add likely/unlikely macro for unsupport c++20 compiler.

[xuhancn]

Issue:

Intel validation team found some low version gcc which not support c++20 will occur below issue:

[2024-04-13T08:03:25.142Z] g++ /tmp/torchinductor_root/vd/cvdytwwwlhi63ofh3pwzqfpjga4w4xe7bjfdoavpblbo5khzf3b2.cpp -shared -fPIC -Wall -std=c++17 -Wno-unused-variable -Wno-unknown-pragmas -D_GLIBCXX_USE_CXX11_ABI=0 -I/root/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/include -I/root/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -I/root/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/include/TH -I/root/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/include/THC -I/root/anaconda3/envs/pytorch/include/python3.8 -L/root/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/lib -L/root/anaconda3/envs/pytorch/lib -L/root/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/lib -ltorch -ltorch_cpu -lgomp -ltorch_python -lc10 -mavx2 -mfma -DCPU_CAPABILITY_AVX2 -O3 -DNDEBUG -ffast-math -fno-finite-math-only -fno-unsafe-math-optimizations -ffp-contract=off -march=native -fopenmp -D C10_USING_CUSTOM_GENERATED_MACROS -o /tmp/torchinductor_root/vd/cvdytwwwlhi63ofh3pwzqfpjga4w4xe7bjfdoavpblbo5khzf3b2.so
[2024-04-13T08:03:25.142Z]
[2024-04-13T08:03:25.142Z] Output:
[2024-04-13T08:03:25.142Z] /tmp/torchinductor_root/vd/cvdytwwwlhi63ofh3pwzqfpjga4w4xe7bjfdoavpblbo5khzf3b2.cpp: In function ‘T parse_arg(PyObject*, size_t) [with T = long int; PyObject = _object; size_t = long unsigned int]’:
[2024-04-13T08:03:25.142Z] /tmp/torchinductor_root/vd/cvdytwwwlhi63ofh3pwzqfpjga4w4xe7bjfdoavpblbo5khzf3b2.cpp:117:10: error: expected identifier before ‘[’ token
[2024-04-13T08:03:25.142Z] [[unlikely]] throw std::runtime_error("expected int arg");
[2024-04-13T08:03:25.142Z] ^

The reason is unlikely need c++20 attribute, ref: https://en.cppreference.com/w/cpp/language/attributes/likely

Solution:

Add MACRO to enable non-c++20 attribute GNU compiler.

cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @peterbell10 @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @ColinPeppler @amjames @desertfire @chauhang

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/124997

• 📄 Preview Python docs built from this PR
• 📄 Preview C++ docs built from this PR
• ❓ Need help or want to give feedback on the CI? Visit the bot commands wiki or our office hours

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit 9e0c46a with merge base b2f521f ():
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[DiweiSun]

Functionality Pass with following case:

import torch
from transformers import BertModel

model = BertModel.from_pretrained("bert-base-uncased")
model.eval()

vocab_size = model.config.vocab_size
batch_size = 128
seq_length = 512
data = torch.randint(vocab_size, size=[batch_size, seq_length])
model = torch.compile(model)

with torch.no_grad():
    model(data)

[xuhancn]

@pytorchbot merge

[pytorchmergebot]

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging

Check the merge workflow status
here

[xuhancn]

@malfet We occurred this issus at pytorch 2.3 test cycle, can you cherry-pick this PR to 2.3 branch for further release?

[matthost]

This is not in the torch 2.3.1 release https://github.com/pytorch/pytorch/blob/v2.3.1/torch/_inductor/codecache.py