1llvm-exegesis - LLVM Machine Instruction Benchmark 2================================================== 3 4.. program:: llvm-exegesis 5 6SYNOPSIS 7-------- 8 9:program:`llvm-exegesis` [*options*] 10 11DESCRIPTION 12----------- 13 14:program:`llvm-exegesis` is a benchmarking tool that uses information available 15in LLVM to measure host machine instruction characteristics like latency, 16throughput, or port decomposition. 17 18Given an LLVM opcode name and a benchmarking mode, :program:`llvm-exegesis` 19generates a code snippet that makes execution as serial (resp. as parallel) as 20possible so that we can measure the latency (resp. inverse throughput/uop decomposition) 21of the instruction. 22The code snippet is jitted and executed on the host subtarget. The time taken 23(resp. resource usage) is measured using hardware performance counters. The 24result is printed out as YAML to the standard output. 25 26The main goal of this tool is to automatically (in)validate the LLVM's TableDef 27scheduling models. To that end, we also provide analysis of the results. 28 29:program:`llvm-exegesis` can also benchmark arbitrary user-provided code 30snippets. 31 32EXAMPLE 1: benchmarking instructions 33------------------------------------ 34 35Assume you have an X86-64 machine. To measure the latency of a single 36instruction, run: 37 38.. code-block:: bash 39 40 $ llvm-exegesis -mode=latency -opcode-name=ADD64rr 41 42Measuring the uop decomposition or inverse throughput of an instruction works similarly: 43 44.. code-block:: bash 45 46 $ llvm-exegesis -mode=uops -opcode-name=ADD64rr 47 $ llvm-exegesis -mode=inverse_throughput -opcode-name=ADD64rr 48 49 50The output is a YAML document (the default is to write to stdout, but you can 51redirect the output to a file using `-benchmarks-file`): 52 53.. code-block:: none 54 55 --- 56 key: 57 opcode_name: ADD64rr 58 mode: latency 59 config: '' 60 cpu_name: haswell 61 llvm_triple: x86_64-unknown-linux-gnu 62 num_repetitions: 10000 63 measurements: 64 - { key: latency, value: 1.0058, debug_string: '' } 65 error: '' 66 info: 'explicit self cycles, selecting one aliasing configuration. 67 Snippet: 68 ADD64rr R8, R8, R10 69 ' 70 ... 71 72To measure the latency of all instructions for the host architecture, run: 73 74.. code-block:: bash 75 76 #!/bin/bash 77 readonly INSTRUCTIONS=$(($(grep INSTRUCTION_LIST_END build/lib/Target/X86/X86GenInstrInfo.inc | cut -f2 -d=) - 1)) 78 for INSTRUCTION in $(seq 1 ${INSTRUCTIONS}); 79 do 80 ./build/bin/llvm-exegesis -mode=latency -opcode-index=${INSTRUCTION} | sed -n '/---/,$p' 81 done 82 83FIXME: Provide an :program:`llvm-exegesis` option to test all instructions. 84 85 86EXAMPLE 2: benchmarking a custom code snippet 87--------------------------------------------- 88 89To measure the latency/uops of a custom piece of code, you can specify the 90`snippets-file` option (`-` reads from standard input). 91 92.. code-block:: bash 93 94 $ echo "vzeroupper" | llvm-exegesis -mode=uops -snippets-file=- 95 96Real-life code snippets typically depend on registers or memory. 97:program:`llvm-exegesis` checks the liveliness of registers (i.e. any register 98use has a corresponding def or is a "live in"). If your code depends on the 99value of some registers, you have two options: 100 101- Mark the register as requiring a definition. :program:`llvm-exegesis` will 102 automatically assign a value to the register. This can be done using the 103 directive `LLVM-EXEGESIS-DEFREG <reg name> <hex_value>`, where `<hex_value>` 104 is a bit pattern used to fill `<reg_name>`. If `<hex_value>` is smaller than 105 the register width, it will be sign-extended. 106- Mark the register as a "live in". :program:`llvm-exegesis` will benchmark 107 using whatever value was in this registers on entry. This can be done using 108 the directive `LLVM-EXEGESIS-LIVEIN <reg name>`. 109 110For example, the following code snippet depends on the values of XMM1 (which 111will be set by the tool) and the memory buffer passed in RDI (live in). 112 113.. code-block:: none 114 115 # LLVM-EXEGESIS-LIVEIN RDI 116 # LLVM-EXEGESIS-DEFREG XMM1 42 117 vmulps (%rdi), %xmm1, %xmm2 118 vhaddps %xmm2, %xmm2, %xmm3 119 addq $0x10, %rdi 120 121 122EXAMPLE 3: analysis 123------------------- 124 125Assuming you have a set of benchmarked instructions (either latency or uops) as 126YAML in file `/tmp/benchmarks.yaml`, you can analyze the results using the 127following command: 128 129.. code-block:: bash 130 131 $ llvm-exegesis -mode=analysis \ 132 -benchmarks-file=/tmp/benchmarks.yaml \ 133 -analysis-clusters-output-file=/tmp/clusters.csv \ 134 -analysis-inconsistencies-output-file=/tmp/inconsistencies.html 135 136This will group the instructions into clusters with the same performance 137characteristics. The clusters will be written out to `/tmp/clusters.csv` in the 138following format: 139 140.. code-block:: none 141 142 cluster_id,opcode_name,config,sched_class 143 ... 144 2,ADD32ri8_DB,,WriteALU,1.00 145 2,ADD32ri_DB,,WriteALU,1.01 146 2,ADD32rr,,WriteALU,1.01 147 2,ADD32rr_DB,,WriteALU,1.00 148 2,ADD32rr_REV,,WriteALU,1.00 149 2,ADD64i32,,WriteALU,1.01 150 2,ADD64ri32,,WriteALU,1.01 151 2,MOVSX64rr32,,BSWAP32r_BSWAP64r_MOVSX64rr32,1.00 152 2,VPADDQYrr,,VPADDBYrr_VPADDDYrr_VPADDQYrr_VPADDWYrr_VPSUBBYrr_VPSUBDYrr_VPSUBQYrr_VPSUBWYrr,1.02 153 2,VPSUBQYrr,,VPADDBYrr_VPADDDYrr_VPADDQYrr_VPADDWYrr_VPSUBBYrr_VPSUBDYrr_VPSUBQYrr_VPSUBWYrr,1.01 154 2,ADD64ri8,,WriteALU,1.00 155 2,SETBr,,WriteSETCC,1.01 156 ... 157 158:program:`llvm-exegesis` will also analyze the clusters to point out 159inconsistencies in the scheduling information. The output is an html file. For 160example, `/tmp/inconsistencies.html` will contain messages like the following : 161 162.. image:: llvm-exegesis-analysis.png 163 :align: center 164 165Note that the scheduling class names will be resolved only when 166:program:`llvm-exegesis` is compiled in debug mode, else only the class id will 167be shown. This does not invalidate any of the analysis results though. 168 169OPTIONS 170------- 171 172.. option:: -help 173 174 Print a summary of command line options. 175 176.. option:: -opcode-index=<LLVM opcode index> 177 178 Specify the opcode to measure, by index. Specifying `-1` will result 179 in measuring every existing opcode. See example 1 for details. 180 Either `opcode-index`, `opcode-name` or `snippets-file` must be set. 181 182.. option:: -opcode-name=<opcode name 1>,<opcode name 2>,... 183 184 Specify the opcode to measure, by name. Several opcodes can be specified as 185 a comma-separated list. See example 1 for details. 186 Either `opcode-index`, `opcode-name` or `snippets-file` must be set. 187 188.. option:: -snippets-file=<filename> 189 190 Specify the custom code snippet to measure. See example 2 for details. 191 Either `opcode-index`, `opcode-name` or `snippets-file` must be set. 192 193.. option:: -mode=[latency|uops|inverse_throughput|analysis] 194 195 Specify the run mode. Note that some modes have additional requirements and options. 196 197 `latency` mode can be make use of either RDTSC or LBR. 198 `latency[LBR]` is only available on X86 (at least `Skylake`). 199 To run in `latency` mode, a positive value must be specified for `x86-lbr-sample-period` and `--repetition-mode=loop`. 200 201 In `analysis` mode, you also need to specify at least one of the 202 `-analysis-clusters-output-file=` and `-analysis-inconsistencies-output-file=`. 203 204.. option:: -x86-lbr-sample-period=<nBranches/sample> 205 206 Specify the LBR sampling period - how many branches before we take a sample. 207 When a positive value is specified for this option and when the mode is `latency`, 208 we will use LBRs for measuring. 209 On choosing the "right" sampling period, a small value is preferred, but throttling 210 could occur if the sampling is too frequent. A prime number should be used to 211 avoid consistently skipping certain blocks. 212 213.. option:: -repetition-mode=[duplicate|loop|min] 214 215 Specify the repetition mode. `duplicate` will create a large, straight line 216 basic block with `num-repetitions` copies of the snippet. `loop` will wrap 217 the snippet in a loop which will be run `num-repetitions` times. The `loop` 218 mode tends to better hide the effects of the CPU frontend on architectures 219 that cache decoded instructions, but consumes a register for counting 220 iterations. If performing an analysis over many opcodes, it may be best 221 to instead use the `min` mode, which will run each other mode, and produce 222 the minimal measured result. 223 224.. option:: -num-repetitions=<Number of repetitions> 225 226 Specify the number of repetitions of the asm snippet. 227 Higher values lead to more accurate measurements but lengthen the benchmark. 228 229.. option:: -max-configs-per-opcode=<value> 230 231 Specify the maximum configurations that can be generated for each opcode. 232 By default this is `1`, meaning that we assume that a single measurement is 233 enough to characterize an opcode. This might not be true of all instructions: 234 for example, the performance characteristics of the LEA instruction on X86 235 depends on the value of assigned registers and immediates. Setting a value of 236 `-max-configs-per-opcode` larger than `1` allows `llvm-exegesis` to explore 237 more configurations to discover if some register or immediate assignments 238 lead to different performance characteristics. 239 240 241.. option:: -benchmarks-file=</path/to/file> 242 243 File to read (`analysis` mode) or write (`latency`/`uops`/`inverse_throughput` 244 modes) benchmark results. "-" uses stdin/stdout. 245 246.. option:: -analysis-clusters-output-file=</path/to/file> 247 248 If provided, write the analysis clusters as CSV to this file. "-" prints to 249 stdout. By default, this analysis is not run. 250 251.. option:: -analysis-inconsistencies-output-file=</path/to/file> 252 253 If non-empty, write inconsistencies found during analysis to this file. `-` 254 prints to stdout. By default, this analysis is not run. 255 256.. option:: -analysis-clustering=[dbscan,naive] 257 258 Specify the clustering algorithm to use. By default DBSCAN will be used. 259 Naive clustering algorithm is better for doing further work on the 260 `-analysis-inconsistencies-output-file=` output, it will create one cluster 261 per opcode, and check that the cluster is stable (all points are neighbours). 262 263.. option:: -analysis-numpoints=<dbscan numPoints parameter> 264 265 Specify the numPoints parameters to be used for DBSCAN clustering 266 (`analysis` mode, DBSCAN only). 267 268.. option:: -analysis-clustering-epsilon=<dbscan epsilon parameter> 269 270 Specify the epsilon parameter used for clustering of benchmark points 271 (`analysis` mode). 272 273.. option:: -analysis-inconsistency-epsilon=<epsilon> 274 275 Specify the epsilon parameter used for detection of when the cluster 276 is different from the LLVM schedule profile values (`analysis` mode). 277 278.. option:: -analysis-display-unstable-clusters 279 280 If there is more than one benchmark for an opcode, said benchmarks may end up 281 not being clustered into the same cluster if the measured performance 282 characteristics are different. by default all such opcodes are filtered out. 283 This flag will instead show only such unstable opcodes. 284 285.. option:: -ignore-invalid-sched-class=false 286 287 If set, ignore instructions that do not have a sched class (class idx = 0). 288 289.. option:: -mcpu=<cpu name> 290 291 If set, measure the cpu characteristics using the counters for this CPU. This 292 is useful when creating new sched models (the host CPU is unknown to LLVM). 293 294.. option:: --dump-object-to-disk=true 295 296 By default, llvm-exegesis will dump the generated code to a temporary file to 297 enable code inspection. You may disable it to speed up the execution and save 298 disk space. 299 300EXIT STATUS 301----------- 302 303:program:`llvm-exegesis` returns 0 on success. Otherwise, an error message is 304printed to standard error, and the tool returns a non 0 value. 305
