92 lines
4.6 KiB
Plaintext
92 lines
4.6 KiB
Plaintext
How to profile R4300 instructions with mupen64plus:
|
|
|
|
Pre-requisites:
|
|
- either 32-bit (x86) or 64-bit (amd64) Linux system
|
|
- Lots of memory. It doesn't cause swapping in a machine with 2GB of ram, but it probably would with 1GB
|
|
- OProfile (http://oprofile.sourceforge.net/)
|
|
- Mupen64plus source code
|
|
|
|
Procedure:
|
|
1. Install OProfile linux tool
|
|
|
|
2. Build r4300 profile tool with "gcc -o r4300prof r4300prof.c"
|
|
|
|
3. Build mupen64plus with "make DBGSYM=1 DBG_PROFILE=1"
|
|
|
|
4. Delete any pre-existing profiling files: "rm instructionaddrs.dat"
|
|
|
|
5. Clear any residual profiling data in oprofile: "sudo opcontrol --reset"
|
|
|
|
6. Make profiling run by typing in console:
|
|
sudo opcontrol --start ; ./mupen64plus --nogui --emumode 2 --audio ./plugins/dummyaudio.so <path-to-n64-rom> ; sudo opcontrol --stop
|
|
|
|
7. Exit emulator with Escape key after running for desired time
|
|
|
|
7. Move data file into tools folder: "mv instructionaddrs.dat tools/"
|
|
|
|
8. Dump instruction-level profiling data for mupen64plus with oprofile:
|
|
opreport -d -l ./mupen64plus > ./tools/prof-mupen64-detail.txt
|
|
|
|
9. Run tool to generate r4300 instruction profile report:
|
|
cd tools ; ./r4300prof instructionaddrs.dat prof-mupen64-detail.txt
|
|
|
|
|
|
Example profile output:
|
|
|
|
Loading instructionaddrs.dat...
|
|
283844 r4300 instruction locations read.
|
|
247911 non-empty MIPS instructions.
|
|
Loading prof-mupen64-detail.txt...
|
|
118181 lines in sample data file.
|
|
Found 117905 profile hits.
|
|
|
|
Instruction time (samples):
|
|
reserved: 00007515 NI: 00000000 J: 00000043 JAL: 00003345 BEQ: 00004498
|
|
BNE: 00003424 BLEZ: 00000222 BGTZ: 00000029 ADDI: 00000585 ADDIU: 00017439
|
|
SLTI: 00000765 SLTIU: 00000089 ANDI: 00003847 ORI: 00000381 XORI: 00000035
|
|
LUI: 00010389 BEQL: 00001873 BNEL: 00002617 BLEZL: 00000013 BGTZL: 00000010
|
|
DADDI: 00000000 DADDIU: 00000000 LDL: 00000000 LDR: 00000000 LB: 00002600
|
|
LH: 00006653 LW: 00024840 LWL: 00000001 LBU: 00003525 LHU: 00004954
|
|
LWU: 00000000 LWR: 00000003 SB: 00003007 SH: 00004133 SW: 00023221
|
|
SWL: 00000000 SWR: 00000000 SDL: 00000000 SDR: 00000000 LWC1: 00012405
|
|
LDC1: 00001855 LD: 00000815 LL: 00000000 SWC1: 00007671 SDC1: 00001326
|
|
SD: 00001233 SC: 00000000 BLTZ: 00000383 BGEZ: 00000331 BLTZL: 00000161
|
|
BGEZL: 00000168 BLTZAL: 00000000 BGEZAL: 00000115 BLTZALL: 00000000 BGEZALL: 00000000
|
|
SLL: 00003376 SRL: 00000604 SRA: 00000686 SLLV: 00000015 SRLV: 00000039
|
|
SRAV: 00000000 JR: 00001358 JALR: 00000004 SYSCALL: 00000000 MFHI: 00000058
|
|
MTHI: 00000002 MFLO: 00000277 MTLO: 00000004 DSLLV: 00000000 DSRLV: 00000000
|
|
DSRAV: 00000000 MULT: 00000000 MULTU: 00000583 DIV: 00000941 DIVU: 00000000
|
|
DMULT: 00000000 DMULTU: 00000000 DDIV: 00000000 DDIVU: 00000000 ADD: 00000170
|
|
ADDU: 00001706 SUB: 00000037 SUBU: 00000833 AND: 00000909 OR: 00006360
|
|
XOR: 00000079 NOR: 00000004 SLT: 00000925 SLTU: 00001401 DADD: 00000000
|
|
DADDU: 00000000 DSUB: 00000000 DSUBU: 00000000 DSLL: 00000000 DSRL: 00000000
|
|
DSRA: 00000000 TEQ: 00000000 DSLL32: 00000000 DSRL32: 00000000 DSRA32: 00000000
|
|
BC1F: 00000286 BC1T: 00000032 BC1FL: 00000775 BC1TL: 00000077 TLBWI: 00000000
|
|
TLBP: 00000000 TLBR: 00000000 TLBWR: 00000000 ERET: 00000036 MFC0: 00000396
|
|
MTC0: 00000034 MFC1: 00000929 DMFC1: 00000000 CFC1: 00000141 MTC1: 00003081
|
|
DMTC1: 00000000 CTC1: 00000163 f.CVT: 00001113 f.CMP: 00002149 f.ADD: 00001387
|
|
f.SUB: 00000947 f.MUL: 00002696 f.DIV: 00000315 f.SQRT: 00000025 f.ABS: 00000000
|
|
f.MOV: 00000631 f.NEG: 00000323 f.ROUND: 00000000 f.TRUNC: 00000867 f.CEIL: 00000000
|
|
f.FLOOR: 00000000
|
|
|
|
Special code samples:
|
|
Regcache flushing: 12371
|
|
Jump wrappers: 15520
|
|
NOTCOMPILED: 33
|
|
block postfix & link samples: 619
|
|
|
|
Unaccounted samples: 19929
|
|
Total accounted instruction samples: 221836
|
|
Load: 35.2% (68040)
|
|
Store: 21.0% (40591)
|
|
Data move/convert: 03.5% (6829)
|
|
32-bit math: 16.0% (30861)
|
|
64-bit math: 05.7% (10948)
|
|
Float Math: 04.5% (8709)
|
|
Jump: 02.5% (4750)
|
|
Branch: 07.8% (15014)
|
|
Exceptions: 00.0% (36)
|
|
Reserved: 03.9% (7515)
|
|
Other: 00.0% (0)
|
|
|