system.cpu2.decode.IdleCycles 8495766 # Number of cycles decode is idle
system.cpu2.decode.BlockedCycles 9778515 # Number of cycles decode is blocked
system.cpu2.decode.RunCycles 8128034 # Number of cycles decode is running
system.cpu2.decode.UnblockCycles 307242 # Number of cycles decode is unblocking
system.cpu2.decode.SquashCycles 380496 # Number of cycles decode is squashing
system.cpu2.decode.BranchResolved 165135 # Number of times decode resolved a branch
system.cpu2.decode.BranchMispred 12538 # Number of times decode detected a branch misprediction
system.cpu2.decode.DecodedInsts 36269918 # Number of instructions handled by decode
system.cpu2.decode.SquashedInsts 39153 # Number of squashed instructions handled by decode
system.cpu2.rename.SquashCycles 380496 # Number of cycles rename is squashing
system.cpu2.rename.IdleCycles 8853799 # Number of cycles rename is idle
system.cpu2.rename.BlockCycles 2797423 # Number of cycles rename is blocking
system.cpu2.rename.serializeStallCycles 5789351 # count of cycles rename stalled for serializing inst
system.cpu2.rename.RunCycles 7998658 # Number of cycles rename is running
system.cpu2.rename.UnblockCycles 1270334 # Number of cycles rename is unblocking
system.cpu2.rename.RenamedInsts 35131949 # Number of instructions processed by rename
system.cpu2.rename.ROBFullEvents 2438 # Number of times rename has blocked due to ROB full
system.cpu2.rename.IQFullEvents 231189 # Number of times rename has blocked due to IQ full
system.cpu2.rename.LSQFullEvents 444117 # Number of times rename has blocked due to LSQ full
system.cpu2.rename.RenamedOperands 23541427 # Number of destination operands rename has renamed
system.cpu2.rename.RenameLookups 43931372 # Number of register rename lookups that rename has made
system.cpu2.rename.int_rename_lookups 43768405 # Number of integer rename lookups
system.cpu2.rename.fp_rename_lookups 162967 # Number of floating rename lookups
system.cpu2.rename.CommittedMaps 21760313 # Number of HB maps that are committed
system.cpu2.rename.UndoneMaps 1781114 # Number of HB maps that are undone due to squashing
system.cpu2.rename.serializingInsts 501831 # count of serializing insts renamed
system.cpu2.rename.tempSerializingInsts 59191 # count of temporary serializing insts renamed
system.cpu2.rename.skidInsts 3719256 # count of insts added to the skid buffer
system.cpu2.memDep0.insertedLoads 3350609 # Number of loads inserted to the mem dependence unit.
system.cpu2.memDep0.insertedStores 2097879 # Number of stores inserted to the mem dependence unit.
system.cpu2.memDep0.conflictingLoads 369762 # Number of conflicting loads.
system.cpu2.memDep0.conflictingStores 260934 # Number of conflicting stores.
system.cpu2.iq.iqInstsAdded 32641753 # Number of instructions added to the IQ (excludes non-spec)
system.cpu2.iq.iqNonSpecInstsAdded 622044 # Number of non-speculative instructions added to the IQ
system.cpu2.iq.iqInstsIssued 32196803 # Number of instructions issued
system.cpu2.iq.iqSquashedInstsIssued 34835 # Number of squashed instructions issued
system.cpu2.iq.iqSquashedInstsExamined 2138258 # Number of squashed instructions iterated over during squash; mainly for profiling
system.cpu2.iq.iqSquashedOperandsExamined 1073109 # Number of squashed operands that are examined and possibly removed from graph
system.cpu2.iq.iqSquashedNonSpecRemoved 438824 # Number of squashed non-spec instructions that were removed
system.cpu2.iq.issued_per_cycle::samples 27335965 # Number of insts issued each cycle
system.cpu2.iq.issued_per_cycle::mean 1.177818 # Number of insts issued each cycle
system.cpu2.iq.issued_per_cycle::stdev 1.573987 # Number of insts issued each cycle
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.iq.issued_per_cycle::underflows 0 0.00% 0.00% # Number of insts issued each cycle
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.iq.issued_per_cycle::0 15167963 55.49% 55.49% # Number of insts issued each cycle
system.cpu2.iq.issued_per_cycle::1 3067850 11.22% 66.71% # Number of insts issued each cycle
system.cpu2.iq.issued_per_cycle::2 1557003 5.70% 72.41% # Number of insts issued each cycle
system.cpu2.iq.issued_per_cycle::3 5712284 20.90% 93.30% # Number of insts issued each cycle
system.cpu2.iq.issued_per_cycle::4 903378 3.30% 96.61% # Number of insts issued each cycle
system.cpu2.iq.issued_per_cycle::5 480833 1.76% 98.37% # Number of insts issued each cycle
system.cpu2.iq.issued_per_cycle::6 285081 1.04% 99.41% # Number of insts issued each cycle
system.cpu2.iq.issued_per_cycle::7 142652 0.52% 99.93% # Number of insts issued each cycle
system.cpu2.iq.issued_per_cycle::8 18921 0.07% 100.00% # Number of insts issued each cycle
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.iq.issued_per_cycle::overflows 0 0.00% 100.00% # Number of insts issued each cycle
system.cpu2.iq.issued_per_cycle::min_value 0 # Number of insts issued each cycle
system.cpu2.iq.issued_per_cycle::max_value 8 # Number of insts issued each cycle
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.iq.issued_per_cycle::total 27335965 # Number of insts issued each cycle
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.iq.fu_full::No_OpClass 0 0.00% 0.00% # attempts to use FU when none available
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.iq.fu_full::IntAlu 33684 13.60% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::IntMult 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::IntDiv 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::FloatAdd 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::FloatCmp 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::FloatCvt 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::FloatMult 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::FloatDiv 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::FloatSqrt 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdAdd 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdAddAcc 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdAlu 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdCmp 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdCvt 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdMisc 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdMult 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdMultAcc 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdShift 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdShiftAcc 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdSqrt 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdFloatAdd 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdFloatAlu 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdFloatCmp 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdFloatCvt 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdFloatDiv 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdFloatMisc 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdFloatMult 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdFloatMultAcc 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::SimdFloatSqrt 0 0.00% 13.60% # attempts to use FU when none available
system.cpu2.iq.fu_full::MemRead 113022 45.64% 59.24% # attempts to use FU when none available
system.cpu2.iq.fu_full::MemWrite 100957 40.76% 100.00% # attempts to use FU when none available
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.iq.fu_full::IprAccess 0 0.00% 100.00% # attempts to use FU when none available
system.cpu2.iq.fu_full::InstPrefetch 0 0.00% 100.00% # attempts to use FU when none available
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.iq.FU_type_0::No_OpClass 2440 0.01% 0.01% # Type of FU issued
system.cpu2.iq.FU_type_0::IntAlu 26526068 82.39% 82.39% # Type of FU issued
system.cpu2.iq.FU_type_0::IntMult 20082 0.06% 82.46% # Type of FU issued
system.cpu2.iq.FU_type_0::IntDiv 0 0.00% 82.46% # Type of FU issued
system.cpu2.iq.FU_type_0::FloatAdd 8432 0.03% 82.48% # Type of FU issued
system.cpu2.iq.FU_type_0::FloatCmp 0 0.00% 82.48% # Type of FU issued
system.cpu2.iq.FU_type_0::FloatCvt 0 0.00% 82.48% # Type of FU issued
system.cpu2.iq.FU_type_0::FloatMult 0 0.00% 82.48% # Type of FU issued
system.cpu2.iq.FU_type_0::FloatDiv 1220 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::FloatSqrt 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdAdd 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdAddAcc 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdAlu 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdCmp 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdCvt 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdMisc 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdMult 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdMultAcc 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdShift 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdShiftAcc 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdSqrt 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdFloatAdd 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdFloatAlu 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdFloatCmp 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdFloatCvt 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdFloatDiv 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdFloatMisc 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdFloatMult 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdFloatMultAcc 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::SimdFloatSqrt 0 0.00% 82.49% # Type of FU issued
system.cpu2.iq.FU_type_0::MemRead 3318552 10.31% 92.79% # Type of FU issued
system.cpu2.iq.FU_type_0::MemWrite 2030927 6.31% 99.10% # Type of FU issued
system.cpu2.iq.FU_type_0::IprAccess 289082 0.90% 100.00% # Type of FU issued
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.iq.FU_type_0::InstPrefetch 0 0.00% 100.00% # Type of FU issued
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.iq.FU_type_0::total 32196803 # Type of FU issued
system.cpu2.iq.rate 1.030460 # Inst issue rate
system.cpu2.iq.fu_busy_cnt 247663 # FU busy when requested
system.cpu2.iq.fu_busy_rate 0.007692 # FU busy rate (busy events/executed inst)
system.cpu2.iq.int_inst_queue_reads 91777621 # Number of integer instruction queue reads
system.cpu2.iq.int_inst_queue_writes 35291242 # Number of integer instruction queue writes
system.cpu2.iq.int_inst_queue_wakeup_accesses 31803164 # Number of integer instruction queue wakeup accesses
system.cpu2.iq.fp_inst_queue_reads 234448 # Number of floating instruction queue reads
system.cpu2.iq.fp_inst_queue_writes 114643 # Number of floating instruction queue writes
system.cpu2.iq.fp_inst_queue_wakeup_accesses 110912 # Number of floating instruction queue wakeup accesses
system.cpu2.iq.int_alu_accesses 32319915 # Number of integer alu accesses
system.cpu2.iq.fp_alu_accesses 122111 # Number of floating point alu accesses
system.cpu2.iew.lsq.thread0.forwLoads 186470 # Number of loads that had data forwarded from stores
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.iew.lsq.thread0.invAddrLoads 0 # Number of loads ignored due to an invalid address
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.iew.lsq.thread0.squashedLoads 409308 # Number of loads squashed
system.cpu2.iew.lsq.thread0.ignoredResponses 1087 # Number of memory responses ignored because the instruction is squashed
system.cpu2.iew.lsq.thread0.memOrderViolation 3940 # Number of memory ordering violations
system.cpu2.iew.lsq.thread0.squashedStores 154806 # Number of stores squashed
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.iew.lsq.thread0.invAddrSwpfs 0 # Number of software prefetches ignored due to an invalid address
system.cpu2.iew.lsq.thread0.blockedLoads 0 # Number of blocked loads due to partial load-store forwarding
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.iew.lsq.thread0.rescheduledLoads 4179 # Number of loads that were rescheduled
system.cpu2.iew.lsq.thread0.cacheBlocked 28515 # Number of times an access to memory failed due to the cache being blocked
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.iew.iewIdleCycles 0 # Number of cycles IEW is idle
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.iew.iewSquashCycles 380496 # Number of cycles IEW is squashing
system.cpu2.iew.iewBlockCycles 2018433 # Number of cycles IEW is blocking
system.cpu2.iew.iewUnblockCycles 205280 # Number of cycles IEW is unblocking
system.cpu2.iew.iewDispatchedInsts 34533473 # Number of instructions dispatched to IQ
system.cpu2.iew.iewDispSquashedInsts 223572 # Number of squashed instructions skipped by dispatch
system.cpu2.iew.iewDispLoadInsts 3350609 # Number of dispatched load instructions
system.cpu2.iew.iewDispStoreInsts 2097879 # Number of dispatched store instructions
system.cpu2.iew.iewDispNonSpecInsts 552418 # Number of dispatched non-speculative instructions
system.cpu2.iew.iewIQFullEvents 143005 # Number of times the IQ has become full, causing a stall
system.cpu2.iew.iewLSQFullEvents 2030 # Number of times the LSQ has become full, causing a stall
system.cpu2.iew.memOrderViolationEvents 3940 # Number of memory order violations
system.cpu2.iew.predictedTakenIncorrect 62474 # Number of branches that were predicted taken incorrectly
system.cpu2.iew.predictedNotTakenIncorrect 127218 # Number of branches that were predicted not taken incorrectly
system.cpu2.iew.branchMispredicts 189692 # Number of branch mispredicts detected at execute
system.cpu2.iew.iewExecutedInsts 32041792 # Number of executed instructions
system.cpu2.iew.iewExecLoadInsts 3211958 # Number of load instructions executed
system.cpu2.iew.iewExecSquashedInsts 155011 # Number of squashed instructions skipped in execute
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.iew.exec_swp 0 # number of swp insts executed
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.iew.exec_nop 1269676 # number of nop insts executed
system.cpu2.iew.exec_refs 5228104 # number of memory reference insts executed
system.cpu2.iew.exec_branches 7451179 # Number of branches executed
system.cpu2.iew.exec_stores 2016146 # Number of stores executed
system.cpu2.iew.exec_rate 1.025499 # Inst execution rate
system.cpu2.iew.wb_sent 31946323 # cumulative count of insts sent to commit
system.cpu2.iew.wb_count 31914076 # cumulative count of insts written-back
system.cpu2.iew.wb_producers 18560539 # num instructions producing a value
system.cpu2.iew.wb_consumers 21756623 # num instructions consuming a value
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.iew.wb_penalized 0 # number of instrctions required to write to 'other' IQ
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.iew.wb_rate 1.021411 # insts written-back per cycle
system.cpu2.iew.wb_fanout 0.853098 # average fanout of values written-back
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.iew.wb_penalized_rate 0 # fraction of instructions written-back that wrote to 'other' IQ
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.commit.commitSquashedInsts 2307107 # The number of squashed insts skipped by commit
system.cpu2.commit.commitNonSpecStalls 183220 # The number of times commit has been forced to stall to communicate backwards
system.cpu2.commit.branchMispredicts 175579 # The number of times a branch was mispredicted
system.cpu2.commit.committed_per_cycle::samples 26955469 # Number of insts commited each cycle
system.cpu2.commit.committed_per_cycle::mean 1.193861 # Number of insts commited each cycle
system.cpu2.commit.committed_per_cycle::stdev 1.846623 # Number of insts commited each cycle
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.commit.committed_per_cycle::underflows 0 0.00% 0.00% # Number of insts commited each cycle
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.commit.committed_per_cycle::0 16175286 60.01% 60.01% # Number of insts commited each cycle
system.cpu2.commit.committed_per_cycle::1 2330504 8.65% 68.65% # Number of insts commited each cycle
system.cpu2.commit.committed_per_cycle::2 1226068 4.55% 73.20% # Number of insts commited each cycle
system.cpu2.commit.committed_per_cycle::3 5456953 20.24% 93.45% # Number of insts commited each cycle
system.cpu2.commit.committed_per_cycle::4 503178 1.87% 95.31% # Number of insts commited each cycle
system.cpu2.commit.committed_per_cycle::5 186113 0.69% 96.00% # Number of insts commited each cycle
system.cpu2.commit.committed_per_cycle::6 177622 0.66% 96.66% # Number of insts commited each cycle
system.cpu2.commit.committed_per_cycle::7 179384 0.67% 97.33% # Number of insts commited each cycle
system.cpu2.commit.committed_per_cycle::8 720361 2.67% 100.00% # Number of insts commited each cycle
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.commit.committed_per_cycle::overflows 0 0.00% 100.00% # Number of insts commited each cycle
system.cpu2.commit.committed_per_cycle::min_value 0 # Number of insts commited each cycle
system.cpu2.commit.committed_per_cycle::max_value 8 # Number of insts commited each cycle
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.commit.committed_per_cycle::total 26955469 # Number of insts commited each cycle
system.cpu2.commit.committedInsts 32181084 # Number of instructions committed
system.cpu2.commit.committedOps 32181084 # Number of ops (including micro ops) committed
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.commit.swp_count 0 # Number of s/w prefetches committed
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.commit.refs 4884374 # Number of memory references committed
system.cpu2.commit.loads 2941301 # Number of loads committed
system.cpu2.commit.membars 64148 # Number of memory barriers committed
system.cpu2.commit.branches 7305681 # Number of branches committed
system.cpu2.commit.fp_insts 109768 # Number of committed floating point instructions.
system.cpu2.commit.int_insts 30735120 # Number of committed integer instructions.
system.cpu2.commit.function_calls 229363 # Number of function calls committed.
system.cpu2.commit.bw_lim_events 720361 # number cycles where commit BW limit reached
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.commit.bw_limited 0 # number of insts not committed due to BW limits
stats: Update stats for monitor, cache and bus changes This patch removes the sparse histogram total from the CommMonitor stats. It also bumps the stats after the unit fixes in the atomic cache access. Lastly, it updates the stats to match the new port ordering. All numbers are the same, and the only thing that changes is which master corresponds to what port index.
2013-06-27 11:49:51 +02:00
system.cpu2.rob.rob_reads 60649307 # The number of ROB reads
system.cpu2.rob.rob_writes 69356385 # The number of ROB writes
system.cpu2.timesIdled 245741 # Number of times that the entire CPU went into an idle state and unscheduled itself
system.cpu2.idleCycles 3909113 # Total number of cycles that the CPU has spent unscheduled due to idling
system.cpu2.quiesceCycles 1746532644 # Total number of cycles that CPU has spent quiesced or waiting for an interrupt
system.cpu2.committedInsts 31016177 # Number of Instructions Simulated
system.cpu2.committedOps 31016177 # Number of Ops (including micro ops) Simulated
system.cpu2.committedInsts_total 31016177 # Number of Instructions Simulated
system.cpu2.cpi 1.007380 # CPI: Cycles Per Instruction
system.cpu2.cpi_total 1.007380 # CPI: Total CPI of All Threads
system.cpu2.ipc 0.992674 # IPC: Instructions Per Cycle
system.cpu2.ipc_total 0.992674 # IPC: Total IPC of All Threads
system.cpu2.int_regfile_reads 42141472 # number of integer regfile reads
system.cpu2.int_regfile_writes 22438304 # number of integer regfile writes
system.cpu2.fp_regfile_reads 67749 # number of floating regfile reads
system.cpu2.fp_regfile_writes 68082 # number of floating regfile writes
system.cpu2.misc_regfile_reads 5235386 # number of misc regfile reads
system.cpu2.misc_regfile_writes 258296 # number of misc regfile writes
tests: Add CPU switching tests This changeset adds a set of tests that stress the CPU switching code. It adds the following test configurations: * tsunami-switcheroo-full -- Alpha system (atomic, timing, O3) * realview-switcheroo-atomic -- ARM system (atomic<->atomic) * realview-switcheroo-timing -- ARM system (timing<->timing) * realview-switcheroo-o3 -- ARM system (O3<->O3) * realview-switcheroo-full -- ARM system (atomic, timing, O3) Reference data is provided for the 10.linux-boot test case. All of the tests trigger a CPU switch once per millisecond during the boot process. The in-order CPU model was not included in any of the tests as it does not support CPU handover.
2013-01-07 19:05:52 +01:00
system.cpu2.kern.inst.arm 0 # number of arm instructions executed
system.cpu2.kern.inst.quiesce 0 # number of quiesce instructions executed
system.cpu2.kern.inst.hwrei 0 # number of hwrei instructions executed
system.cpu2.kern.mode_switch::kernel 0 # number of protection mode switches
system.cpu2.kern.mode_switch::user 0 # number of protection mode switches
system.cpu2.kern.mode_switch::idle 0 # number of protection mode switches
system.cpu2.kern.mode_good::kernel 0
system.cpu2.kern.mode_good::user 0
system.cpu2.kern.mode_good::idle 0
system.cpu2.kern.mode_switch_good::kernel nan # fraction of useful protection mode switches
system.cpu2.kern.mode_switch_good::user nan # fraction of useful protection mode switches
system.cpu2.kern.mode_switch_good::idle nan # fraction of useful protection mode switches
system.cpu2.kern.mode_switch_good::total nan # fraction of useful protection mode switches
system.cpu2.kern.mode_ticks::kernel 0 # number of ticks spent at the given mode
system.cpu2.kern.mode_ticks::user 0 # number of ticks spent at the given mode
system.cpu2.kern.mode_ticks::idle 0 # number of ticks spent at the given mode
system.cpu2.kern.swap_context 0 # number of times the context was actually changed
---------- End Simulation Statistics ----------
Reference in a new issue Copy permalink