[Help Wanted] Needing corpus for research purpose

[Vincent]

Hi all,

We are trying to use transfomer to learn the relationship between syscalls and return high-quality answers (syscalls) based on the given program to improve the syzkaller efficiency.

We are collecting data sets and could you share us your corpus?

Regards,

Wenzhi (Vincent)

[Dmitry Vyukov]

Hi Wenzhi,

You can download syzbot corpus here (and for other manager instances
similar links):
https://storage.googleapis.com/syzkaller/corpus/ci-upstream-kasan-gce-root-corpus.db

[Dmitry Vyukov]

On Sun, 13 Oct 2024 at 21:19, Abagail ren <renzezh...@gmail.com> wrote:
>
> Hi all,
>
> I am researching the life cycle of a corpus.
>
> Can I get more details about the corpus, such as the version of the targeted Linux kernel for ci-qemu-upstream-corpus.db ?

+syzkaller mailing list

Hi Abagail ren,

The corpus is almost always collected on the tree HEAD commit (that's
what syzbot tests).
You can find the exact revision an instance tests in the Instances table here:
https://syzkaller.appspot.com/upstream


> Best Regards,
>
> RenZezhong