Hannes Hauswedell

> Well... that's not exactly what I meant. This assumes, that you already downloaded seqan to some place in your filesystem. There is a possibility to define external projects in...

> The only I thing I could not find out is how to tell cmake to clone seqan recursively ("GIT_CONFIG "--recursive" did not work). I still think we should figure...

Note that you need to get out of the realm of CPU caches for the benchmark to be meaningful!

> Is that something we have to do in every single benchmark (like this one) or some over all adaptation? I am not sure. Maybe try with sizes similar to...

> Otherwise, I would just consider it a research task for this issue The cache sizes vary a lot. I just wanted to point out that I have seen very...

> I would just test a tiny size (let's say 1000) and a big size (8*1024*1024) and see if there are significant differences in the timing for a fixed number...

Thank you for the long answer. I have tried to find out more regarding FreeBSD's state of i2c support / exposure, but I don't know yet what to look for...

Thank you so much for looking into this! Interestingly I do not even have `iicdev.ko` on my system. In any case, I will open an issue in the FreeBSD bug...

Here is the issue in the FreeBSD tracker: https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=251844 > This is a shortcoming in the Nvidia driver, but one that we could theoretically address by porting over logic from...

Hi Jeremiah, thanks for the quick reply and rerunning the benchmark! Is this for the jumping test or reading test or combined? And what does "core only" mean, is that...