This is an archive of the discontinued LLVM Phabricator instance.

[exegesis][X86] ParallelSnippetGenerator: don't accidentally create serialized instructions
ClosedPublic

Authored by lebedev.ri on Sep 4 2021, 11:30 AM.

Details

Summary

In the case of no tied variables, we pick random defs, and then random uses that don't alias with defs we just picked.
Sounds good, except that an X86 instruction may have implicit reg uses,
e.g. for MULX it's EDX/RDX: Intel SDM, 4-162 Vol. 2B MULX — Unsigned Multiply Without Affecting Flags

Performs an unsigned multiplication of the implicit source operand (EDX/RDX) and the specified source operand
(the third operand) and stores the low half of the result in the second destination (second operand), the high half
of the result in the first destination operand (first operand), without reading or writing the arithmetic flags.

And indeed, every once in a while llvm-exegesis happened to pick EDX as a def while measuring throughput,
and producing garbage output:

$ ./bin/llvm-exegesis -num-repetitions=1000000 -mode=inverse_throughput -repetition-mode=min --loop-body-size=4096 -dump-object-to-disk=false -opcode-name=MULX32rr --max-configs-per-opcode=65536
---
mode:            inverse_throughput
key:
  instructions:
    - 'MULX32rr EDX R11D R12D'
  config:          ''
  register_initial_values:
    - 'R12D=0x0'
    - 'EDX=0x0'
cpu_name:        znver3
llvm_triple:     x86_64-unknown-linux-gnu
num_repetitions: 1000000
measurements:
  - { key: inverse_throughput, value: 4.00014, per_snippet_value: 4.00014 }
error:           ''
info:            instruction has no tied variables picking Uses different from defs
assembled_snippet: 415441BC00000000BA00000000C4C223F6D4C4C223F6D4C4C223F6D4C4C223F6D4415CC3415441BC00000000BA0000000049B80200000000000000C4C223F6D4C4C223F6D44983C0FF75F0415CC3
...
$ ./bin/llvm-exegesis -num-repetitions=1000000 -mode=inverse_throughput -repetition-mode=min --loop-body-size=4096 -dump-object-to-disk=false -opcode-name=MULX32rr --max-configs-per-opcode=65536
---
mode:            inverse_throughput
key:
  instructions:
    - 'MULX32rr R13D EDX ECX'
  config:          ''
  register_initial_values:
    - 'ECX=0x0'
    - 'EDX=0x0'
cpu_name:        znver3
llvm_triple:     x86_64-unknown-linux-gnu
num_repetitions: 1000000
measurements:
  - { key: inverse_throughput, value: 3.00013, per_snippet_value: 3.00013 }
error:           ''
info:            instruction has no tied variables picking Uses different from defs
assembled_snippet: 4155B900000000BA00000000C4626BF6E9C4626BF6E9C4626BF6E9C4626BF6E9415DC34155B900000000BA0000000049B80200000000000000C4626BF6E9C4626BF6E94983C0FF75F0415DC3
...

Oops! Not only does that not look fun, i did hit that pitfail during AMD Zen 3 enablement.
While i have since then addressed this in rGd4d459e7475b4bb0d15280f12ed669342fa5edcd,
i suspect there may be other buggy results lying around, so we should at least stop producing them.

Diff Detail

Event Timeline

lebedev.ri created this revision.Sep 4 2021, 11:30 AM
lebedev.ri requested review of this revision.Sep 4 2021, 11:30 AM
lebedev.ri edited the summary of this revision. (Show Details)

Thanks for the patch. Give me some time to add a unit test that this would break, I'll ping here when submitted.

courbet accepted this revision.Sep 7 2021, 12:08 AM

Done, please rebase to rG131f7bac63b8 and you should now have a failing test.

This revision is now accepted and ready to land.Sep 7 2021, 12:08 AM

Forgot to say, thank you for the review!