AMD Confirms Linux Performance Marginality Problem Affecting Some, Doesn't Affect Epyc / TR

[littleowl]

As a software developer do you not need to know how a CPU works internally. You only need to write your software in conformance to its specification.

If the system is not stable and the application keeps crashing, the software developer is usually blamed first and he/she has to prove that the issue is in the hardware - and we can see now that it is not so easy when it comes to hardware race conditions. The internal knowledge of CPU/HW architecture is extremely important when you want to optimise the application and get maximum out of it. bug77 is right, a "magic vanishing" of the issue would be the worst possible outcome and put me off buying a Ryzen CPU.

[sdack]

Obviously. AMD acknowledged that yesterday. It's part of history now. Whats done is done. The only thing pathetic is you weak ass excuses. I can't wait for Intel to release their next horribly bugged product. Then we can listen to all the Intel fanboys like you bitch about how it's OK and perfectly normal.

Here, read this: http://support.amd.com/TechDocs/5537...Processors.pdf

All CPUs are like this. These issues don't get a fix, but are considered features. See the document. Most of them say "No fix planned". Why? Because it's called hardware for a good reason. Most hardware cannot be changed easily (with the exception of microcode), but it's rather the job of the software developers to work with it, and not against it.

So this is still just a tantrum.

[duby229]

Here, read this: http://support.amd.com/TechDocs/5537...Processors.pdf

All CPUs are like this. These issues don't get a fix, but are considered features. See the document. Most of them say "No fix planned". Why? Because it's called hardware for a good reason. Most hardware cannot be changed easily (with the exception of microcode), but it's rather the job of the software developers to work with it, and not against it.

So this is still just a tantrum.

And this post is exactly what I said. You're looking for an excuse to say that bugged products are perfectly normal. I can't wait for Intel's next buggy CPU so we can hear all the fanboys like you claim how awesome it is. It'll be the best buggy product evar!

[sdack]

W-T-F?

Edit: Oh, I see now. You're blaming this on gcc. Again, you can't do that just yet, you still need the root cause to sort this out.

No, not on gcc. I'm blaming it on some people being fanboys, calling themselves software developers and what not, but not having a clue about reality and thus making a tantrum, calling it a "horribly bugged CPU" for instance. I wish there was a fix for these people, and yet do we have to accept them together with their idiocy. So excuse me when I'm having a laugh at them.

[sdack]

And this post is exactly what I said. You're looking for an excuse to say that bugged products are perfectly normal. ...

You really still believe to be living in a perfect world? Well, you can't get more pathetic than this! (But I dare you to teach me otherwise. *lol*)

[duby229]

No, not on gcc. I'm blaming it on some people being fanboys, calling themselves software developers and what not, but not having a clue about reality and thus making a tantrum, calling it a "horribly bugged CPU" for instance. I wish there was a fix for these people, and yet do we have to accept them together with their idiocy. So excuse me when I'm having a laugh at them.

Seriously, the fanboy is strongest in you. You're going so far as to say that buugy CPU'S are ok and normal and we should all just be happy and except them for their awesomeness or some bullshit. And it's all so that you can have an excuse for Intel's next buggy crap. That's fanboyism to an extreme.

[sdack]

Seriously, the fanboy is strongest in you. You're going so far as to say that buugy CPU'S are ok and normal and we should all just be happy and except them for their awesomeness or some bullshit. And it's all so that you can have an excuse for Intel's next buggy crap. That's fanboyism to an extreme.

Don't cry, duby. That's just the world we're living in.

[bug77]

No, not on gcc. I'm blaming it on some people being fanboys, calling themselves software developers and what not, but not having a clue about reality and thus making a tantrum, calling it a "horribly bugged CPU" for instance. I wish there was a fix for these people, and yet do we have to accept them together with their idiocy. So excuse me when I'm having a laugh at them.

So you're not disputing there is a bug that hasn't been tracked down and as such it's still impossible to ascertain its real-world impact. Then I guess we're good.

[sdack]

If the system is not stable ...

No. The system is stable with Windows. But according to you fanboys is it a fault with the hardware and not with Linux. So when you're trying to run Linux on an new Xeon CPU, but it's not working, then it's Intel's fault for not having developed their CPU to work with Linux? No.

The truth is rather that Linux is the most-flexible OS you can find. So if it's anyone's fault then it's Linux's fault - or it just isn't as good as it claims to be.

I however don't blame AMD or Linux. Linux will adapt to it in due time. The fault lies with fanboys throwing a tantrum, because of their delusional expectations and lack of understanding on what makes a system work.

[duby229]

No. The system is stable with Windows. But according to you fanboys is it a fault with the hardware and not with Linux. So when you're trying to run Linux on an new Xeon CPU, but it's not working, then it's Intel's fault for not having developed their CPU to work with Linux? No.

The truth is rather that Linux is the most-flexible OS you can find. So if it's anyone's fault then it's Linux's fault - or it just isn't as good as it claims to be.

I however don't blame AMD or Linux. Linux will adapt to it in due time. The fault lies with fanboys throwing a tantrum, because their delusional in their expectations and lack an understanding on what makes a system work.

No it isn't. Linux is a kernel and WSL shows the same bug while running the windows kernel. You might be able to say that it's bug in the userspace, but that hasn't been shown yet. You say that it hasn't been demonstrated on Windows, but you forgot to use the word "yet" The only one delusional here is you. At every point you misunderstood the facts.