Korona wrote:
you make it sound as if the author claims that RAM is slower than disk
I never said that. What the author claims is that the RAM has the same speed as the NVMe, here's the quote (again):
Quote:
Those operations [page fault, interrupts, copies or virtual memory mapping update] are now in the same order of magnitude of the I/O operation itself.
Korona wrote:
that your tone is not appropriate
What would you say about the tone of "Or maybe, maybe, maybe, that guy is an expert in his field and your understanding of the blog post is flawed."? Do you honestly think that this is not a personal insult? Do you really think that writing something like that is within the scope of a "nice technical discussion"?
You also wrote "That's something that people cannot really verify since you are posting under a pseudonym here." which is essentially calling me a liar. No, I don't lie, I really do have that certificate, and btw I don't use pseudonym, I use a monogram.
Korona wrote:
Claiming that a substaintial contributor to the Linux readahead path does "not know" how readahead works is just ridiculous.
Why would it be ridiculous? Do you really think that Linux source is only written by genius experts? Reading through the Linux source and looking at the
insane number of bugs I'm certain it was written mostly by incompetent developers, and the real experienced developers are having hard time cleaning up the mess after them. Just for the records,
Con Kolivas and Torvalds agrees (
here and
here and
here and
here). Unlike you, I back up my claim, because that's what makes a discussion civilized.
Look, from your posts it's obvious that you have put the blog's author on a pedestal and I'm sorry that you were mistaken about that. But that's still no reason for you to make personal insults the way you did.
Cheers,
bzt