[OpenIndiana-discuss] What's HASWELL support like on OI?

[Toomas Soome]

> On 29. Jan 2021, at 07:48, Chris <oidev at bsdos.info> wrote:
> 
> On 2021-01-28 19:48, Gary Mills wrote:
>> On Thu, Jan 28, 2021 at 05:35:38PM -0800, Chris wrote:
>>> I'm trying to find some combination of hardware that will
>>> produce reasonable performance at the console.
>> By console, do you mean the raw console without a window manager?  If
>> that's the case, it's no wonder that you are finding it slow.  Most
>> people never see this, except during boot.  The raw console is known
>> to be quite slow.  It's never been a problem because the terminal
>> windows you get with a window manager are quite fast.  I have five
>> low-end systems with a variety of video cards all running OI here.
>> All I ever did was to do the initial install from the live USB image,
>> and do updates afterwards with the pkg command.
> Well *that's* depressing. Hmm. I guess my first task is going to be
> *fixing* that. Frustrating; as I do a great deal of my initial work
> from the console. Maybe I'm spoiled, because the (Free)BSD console(s)
> are great. I can CTRL+ALT+F(1-12) for a new session, and attack
> several different tasks simultaneously.
> Guess I'm going to have to build all that into OI. Looks like OI uses
> wcons. I'll see if I can coerce syscons(4) to replace it. Unless
> someone else has a better idea/option. :-)
> 
> Thanks for taking the time to reply, Gary. Greatly appreciated!
> 
> —Chris
> 


You do not really want to start with replacing console with syscons, syscons itself is already obsolete and freebsd got itself in situation with two competing (and broken in parts) console implementations:)

For virtual consoles, you want to check VT(7I), I do not know why we do not enable vt’s bt default:

tsoome at beastie:~$ svcs -a | grep vt
disabled       11:34:57 svc:/system/vtdaemon:default
disabled       11:34:57 svc:/system/console-login:vt2
disabled       11:34:57 svc:/system/console-login:vt3
disabled       11:34:57 svc:/system/console-login:vt4
disabled       11:34:57 svc:/system/console-login:vt5
disabled       11:34:57 svc:/system/console-login:vt6

IMO this is a bug. In any case, there is not need to start with replacing random components with another kind of random components when there is perfectly good option about making existing components better;)

rgds,
toomas