Noisytoot has quit [Remote host closed the connection]
Noisytoot has joined #commonlisp
Noisytoot has quit [Ping timeout: 246 seconds]
Noisytoot has joined #commonlisp
kiyafirs has quit [Remote host closed the connection]
King_julian has joined #commonlisp
msavoritias has joined #commonlisp
zxcvz has joined #commonlisp
zxcvz has quit [Quit: zxcvz]
shka has joined #commonlisp
amb007 has quit [Ping timeout: 240 seconds]
amb007 has joined #commonlisp
Th30n has joined #commonlisp
zxcvz has joined #commonlisp
King_julian has quit [Ping timeout: 246 seconds]
kyborg2011 has quit [Ping timeout: 256 seconds]
kyborg2011 has joined #commonlisp
skeemer has quit [Ping timeout: 252 seconds]
pve has joined #commonlisp
Cymew has joined #commonlisp
nybble has quit [Ping timeout: 255 seconds]
danse-nr3 has joined #commonlisp
jonatack has quit [Ping timeout: 264 seconds]
dino_tutter has joined #commonlisp
Cymew has quit [Client Quit]
jonatack has joined #commonlisp
manwithluck has joined #commonlisp
msavoritias has quit [Ping timeout: 268 seconds]
amb007 has quit [Read error: Connection reset by peer]
varjag has joined #commonlisp
amb007 has joined #commonlisp
mgl has joined #commonlisp
kg7ski has quit [Ping timeout: 268 seconds]
attila_lendvai has joined #commonlisp
zxcvz has quit [Quit: zxcvz]
zxcvz has joined #commonlisp
kg7ski has joined #commonlisp
msavoritias has joined #commonlisp
dlowe has quit [Ping timeout: 264 seconds]
dlowe has joined #commonlisp
Cymew has joined #commonlisp
donleo has joined #commonlisp
inline-_ has quit [Quit: Leaving]
Inline has joined #commonlisp
zxcvz has quit [Quit: zxcvz]
Cymew has quit [Quit: Konversation terminated!]
Cymew has joined #commonlisp
King_julian has joined #commonlisp
Cymew has quit [Quit: Konversation terminated!]
Inline has quit [Remote host closed the connection]
Inline has joined #commonlisp
msavoritias has quit [Remote host closed the connection]
Cymew has joined #commonlisp
rainthree has quit [Remote host closed the connection]
rainthree has joined #commonlisp
King_julian has quit [Ping timeout: 264 seconds]
random-nick has joined #commonlisp
kyborg2011 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
mwnaylor has quit [Ping timeout: 256 seconds]
amb007 has quit [Ping timeout: 255 seconds]
amb007 has joined #commonlisp
amb007 has quit [Read error: Connection reset by peer]
amb007 has joined #commonlisp
danse-nr3 has quit [Ping timeout: 268 seconds]
yitzi has joined #commonlisp
rainthree has quit [Remote host closed the connection]
dawids has joined #commonlisp
danza has joined #commonlisp
dawids has quit [Client Quit]
Th30n has quit [Ping timeout: 255 seconds]
Th30n has joined #commonlisp
rtypo has joined #commonlisp
danza has quit [Ping timeout: 240 seconds]
skeemer has joined #commonlisp
jonatack has quit [Ping timeout: 256 seconds]
Posterdati has joined #commonlisp
jonatack has joined #commonlisp
danse-nr3 has joined #commonlisp
King_julian has joined #commonlisp
Cymew has quit [Quit: Konversation terminated!]
King_julian has quit [Ping timeout: 272 seconds]
King_julian has joined #commonlisp
kyborg2011 has joined #commonlisp
msavoritias has joined #commonlisp
jon_atack has joined #commonlisp
awlygj has joined #commonlisp
jonatack has quit [Ping timeout: 272 seconds]
jon_atack has quit [Quit: WeeChat 4.3.2]
<cdegroot>
I wish I had a game to write so I could test Trial. Love the idea. Poor people in the industry that still think that C++ is the bee's knees ;-)
<cdegroot>
(a game with a website so I could test CLOG as well lol)
<beach>
Maybe Shinmera could suggest something for you.
<Shinmera>
iunno just clone an existing game you like
<cdegroot>
(and bought Kandria, I still had a tab open from a couple of weeks ago when I stumbled upon Trial, sorry for the delay, hth etc)
<Shinmera>
thanks
<cdegroot>
Well, there's "I have an idea" and there's "I have time to do it" (I'd love to do a riff on evoland but then you start with pong and end up in an MMO or something), there's this book that wants to be finished. Well, not the book that wants it, but the editor is getting insistent :)
<cdegroot>
(don't worry, it's a Lisp book, so I'm not straying from the One True Path :P )
<cdegroot>
But yeah, developing a game in CL makes so much sense.
<Shinmera>
Not really if you actually want to make a profit
<cdegroot>
I have a warning letter left with friends, "open in case he gets mad enough to want to make money with a game" (jk, but yeah... It happens, though; I've got a bunch of indie games where the creators should be able to make a reasonable living off it, and none of them should be out of reach for an engine like Trial. In fact, that's what I like about indie games - they focus more on gameplay than on the shinies)
<cdegroot>
(I grew up playing text adventures; totally enthralling, zero shinies. It went downhilll from there lol)
<Shinmera>
I think you're vastly overestimating the accessibility of something like Trial
jonatack has joined #commonlisp
josrr has joined #commonlisp
jonatack has quit [Read error: Connection reset by peer]
jonatack has joined #commonlisp
kevingal has joined #commonlisp
attila_lendvai has quit [Ping timeout: 256 seconds]
<cdegroot>
I
<cdegroot>
I'll let you know once I, err, trialed it :)
kg7ski- has joined #commonlisp
kg7ski has quit [Ping timeout: 240 seconds]
skeemer has quit [Ping timeout: 264 seconds]
<Shinmera>
It's not that I don't care about the accessibility of Trial, it's just that it's not conducive to my current goal of making games myself
<Shinmera>
And I'm not being paid (enough) to just develop the engine
ocra8 has joined #commonlisp
<younder>
Ironically your talk is followed by a talk on the Godot game engine
<Shinmera>
Not sure how that's ironic
<Shinmera>
There were 9 presentations on different engines that evening
<pranav>
Shinmera: Were these invited talks? Or do your coworking space really has that many game engine devs?
<Shinmera>
folks were asked to present on something they knew, but aside from myself nobody here was actually an engine dev
<Shinmera>
they just knew how to use it
amb007 has quit [Read error: Connection reset by peer]
<younder>
How on earth do you get the idea 'I want to develop a game engine in Common Lisp'?
<pranav>
I see. So all were happy (potential) customers.
<Shinmera>
I've been making games all my life. I just wanted to try doing it.
<Shinmera>
Hence: Trial
<dlowe>
It doesn't seem that weird to me. CL is cool, game engines are cool.
<younder>
Like I'm a Lisper, but I would grab for C
<Shinmera>
now *that* is an astoundingly terrible idea
<Shinmera>
I know it's a frequent one, but still terrible
amb007 has joined #commonlisp
<dlowe>
C# seems to be most common now due to Unity
<dlowe>
I would reach for CL first if I were going to make a game.
<dlowe>
I mean, I did do that. I just didn't follow through once I got the coding part done :p
<younder>
shinmmey why is it terrible to develop in c?
<Shinmera>
wow, that's a new butchering of my name
<dlowe>
Then all the necessary assets and game building started to get overwhelming
<dlowe>
Shinmera deserves all the respect and love for following through
<Shinmera>
because C is an awfully slow language to develop in, especially for games which require frequent and fast iterative development
jon_atack has joined #commonlisp
<younder>
sorry not sure how that happened shinmera
<dlowe>
C can also BE slow due to using inappropriate logic to avoid code-level complexity
<cdegroot>
It's not even king of the hill w.r.t. speed anymore. Worse: half our CPUs have kludges to make C benchmarks go fast. I've been paid for the better part of a decade to write C code, and I like the language, but it should go the way of the dodo.
jonatack has quit [Ping timeout: 272 seconds]
<dlowe>
So long and thanks for the aBI
<hayley>
The great thing about C ABI standards is that there's so many interpretations to choose from.
<Shinmera>
thinken bout that SBCL struct-by-value PR
<dlowe>
C++ ABIs, sure. I thought C's was pretty stable
eddof13 has joined #commonlisp
<Shinmera>
stable sure, but there's still multiple
<cdegroot>
C classically had two calling conventions (well, I know more, but the main ones were just these): C mode where the caller cleaned up the stack and "Pascal" mode where the callee cleaned up the stack. The latter, In Ye Olde Days, could be faster but less flexible for varargs and fun like that.
<hayley>
(Not that you use 128-bit integers much in C, the ABIs do have 'em.)
jon_atack has quit [Ping timeout: 264 seconds]
<pranav>
Shinmera: Do you think Trial would benefit significantly from using fast-generic-functions/sealable metaobjects?
<Shinmera>
iunno, never tried
<pranav>
Okay
<Shinmera>
we were planning on doing something like that for he Switch port, since it's necessary there
<Shinmera>
but atm that's kinda on hold because I don't know that I should spend more money on it, with my finances already the way they are
<pranav>
Although inlining generic functions would block any use of block compilation, I suppose.
<younder>
The biggest flaw in C as I see it is the insistence of defaulting to 32 bit integers and disallowing anything more than SSE2. Not that is not a language problem but a implementation problem.
<younder>
Yes, I know I can override that with a lang=native or something
<cdegroot>
The biggest flaw in C is essentially everything that Rust added to C to make it workable ;-)
King_julian has quit [Ping timeout: 255 seconds]
<hayley>
Perhaps I too should start doing drugs.
<younder>
Please do. It would cut you down to our size ;)
dawids has joined #commonlisp
<cdegroot>
(in all honesty: C is a pretty great language for what it should be doing: gluing the low level system bits to the higher level languages that do the heavy lifting; it can't help that it became an application language and then sprouted a very ugly child with some bad ideas about OO. But Rust is a good cleanup and replacement)
<cdegroot>
(with a community making the same mistake: using a very low level close to the hardware language for everything and the kitchen sink)
King_julian has joined #commonlisp
<bike>
i dunno, this article hayley linked makes a pretty compelling case otherwise.
<younder>
Quite frankly your obvious intelligence can be quite unnerving!
dawids has quit [Client Quit]
<Shinmera>
Anyway, please give me money on patreon so I can continue to waste my life with Lisp
gone123 has joined #commonlisp
gone123 has quit [Changing host]
gone123 has joined #commonlisp
chomwitt has joined #commonlisp
eddof13 has quit [Quit: eddof13]
NotThatRPG has joined #commonlisp
jon_atack has joined #commonlisp
jonatack has quit [Ping timeout: 252 seconds]
hayley has left #commonlisp [#commonlisp]
eddof13 has joined #commonlisp
gone123 has quit [Remote host closed the connection]
gone123 has joined #commonlisp
gone123 has quit [Remote host closed the connection]
gone123 has joined #commonlisp
lucasta has joined #commonlisp
NotThatRPG has quit [Ping timeout: 272 seconds]
King_julian has quit [Ping timeout: 268 seconds]
varjag has quit [Quit: ERC (IRC client for Emacs 27.1)]
King_julian has joined #commonlisp
jonatack has joined #commonlisp
skeemer has joined #commonlisp
chiselfuse has quit [Remote host closed the connection]
chiselfuse has joined #commonlisp
eddof13 has quit [Quit: eddof13]
jon_atack has quit [Ping timeout: 252 seconds]
msavoritias has quit [Ping timeout: 255 seconds]
jonatack has quit [Ping timeout: 264 seconds]
danse-nr3 has quit [Read error: Connection reset by peer]
jonatack has joined #commonlisp
X-Scale has joined #commonlisp
danse-nr3 has joined #commonlisp
King_julian has quit [Ping timeout: 264 seconds]
markb1 has quit [Read error: Connection reset by peer]
eddof13 has joined #commonlisp
jonatack has quit [Ping timeout: 256 seconds]
X-Scale has quit [Ping timeout: 250 seconds]
random-nick has quit [Ping timeout: 268 seconds]
markb1 has joined #commonlisp
gone123 has quit [Remote host closed the connection]
NotThatRPG has joined #commonlisp
random-nick has joined #commonlisp
markb1 has quit [Read error: Connection reset by peer]
jonatack has joined #commonlisp
istewart has joined #commonlisp
edgar-rft_ has joined #commonlisp
markb1 has joined #commonlisp
edgar-rft has quit [Ping timeout: 268 seconds]
pfdietz has joined #commonlisp
kyborg2011 has quit [Ping timeout: 255 seconds]
kyborg2011 has joined #commonlisp
edgar-rft_ is now known as edgar-rft
wacki has quit [Quit: My iMac has gone to sleep. ZZZzzz…]
istewart has quit [Quit: Konversation terminated!]
danse-nr3 has joined #commonlisp
mgl has quit [Ping timeout: 256 seconds]
mgl_ has quit [Ping timeout: 264 seconds]
lucasta has quit [Quit: Leaving]
nybble has joined #commonlisp
pranavats has joined #commonlisp
mgl_ has joined #commonlisp
cage has joined #commonlisp
Inline has quit [Remote host closed the connection]
Inline has joined #commonlisp
amb007 has quit [Ping timeout: 264 seconds]
mgl_ has quit [Ping timeout: 272 seconds]
amb007 has joined #commonlisp
simendsjo has joined #commonlisp
danse-nr3 has quit [Quit: Leaving]
markb1 has quit [Read error: Connection reset by peer]
eddof13 has quit [Quit: eddof13]
pfdietz has quit [Ping timeout: 250 seconds]
Lord_of_Life_ has joined #commonlisp
mgl_ has joined #commonlisp
Lord_of_Life has quit [Ping timeout: 272 seconds]
Lord_of_Life_ is now known as Lord_of_Life
markb1 has joined #commonlisp
chomwitt has quit [Ping timeout: 264 seconds]
ym has joined #commonlisp
kevingal has quit [Ping timeout: 268 seconds]
rtypo has quit [Quit: WeeChat 4.2.2]
ocra8 has quit [Quit: WeeChat 4.3.2]
eddof13 has joined #commonlisp
josrr has quit [Ping timeout: 268 seconds]
attila_lendvai has joined #commonlisp
markb1 has quit [Read error: Connection reset by peer]
jrm has quit [Ping timeout: 268 seconds]
mgl_ has quit [Ping timeout: 264 seconds]
waleee has joined #commonlisp
jrm has joined #commonlisp
kevingal has joined #commonlisp
ym has quit [Ping timeout: 256 seconds]
jonatack has quit [Ping timeout: 268 seconds]
jonatack has joined #commonlisp
josrr has joined #commonlisp
markb1 has joined #commonlisp
ym has joined #commonlisp
King_julian has joined #commonlisp
attila_lendvai has quit [Ping timeout: 272 seconds]
attila_lendvai has joined #commonlisp
vats has joined #commonlisp
waleee has quit [Ping timeout: 268 seconds]
eddof13 has quit [Quit: eddof13]
King_julian has quit [Ping timeout: 268 seconds]
eddof13 has joined #commonlisp
<NotThatRPG>
Indirectly a CL question: how does one read a CL s-expression using emacs-lisp? I was trying to use elisp's `read`, but it has immutable syntax that differs from CL syntax (notably the question mark character is not an acceptable first character for a symbol). Hoping someone knows a quick answer to this
<paulapatience>
I don't think there's a way.
<paulapatience>
Outside of writing your own reader
eddof13 has quit [Client Quit]
<pl>
cdegroot: C is pretty shitty language as glue language, which is partly why it's so hard to glue things together in binary on unix
<pl>
a lot of C early, unmoveable design decisions stem from it having to fit both the compiler and working set in 32 kB
jon_atack has joined #commonlisp
jonatack has quit [Ping timeout: 264 seconds]
attila_lendvai has quit [Ping timeout: 256 seconds]
<cdegroot>
Nothing is every unmovable, but yes, the origin of the language is far removed from today's hardware and that hasn't helped. But not really an argument if we're in a channel that talks about a language that is nearing its 70th birthday ;-)
<gilberth>
Our hardware doesn't look much different. Or the ISA rather. Words got wider and address space is plenty. Otherwise? All the same.
<NotThatRPG>
paulapatience: I think that's right. Was hoping someone could point me at a reader already out there. I figure there must be one in Slime, Sly, or CL mode...
m5zs7k has quit [Ping timeout: 252 seconds]
<gilberth>
And I doubt the there has been much design in C. I see C as a hack on B needed because the PDP-11 had poiners to bytes and was not word oriented. That gave us 'char'. And it is exactly those byte pointers that your current hardware has as well. Or the ISA rather.
mgl_ has joined #commonlisp
<gilberth>
From B came that there are no arrays as rvalues in C. It's surprising that structs are. But they came later.
m5zs7k has joined #commonlisp
wacki has quit [Ping timeout: 256 seconds]
jon_atack has quit [Read error: Connection reset by peer]
wacki has joined #commonlisp
attila_lendvai has joined #commonlisp
jonatack has joined #commonlisp
mgl_ has quit [Ping timeout: 264 seconds]
<pl>
it took 15 years for C to accept identifiers longer than 6 characters because of the original 32kB size limit
kevingal has quit [Ping timeout: 264 seconds]
<pl>
B and C both had design applied based on "we need to do this in as little amount of passes as possible"
<gilberth>
Doesn't six come from six chars = 36 bit? I always thought so.
<gilberth>
pl: Yes.
<gilberth>
And it still is impressive what they did in those 32KB.
<gilberth>
The C compiler won't care for types. It would be ok with: struct foo { int a, b; }; .... int p; p->a = 42; e.g. UNIX sources are full of this.
<pl>
gilberth: six chars in RADIX50, iirc
<pl>
it's also why "old" C required unique names in struct fields
<pl>
there was no space given to differentiate symbols in structs from those outside them
<gilberth>
I am aware.
cage has quit [Quit: rcirc on GNU Emacs 29.3]
<pl>
RADIX50 in 16bit version encoded 3 characters per 16bit word
vats has quit [Ping timeout: 240 seconds]
<pl>
similarly it encoded 3 characters per word on 18bit systems, from which early unix came
<gilberth>
And I always wondered if with B it was custom to say e.g. #define x 0 \n #define y 1 \n ... foo(p) { return x[p] + y[p]; }
mgl_ has joined #commonlisp
<gilberth>
Or in modern C: enum { x, y }; int foo (int *p) { return x[p] + y[p]; } perhaps with saying e.g. pt_x instead of just x.
<pl>
I'd like also to add that reading Unix as "eunuch" implying it was castrated Multics was a pun that was used in Bell Labs
attila_lendvai has quit [Ping timeout: 256 seconds]
josrr has quit [Ping timeout: 264 seconds]
<gilberth>
Still, I invite people that blame UNIX too much to come up with an as usable and practical system in 32KB.
chrcav has quit [Ping timeout: 255 seconds]
kevingal has joined #commonlisp
chrcav has joined #commonlisp
<yitzi>
NotThatRPG: load emacs CL emulation and attempt to load Eclector?
<NotThatRPG>
yitzi: woof! That's a pretty big lift!
<yitzi>
Well, a CL reader with character traits, a readtable, symbol macros isn't?
<NotThatRPG>
If emacs has CL emulation, can't I just use that? I didn't know there was such a beast?
<contrapunctus>
I'm using `quri:make-uri` and if I pass it something like `:params '(("key" . "val1,val2"))`, it URL-encodes the comma into %2C, and if I use that URI with `dex:request`, the request is rejected by the server (400 bad request). How do I prevent `quri` from URL-encoding the parameter value?
<NotThatRPG>
yitzi: Oh, that's just the cl-macs.
<NotThatRPG>
no cl-reader.
<NotThatRPG>
I can't imagine that would be enough to load Eclector. Probably best to have a companion CL process.
<NotThatRPG>
Man, that's stinky
pranavats has left #commonlisp [Error from remote client]
<gilberth>
contrapunctus: The question should be rather how to fix dex:request. That's the part where the error is.
<contrapunctus>
gilberth: What's the issue with `dex:request`?
<gilberth>
It should unescape that %2c.
yuki74 has joined #commonlisp
ym has quit [Ping timeout: 256 seconds]
<cdegroot>
Where does the 32kB come from? I fool around with a PDP-10 emulator from time-to-time, same era, but that thing had more like a megabyte.
awlygj has quit [Quit: leaving]
<pl>
cdegroot: PDP-11
<pl>
UNIX on PDP-11 had 32kB for kernel space and 32kB for application
wacki has quit [Quit: My iMac has gone to sleep. ZZZzzz…]
<cdegroot>
So the -11 was smaller than the -10? (I must confess, I only dug in as far as I needed for some book research, my to do list was too long for a deep dive into retrocomputing, no matter how tempting it was). DEC model numbers confuse me...
<pl>
yes
<pl>
PDP-11 was the cheap mini
<pl>
PDP-10 was huge multiprocessor mainframe
<pl>
that was nearly killed because PDP-6 was such a flop, so first PDP-10 had technically possible "super small" configuration that was never sold but which was used to bypass scrutiny at Digital
<pl>
late model PDP-10 used a PDP-11 as front end processor
<pl>
there exists extended addressing for PDP-11, but it arrived pretty late
<pl>
and even Interdata and VAX ports didn't have all that much space early on
<cdegroot>
We had a PDP-11/44 at my university when I was a student. No clue about the specs. Just that it was not very well secured :P
<pl>
meanwhile as early as KA-10, the memory could go to at least 128kW (a little above half megabyte), and that's with only single memory cabinet. the smallest build used single 16kW which was equivalent of more memory than PDP-11 could address
simendsjo has quit [Ping timeout: 264 seconds]
<pl>
PDP-10 *as produced by Digital* ultimately were capped at 4MW, it's possible to modify last KL10-B models to support 6MW afaik and some PDP-10 from other vendors support more memory
yuki74 has quit [Quit: Client closed]
yitzi has quit [Remote host closed the connection]
mgl_ has quit [Ping timeout: 252 seconds]
<cdegroot>
That's.... a lot. I mean, my first *nix box in the '90s had much less.
waleee has joined #commonlisp
<pl>
the canceled Jupiter (later realized in XKL TOAD-1) supported up to 1GW, or about 4.5 GB of RAM, at least theoretically
eddof13 has joined #commonlisp
<pl>
so yeah, UNIX was built on *very* limited hardware in comparison
<pl>
there's a reason why Xenix worked so well on 286 :P
<aeth>
meanwhile, just about anything with a processor is probably more powerful than a PDP-10 these days
<cdegroot>
yeah, have you seen the PiDP-10? ;-)
<pl>
(in practice XKL-1 could physically use 128MW of RAM, SC40 did 64MW but were made I think only for CompuServe internal use)
<aeth>
cdegroot: and emulating is always way more processor intense than the original thing
<pl>
don't recall if Multics used word-addressed CPU or byte addressed
<cdegroot>
(I only used Xenix on 386, and with a whopping 4MB)
<pl>
but I think word-addressed?
<pl>
anyway, the moment 32bit address space opened up, some good tricks came out with lowtags instead of hightags
<cdegroot>
ITS I know, 36 bit words for everything. 6 bit characters, so also a 6-char filename limit (well, filenames had two parts, not really "name" and "extension", and an interesting versioning scheme but the tl&dr was "2 words")
<pl>
cdegroot: ITS file name scheme evolved from CTSS (whose ancestry it sorta shares with Multics, so arguably Unix and ITS are sorta related)
<cdegroot>
(and great for lisp of course, 18bits for CAR and CDR. When I read about that I was like "wow")
<pl>
yeah, but 18bit address space turned out to be also its greatest limiter later on
<cdegroot>
20/20 hindsight, the first thing I should've done when I got an AlphaStation in the '90s was to implement Lisp on it using 2x32bits in a machine word :)
<pl>
32bit machines with lowtag allowed type-agnostic loads in 2 instructions, type-aware loads in 1 instructions, and storing pairs as two words turned out to be pretty great
<pl>
cdegroot: fortunately Alpha allows 4 byte access granularity, otherwise you'd be in a pickle at times ;)
<pl>
(older Alpha CPUs don't support lower granularity than 4 byte)
<pl>
anyway, going back to origin of this discussion, C sucks because it has shit for ABI etc.
<pl>
it's not a good glue
<pl>
on systems with "good" glue approach, C was sometimes the one language that was shittiest to write in
<cdegroot>
true. But as a lingua franca of sorts, often the glue at hand.
<pl>
it's accidental lingua franca of most unices, yes
<pl>
but not, for example, of Windows
<pl>
though Windows did fall into trap of having a lot of crap defined in C includes
<cdegroot>
I wrote C on VMS and it was terrible. Switched to Pascal of all languages, and life became much simpler because VAX Pascal grokked VMS system calls (which were awesome, way ahead of their time, and something C had huge trouble dealing with)
<pl>
recently while walking the dog I was thinking if it wouldn't be possible to get Linux kernel to use some form of IDL for externally-accessible interfaces (one that was C-familiar) that would be unambiguous in parsing
<pl>
i.e. the kernel could then offer something similar to /proc/config.gz with equivalent of all userland kernel includes
<cdegroot>
Wasn't that what the message-passing microkernels of the '90s essentially tried?
<pl>
maybe reuse BTF for that
<pl>
cdegroot: no
<pl>
cdegroot: orthogonal issue
<pl>
you can write message passing microkernel that still has horrible C headers as interface "spec"
<pl>
you can make a monolithic kernel in c that has IDL
<cdegroot>
I remember sniffing at an experimental one (by the free university of Amsterdam, maybe?) that reused a ton of Sun's networking for IPC. Which made total sense to me back then.
<pl>
meanwhile linux kernel does sorta already introduce an "IDL" in the form of BTF data
<pl>
which is used for BPF compilers
<pl>
my issue, to stay on topic, is that I want a good way to access kernel userland ABI without polluting the image with C-isms
<pl>
and if I could, for example, dynamically query all possible address families provided by kernel without parsing C headers, even better
<pl>
(this in turn was spurned by idea of implementing OSI networking in kernel)
<cdegroot>
Yeah, exposing an IDL would make sense.
<cdegroot>
I have been in "parse C kernel headers land" and I do not want to visit that place again :)
<cdegroot>
I'm frankly surprised it's not there. Every syscall is pretty much composed by a macro.
<pl>
cdegroot: Linux kernel is a bit unusual among open source unices because it maintains that the syscall interface is stable ABI
<thuna`>
A dispatch macro's main-char is always terminating, right?
eddof13 has quit [Quit: eddof13]
chomwitt has joined #commonlisp
dino_tutter has quit [Ping timeout: 256 seconds]
eddof13 has joined #commonlisp
NotThatRPG has quit [Ping timeout: 240 seconds]
donleo has quit [Ping timeout: 240 seconds]
amb007 has quit [Ping timeout: 264 seconds]
Alfr has quit [Quit: Leaving]
char has joined #commonlisp
mala has quit [Ping timeout: 264 seconds]
shka has quit [Quit: Konversation terminated!]
istewart has joined #commonlisp
NotThatRPG has joined #commonlisp
eddof13 has quit [Quit: eddof13]
yitzi has joined #commonlisp
pve has quit [Quit: leaving]
eddof13 has joined #commonlisp
Pixel_Outlaw has joined #commonlisp
eddof13 has quit [Client Quit]
Noisytoot has quit [Remote host closed the connection]