Intel Spots A 3888.9% Performance Improvement In The Linux Kernel From One Line Of Code

[Phil995511]

Does AMD have a kernel robot?

Intel is developing Clear Linux, most likely why they have a Linux kernel testing bot...

[Michael]

Intel is developing Clear Linux, most likely why they have a Linux kernel testing bot...

No, it's unrelated. The kernel test robot isn't even running Clear.

[S.Pam]

Isn't this just a fix for the regression mentioned in the comment?

[npwx]

So basically it restores the performance back to what it was before the regression by limiting the optimization to the cases where it actually makes sense. Very misleading description.

[Phil995511]

No, it's unrelated. The kernel test robot isn't even running Clear.

If you have a distribution you have to test the kernels.

After that, the way you do your tests is up to you. It may be that in parallel they are also aiming for a better efficiency of their drivers for Linux.

[npwx]

So basically it restores the performance back to what it was before the regression by limiting the optimization to the cases where it actually makes sense. Very misleading description.

Adding to that, I read Phoronix (among other things) to read about *real* optimizations and cherry pick them if they are applicable to workloads I care about. This spammy clickbait is really annoying me and I find it quite unprofessional.

[blackshard]

Adding to that, I read Phoronix (among other things) to read about *real* optimizations and cherry pick them if they are applicable to workloads I care about. This spammy clickbait is really annoying me and I find it quite unprofessional.

Tend to agree... news like this are mostly curiosity rather than anything serious

[bug77]

Adding to that, I read Phoronix (among other things) to read about *real* optimizations and cherry pick them if they are applicable to workloads I care about. This spammy clickbait is really annoying me and I find it quite unprofessional.

This could be valuable info for someone exercising that code path very often. Or for someone who developed a workaround specifically because that code path was too slow. Granted, that would be a limited audience. Still...

[moonwalker]

This could be valuable info for someone exercising that code path very often. Or for someone who developed a workaround specifically because that code path was too slow. Granted, that would be a limited audience. Still...

...the headline IS rather clickbait-y...

[Developer12]

So intel didn't have any hand in finding the fix or authoring the change, and only found the results the authors expected.

Why is this news again?