mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-25 20:23:11 +01:00
5d534d8259
I really needed this, like, factually, yesterday,
when verifying dependency breaking idioms for AMD Zen 3 scheduler model.
Consider the following example:
```
$ ./bin/llvm-exegesis --mode=inverse_throughput --snippets-file=/tmp/snippet.s --num-repetitions=1000000 --repetition-mode=duplicate
Check generated assembly with: /usr/bin/objdump -d /tmp/snippet-4a7e50.o
---
mode: inverse_throughput
key:
instructions:
- 'VPXORYrr YMM0 YMM0 YMM0'
config: ''
register_initial_values: []
cpu_name: znver3
llvm_triple: x86_64-unknown-linux-gnu
num_repetitions: 1000000
measurements:
- { key: inverse_throughput, value: 0.31025, per_snippet_value: 0.31025 }
error: ''
info: ''
assembled_snippet: C5FDEFC0C5FDEFC0C5FDEFC0C5FDEFC0C5FDEFC0C5FDEFC0C5FDEFC0C5FDEFC0C5FDEFC0C5FDEFC0C5FDEFC0C5FDEFC0C5FDEFC0C5FDEFC0C5FDEFC0C5FDEFC0C3
...
```
What does it tell us?
So wait, it can only execute ~3 x86 AVX YMM PXOR zero-idioms per cycle?
That doesn't seem right. That's even less than there are pipes supporting this type of op.
Now, second example:
```
$ ./bin/llvm-exegesis --mode=inverse_throughput --snippets-file=/tmp/snippet.s --num-repetitions=1000000 --repetition-mode=loop
Check generated assembly with: /usr/bin/objdump -d /tmp/snippet-2418b5.o
---
mode: inverse_throughput
key:
instructions:
- 'VPXORYrr YMM0 YMM0 YMM0'
config: ''
register_initial_values: []
cpu_name: znver3
llvm_triple: x86_64-unknown-linux-gnu
num_repetitions: 1000000
measurements:
- { key: inverse_throughput, value: 1.00011, per_snippet_value: 1.00011 }
error: ''
info: ''
assembled_snippet: 49B80800000000000000C5FDEFC0C5FDEFC04983C0FF75F2C3
...
```
Now that's just worse. Due to the looping, the throughput completely plummeted,
and now we can only do a single instruction/cycle!?
That's not great.
And final example:
```
$ ./bin/llvm-exegesis --mode=inverse_throughput --snippets-file=/tmp/snippet.s --num-repetitions=1000000 --repetition-mode=loop --loop-body-size=1000
Check generated assembly with: /usr/bin/objdump -d /tmp/snippet-c402e2.o
---
mode: inverse_throughput
key:
instructions:
- 'VPXORYrr YMM0 YMM0 YMM0'
config: ''
register_initial_values: []
cpu_name: znver3
llvm_triple: x86_64-unknown-linux-gnu
num_repetitions: 1000000
measurements:
- { key: inverse_throughput, value: 0.167087, per_snippet_value: 0.167087 }
error: ''
info: ''
assembled_snippet: 49B80800000000000000C5FDEFC0C5FDEFC04983C0FF75F2C3
...
```
So if we merge the previous two approaches, do duplicate this single-instruction snippet 1000x
(loop-body-size/instruction count in snippet), and run a loop with 1000 iterations
over that duplicated/unrolled snippet, the measured throughput goes through the roof,
up to 5.9 instructions/cycle, which finally tells us that this idiom is zero-cycle!
Reviewed By: courbet
Differential Revision: https://reviews.llvm.org/D102522
|
||
|---|---|---|
| .. | ||
| bugpoint | ||
| bugpoint-passes | ||
| dsymutil | ||
| gold | ||
| llc | ||
| lli | ||
| llvm-ar | ||
| llvm-as | ||
| llvm-as-fuzzer | ||
| llvm-bcanalyzer | ||
| llvm-c-test | ||
| llvm-cat | ||
| llvm-cfi-verify | ||
| llvm-config | ||
| llvm-cov | ||
| llvm-cvtres | ||
| llvm-cxxdump | ||
| llvm-cxxfilt | ||
| llvm-cxxmap | ||
| llvm-diff | ||
| llvm-dis | ||
| llvm-dwarfdump | ||
| llvm-dwp | ||
| llvm-elfabi | ||
| llvm-exegesis | ||
| llvm-extract | ||
| llvm-go | ||
| llvm-gsymutil | ||
| llvm-ifs | ||
| llvm-isel-fuzzer | ||
| llvm-itanium-demangle-fuzzer | ||
| llvm-jitlink | ||
| llvm-jitlistener | ||
| llvm-libtool-darwin | ||
| llvm-link | ||
| llvm-lipo | ||
| llvm-lto | ||
| llvm-lto2 | ||
| llvm-mc | ||
| llvm-mc-assemble-fuzzer | ||
| llvm-mc-disassemble-fuzzer | ||
| llvm-mca | ||
| llvm-microsoft-demangle-fuzzer | ||
| llvm-ml | ||
| llvm-modextract | ||
| llvm-mt | ||
| llvm-nm | ||
| llvm-objcopy | ||
| llvm-objdump | ||
| llvm-opt-fuzzer | ||
| llvm-opt-report | ||
| llvm-pdbutil | ||
| llvm-profdata | ||
| llvm-profgen | ||
| llvm-rc | ||
| llvm-readobj | ||
| llvm-reduce | ||
| llvm-rtdyld | ||
| llvm-rust-demangle-fuzzer | ||
| llvm-shlib | ||
| llvm-size | ||
| llvm-special-case-list-fuzzer | ||
| llvm-split | ||
| llvm-stress | ||
| llvm-strings | ||
| llvm-symbolizer | ||
| llvm-undname | ||
| llvm-xray | ||
| llvm-yaml-numeric-parser-fuzzer | ||
| llvm-yaml-parser-fuzzer | ||
| lto | ||
| msbuild | ||
| obj2yaml | ||
| opt | ||
| opt-viewer | ||
| remarks-shlib | ||
| sancov | ||
| sanstats | ||
| split-file | ||
| verify-uselistorder | ||
| vfabi-demangle-fuzzer | ||
| xcode-toolchain | ||
| yaml2obj | ||
| CMakeLists.txt | ||