By: Maynard Handley (name99.delete@this.name99.org), December 19, 2020 10:23 am
Room: Moderated Discussions
anon2 (anon.delete@this.anon.com) on December 19, 2020 1:00 am wrote:
> Maynard Handley (name99.delete@this.name99.org) on December 18, 2020 9:30 am wrote:
> > Maynard Handley (name99.delete@this.name99.org) on December 17, 2020 9:43 pm wrote:
>
> [...]
>
> > > My larger point is that if you ask "suppose I were willing to make the effort to provide metadata bits
> > > for each line in DRAM. What interesting things could I do with that?" opportunities do suggest themselves.
>
> [...]
>
> > This twitter thread (go up and down from the link I posted) is interesting insofar as it confirms
> > my point.
>
> It does not confirm any such thing.
>
It confirms the point that there are many systems out there with dodgy RAM, and that the users appear not to know this, only that their machines occasionally crash for random reasons.
Such systems would be improved if the users were informed by the OS that their RAM was dodgy, even if that's the limit of the ECC's capabilities.
> Maynard Handley (name99.delete@this.name99.org) on December 18, 2020 9:30 am wrote:
> > Maynard Handley (name99.delete@this.name99.org) on December 17, 2020 9:43 pm wrote:
>
> [...]
>
> > > My larger point is that if you ask "suppose I were willing to make the effort to provide metadata bits
> > > for each line in DRAM. What interesting things could I do with that?" opportunities do suggest themselves.
>
> [...]
>
> > This twitter thread (go up and down from the link I posted) is interesting insofar as it confirms
> > my point.
>
> It does not confirm any such thing.
>
It confirms the point that there are many systems out there with dodgy RAM, and that the users appear not to know this, only that their machines occasionally crash for random reasons.
Such systems would be improved if the users were informed by the OS that their RAM was dodgy, even if that's the limit of the ECC's capabilities.