PERF-AMD-IBS(1) perf Manual PERF-AMD-IBS(1) (NAME) perf-amd-ibs - AMD (IBS) perf (SYNOPSIS) perf record -e ibs_op// perf record -e ibs_fetch// (DESCRIPTION) (IBS Instruction-Based Sampling) () (IP) AMD . IBS : IBS Op IBS Fetch. IBS Op ( micro-op) / d-cache / d-TLB / (load/store) (branch behavior) . IBS Fetch / i-cache / i-TLB . IBS SMT (per-smt-thread) SMT IBS . IBS Op IBS Fetch PMU perf . IBS (boot) . /sys/bus/event_source/devices/ibs_op/ /sys/bus/event_source/devices/ibs_fetch/ IBS Op PMU : cycles micro ops. IBS Fetch PMU : fetch ops. IBS PMU / (user/kernel filtering) CAP_SYS_ADMIN CAP_PERFMON . IBS PMU (IBS VS. REGULAR CORE PMU) IBS IP IP IBS (skid) IP PMU ( IP X perf IP X+n ). PMU . IBS . PMU ( ) . core PMU precise_ip IBS Op PMU : -e cpu-cycles:p -e ibs_op// -e r076:p -e ibs_op// -e r0C1:p -e ibs_op/cnt_ctl=1/ (EXAMPLES) IBS Op PMU cycles : 100000 # perf record -e ibs_op// -c 100000 -a (cpu10) cycles : 100000 # perf record -e ibs_op// -c 100000 -C 10 (cpu10) cycles : 100000 Zen6 ( ): # perf record -e ibs_op//u -c 100000 -C 10 Zen5: # perf record -e ibs_op/swfilt=1/u -c 100000 -C 10 (cpu10) cycles : 1000 # perf record -e ibs_op// -F 1000 -C 10 uOps : 100000 # perf record -e ibs_op/cnt_ctl=1/ -c 100000 -a IBS perf: # perf record -e ibs_op/cnt_ctl=1/ -c 100000 -a --raw-samples uOps : 100000 L3MissOnly ( Zen4 ) # perf record -e ibs_op/cnt_ctl=1,l3missonly=1/ -c 100000 -a cycles : 100000 LdLat ( Zen5 ) # perf record -e ibs_op/ldlat=128/ -c 100000 -a (load latency) 128 2048 ( ) . 128 (overhead) . cycles : 100000 streaming store ( Zen6 ) # perf record -e ibs_op/strmst=1/ -c 100000 -a (upstream v6.2 ) uOps : 100000 # perf record -e ibs_op/cnt_ctl=1/ -c 100000 -p 1234 (upstream v6.2 ) uOps : 100000 # perf record -e ibs_op/cnt_ctl=1/ -c 100000 -- ls # perf report /* 'a' . */ # perf script IBS --raw-samples # perf report -D /* PERF_RECORD_SAMPLE */ : ibs_op_ctl: 000002c30006186a MaxCnt 100000 L3MissOnly 0 En 1 Val 1 CntCtl 0=cycles CurCnt 707 IbsOpRip: ffffffff8204aea7 ibs_op_data: 0000010002550001 CompToRetCtr 1 TagToRetCtr 597 BrnRet 0 RipInvalid 0 BrnFuse 0 Microcode 1 ibs_op_data2: 0000000000000013 RmtNode 1 DataSrc 3=DRAM ibs_op_data3: 0000000031960092 LdOp 0 StOp 1 DcL1TlbMiss 0 DcL2TlbMiss 0 DcL1TlbHit2M 1 DcL1TlbHit1G 0 DcL2TlbHit2M 0 DcMiss 1 DcMisAcc 0 DcWcMemAcc 0 DcUcMemAcc 0 DcLockedOp 0 DcMissNoMabAlloc 0 DcLinAddrValid 1 DcPhyAddrValid 1 DcL2TlbHit1G 0 L2Miss 1 SwPf 0 OpMemWidth 32 bytes OpDcMissOpenMemReqs 12 DcMissLat 0 TlbRefillLat 0 IbsDCLinAd: ff110008a5398920 IbsDCPhysAd: 00000008a5398920 IBS 90% tbench scheduler hint . IBS perf : https://lore.kernel.org/r/20220921063638.2489-1-kprateek.nayak@amd.com Fetch PMU . fetch ops : 100000 # perf record -e ibs_fetch// -c 100000 -a fetch ops : 100000 Zen6 ( NOTES ): # perf record -e ibs_fetch//u -c 100000 -a Zen5: # perf record -e ibs_fetch/swfilt=1/u -c 100000 -a fetch ops : 100000 Random enable # perf record -e ibs_fetch/rand_en=1/ -c 100000 -a Random enable . PMU . fetch ops : 10000 ( Zen6 ) # perf record -e ibs_fetch/fetchlat=128/ -c 10000 -a 128 1920 ( ) . 128 . ..SH "PERF MEM PERF C2C (PERF MEM AND PERF C2C)" perf mem () perf c2c (cacheline) . AMD IBS Op PMU . perf mem . # perf mem record -c 100000 -- make # perf mem report perf mem report . . : # perf mem report -F overhead,cache,snoop,comm ... # Samples: 92K of event 'ibs_op//' # Total weight : 531104 # # ---------- Cache ----------- --- Snoop ---- # Overhead L1 L2 L1-buf Other HitM Other Command # ........ ............................ .............. .......... # 76.07% 5.8% 35.7% 0.0% 34.6% 23.3% 52.8% cc1 5.79% 0.2% 0.0% 0.0% 5.6% 0.1% 5.7% make 5.78% 0.1% 4.4% 0.0% 1.2% 0.5% 5.3% gcc 5.33% 0.3% 3.9% 0.0% 1.1% 0.2% 5.2% as 5.00% 0.1% 3.8% 0.0% 1.0% 0.3% 4.7% sh 1.56% 0.1% 0.1% 0.0% 1.4% 0.6% 0.9% ld 0.28% 0.1% 0.0% 0.0% 0.2% 0.1% 0.2% pkg-config 0.09% 0.0% 0.0% 0.0% 0.1% 0.0% 0.1% git 0.03% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% rm ... . : # perf mem report -s mem,snoop ... # Samples: 92K of event 'ibs_op//' # Total weight : 531104 # Sort order : mem,snoop # # Overhead Samples Memory access Snoop # ........ ............ ....................................... ............ # 47.99% 1509 L2 hit N/A 25.08% 338 core, same node Any cache hit HitM 10.24% 54374 N/A N/A 6.77% 35938 L1 hit N/A 6.39% 101 core, same node Any cache hit N/A 3.50% 69 RAM hit N/A 0.03% 158 LFB/MAB hit N/A 0.00% 2 Uncached hit N/A (man page) . (NOTES) (kernel) (userspace) . 32-bit . IBS (host) 64-bit (guest) 32-bit IBS 64-bit 64-bit . (SEE ALSO) perf-record(1), perf-script(1), perf-report(1), perf-mem(1), perf- c2c(1) perf 09/04/2026 PERF-AMD-IBS(1)