Quote of the week [LWN.net]

I understand that this situation could be quite frustrating, but we can only expect a memory model to model memory. Its job is to help us understand what can work and what will not work from a memory-ordering perspective, which at best will provide you with the options that you seem to be so dissatisfied with. The memory model is quite incapable of browbeating intransigent human beings into agreeing on which option should be used in a given situation. This last never was a requirement of the LKMM project. Please rest assured that it will remain a non-requirement.

— Paul McKenney

Quote of the week

Posted Oct 15, 2020 11:02 UTC (Thu) by smoogen (subscriber, #97) [Link]

The following quote in the article was equally enlightening as it is something I have to remember daily.

> Are you disputing that that's a reasonable thing to do? There's a lot of kernel
> code that does something like this.

If you wish to actually solve this problem, you are going to have to face
up to the fact that you are going to have to make some choices, and no
matter what set of choices you make, there will be people who are unhappy.
You will then either need to convince a critical mass of people of the
wisdom of your choices or adjust your choices to get a critical mass
on board. If you do not wish to take on this particular challenge, your
best strategy is of course to stop complaining about it and take up other
challenges that are more technical and less people-oriented in nature.

Quote of the week

Posted Nov 18, 2020 12:01 UTC (Wed) by mathieu_lacage (guest, #3967) [Link]

I wanted to point out, like the two previous comments, that the actual email is a lot more interesting than the sarcastic quote taken out of context here. I urge others to read the email in its entirety.