ynnn

8 ms·

1.
by ynnn 7y ago
If you run a buggy program on a modern OS, it won't crash the system or impact other processes. If you run a buggy program on DOS, it will write to random physical addresses, probably clobbering the state of other processes and of DOS.
2.
by ynnn 7y ago
TIL! Thanks.
3.
by ynnn 7y ago
Selling what to Lenovo? Thinkpad? That's hardly a supercomputer. I'm not aware of any supercomputers built by Lenovo. IBM has built a lot.
4.
by ynnn 7y ago
Where did 'StaticRedux say anything about demanding anything? It's perfectly possible to report a bug without demanding that it be fixed.
5.
by ynnn 7y ago
The ford logo is a nice touch.
6.
by ynnn 7y ago
No, it's like telling the candy maker that their candy made them sick or otherwise left something to be desired. If the candy maker cares about making good candy, that can be useful information, even if they may not have time to do som
7.
by ynnn 7y ago
IBM.
8.
by ynnn 7y ago
Sadly, that's what I've been hearing for at least 5 years.
9.
by ynnn 7y ago
Yeah, if you choose to turn off hyperthreading. Pretty expected tbh - hyperthreading helps quite a bit for some things.
10.
by ynnn 7y ago
Welcomed and encouraged to submit their driver for consideration . I would guess that the vast majority of vendor-proprietary drivers would get rejected from upstream, so for most cases the viable options for vendors wishing have an driver
11.
by ynnn 7y ago
There are a number of things that could be done to improve the experience with proprietary drivers, but all of them undermine the Linux driver philosophy. For the kernel community, any driver that is not mainlined is doing it wrong (and the
12.
by ynnn 7y ago
There's basically no chance that the current proprietary drivers will ever end up upstream, so if ARM / nvidia drivers were open source they would be out of tree. Given that the most frequently cited reason that these drivers shou