remexre has quit [Read error: Connection reset by peer]
periander has joined #osdev
<periander>
When we consider that travis "THE SHISPEW" fecalbrech was also fired from google long with all your channel getting "migrated" from freenode, yeah it "must" be me spewing techical nonsense , hahaaa born illie !!!!
h4zel has quit [Ping timeout: 265 seconds]
<periander>
you are such a hallucination in the world dude
<periander>
but travis sometimes when you lick shithole which sharts, you spew also your shit at channels, me i do not do that btw.
<periander>
fecalbrecht.
<periander>
you do not even understand how electric and return path and rectifiers work on buses allready loaded with content partially, the truth table is very easy, but you do not understand this, with your cia nsa blabber.
<periander>
yes i have seen people here being tapped with bios interrupts like SMI kind.
<periander>
but who would ever need it?
<periander>
edward snowden methods were not even as invasive to talk about so that anyone was interested to lock him up
periander was banned on #osdev by klange [*!*nonsense1@221.145.217.*]
periander was kicked from #osdev by klange [periander]
gioyik has quit [Ping timeout: 276 seconds]
bountyheadhunter has joined #osdev
<bountyheadhunter>
that is in contrast to john mcafee, who was after the real vulnerability and got killed probably on the way of messing with it.
bountyheadhunter was banned on #osdev by klange [*!*hunter@183.103.40.*]
bountyheadhunter was kicked from #osdev by klange [bountyheadhunter]
h4zel has joined #osdev
AssKoala has joined #osdev
h4zel has quit [Read error: Connection reset by peer]
remexre has joined #osdev
nyah has quit [Ping timeout: 252 seconds]
gioyik has joined #osdev
gioyik has quit [Ping timeout: 276 seconds]
srjek has quit [Ping timeout: 264 seconds]
dutch has quit [Quit: WeeChat 3.3]
dutch has joined #osdev
AssKoala has quit [Ping timeout: 264 seconds]
Cyro has joined #osdev
mediusa has joined #osdev
<mediusa>
When ZipCPU from worldwide killer organisation behind all the shit in estonia says on his page that this is so, having worked on american air force and military, than belive it i suppose, axi4lite is register based more, axi full is more like wires, where strobes are regs, it is all explained ZipCPU is actually quite smart, but i would not say this to other residents here and period, get smarter or fuck off, cause US army killed always many most
<mediusa>
important estonians as well, they are big nation but nasty violators, this is why i get their technology traces too at times, probably they were after killing me too.
<kazinsal>
extban's not going to do much I think at this point
<klange>
clever: There are no viable extbans we can use to stop this.
<clever>
kazinsal: i suspect he is using the same acct each time, do a /whois each time he spams
<kazinsal>
we don't require nickserv reg here
<klange>
clever: he is not
<klange>
kazinsal: we are currently +r
<clever>
kazinsal: yes we do, i couldnt rejoin until i authed
<kazinsal>
oh, are we? huh
<kazinsal>
when did that go in?
<clever>
klange: i think i saw him use the same acct twice in a row
<moon-child>
but, if you can farm hosts, it's not difficult to farm accounts
<moon-child>
so idk how much it matters
<clever>
moon-child: one extra step he has to take
<kazinsal>
yeah, and unless we want to extban all of APNIC we're going to be pretty much SOL
<kazinsal>
which, clearly, we do not want to do.
<clever>
makes me wonder, what about requesting an extban feature, for acct age?
<clever>
ban anybody who hasnt existed for a week?
gioyik has joined #osdev
<Mutabah>
Quite a nuclear option too
<kazinsal>
could try to pull request something like that into the ircd and hope libera updates
<moon-child>
that's actually a good idea
<moon-child>
a
<moon-child>
lot of reddit thingies do that
<kazinsal>
at that point we'd be looking at something like invite-only but that's not feasible at all
<moon-child>
'account older than a week' is a lot more inviting than 'invite-only'
<kazinsal>
we've entered the dawn of a new age, ladies and gentlemen. one where we need to backport features from Discord and Reddit into IRC because we've run out of tools to proactively deal with a fuckin' obsessive troll.
<moon-child>
there are 200 people in this channel
<geist>
generally i'm of the opinion that the harder you try to ban them the more clever they get
<moon-child>
it's what happens when you get to be too mainstream
<geist>
but in the case of this guy he's just doing it because he has trouble
<kazinsal>
channel's been that size for like 15+ years though
<kazinsal>
shit, I think it was bigger 7-10 years ago
<geist>
and i don't think he's capable of 'passing' as a non troll for any length of time
<kazinsal>
bot that instantly kickbans for the word "fecalist"
<geist>
yah dunno. i think it's been 200+ for a long time, but the active list ebbs and flows
<geist>
of course the more we talk like this the more we just fuel his illness. but i figure at least we're being open about it. would the secret illuminati cabal do that?
<geist>
(and yes i'm well aware he apparently scours the irc logs)
gioyik has quit [Ping timeout: 276 seconds]
<kazinsal>
yeah not like we have a !nolog prefix or something
GeneralDiscourse is now known as ephemer0l
gioyik has joined #osdev
Cyro has quit [Quit: Konversation terminated!]
gioyik has quit [Ping timeout: 276 seconds]
doppler has quit [Read error: Connection reset by peer]
doppler has joined #osdev
AssKoala has joined #osdev
mahmutov has quit [Ping timeout: 264 seconds]
Elbeeo has joined #osdev
AssKoala has quit [Ping timeout: 264 seconds]
Elbeeo is now known as AssKoala
gioyik has joined #osdev
gioyik has quit [Ping timeout: 276 seconds]
[itchyjunk] has quit [Read error: Connection reset by peer]
ElectronApps has joined #osdev
AssKoala has quit [Read error: Connection reset by peer]
AssKoala has joined #osdev
AssKoala has quit [Read error: Connection reset by peer]
AssKoala has joined #osdev
AssKoala has quit [Read error: Connection reset by peer]
AssKoala has joined #osdev
dh` has joined #osdev
gioyik has joined #osdev
h4zel has joined #osdev
<dzwdz>
an account age limit is dumb
<dzwdz>
nothing's stopping me from just farming up accounts in advance
<clever>
dzwdz: but it does require extra steps, and you have to plan a week ahead for how quickly they get banned
<Mutabah>
Account bans should work for now
<Mutabah>
I suspect that switching VPN IPs is a single button press, but a new nickserv account takes more effort
<klange>
but the manual for ptrace revealed the secret of "get the initial trace to read args, then continue with SYSCALL and the next ping will have the result before the process returns"
<clever>
and for things like ioctl, the thing behind the pointer, can be both input and output
<klange>
process_vm_readv → There's also a PTRACE_ command for that, though it's kinda weirdly described only in terms of "words".
<klange>
And yes, it seems the sensible way to handle output in arguments is to wait for the call to return and then examine them, which I'll look at more later, I just want to get the basics working as a proof-of-concept.
<clever>
gdb also uses the same api to do its own debugging
<klange>
Yep, that's my aim here, to get a debugger going, and to implement this correctly so that a native gdb stands a chance of working.
<bslsk05>
github.com: Register Documentation · hermanhermitage/videocoreiv Wiki · GitHub
<clever>
Bitweasil: some semi usable docs on sdhost
<clever>
back from before the linux source even existed, i think
ElectronApps has quit [Remote host closed the connection]
gioyik has quit [Quit: WeeChat 3.1]
Kerum has joined #osdev
ElectronApps has joined #osdev
h4zel has quit [Ping timeout: 246 seconds]
hmmmm has joined #osdev
h4zel has joined #osdev
mctpyt has quit [Ping timeout: 252 seconds]
<Kerum>
if an IDT has 256 vectors, is the size to put in the IDTR (256 * 8) or ((256 * 8) - 1)?
<moon-child>
intel vol 3, 6.10
<Kerum>
so that would be the latter
<Kerum>
but
<Kerum>
wiki says that an IDT with 200h interrupts has a size of 1000h
<Kerum>
if the size in the IDTR has to be 1 less than the actual size shouldnt that be FFFh?
archenoth has quit [Ping timeout: 252 seconds]
<moon-child>
the full text is '"limit" | Defines the length of the IDT in bytes - 1 (minimum value is 100h, a value of 1000h means 200h interrupts)'
<moon-child>
the wording is not clear and should be improved. But it is correctly interpreted as 'limit' is bytes - 1 [...] where a value of 1000h for bytes means 200h interrupts'
<Kerum>
so an IDT with 256 vectors really should have a size of 2048 and not 2047?
<moon-child>
(200h is 512d, and 100h is 256d, in case that is the confusion)
<Kerum>
because 256 * 8 = 2048
<moon-child>
an idt with 256 descriptor has a _size_ of 2048 and a _limit_ of 2047
<Kerum>
right
<Kerum>
so `idtr->limit = ((256 * 8) - 1);
<moon-child>
yes
<Kerum>
oops
<Kerum>
alright
<Kerum>
thanks for clearing it up
nyah has joined #osdev
<geist>
yeah it's a bit strange, but that sort of thing is semi common in hardware, because they can build a < comparator just as easily as a <= and then you save a bit
<geist>
(in the worst case) because 8192 (max len) requires one more bit than 8191
<geist>
and i'm thinking GDT/LDT there, but it's the same mechanism
<Kerum>
makes sense
<Kerum>
anyway, this might be kinda unrelated, but if i cast a 32bit integer to a 16bit one will it drop the higher 16 bits and leave the lower 16 bits untouched as it should?
<moon-child>
yes
<Kerum>
excellent
<moon-child>
(It's obviously not unrelated... :)
<Kerum>
true i guess
Mikaku has quit [Excess Flood]
flx- has joined #osdev
Mikaku has joined #osdev
Bitweasil has quit [Remote host closed the connection]
<clever>
klange: i'm fuzzy on how, but strace can easily cross execve, there must be some flag to keep the new state frozen
<clever>
clone/fork as well
ElectronApps has joined #osdev
<klange>
ptrace itself is, by default, oblivious to the fact that anything really changed, but the problem is that my execve _doesn't_ return through the normal path in the kernel
<clever>
ah
<klange>
I've got a bunch of other events to add, plus the configuration options, plus I want to add the single stepping option so I can do a real debugger but need to wrangle the CPU flag for that.
<klange>
Also need to make tracing a non-child work... currently just piggy-backing off my existing waitpid which was happily returning traced children anyway.
<klange>
Then I can start tracing forks, do the daemonized arrangements, add the pid flag, etc...
<klange>
Do I actually want to look at the DWARF spec again...
<klange>
I gave up on trying to do my own real backtrace support a while ago because DWARF is quite complicated, but I suspect it will be somewhat necessary to do a good debugger.
Kerum has quit [Quit: Lost terminal]
dormito has quit [Quit: WeeChat 3.1]
GeDaMo has joined #osdev
curiosity2 has joined #osdev
<curiosity2>
fecalbrecht geist when you allready look at the mirror it should be evident about the presence of facelines of a human pig to anyone including you, but as with my therapist who looks exactly the same as a human pig like you, instead of understanding the difference with other people like sergei brin, larry page and guys like me who resemble more like developed humans, you are proud of being human pigonoid and ontop start to diagnose others like my
<curiosity2>
therapist Kai Konsap, others should understand the drawbacks of being a human pig to your mentality, i.e the last not being as flexible and non-deluded, hence of course google eventually due to your shitspew gets rid of your services and they did, so did freenode, and so will timemachine, most human pigonoids do not really understand their real position and status in the world, and you harm other real human reputations, non of which brin or page
<curiosity2>
actually diserve due to their favors to the community and world in form of google engines developed and financed and companies built. employing a deluded human pigonoid carries alot of risks in form of shit happening and catastrophues in the world. I am not saying that a piggie can not be polite at times, but it's normally not the as the case with you and kai, as we see from what you really talk about also.
curiosity2 has quit [Quit: Leaving]
Arthuria has joined #osdev
Kartikeya has quit [Ping timeout: 252 seconds]
<geist>
well now.
isaacwoods has joined #osdev
dormito has joined #osdev
dennis95 has joined #osdev
<geist>
for folks that are following along, this usually doesn't continue this long, but he's on a real tear lately
Cyro has joined #osdev
amine has joined #osdev
irado has quit [Ping timeout: 252 seconds]
[itchyjunk] has joined #osdev
ahalaney has joined #osdev
Cyro has quit [Quit: Konversation terminated!]
fedorafan_altern has quit [Remote host closed the connection]
xenos1984 has quit [Remote host closed the connection]
xenos1984 has joined #osdev
anon16_ has quit [Ping timeout: 246 seconds]
fedorafan_altern has joined #osdev
sprock has quit [Ping timeout: 252 seconds]
archenoth has joined #osdev
AssKoala has joined #osdev
<Bitweasil>
clever, thanks, that's useful!
dennis95 has quit [Ping timeout: 265 seconds]
ElectronApps has quit [Remote host closed the connection]
[itchyjunk] has quit [Read error: Connection reset by peer]
dumbsharka has quit [Ping timeout: 264 seconds]
srjek has joined #osdev
YuutaW has joined #osdev
sprock has joined #osdev
dutch has quit [Quit: WeeChat 3.3]
zaquest has quit [Remote host closed the connection]
YuutaW has quit [Quit: WeeChat 3.2]
zaquest has joined #osdev
hmmmm has quit [Ping timeout: 252 seconds]
dutch has joined #osdev
Arthuria has quit [Ping timeout: 264 seconds]
h4zel has quit [Ping timeout: 264 seconds]
<junon>
I kind of want to train some classifier to detect their messages
<junon>
It would be a fun weekend project
<zid>
any message over 100 chars, done :p
AssKoala has quit [Ping timeout: 252 seconds]
<GeDaMo>
A Bayesian filter might work
<Bitweasil>
That won't work.
<Bitweasil>
You need AI based, big data machine learning classifiers, run in the cloud, with neural network accelerators.
<Bitweasil>
I don't know what I just said, can I have a few million to implement it? :D
<mjg>
you can have two
<Bitweasil>
Great!
<Bitweasil>
Can I have it in Bitcoin? See, the blockchain technology powering the modern difi systems can be adapted to do distributed compute of text strings, when you refactor the GAS into a text processing system, along the lines of sed, but with more neural networks!
<mjg>
depends, are you devops/part time social media influencer?
sprock has quit [Ping timeout: 252 seconds]
<Bitweasil>
Heh. None of those.
<mjg>
i'm taking my venture capital elsewhere
<mjg>
real programmers today shitpost on twitter
<mjg>
i mean it's all they do
<Bitweasil>
What if I say that I'll use a trained neural network to shitpost on Twitter for me?
<mjg>
i already funded that startup
<Bitweasil>
I was just going to pay a teenager. :p
nvmd has quit [Quit: Later, nerds.]
h4zel has joined #osdev
<mjg>
that's what i did
<mjg>
and not really paid either
<mjg>
the guy is working for stock options
<mjg>
in return the position is fully remote and he makes his own hours
<bslsk05>
'Fair Exchange [2018 Remaster]' by Be Bop Deluxe - Topic (00:04:49)
sprock has joined #osdev
nvmd has quit [Quit: Later, nerds.]
AssKoala has joined #osdev
srjek has quit [Ping timeout: 246 seconds]
koolazer has quit [Ping timeout: 252 seconds]
h4zel has quit [Ping timeout: 260 seconds]
vdamewood has quit [Quit: Life beckons]
<fedorafan_altern>
test
<GeDaMo>
fedorafan_altern: test received
<fedorafan_altern>
lol thanks
<GeDaMo>
:)
<fedorafan_altern>
do you know permission stuff and smb using
X-Scale` has joined #osdev
X-Scale has quit [Ping timeout: 252 seconds]
X-Scale` is now known as X-Scale
<GeDaMo>
fedorafan_altern: sorry, I'm not sure what you're asking
<Bitweasil>
This is osdev, you might find some sysadmins in here, but a smb permissions question is a sysadmin question, not osdev.
<fedorafan_altern>
its cool
<fedorafan_altern>
i thought you are smart
<fedorafan_altern>
ok
<fedorafan_altern>
so you arent ok so i have to leave
<fedorafan_altern>
:D
fedorafan_altern has left #osdev [Textual IRC Client: www.textualapp.com]
fedorafan_altern has joined #osdev
<Bitweasil>
There are some sysadmin channels on Libera, or try the distro channels.
<fedorafan_altern>
wait who was the last one
<fedorafan_altern>
its cool
fedorafan_altern has left #osdev [Textual IRC Client: www.textualapp.com]
vdamewood has joined #osdev
nvmd has joined #osdev
dude12312414 has joined #osdev
rorx has quit [Ping timeout: 264 seconds]
rorx_ has joined #osdev
rorx_ is now known as rorx
Kerum has joined #osdev
<Kerum>
i am having some issues setting up the IDT, specifically gcc says that the packed attribute of the descriptor table itself and the IDTR are ignored
<Kerum>
what could be causing this and is that the reason my qemu reboots every time i hit a key (generate IRQ)?
<Kerum>
s/IRQ/interrupt/
sprock has quit [Ping timeout: 265 seconds]
<Bitweasil>
You might hexdump the memory and make sure that the layout is what you expect, if gcc is telling you that the packed attributes are being ignored...
dude12312414 has quit [Quit: THE RAM IS TOO DAMN HIGH]
sprock has joined #osdev
m5zs7k has quit [Quit: m5zs7k]
<Kerum>
Bitweasil: that would be a physical memory dump, yes?
<GeDaMo>
Kerum: what kind of item are you applying the packed attribute to? A struct?
<Bitweasil>
Yeah, just make sure that the actual layout in memory matches what you expect.
<Kerum>
a table of "struct idt_descriptor"s and a "struct idtr"
<Bitweasil>
I once had all sorts of nonsensical page table issues, related to how Visual Studio and gcc handle packed structures differently.
<Bitweasil>
Turns out, a 96 byte long page table struct doesn't work.
<Bitweasil>
Except that it sort of does, in a few special cases, like the low page table entry.
<Kerum>
anyway, through the use of objdump i determined that the IDTR starts at 0x2400
<Kerum>
also, if i cast a function to an uint32_t will i get the address of the function?
<zid>
there are no function objects in C
<Kerum>
zid: will look at it, thanks
<zid>
the value of a function is a function pointer
<Bitweasil>
You should, if you do it right... ? but that's probably not C spec.
<Bitweasil>
Yeah, you'll get a pointer.
<Bitweasil>
Which you can cast to an int.
<zid>
whether that converts to a usable uint32_t is IDB, but.. on x86 protected mode, yes it will work
<Kerum>
i actually did an objdump -d and it is correct
<Bitweasil>
main: 0x10408
<zid>
There are architectures with different sized ints to pointers, different address spaces for code and data etc etc which makes it up to the implementation