IRC channel logs

<damo22>azert: there are many ways to write drivers, the point being we want to minimise what we need to implement from scratch, reusing what is already built and maintained is vital for our success

<azert>Clear

<damo22>even now, i wonder if we can just link all the peripheral drivers into rumpusb

<damo22>until we have a way to separate them

<azert>I’d just do that, as a temporary solution

<damo22>although, treating some as non-disk will be tricky

<azert>Some will be excluded of course

<damo22>ie how will the translator know which device kind was attached

<damo22>and so which interface to expose

<azert>It could just expose the netbsd interface

<azert>Whatever that is

<azert>Of course some thing might just not work unless integrated and supported in other servers, thinking about an usb keyboard and the console

<damo22>right

<damo22>it appears usb peripheral driver opens a private pointer from the kernel

<damo22>device_lookup_private

<damo22>so i dont know how that can work without the two drivers being in the same process

<azert>Does Hurd devnode supports ioctls?

<azert>Like if you do device_open on a Hurd devnode, I guess you can send it ioctls right?

<damo22>i dont know if that works

<damo22>does the mach device interface support ioctls

<youpi>it doesn't need to

<youpi>the device translator can just also implement the ioctl interface

<damo22>ok

<gnucode>hmmm. I really need to try to get the hurd-console to start again on my T43.

<gnucode>I miss being able to edit the wiki.

<Pellescours>gnucode: updating hurd packet is not fixing it?

<gnucode>I'll try updating my laptop again tonight. But i did an update maybe a week and a half ago. Still had issues.

IRC channel logs

2025-04-16.log