Posts
8819
Following
221
Followers
648
Kernel MM hacker. Work at Oracle, my views on here are however solely my own. Brit.

Wrote a book about kernel mm (in editing phase) - https://linuxmemory.org/

Occasionally make ambient music - https://soundcloud.com/distal_music/

Website at https://ljs.io/

Motto: Gainz for Brainz

"I'm afraid it's way easier to re-hinge a massive door than Lorenzo"

Lorenzo Stoakes

Linus Clawvalds is always watching.
0
1
2
@oblomov @arnd Not criticsing this to be clear that's a fine suggestion [text can be shit for conveying this] :)

But in my tests I found that it's not just xen that's the problem, so you'd have to do this everywhere, which isn't really practical.

I mean it's worth whacking this egregious case but it's a band aid, we definitely need a general solution.

I am pretty certain people have also whacked other egregious cases individually too, got major de ja vous over this.
1
0
3
@arnd @oblomov well this solution seems viable then? maybe have a cmin()/cmax() for these scenarios or something like that.
0
0
2
@arnd @oblomov ah right yeah.

The problem is we have tons of callers so actually doing that would be a giant pain.

I feel like a way forward may be to have a specific macro for the extremely strict case (type decl) and a looser one for everywhere else that eliminates this.

But question is - how often are we using min()/max() + friends in places like array lengths where we must strictly be const?
1
0
3

There's so many goobers falling for Microsoft's claim that EU regulations were responsible for the Crowdstrike crash. It's extremely obviously a political play to try and undermine EU's anti-trust case against them, but people are eating it up like they don't have two brain cells to rub together.

7
10
2
@oblomov @osandov @arnd I'm no expert on the behaviour of the pre-processor or C const behaviour + don't really have time to dig deep but based on discussions on IRC - there's some really subtle complexity around const behaviour (esp. in scenarios like array indices where it truly has to be const) that I believe makes a relatively straight forward solution trickier.

There are bits of existing code that do this but again, seems super subtle.

Anyway I will leave the details of a solution to others :)
1
0
3
@osandov @arnd For those following, posted https://lore.kernel.org/linux-mm/c83c17bb-be75-4c67-979d-54eee38774c6@lucifer.local/T/#u summarising situation + showing some results on my local box (ignore cringe hostname of said box)
0
2
11

Sometimes I still sit here and think about how 4 years ago, scientists pulled off a fucking miracle of developing 3 brand new vaccines for a novel virus that had never had a vaccine for it before in like 6 months, scaled it out to produce billions of doses, and then like 40% of the country just said "nah".

2
15
4
@osandov @arnd Regardless, this is an unacceptable build regression.

Chatted to @arnd on IRC who suggests he is tied up with stuff to dig into this more so I'll have a wee look :)

EDIT: Think that'll be more likely a report to list + leave others to solve as this seems quite involved and I have a ton of things to do :>)
3
1
7
@arnd worth posting to the linux-mm ML at least?

I mean this is unacceptable
0
0
3
@arnd we going to revert that series then?

I feel there is pedantry at play...
2
0
1
@oleksandr @filabrazilska @lkundrak @vbabka I thought the scale was from Ostrava to Brno?
1
0
1

Lorenzo Stoakes

Another incredible video about the cutting edge of dark matter experimentation from the criminally underrated Sam Gregson.

Check him out if you're in any way interested in physics and science in general, top quality content.

https://www.youtube.com/watch?v=dbP6DvBUKHs
0
0
0
@lkundrak @filabrazilska @oleksandr sorry where do I rate on this scale? And @vbabka ?
2
0
1
@dougmerritt I went as far as to turn the lights out and put headphones on for the latest one and watch it on my not-stupid 49" ultrawide.

(Note I said NOT stupid)
0
0
1
@dougmerritt we'll both be fine bro. Definitely. Definitely definitely definitely.

Well, probably anyway.
1
0
1
Show older