<dovsienko>
anyway, let's focus on problems that have practicable solutions, and maybe with a smaller problem space a small budget will look a bit better
bjork1intosh has joined #haiku
bjorkint0sh has quit [Ping timeout: 480 seconds]
dalme has quit []
ablyss has joined #haiku
_justin_kelly166686824 has joined #haiku
_justin_kelly16668682 has quit [Ping timeout: 480 seconds]
ablyss has quit [Quit: Konversation terminated!]
erysdren has joined #haiku
nephele_xmpp has left #haiku [Disconnected: Hibernating too long]
nephele_xmpp has left #haiku [Error from remote client]
<mbrumbelow>
Ok, good to know. Thanks
<coolcoder613>
I just got a 486 laptop!
MisthaLu has quit [Quit: Leaving]
<Begasus>
nice coolcoder613, started out with that :)
<Begasus>
biab
Begasus has quit [Quit: Vision[]: i've been blurred!]
Begasus has joined #haiku
HaikuUser has joined #haiku
HaikuUser is now known as Begas_
HaikuUser has joined #haiku
Begas_ has quit []
<Begasus>
coolcoder613, trying Haiku on that?
<coolcoder613>
lol
<Begasus>
:)
<coolcoder613>
Haiku on 4MB RAM
<Begasus>
would be a challenge :P
nephele_xmpp has joined #haiku
<nephele_xmpp>
maybe a replication issue again
<nephele_xmpp>
Begasus: Could be nicer if you enable text shadows on the desktop :)
oevl has quit [Ping timeout: 480 seconds]
Anarchos has joined #haiku
<Anarchos>
hello
nexus6 has joined #haiku
Halian|Jardin has joined #haiku
Begasus has quit [Quit: Vision[]: i've been blurred!]
x10z has joined #haiku
Halian|Jardin has quit [Quit: I'll be back.]
Halian|Jardin has joined #haiku
Halian|Jardin has quit []
x10z has quit [Quit: My Unrecognized Mac has gone to sleep. ZZZzzz…]
rann has quit [Quit: Connection closed for inactivity]
<dovsienko>
coolcoder613: That could be an OK DOS game machine
<dovsienko>
my very first [second-hand] laptop had something between 1 and 4 MB RAM, I managed to run Linux on it (either text mode or an X server for network clients) and connect to Internet
<dovsienko>
However, to install it in the first place it took to plug the hard drive into a more modern desktop PC and to strip the resulting system to the minimum
<coolcoder613>
monochrome screen, no sound
Halian|Jardin has joined #haiku
mikezt has joined #haiku
<mikezt>
hello
<coolcoder613>
The hard drive seems to use some kind of three pin power connector instead of the standard IDE power
<phschafft>
three? not four?
<dovsienko>
coolcoder613: could be the LED rather than power, the old PATA 2.5" and 1.8" drives ran the power through the double-row data header IIRC
<phschafft>
dovsienko++
<dovsienko>
well, 12V, 5V and ground do fit into three pins, right? a multimeter could tell quickly
<phschafft>
or a M/S jumper. sometimes they had three pins so you don't loose th jumper itself.
<phschafft>
dovsienko: that is true. but even the smaller ones I have seen had four contacts. but clearly there is more than anyone of us could ever see.
<dovsienko>
coolcoder613: now you've talked yourself into identifying it properly and reporting the details :-)
<phschafft>
while you're at it. I have an floppy drive that has a connect just like the standard one but two pins (= 1 row) short. and it also has power on that connector. it works fine but seems read-only.
<coolcoder613>
I'm not sure if my new drive would work anyway. It's an 80GB drive.
<phschafft>
I had systems that only worked up to 4GB but once Linux was running they supported 80GB.
<coolcoder613>
WOuld it be able to use 4Gb of the 80GB drive?
<phschafft>
so you needed to select parameters just right that the BIOS found some boot image and then it was fine.
<phschafft>
yes.
<coolcoder613>
Oh, that would be fine
<phschafft>
but I think I needed to override the settings in the HDD (not the BIOS) and then let Linux figure it out later on.
<phschafft>
something like that.
<phschafft>
the simpler solution would be a second boot device (floppy or second HDD) and only use it for data.
<coolcoder613>
My main problem now is how to powerr this hdd
* phschafft
nods.
<phschafft>
I think that one had a BIOS that was half in ROM and half on HDD. maybe that was the reason for the limitation.
<coolcoder613>
this hdd seems to be bad, dos install has write errors
<coolcoder613>
thats why im trying to replace
<phschafft>
:(
<phschafft>
that's bad.
<Halian|Jardin>
phschafft that's a weird setup
<phschafft>
hm?
<Anarchos>
Halian|Jardin many peripheral have their own bios (graphics cards, network cards...) and it is intended like that, so your motherboard bios can call the peripheral bios to initialize themself in the POST. Th'at's why you get graphics activity during bios POST.
<Halian|Jardin>
I hadn't read the scrollback, and so thought it was the computer's main BIOS that was half on the hard drive ;w;
<dovsienko>
coolcoder613: that's a 3.5" HDD, are you sure it is a laptop?
<dovsienko>
the thing with PC BIOS and HDD size is a problem from old times, which is why I still make a small /boot partition below 1GB on my Linux hosts
<dovsienko>
sometimes the easiest way to set it up is to have a simple boot device for the bootloader and /boot (SD card or a USB storage or a plain HDD) and then the kernel can deal with the root FS being on hardware RAID or who knows where
zard has joined #haiku
<phschafft>
Halian|Jardin: in my case yes, half of the bios is in ROM and the other half is loaded from the HDD before the real OS is loaded. it used to be like that on some older high end machones.
<phschafft>
*machines
<phschafft>
basically when the BIOS did not fit into the ROM.
myriad has joined #haiku
<Halian|Jardin>
Ahh
<phschafft>
it was a time when people where happy to experiment. some of those machines had windowed, multi-tasking BIOS (like you could browse the system info while you formatted a disk in background).
<phschafft>
but then in the 2000s that all went away. and then flash sizes became bigger they added more graphical (and sometimes sound) elements again.
<phschafft>
but it was a different style.
<andreasdr[m]>
Hi there.
* phschafft
waves.
myriad has left #haiku [#haiku]
<coolcoder613>
hi zard
<coolcoder613>
I got a 486 laptop
<zard>
Hello coolcoder613 :)
* coolcoder613
heads to bed, it's almost midnight
<zard>
Goodnight coolcoder613 :)
Anarchos has quit [Quit: Vision[]: i've been blurred!]
SLema has quit [Ping timeout: 480 seconds]
Begasus has joined #haiku
<Begasus>
k, that didn't went too wel :P
MisthaLu has joined #haiku
Halian|Jardin has quit [Quit: My MacBook Air has gone to sleep. ZZZzzz…]
<nephele_xmpp>
hi
<Begasus>
hi nephele_xmpp
<nephele_xmpp>
my freewear order is on the way
<nephele_xmpp>
they were quite nice, i ordered 2 things as 2 orders, and asked for the second order that they be shipped together. And they do! :D also gave me the adjusted price for postage only once
<Begasus>
nice!
<nephele_xmpp>
so far, great customer service
oevl has joined #haiku
Halian|Jardin has joined #haiku
Halian|Jardin has quit []
<Habbie>
will the haiku swag be available to buy at FOSDEM? :)
<Begasus>
would be nice, but I doubt it, don't know what PulkoMandy still has around :)
<Begasus>
a nice sticker for the laptop would do, now the one from KDE is blinking on it :P
<nekobot>
[haiku/haiku] 04bf57110ca4 - Tracker: Disallow control characters in file name edit
yann64 has quit [Quit: Vision[]: i've been blurred!]
<dru_satori>
so how can I remove a bad patchset from Gerrit ?
<waddlesplash>
you just push a new one
<waddlesplash>
with the same Change-Id
<dru_satori>
when I try, it rejects it as having no changes
<waddlesplash>
then that means you didn't modify anything
<waddlesplash>
and the commit hash is identical
<dru_satori>
I'm trying to update patchset 4, because 5 includes files unintended
<waddlesplash>
the patchset numberings are just what Gerrit generates
<waddlesplash>
you can't update any particular patchset, you can only add new ones
<waddlesplash>
a "patchset" on Gerrit is just a commit, that's all, nothing more or less
<dru_satori>
based upon comments, I cherry picked, made the requested changes, now it is rejecting as identical. I
<waddlesplash>
I basically never use cherry-pick
<waddlesplash>
amending commits is all you need here
<dru_satori>
I'm to the point of punting and simply saying it is not worth the effort.
<waddlesplash>
I'm confused as to what you are confused about, or what is difficult about this
<dru_satori>
and yet, pushing the commits rejects, so yeah.
<waddlesplash>
git has a lot of features and so can get confusing
<waddlesplash>
but once you understand the basic concepts it's mostly understandable from there
<waddlesplash>
what is the most recent commit you have locally?
<dru_satori>
I use git all the time. this is the combo of gerrit and git, trying to keep things in it's sync
<Begasus>
ow bugger, updated samba4 to see it's using : devel:libicuuc$secondaryArchSuffix >= 75 :P
<waddlesplash>
dru_satori: Gerrit just takes commits, and replaces commits on push by matching Change-IDs. that's it
<waddlesplash>
one "change
<waddlesplash>
" on Gerrit always 1 commit
<waddlesplash>
Gerrit knows nothing about what makes commits the same or different besides the commit hash and the Change-Id
<waddlesplash>
you could push a change with NOTHING in common with the previous version of the change with the same Change-Id and gerrit will happily accept that
<dru_satori>
which is how I got to this point :( I did a push that included bad files, and then made the mistake of recerting.
<waddlesplash>
that's fine, just make the Change-Ids match
<waddlesplash>
easy
<waddlesplash>
so again, what's the top commit you have locally?
<dru_satori>
anyways, at this point, all of the relevant code is in, the id's match, the current head reflects what I changed.
<dru_satori>
so, it either works or it doesn't.
<waddlesplash>
do you want help with fixing your git situation or not
<waddlesplash>
because I'm trying to walk you through it here, but I can't do that if you don't answer my questions
janking has quit [Quit: Vision[]: i've been blurred!]
<dru_satori>
I'm not sure what to answer. Gerrit shows the correct intel_extreme.h header. it just included the mt7921 files that I was also working.
HaikuUser2 has joined #haiku
<waddlesplash>
dru_satori: yes, so I want to help you fix that
<dru_satori>
I can git rm tthose files and resubmit.
<waddlesplash>
no
<waddlesplash>
don't git rm
<waddlesplash>
and don't "resubmit". We can fix this in the working copy directly
<waddlesplash>
so once again: what is the top commit locally?
<dru_satori>
the one I abandoed because the change_ig mismatched
<waddlesplash>
OK, so we don't want that commit, BUT we want its changes in the original commit
<dru_satori>
the prior one commit 9b03bb5a018c6d216167b3237ca86b18d95e3c89 is the one that included
<waddlesplash>
so: "git reset HEAD~1"
erysdren has quit [Quit: Konversation terminated!]
<waddlesplash>
this undoes the top commit but *without* modifying your working tree
<waddlesplash>
so the changes made will still be in the working tree
<waddlesplash>
once you have done that, run "git gui"
<waddlesplash>
and then check the "Amend last commit" box. That commit's message and changes will then appear in Git Gui's view
<waddlesplash>
Then what you want to do is de-stage the files you don't want: In the box in the lower left of the view, click the icons to the left of the file names. This will remove those files from the commit
<waddlesplash>
Then, add the files you do want: in the box in the upper left of the window, click the icons to the left of the file names for just those files
<dru_satori>
this is all what was done before
<waddlesplash>
well, it definitely was not, because we got a different result on Gerrit.
<waddlesplash>
after you did that, click "Commit"
<waddlesplash>
that's what actually *amends* the change, if the "Amend" box is checked
<waddlesplash>
otherwise, nothing happens
<waddlesplash>
once you've done that, just "git push origin master:refs/for/master" as before
<waddlesplash>
the change on Gerrit will then get updated
<dru_satori>
hopefully that worked. now I can go back to working through the wlan driver
thesynthax has quit [Ping timeout: 480 seconds]
HaikuUser2 has quit [Quit: Vision[]: i've been blurred!]
<nekobot>
[haiku/haiku] 0db74e1ae12a - intel_extreme: add support for Alder Lake chipsets.
<dovsienko>
waddlesplash: it may be a good idea to make RAMFS an off-by-default feature and to prioritise defects that spoil everyone's experience with the OS for a while, what do you think?
<waddlesplash>
why?
<waddlesplash>
in point of fact RAMFS is used daily by most users and there are no crashes
<dovsienko>
the last time I ran benchmarks with and without RAMFS to support my argument about it, it surprised me that BFS actually performed better (in a VM on a host with lots of RAM and SSD)
<waddlesplash>
because it's used by the wayland compatibility layer and thus Iceweasel etc. for shared memory
<waddlesplash>
and this is not a small performance improvement, it's a huge one actually
<waddlesplash>
but this is a very different case than your benchmarks I'd guess, because this uses a small number of files, but very large ones with frequent writes
<waddlesplash>
so if we used BFS as backing for that it'd create tons of disk thrashing
<phschafft>
generally it's not a good idea to have stuff like shared memory physically write on an SSD...
<dovsienko>
I know it is useful in principle and use equivalents on other OSes, but it does not make sense how it could work well for everyone and keep breaking in my tests
<waddlesplash>
because your tests do things nobody else is doing lol
<waddlesplash>
ramfs isn't used as tmpfs by default
<waddlesplash>
so, that's the difference
<dovsienko>
such as compiling C files
<dovsienko>
and crashing Clang
<waddlesplash>
the Clang crashes are all fixed now?
<phschafft>
dovsienko: how does it do that?
<waddlesplash>
and again very few people use ramfs to do things like that, the default case is just the shared memory directory
<dovsienko>
never heard back from my bug report about it
<waddlesplash>
phschafft: didn't clear tails of pages
<waddlesplash>
dovsienko: ? the bugs are fixed?
<phschafft>
waddlesplash: ah. thanks. so at least it's an identified problem.
<waddlesplash>
the description it's running on SUSE Linux
<dovsienko>
#11133 indeed does not mention it, so it may be unrelated (that said, my TMPDIR is set to RAMFS)
<dovsienko>
90993 is the same problem, as far as I can see from the error messages, but it would be more appropriate for a Clang developer to debug
<Skipp_OSX>
I'm pretty sure that if I disallowed B_DELETE you wouldn't be able to use the delete key in a filename edit so that would not be good. But I didn't push the submit button...
<dovsienko>
anyway, the point is it keeps chasing bugs out at a rate that draws precious time from other bug fixing
<Skipp_OSX>
Delete is allowed for good reason, so you did the right thing I think but submitting.
<Skipp_OSX>
*by submitting
<dovsienko>
so just in case you thought without RAMFS I would not be able to do what I need to do, that's not the case
<dovsienko>
one potential way to put RAMFS under control could be making it a ramdisk device instead and putting an ordinary BFS on top, this way there would be one fewer filesystem to debug
<phschafft>
das the BFS driver support trim?
<dovsienko>
there are some messages about TRIM in the boot log, but I do not know how complete that is
<phschafft>
I personally like when a system supports having some flat ramdisk device that one can use e.g. for a normal filesystem. however I don't think that is the most useful for stuff that has a lot of structural changes (changes in size, delete, create, appending, ...).
<dovsienko>
you are welcome to consider the idea
<nephele_xmpp>
Skipp_OSX: huh, is that mapped to B_DELETE? IN that case nevermind :)
<phschafft>
dovsienko: I just fear it would end up using shared memory backed by some random filesystem as storage for the ramdisk...
<Skipp_OSX>
short answer, yes.
nexus6 has quit [Quit: Vision[]: i've been blurred!]
HaikuUser has quit [Quit: Vision[]: i've been blurred!]
<dru_satori>
before I file it as a bug, is it just me or is mouse click behavior on context menus sometimes inconsistant or just exceptionally slow ?
<nephele_xmpp>
depends on if you are talking about tracker ;)
<dru_satori>
tracker is where I notice it the most, but it does seem to crop up in other places as well
<dru_satori>
Deskbar is particularly prone as well ( though I imagine that is shared code )
Begasus has quit [Quit: Vision[]: i've been blurred!]
<dru_satori>
I see. an interesting conundrum as to how to handle it though. I am not sure there is a great fix, and making it even more interesting. With Tracker, I'd bet that it is more prevalent in I/O bound conditions ( running from a USB drive ) where it has to iterate the filesystem
<dru_satori>
same reason it would show up in emulated / non-bare metal configurations
<Skipp_OSX>
Well, for everyone that's true, but for you personally you can fix the issue by raising the double click speed in Input prefs
<Skipp_OSX>
At least make the issue less noticeable
<dru_satori>
oh certainly, but I try to stay more or less to defaults in order to be able to simulate other users conditions
<Skipp_OSX>
Well then you got the bug and the code where the fix is needed, have at it.
thesynthax has joined #haiku
<Skipp_OSX>
one issue is in Tracker we remove and add the Nav menus from the right click menu, and if you go quick you can mess that up by having two menus open at the same time.
<dru_satori>
I'll get there eventually. trying to stay out of core tools until I am more comfortable with this particular process. wanna get to a point where I can point to a single machine that can be 'just works'
<dru_satori>
don't get me started on the mess that Explorer in the early days of the Windows 95 shell was when implementing any type of menu extensions
<Skipp_OSX>
ok I won't
<dru_satori>
nest pointers to item_id lists 8 and 10 levels deep
<dru_satori>
what you are describing is the same core problem they tried to solve, and ultimately failed.
<dru_satori>
by release, it had changed to a model that handled the menus immediately, and deferred any iteration for expansion until the item was clicked ( or mouse over )
mmu_man has joined #haiku
<dru_satori>
and even then, it would default to cached instance
tqh has joined #haiku
janking has joined #haiku
<janking>
hi i cant login on my BeAIM can you nipos
<janking>
is nipos here?
zard has quit [Quit: leaving]
janking has quit [Quit: Vision[]: i've been blurred!]
thesynthax has quit [Ping timeout: 480 seconds]
janking has joined #haiku
DKnoto has quit [Ping timeout: 480 seconds]
DKnoto has joined #haiku
jmairboeck has joined #haiku
<waddlesplash>
dovsienko: we already do support that, but the ramfs bugs thus far were also bugs in other subsystems or at least related to them for the most part
<waddlesplash>
so I think we should keep at it here
<dovsienko>
alright
<waddlesplash>
the new bug you have found is totally unrelated to ramfs actually
<waddlesplash>
it's not even in ramfs data structures
<waddlesplash>
it's probable that it may only happen with ramfs involved because on BFS there's more locks and things are slower, idk
<waddlesplash>
but the fault in question is not ramfs-related
<nephele_xmpp>
We can decide for a release wether to have something enabled/disabled per default, but for the nightlies this makes no sense. We should keep stuff enabled even if it is buggy to be able to test it
<janking>
nipos?
<waddlesplash>
dovsienko: also, your benchmark may give different results now, after my recent concurrency-related changes in ramfs
<nekobot>
• lonemadmax (7ad722d2): qt6_base: get latest qt6-haikuplugins (#11669)…
janking has quit [Quit: Vision[]: i've been blurred!]
<dovsienko>
waddlesplash: that's a fair point, let's dismiss that argument until I have new data
janking has joined #haiku
<janking>
nipos?
SLema has joined #haiku
vadique has joined #haiku
<vadique>
aloha
<vadique>
I was used haiku in qemu, can I copy that image directly to usb flash and boot it natively?
<vadique>
and if can, how to do it
janking has quit [Quit: Vision[]: i've been blurred!]
janking has joined #haiku
<nephele_xmpp>
Depends, if your qemu install is raw you can just copy it to a usb directly, if you used a compressed format like qcow2 you have to do some extra steps (which i don’t know how :D)
<nephele_xmpp>
alternatively attach a thumb drive to the VM and use Installer from the running system to install it to the thumb stick
janking has quit []
<vadique>
I just did cp haiku.qcow2 dumpme.qcow2 ; qemu-img resize --shrink dumpme.qcow2 1932735488 ; sudo qemu-img convert dumpme.qcow2 -O raw /dev/sdb
janking has joined #haiku
* vadique
waits for the last thing to finish
<Skipp_OSX>
You could open up qemu, load up your usb stick, format it and run Installer to install to your thumb drive
<vadique>
Skipp_OSX, I really don't wish to redo all the thing I've already done before after installing
<vadique>
before after... lol, since installing
<Skipp_OSX>
That's the beauty of it, you shouldn't have to because the installer will copy your system over including modifications
<vadique>
hmm, really?
<nephele_xmpp>
the only thing you have to take care off is partitioning and the EFI loader, other than that Installer will just replicate your system
janking has quit [Quit: Vision[]: i've been blurred!]
<Skipp_OSX>
Haiku: Using Installer and DriveSetup section
<Skipp_OSX>
skip the part about anyboot image since you're not using that, but the rest
diver has quit [Quit: Leaving.]
jmairboeck has quit [Quit: Konversation terminated!]
<vadique>
okay, I'll try to boot natively now, see you
janking has joined #haiku
vadique has left #haiku [bye]
janking has quit []
<waddlesplash>
dru_satori: m_defrag on the various BSDs are incompatible, there are ifdefs for it in other OpenBSD driver ports.
<waddlesplash>
you will need to look at those and see what to use instead
HaikuUser has joined #haiku
HaikuUser has quit []
<dru_satori>
@waddlesplash thanks. at the moment, I'm more concerned with getting the driver to even load and init. once there, I can start the real debugging and fixing of that stuff
Halian|Jardin has joined #haiku
<dru_satori>
which is entertaining, bouncing between two machines to get this working
<Skipp_OSX>
waddlesplash do you think you could look into Power saving mode at some point? It involves the scheduler, and you're the only one who could possibly make it work.
<waddlesplash>
what do you mean?
<Skipp_OSX>
In ProcessController => Power savings mode
<dru_satori>
off to boot the other machine
dru_satori has quit [Quit: Vision[]: i've been blurred!]
<waddlesplash>
Skipp_OSX: yes, what do you mean, does it not work?
<Skipp_OSX>
correct, does not work, also crashes apparently
<Halian|Jardin>
Skipp_OSX: henlo fellow OS X user (though I'm on Asahi Linux ALARM right now)
<Skipp_OSX>
heyo
x10z has joined #haiku
<Skipp_OSX>
The idea is that it puts the scheduler in a mode that uses as few cores as possible so the other cores can be turned off to save power.
<Skipp_OSX>
mmlr's idea... but the rest is beyond my understanding.
<waddlesplash>
it crashes? this is news to me
<waddlesplash>
there is a bug that it locks up PS/2 input on somehardware, yes
<Skipp_OSX>
I think the PS/2 input lockup is the actual problem afterall
BrunoSpr has quit [Quit: Vision[]: Ich wurde eingeweicht!]
<Skipp_OSX>
It the problem is outside of the scheduler/kernel then nvm, somebody else can tackle it, I was under the impression it was a scheduler issue.
Halian|Jardin has quit [Quit: I'll be back.]
DKnoto has quit [Ping timeout: 480 seconds]
drusatori_alt has joined #haiku
Anarchos has joined #haiku
DKnoto has joined #haiku
<nipos>
janking: Sorry, I wasn't here, just came home. If BeAIM surprisingly doesn't work without installing an update first, it usually means it's a server issue and that's not something I can help with. You can message Wildman on the forum, he operates the server, I only write the client software.
<drusatori_alt>
@waddlesplash before I wander into a rabbit hole. I am assuming that the best way to debug the WLAN interfaces DPRINTF() liberally applied ?
B2IA has quit [Quit: Vision[]: i've been blurred!]
B2IA has joined #haiku
AlienSoldier has joined #haiku
<waddlesplash>
Skipp_OSX: I mean it may still be a scheduler issue, I haven't investigated
<waddlesplash>
drusatori_alt: probably
Halian|Jardin has joined #haiku
Halian|Jardin has quit []
<drusatori_alt>
you know you have issues when getting a kernel panic is good, because it means the driver loaded :)
<Anarchos>
drusatori_alt \o/
<Anarchos>
pkgman update still broken ?
<Skipp_OSX>
On to the next step
B2IA has quit [Quit: Vision[]: i've been blurred!]
x10z has quit [Quit: My Unrecognized Mac has gone to sleep. ZZZzzz…]
<Skipp_OSX>
No, he's working on wifi driver support it seems
<drusatori_alt>
yeah, I'm trying to add support for the MediaTek MT7921, MT7921k and MT7922 WiFi chips on some of the AlderLake - N motherboards
B2IA has joined #haiku
<Skipp_OSX>
are you porting over the existing FreeBSD 14.x/OpenBSD driver?
<drusatori_alt>
yes
<Skipp_OSX>
ok carry on
tqh has quit [Quit: Leaving]
<drusatori_alt>
it's an OpenBSD driver, but based upon the code, I'm fairly certain it is really some sort of ugly hybrid of both
<Skipp_OSX>
As I understand it they're the same driver, but my understanding is pretty shallow
<drusatori_alt>
just some really odd mishmash of includes and linked libraries
Anarchos has quit [Quit: Vision[]: i've been blurred!]
<drusatori_alt>
and right now I'm hitting a Kernel Debug Land crash on fbsd uninit, which says that it found the device, tried to load the driver and then fails during the uninit when it cannot properly startup
Halian|Jardin has joined #haiku
janking has joined #haiku
erysdren has joined #haiku
Halian|Jardin has quit [Quit: I'll be back.]
thesynthax has joined #haiku
<drusatori_alt>
interesting. looks like these devices have to load the rom and firmware at load time. time to go digging or those files and determine if they are usable in this context, or have to be user installed.
ablyss has joined #haiku
nephele_xmpp has left #haiku [Error from remote client]
Halian has quit [Quit: I'll be back.]
nephele_xmpp has joined #haiku
janking has quit [Quit: Vision[]: i've been blurred!]