xenos1984 has quit [Remote host closed the connection]
xenos1984 has joined #osdev
bauen1 has joined #osdev
kspalaiologos has joined #osdev
doorzan has quit [Remote host closed the connection]
the_lanetly_052_ has joined #osdev
GeDaMo has joined #osdev
Burgundy has joined #osdev
DanDan has joined #osdev
heat has joined #osdev
the_lanetly_052 has joined #osdev
the_lanetly_052_ has quit [Ping timeout: 246 seconds]
gog has joined #osdev
<gog>
mew
Vercas has quit [Remote host closed the connection]
Vercas has joined #osdev
<heat>
my vterm takes 5ms to scroll a line
<heat>
ouch
<zid>
not bad
<zid>
that's 11ms too quick for 60fps
<heat>
a 5ms trip to the kernel that doesn't involve actual IO is like at best 4ms too long
<zid>
just move it all into the kernel, then the syscall overhead is less, brilliant
<heat>
what if we move it all to userspace
<heat>
like
<heat>
everything
<gog>
no
<gog>
anti-microkernel aktion
<heat>
nooo
<heat>
it's not a microkernel
<heat>
there's no kernel
<heat>
it's just micro
<gog>
i don't believe you
<heat>
why
nyah has joined #osdev
<gog>
because
<heat>
okok seems fair
<heat>
i wouldn't trust me either
* Ermine
called microkernel police
<heat>
hope it was an RPC
<heat>
else you're going down with us
<heat>
i think the issue is that it's aggressively ioexiting under qemu
<heat>
linux's console is faster but not by that much
<heat>
actual bookkeeping with the console cells takes a negligible amount of time
<heat>
it's all font drawing
<heat>
would a rep movsb cause a single vmexit instead?
<heat>
I also meant vmexiting instead of ioexiting, my mistake
kspalaiologos has quit [Quit: Leaving]
sortie has quit [Ping timeout: 276 seconds]
* gog
vmexits
sortie has joined #osdev
<heat>
no
Likorn has joined #osdev
rwb is now known as rb
Likorn has quit [Quit: WeeChat 3.4.1]
Vercas has quit [Remote host closed the connection]
Vercas has joined #osdev
xenos1984 has quit [Read error: Connection reset by peer]
dude12312414 has joined #osdev
xenos1984 has joined #osdev
Vercas has quit [Remote host closed the connection]
Vercas has joined #osdev
vimal has quit [Ping timeout: 248 seconds]
xenos1984 has quit [Remote host closed the connection]
xenos1984 has joined #osdev
heat has quit [Read error: Connection reset by peer]
heat has joined #osdev
immibis has quit [Ping timeout: 240 seconds]
Likorn has joined #osdev
dude12312414 has quit [Remote host closed the connection]
dude12312414 has joined #osdev
dude12312414 has quit [Remote host closed the connection]
Terlisimo has quit [Quit: Connection reset by beer]
Terlisimo has joined #osdev
doorzan has joined #osdev
<FatalNIX>
I just realized that if I use FSF's multiboot.h it has no magic number in the structure itself.
vimal has joined #osdev
<FatalNIX>
it's in ebx :P
<FatalNIX>
i was being dumb
<mrvn>
Has to be in a register so you know the address of the struct is valid to read at all.
<mrvn>
heat: when I scroll a tty on full HD you can watch it go top to bottom while the FB is uncached.
<mrvn>
It's often faster to just redraw the whole thing than to scroll.
<FatalNIX>
Ok wth
<FatalNIX>
I just noticed though, after all this time
<FatalNIX>
the magic of multiboot 1 says "Too Bad Boo" 2
<mrvn>
...
doorzan has quit [Ping timeout: 256 seconds]
<FatalNIX>
0x2badb002
<mrvn>
Just wait for version 7
<FatalNIX>
heh
<gog>
bad boot
Burgundy has quit [Ping timeout: 250 seconds]
<gog>
big bad boot
<zid>
gog anti footwear fetish
<gog>
nah i like a good pair of boots
<zid>
I like women in boots
<zid>
but this isn't the time nor place
<geist>
hmm, too bad boo. i wonder what they were getting at
<geist>
aside from just that
<zid>
They just really like the number 2
<zid>
so it's 2ba2db202072 but that didn't fit so they removed some stuff
<geist>
0x600db007 maybe
<geist>
0x2bad2b007?
Dominika has joined #osdev
<gog>
it is impossible to know
<zid>
grub is older than git or we might be able to ask who to blame
<zid>
as it's obviously impossible without git
<geist>
totally. since no form of source control existed prior to git
<zid>
Correct.
<gog>
yes
<gog>
git was the very first
<zid>
I think someone tried a failed experiment called mercurial though, which was like a test run for whether source control was even possible
<zid>
it showed promise but didn't work
<gog>
even with git i don't think it is
<gog>
the notion that we, an insignificant species on a speck of dust in the vast cosmos that may or may not be deterministic can have control over anything in particular is laughable
<zid>
If you want to commit some code, you must first construct a universe
heat_ has joined #osdev
<zid>
hopefully one in which it's actually possible
heat has quit [Read error: Connection reset by peer]
<bslsk05>
andryblack/fontbuilder - Bitmap font generator (95 forks/409 stargazers/MIT)
vimal has quit [Remote host closed the connection]
<klange>
heat: there are a lot of different bitmap font formats; I have a tool that uses FreeType to blit out stuff for my own uses that I made before building my own TrueType library...
<klange>
I should port that over to use mine instead...
<klange>
(freetype does a better job, though, since it hints things and that makes for better bitmap presentation; I don't even have a non-antialiased rendering mode at the moment, but could make one pretty easily...)
<heat>
actually all these fonts look a bit like trash when not anti aliased :/
<klange>
That's going to be the benefit with unifont, it's natively a bitmap font
<heat>
i can't actually keep them antialiased since I can't just blit them like that