Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo:
- Fixes for problems experienced with new GCC 8 warnings, that treated
as errors, broke the build, related to snprintf and casting issues.
(Arnaldo Carvalho de Melo, Jiri Olsa, Josh Poinboeuf)
- Fix build of new breakpoint 'perf test' entry with clang < 6, noticed
on fedora 25, 26 and 27 (Arnaldo Carvalho de Melo)
- Workaround problem with symbol resolution in 'perf annotate', using
the symbol name already present in the objdump output (Arnaldo Carvalho de Melo)
- Document 'perf top --ignore-vmlinux' (Arnaldo Carvalho de Melo)
- Fix out of bounds access on array fd when cnt is 100 in one of the
'perf test' entries, detected using 'cpptest' (Colin Ian King)
- Add support for the forced leader feature, i.e. 'perf report --group'
for a group of events not really grouped when scheduled (without using
{} to enclose the list of events in the command line) in pipe mode,
e.g.:
+ "PublicDescription": "Instructions can receive, but not send",
+ "EventCode": "0x2014",
+ "EventName": "FETCH_BUBBLE",
+ "BriefDescription": "Instructions can receive, but not send",
+ },
+ {
+ "PublicDescription": "Prefetch request from LSU",
+ "EventCode": "0x6013",
+ "EventName": "PRF_REQ",
+ "BriefDescription": "Prefetch request from LSU",
+ },
+ {
+ "PublicDescription": "Hit on prefetched data",
+ "EventCode": "0x6014",
+ "EventName": "HIT_ON_PRF",
+ "BriefDescription": "Hit on prefetched data",
+ },
+ {
+ "PublicDescription": "Cycles of that the number of issuing micro operations are less than 4",
+ "EventCode": "0x7001",
+ "EventName": "EXE_STALL_CYCLE",
+ "BriefDescription": "Cycles of that the number of issue ups are less than 4",
+ },
+ {
+ "PublicDescription": "No any micro operation is issued and meanwhile any load operation is not resolved",
+ "EventCode": "0x7004",
+ "EventName": "MEM_STALL_ANYLOAD",
+ "BriefDescription": "No any micro operation is issued and meanwhile any load operation is not resolved",
+ },
+ {
+ "PublicDescription": "No any micro operation is issued and meanwhile there is any load operation missing L1 cache and pending data refill",
+ "EventCode": "0x7006",
+ "EventName": "MEM_STALL_L1MISS",
+ "BriefDescription": "No any micro operation is issued and meanwhile there is any load operation missing L1 cache and pending data refill",
+ },
+ {
+ "PublicDescription": "No any micro operation is issued and meanwhile there is any load operation missing both L1 and L2 cache and pending data refill from L3 cache",
+ "EventCode": "0x7007",
+ "EventName": "MEM_STALL_L2MISS",
+ "BriefDescription": "No any micro operation is issued and meanwhile there is any load operation missing both L1 and L2 cache and pending data refill from L3 cache",
"BriefDescription": "Finish stall due to a scalar fixed point or CR instruction in the execution pipeline. These instructions get routed to the ALU, ALU2, and DIV pipes"
},
- {,
- "EventCode": "0x1D15C",
- "EventName": "PM_MRK_DTLB_MISS_1G",
- "BriefDescription": "Marked Data TLB reload (after a miss) page size 2M. Implies radix translation was used"
"BriefDescription": "Duration in cycles to reload with Shared (S) data from another chip's L2 or L3 on the same Node or Group (Remote), as this chip due to a marked load"
},
- {,
- "EventCode": "0x3C056",
- "EventName": "PM_DTLB_MISS_64K",
- "BriefDescription": "Data TLB Miss page size 64K"
"BriefDescription": "Duration in cycles to reload with Modified (M) data from another core's ECO L3 on the same chip due to a marked load"
},
+ {,
+ "EventCode": "0xF0B4",
+ "EventName": "PM_DC_PREF_CONS_ALLOC",
+ "BriefDescription": "Prefetch stream allocated in the conservative phase by either the hardware prefetch mechanism or software prefetch. The sum of this pair subtracted from the total number of allocs will give the total allocs in normal phase"
"BriefDescription": "Prefetch stream allocated by the hardware prefetch mechanism"
},
+ {,
+ "EventCode": "0xF0BC",
+ "EventName": "PM_LS2_UNALIGNED_ST",
+ "BriefDescription": "Store instructions whose data crosses a double-word boundary, which causes it to require an additional slice than than what normally would be required of the Store of that size. If the Store wraps from slice 3 to slice 0, thee is an additional 3-cycle penalty"
"BriefDescription": "Marked instruction was reloaded from a location beyond the local chiplet"
},
+ {,
+ "EventCode": "0x58A8",
+ "EventName": "PM_DECODE_HOLD_ICT_FULL",
+ "BriefDescription": "Counts the number of cycles in which the IFU was not able to decode and transmit one or more instructions because all itags were in use. This means the ICT is full for this thread"
"BriefDescription": "Duration in cycles to reload either shared or modified data from another core's L2/L3 on a different chip (remote or distant) due to a marked load"
},
+ {,
+ "EventCode": "0xF888",
+ "EventName": "PM_LSU1_STORE_REJECT",
+ "BriefDescription": "All internal store rejects cause the instruction to go back to the SRQ and go to sleep until woken up to try again after the condition has been met"
+ },
+ {,
+ "EventCode": "0xC098",
+ "EventName": "PM_LS2_UNALIGNED_LD",
+ "BriefDescription": "Load instructions whose data crosses a double-word boundary, which causes it to require an additional slice than than what normally would be required of the load of that size. If the load wraps from slice 3 to slice 0, thee is an additional 3-cycle penalty"
"BriefDescription": "A data line was written to the L1 due to a hardware or software prefetch"
},
{,
- "EventCode": "0xF888",
- "EventName": "PM_LSU1_STORE_REJECT",
- "BriefDescription": "All internal store rejects cause the instruction to go back to the SRQ and go to sleep until woken up to try again after the condition has been met"
+ "EventCode": "0x2608E",
+ "EventName": "PM_TM_LD_CONF",
+ "BriefDescription": "TM Load (fav or non-fav) ran into conflict (failed)"
"BriefDescription": "XL-form branch was mispredicted due to the predicted target address missing from EAT. The EAT forces a mispredict in this case since there is no predicated target to validate. This is a rare case that may occur when the EAT is full and a branch is issued"
},
+ {,
+ "EventCode": "0xC094",
+ "EventName": "PM_LS0_UNALIGNED_LD",
+ "BriefDescription": "Load instructions whose data crosses a double-word boundary, which causes it to require an additional slice than than what normally would be required of the load of that size. If the load wraps from slice 3 to slice 0, thee is an additional 3-cycle penalty"
+ },
+ {,
+ "EventCode": "0xF8BC",
+ "EventName": "PM_LS3_UNALIGNED_ST",
+ "BriefDescription": "Store instructions whose data crosses a double-word boundary, which causes it to require an additional slice than than what normally would be required of the Store of that size. If the Store wraps from slice 3 to slice 0, thee is an additional 3-cycle penalty"
- "BriefDescription": "All I-or-D side load dispatch attempts for this thread that failed due to reason other than address collision (excludes i_l2mru_tch_reqs)"
+ "BriefDescription": "All D-side-Ld or I-side-instruction-fetch dispatch attempts for this thread that failed due to reasons other than an address collision conflicts with an L2 machines (e.g. Read-Claim/Snoop machine not available)"
- "BriefDescription": "All successful Ld/St dispatches for this thread that were an L2 miss (excludes i_l2mru_tch_reqs)"
+ "BriefDescription": "All successful D-side-Ld/St or I-side-instruction-fetch dispatches for this thread that were an L2 miss"
},
{,
- "EventCode": "0x160A0",
- "EventName": "PM_L3_PF_MISS_L3",
- "BriefDescription": "L3 PF missed in L3"
+ "EventCode": "0xF8B8",
+ "EventName": "PM_LS1_UNALIGNED_ST",
+ "BriefDescription": "Store instructions whose data crosses a double-word boundary, which causes it to require an additional slice than than what normally would be required of the Store of that size. If the Store wraps from slice 3 to slice 0, thee is an additional 3-cycle penalty"
- "BriefDescription": "All I-od-D side load dispatch attempts for this thread that failed due to address collision with RC/CO/SN/SQ machine (excludes i_l2mru_tch_reqs)"
+ "BriefDescription": "All D-side-Ld or I-side-instruction-fetch dispatch attempts for this thread that failed due to an address collision conflicts with an L2 machines already working on this line (e.g. ld-hit-stq or Read-claim/Castout/Snoop machines)"
+ "BriefDescription": "Load instructions whose data crosses a double-word boundary, which causes it to require an additional slice than than what normally would be required of the load of that size. If the load wraps from slice 3 to slice 0, thee is an additional 3-cycle penalty"
- "BriefDescription": "The LSU requested a line from L2 for translation. It may be satisfied from any source beyond L2. Includes speculative instructions"
+ "BriefDescription": "The LSU requested a line from L2 for translation. It may be satisfied from any source beyond L2. Includes speculative instructions. Includes instruction, prefetch and demand"
- "BriefDescription": "All I-side dispatch attempts for this thread that failed due to a addr collision with another machine (excludes i_l2mru_tch_reqs)"
+ "BriefDescription": "All I-side-instruction-fetch dispatch attempts for this thread that failed due to an address collision conflict with an L2 machine already working on this line (e.g. ld-hit-stq or RC/CO/SN machines)"
"BriefDescription": "Ict empty for this thread due to dispatch holds because the History Buffer was full. Could be GPR/VSR/VMR/FPR/CR/XVF; CR; XVF (XER/VSCR/FPSCR)"
},
+ {,
+ "EventCode": "0xC894",
+ "EventName": "PM_LS1_UNALIGNED_LD",
+ "BriefDescription": "Load instructions whose data crosses a double-word boundary, which causes it to require an additional slice than than what normally would be required of the load of that size. If the load wraps from slice 3 to slice 0, thee is an additional 3-cycle penalty"
"BriefDescription": "Continuous 16 cycle (2to1) window where this signals rotates thru sampling each SN machine busy. PMU uses this wave to then do 16 cyc count to sample total number of machs running"
},
+ {,
+ "EventCode": "0x36084",
+ "EventName": "PM_L2_RCST_DISP",
+ "BriefDescription": "All D-side store dispatch attempts for this thread"
"BriefDescription": "A demand load referenced a line in an active strided prefetch stream. The stream could have been allocated through the hardware prefetch mechanism or through software."
},
- {,
- "EventCode": "0x36084",
- "EventName": "PM_L2_RCST_DISP",
- "BriefDescription": "All D-side store dispatch attempts for this thread"
- "BriefDescription": "Counts the number of cycles in which the IFU was not able to decode and transmit one or more instructions because all itags were in use. This means the ICT is full for this thread"
- "BriefDescription": "All I-side dispatch attempts for this thread that failed due to a reason other than addrs collision (excludes i_l2mru_tch_reqs)"
+ "BriefDescription": "All I-side-instruction-fetch dispatch attempts for this thread that failed due to reasons other than an address collision conflict with an L2 machine (e.g. no available RC/CO machines)"
"BriefDescription": "Conditional Branch Completed in which the HW correctly predicted the direction as taken. Counted at completion time"
},
+ {,
+ "EventCode": "0xF0B8",
+ "EventName": "PM_LS0_UNALIGNED_ST",
+ "BriefDescription": "Store instructions whose data crosses a double-word boundary, which causes it to require an additional slice than than what normally would be required of the Store of that size. If the Store wraps from slice 3 to slice 0, thee is an additional 3-cycle penalty"
"BriefDescription": "A Page Table Entry was loaded into the TLB from local core's L3 without dispatch conflicts hit on Mepf state. due to a marked data side request. When using Radix Page Translation, this count excludes PDE reloads. Only PTE reloads are included"
},
+ {,
+ "EventCode": "0x4C15C",
+ "EventName": "PM_MRK_DERAT_MISS_16G_1G",
+ "BriefDescription": "Marked Data ERAT Miss (Data TLB Access) page size 16G (hpt mode) and 1G (radix mode)"
"BriefDescription": "A Page Table Entry was loaded into the TLB from local core's L2 without conflict due to a marked data side request. When using Radix Page Translation, this count excludes PDE reloads. Only PTE reloads are included"
},
- {,
- "EventCode": "0x2C05A",
- "EventName": "PM_DERAT_MISS_1G",
- "BriefDescription": "Data ERAT Miss (Data TLB Access) page size 1G. Implies radix translation"
"BriefDescription": "Finish stall because the NTF instruction was issued to the Decimal Floating Point execution pipe and waiting to finish. Includes decimal floating point instructions + 128 bit binary floating point instructions. Not qualified by multicycle"
},
+ {,
+ "EventCode": "0x3C054",
+ "EventName": "PM_DERAT_MISS_16M_2M",
+ "BriefDescription": "Data ERAT Miss (Data TLB Access) page size 16M (HPT mode) or 2M (Radix mode)"
"BriefDescription": "The processor's Instruction cache was reloaded from a memory location including L4 from local remote or distant due to an instruction fetch (not prefetch)"
},
- {,
- "EventCode": "0x1C05A",
- "EventName": "PM_DERAT_MISS_2M",
- "BriefDescription": "Data ERAT Miss (Data TLB Access) page size 2M. Implies radix translation"
"BriefDescription": "A Page Table Entry was loaded into the TLB with Modified (M) data from another core's L3 on the same chip due to a marked data side request. When using Radix Page Translation, this count excludes PDE reloads. Only PTE reloads are included"
},
- {,
- "EventCode": "0x4C15C",
- "EventName": "PM_MRK_DERAT_MISS_16G",
- "BriefDescription": "Marked Data ERAT Miss (Data TLB Access) page size 16G"