Some errors occurred when executing "file" project
Closed this issue · 4 comments
I use the following command：
python run_all_experiments.py --model=gpt-3.5-turbo --context -y ./benchmark-sets/all/file.yaml
According to the execution results, for all APIs in the "file" project, the executed driver was not the generated driver. It seems to be the driver in oss-fuzz, magic_fuzzer.cc
An example is attached.
output-file-magic_load_buffers.tar.gz
sorry for the previous mistake, comments updated.
@fdt622 Shall we close this?
This was because OFG using the wrong fuzz target in the file project, and should have been fixed by:
#308
and it remained the same when updating benchmarks:
https://github.com/google/oss-fuzz-gen/pull/431/files#diff-5e5a0a8bb6f2aa36541f86fd3a1a1276cd54e0ecb2daaedeb5d8674fef5cbc84L59
@fdt622 Shall we close this? This was because OFG using the wrong fuzz target in the
fileproject, and should have been fixed by: #308 and it remained the same when updating benchmarks: https://github.com/google/oss-fuzz-gen/pull/431/files#diff-5e5a0a8bb6f2aa36541f86fd3a1a1276cd54e0ecb2daaedeb5d8674fef5cbc84L59
Ok, thanks for your attempt to fix. I would close this. I remember that there are other projects with this problem. When I manually check the experimental results in the future, I will try to sort out projects with similar situations, which may help with subsequent analysis of the reasons.
Yep thanks!
Please let us know if you find more cases like this : )