easye has quit [Read error: Connection reset by peer]
Farooq has quit [Remote host closed the connection]
Farooq has joined #commonlisp
micro has quit [Ping timeout: 264 seconds]
micro has joined #commonlisp
msv has quit [Ping timeout: 268 seconds]
<nij->
Is there an implementation that supports: redefine a function in the ldb (without using slime/sly) on the fly, in the current frame, and let it the program run from then on, see what happens, and jump back to the beginning to the time where redefinition and funcall didn't happen?
<nij->
>
slyrus has joined #commonlisp
varjag has quit [Ping timeout: 268 seconds]
didi has left #commonlisp [O bella ciao bella ciao bella ciao, ciao, ciao.]
<bike>
nij-: ldb? like, the sbcl low level debugger?
<jmercouris>
you could make your own macro to allow this
<jmercouris>
that's what i would do
mgl has quit [Ping timeout: 256 seconds]
msv has joined #commonlisp
edgar-rft has quit [Ping timeout: 264 seconds]
kevingal has quit [Ping timeout: 256 seconds]
donleo has quit [Ping timeout: 268 seconds]
slyrus has quit [Ping timeout: 255 seconds]
edgar-rft has joined #commonlisp
nij- has quit [Ping timeout: 268 seconds]
slyrus has joined #commonlisp
waleee has quit [Ping timeout: 272 seconds]
slyrus has quit [Read error: Connection reset by peer]
slyrus_ has joined #commonlisp
msv has quit [Ping timeout: 268 seconds]
ronald has quit [Ping timeout: 252 seconds]
slyrus_ has quit [Ping timeout: 264 seconds]
ronald has joined #commonlisp
<Bubblegumdrop>
Kingsy I use drakma for http reqs yes
<Bubblegumdrop>
chsasank Computing with Register Machines iirc
<aeth>
you want something that more or less looks like the source, and then you want something that just looks like a list of in-order execution (though the presence of IF's branching somewhat complicates this)
<aeth>
that sort of fantasy IR implies that the former is held quite late, basically getting you to asm before doing the latter
<aeth>
but you really do, imo, want to turn it all into effectively one big LET* long before it gets to something approaching the asm
st_aldini1 has joined #commonlisp
slyrus has joined #commonlisp
st_aldini has quit [Ping timeout: 268 seconds]
st_aldini1 is now known as st_aldini
slyrus has quit [Ping timeout: 268 seconds]
<chsasank>
One big let* is certainly an option but it has implicit naming. I'm looking for options without names.
piglet has joined #commonlisp
<chsasank>
Kinda like what I posted above for x86
<Bubblegumdrop>
gensym?
<Bubblegumdrop>
iirc there is a "generate gensym for lists" function, i don't remember the name
vyrsh has quit [Ping timeout: 268 seconds]
<Bubblegumdrop>
I guess that's still naming things
<aeth>
chsasank: the names are gensym
<aeth>
it helps you see the control flow because you can see e.g. (#:g527 (+ #:g525 #:g526))
<aeth>
There does not need to be anything other than functions (expand the macros before you get to this point!) and constants, where the constants are assigned a gensym name. If there are named variables, you can unname them (i.e. look them up in their proper scope to what they actually are) by this point. Maybe you can have global variables still.
<aeth>
So even though you don't have any names, once you get to the IR, you have "names". Sort of like how almost all functional programming eventually has mutation somewhere, but not exposing the mutation to the level that the programmer can use lets you make optimizations and stuff.
akoana has quit [Quit: leaving]
<aeth>
but you can just have e.g. (f '(+ 1 2)) => ((#:g525 1) (#:g526 2) (#:g527 (+ #:g525 #:g526)))
<aeth>
Conceptually kind of halfway between SSA and CPS
<Bubblegumdrop>
Somehow I think "With great power comes great responsibility" also applies to (ab)use of overly powerful programming language features.
<Bubblegumdrop>
accurate
<aeth>
chsasank: I think what's confusing you is your commitment to tacit/point-free programming in your design. You don't keep a paradigm the whole way through. You keep it at the interface (the language) and implement it however seems easiest or best (which is strongly suggesting something like A-normal).
<aeth>
(mapcar #'+ (list 1 2 3) (list 4 5 6)) fits your restrictions, but it may just use a LOOP with named variables inside of the actual implementation of MAPCAR
<beach>
chsasank: It would be very strange to represent an SSA program as a list.
skeemer has quit [Ping timeout: 260 seconds]
nij- has quit [Ping timeout: 255 seconds]
pve has joined #commonlisp
<hayley>
"we add lambdas, classes and what not and things get unclear." Why?
<hayley>
A standard instance indicates a lot more structure than a list.
<beach>
The only reason I can see is that the person is unfamiliar with more complex data structures.
<hayley>
And that x86 example has lots of names, I see EAX, ESP and EBP at the least.
<aeth>
You can represent SSA as a list (in particular, the bindings part of a LET*, blurring the line with A-normal). That particular IR probably won't survive the final form of the program, except perhaps as a printable simplification of what's going on.
<hayley>
aeth: Now extend that to multiple basic blocks. I'll wait.
<aeth>
"now extend that" right, it probably won't stay around in your final compiler because it has many limitations
<aeth>
but starting with something is better than endlessly debating what to do
<aeth>
kinda.
<aeth>
sometimes it's messy to try to do things the proper way after a list prototype
<hayley>
One can have no names in the IR, and SSA is pretty good for name-less IR, but lists aren't well-suited unless you want to rely on EQ in somewhat non-obvious ways.
<hayley>
"you could make your own macro to allow this [redefining functions in ldb in SBCL]" No you couldn't, ldb doesn't evaluate arbitrary Lisp code (there is a "call" command) and you're not expected to be able to return to Lisp from ldb.
<hayley>
Unless ldb doesn't mean the low-level debugger in SBCL, which wouldn't shock me to see the name be misused like that. (Now I sound like beach.)
<beach>
Heh! Thanks, I guess. :)
<Mondenkind>
I'm not sure what 'ssa' or 'using lists' rightfully means. If the latter refers to a nested structure of proper lists, then that is a problem because you cannot represent cycles directly, so must employ weak pointers (like names). That is not so nice. If lists means cons cells, then you can use them as ad-hoc data structures, and it's not particularly different from non-ad-hoc data
<Mondenkind>
structures
<aeth>
it's messier to use lists because you're going to eventually rely on destructuring-bind (ideally) or cadadaddr (not ideal)
<aeth>
but, hey, that's fine because the destructuring-bind tells you what you should've started with in the first place if you had structured your data
<aeth>
so it seems fine to me
<aeth>
though in the very, very particular case of SSA you can pretend that it's the bindings part of one giant LET* if you insist on using lists, at least to start with
istewart has quit [Quit: Konversation terminated!]
decweb has quit [Ping timeout: 256 seconds]
shka has joined #commonlisp
wacki has joined #commonlisp
vyrsh has joined #commonlisp
piglet has quit [Ping timeout: 256 seconds]
son0p has quit [Ping timeout: 264 seconds]
vyrsh has quit [Ping timeout: 268 seconds]
chomwitt has joined #commonlisp
Pixel_Outlaw has quit [Remote host closed the connection]
mgl has joined #commonlisp
rgherdt has joined #commonlisp
iNomad has joined #commonlisp
amb007 has quit [Ping timeout: 255 seconds]
ryanbw has joined #commonlisp
ryanbw has quit [Client Quit]
ryanbw has joined #commonlisp
ryanbw has quit [Client Quit]
ryanbw has joined #commonlisp
semarie has quit [Quit: WeeChat 4.2.1]
semarie has joined #commonlisp
ryanbw has quit [Quit: I'll be back.]
ryanbw has joined #commonlisp
amb007 has joined #commonlisp
chomwitt has quit [Ping timeout: 272 seconds]
awlygj has joined #commonlisp
bendersteed has joined #commonlisp
danza has joined #commonlisp
beach` has joined #commonlisp
alcor has joined #commonlisp
bendersteed has quit [Remote host closed the connection]
beach has quit [Killed (NickServ (GHOST command used by beach`!~user@2a01:cb19:682:4600:82e6:313a:3bf9:e547))]
beach` is now known as beach
bendersteed has joined #commonlisp
mgl has quit [Ping timeout: 268 seconds]
occ has quit [Ping timeout: 272 seconds]
occ has joined #commonlisp
occ has quit [Ping timeout: 268 seconds]
zetef has joined #commonlisp
zetef has quit [Remote host closed the connection]
chomwitt has joined #commonlisp
occ has joined #commonlisp
zetef has joined #commonlisp
wacki has quit [Quit: My iMac has gone to sleep. ZZZzzz…]
dino_tutter has joined #commonlisp
bendersteed has quit [Quit: bendersteed]
mgl has joined #commonlisp
son0p has joined #commonlisp
vyrsh has joined #commonlisp
igemnace has quit [Read error: Connection reset by peer]
zetef has quit [Remote host closed the connection]
donleo has joined #commonlisp
danza has quit [Ping timeout: 260 seconds]
igemnace has joined #commonlisp
tibfulv_ is now known as tibfulv
Guest52 has quit [Quit: Client closed]
chomwitt has quit [Ping timeout: 272 seconds]
dra has joined #commonlisp
piglet has joined #commonlisp
danse-nr3 has joined #commonlisp
vyrsh has quit [Ping timeout: 264 seconds]
danse-nr3 has quit [Remote host closed the connection]
attila_lendvai has joined #commonlisp
danse-nr3 has joined #commonlisp
mariari has quit [Quit: WeeChat 4.2.1]
mariari has joined #commonlisp
reb has quit [Ping timeout: 256 seconds]
Lycurgus has quit [Quit: leaving]
King_julian has joined #commonlisp
Lord_of_Life has quit [Ping timeout: 256 seconds]
Lord_of_Life_ has joined #commonlisp
Lord_of_Life_ is now known as Lord_of_Life
King_julian has quit [Ping timeout: 260 seconds]
waleee has joined #commonlisp
zetef has joined #commonlisp
waleee has quit [Ping timeout: 268 seconds]
dra has quit [Ping timeout: 256 seconds]
decweb has joined #commonlisp
piglet has quit [Quit: Konversation terminated!]
waleee has joined #commonlisp
Lord_Nightmare has quit [Ping timeout: 255 seconds]
King_julian has joined #commonlisp
yitzi has joined #commonlisp
zephyr has quit [Quit: Ping timeout (120 seconds)]
cayley5 has quit [Quit: Ping timeout (120 seconds)]
danse-nr3 has quit [Read error: Connection reset by peer]
danse-nr3 has joined #commonlisp
random-nick has joined #commonlisp
tok has joined #commonlisp
King_julian has quit [Ping timeout: 268 seconds]
danse-nr3 has quit [Ping timeout: 246 seconds]
dra has joined #commonlisp
tyson2 has joined #commonlisp
ebrasca has joined #commonlisp
danse-nr3 has joined #commonlisp
josrr has joined #commonlisp
attila_lendvai has quit [Ping timeout: 272 seconds]
attila_lendvai has joined #commonlisp
nij- has joined #commonlisp
<nij->
Is there anyway to debug with watchpoints and catchpoints (like those in gdb)?
alcor has quit [Ping timeout: 268 seconds]
attila_lendvai has quit [Ping timeout: 255 seconds]
<scymtym>
*BREAK-ON-SIGNALS* seems similar to gdb catchpoints. watchpoints may be supported by some implementations but i don't think there is anything in the standard
<beach>
nij-: That would depend on the Common Lisp implementation you are using. But, as my paper describes, the free Common Lisp implementations we use are not very good with supporting debugging features.
<nij->
I see. Thank you both!
piglet has joined #commonlisp
<yitzi>
SBCL has breakpoints, but they seem to sort of unreliable in my experience.
alcor has joined #commonlisp
<yitzi>
I don't recall if they are actually documented anywhere.
<yitzi>
Its sb-di:activate-breakpoint .... if you care.
<beach>
yitzi: It does have breakpoints, but they are not exposed in any higher-level protocol.
<yitzi>
Yeah, I've used them and they don't always work.
<beach>
They are used in one built-in feature, maybe tracing?
<yitzi>
The single stepping in SBCL seems to be better.
<beach>
I forget. I documented it in my paper.
King_julian has joined #commonlisp
<beach>
Well, to me stepping is basically useless unless you can do it from a breakpoint.
fitzsim has quit [Read error: Connection reset by peer]
fitzsim has joined #commonlisp
<nij->
yitzi breakpoints? Are sbcl breakpoints similar to watchpoints (meaning, it breaks whenever it's assigning a value to a variable.)
<beach>
nij-: No, they are control breakpoints.
<beach>
... i.e., they break at a certain point in the control flow.
<beach>
I am not even sure how I would implement watchpoints for lexical variables, at least not in a way that would sort of preserve performance.
<scymtym>
gdb watchpoints seem to monitor memory locations for read and/or write access by the program being debugged. this is not a good fit for CL semantics
<beach>
scymtym: So how does it do that for automatic variables?
<beach>
I can see how it could do it for global variables.
<beach>
For Common Lisp, one might have to make the compiler generate code for each assignment so that the code checks whether the variable has a watchpoint associated with it.
<beach>
I guess that would be possible with a high DEBUG quality.
<scymtym>
gdb can use hardware watchpoints for memory locations and it can switch to single stepping and checking the watched expression in software. the latte mode may support stack location and more complex expressions, but i haven't tried it
<scymtym>
*latter
<beach>
Yes, maybe so.
<beach>
As I recall from my paper, for breakpoints, ACL modifies the code just like GDB does, except it does it in the same process.
<scymtym>
gdb hardware watchpoints work for stack locations. i guess they activate/deactivate the hardware watchpoints as threads enter/leave the surrounding function?
zetef has quit [Remote host closed the connection]
<rgherdt>
thanks!
<beach>
Sure. We might not use the exact technique in that paper, but the "previous work" section is still a good overview of what exists.
Lycurgus has joined #commonlisp
<nij->
Is there anyway to define primitive operators in CL?
<bike>
I don't know what exactly you mean by "primitive", but the answer to your question is probably "no" regardless.
<beach>
No. How would you do that? I mean what would the "body" of the definition contain?
<nij->
s/primitive/special/
<nij->
Hmm.. maybe a bit off-topic, but is there any lisp that allows this?
<beach>
nij-: Macros are essentially special operators that you can define yourself. In fact, the standard allows for special operators to be defined as macros and vice versa.
<beach>
nij-: Please answer my question about what the "body" of such a definition would look like.
<bike>
nij-: we still don't know what "this" is.
<bike>
nij-: you presumably have some idea of what "primitive" means, but it is not obvious to others.
<nij->
I understand the compiler as a function that takes an input AST and generates the compiled code.
<nij->
In many lisp, this compiler first looks at the 0th element in the current AST, and see if it's a special op, a macro, or neither.
<nij->
If it's neither, then it generates code that represent a funcall. If it's a macro, it calls the corresponding macro function that transforms the AST, and feeds the resulting AST back.
<nij->
Finally if it's a special op, then it can do whatever it wants.
<nij->
It depends on the host language, but it may not be easy to define a special op in s-exprs, unless the host language allows it.
<bike>
sure, and "primitive"?
<nij->
Sorry, please remove "primitive" and think of "special".
<beach>
But the compiler generates code for special operators according to the semantics of each one, so if you were allowed to define your own, then you would need a way to inform the compiler about the semantics. How would you do that?
<nij->
s/primitive/special/
<nij->
I used the wrong word.
<bike>
oky, special operators. sure. back in the day some lisps allowed you to define fexprs, which are more or less the same. there is a more recent scheme-like language called kernel that elaborates on them.
<beach>
nij-: Common Lisp allows you to define news such operators by explaining the semantics in terms of existing operators. It is called "macro".
<bike>
but they make it very hard to do any compilation.
<nij->
beach It depends on the host lang. Probably the user just have to write in the host lang, or the host lang should support a way that transforms s-expr to it.
<bike>
i think you should stop thinking in terms of hosts. fexprs are almost entirely unrelated to any notion of hosts.
<nij->
Oh yes, first-class special forms! That's the term I should have said.
<beach>
I don't see how FEXPRs eliminates the necessity of explaining the semantics of the new operator.
<beach>
*eliminate
<bike>
it does not.
<bike>
er, "they don't" rather.
<beach>
So my question still stands. What would the "body" of such a definition look like?
<bike>
for my compiler you have to teach the compiler about every fexpr you want to compile individually.
<beach>
Exactly.
<bike>
and if there isn't any compiler method, it gives up and compiles a call to eval. this is part of why i said it's really hard to compile anything.
<bike>
so you basically need two definitions- an interpreter definition, which is relatively straightforward, and a compiler operation which is less so
<beach>
Yes, I see.
eddof13 has joined #commonlisp
bendersteed has joined #commonlisp
<masinter>
Medley Common Lisp and Interlisp NLAMBDAs: Th C
eddof13 has quit [Client Quit]
piglet has quit [Ping timeout: 264 seconds]
King_julian has quit [Ping timeout: 260 seconds]
Lycurgus has quit [Quit: leaving]
yitzi has quit [Remote host closed the connection]
dra has quit [Ping timeout: 268 seconds]
reb has joined #commonlisp
piglet has joined #commonlisp
Lord_Nightmare has joined #commonlisp
X-Scale has joined #commonlisp
nij- has quit [Ping timeout: 268 seconds]
nij- has joined #commonlisp
danse-nr3 has quit [Ping timeout: 268 seconds]
dra has joined #commonlisp
danse-nr3 has joined #commonlisp
nij- has quit [Ping timeout: 268 seconds]
occ has quit [Read error: Connection reset by peer]
chomwitt has joined #commonlisp
dra has quit [Ping timeout: 268 seconds]
danse-nr3 has quit [Ping timeout: 268 seconds]
danse-nr3 has joined #commonlisp
bendersteed has quit [Quit: bendersteed]
occ has joined #commonlisp
alcor has quit [Read error: Connection reset by peer]
alcor has joined #commonlisp
waleee has quit [Quit: WeeChat 4.1.2]
danse-nr3 has quit [Quit: Leaving]
danse-nr3 has joined #commonlisp
nij- has joined #commonlisp
wacki has joined #commonlisp
igemnace has quit [Read error: Connection reset by peer]
occ has quit [Ping timeout: 268 seconds]
chiselfuse has quit [Ping timeout: 260 seconds]
reb has quit [Remote host closed the connection]
chiselfuse has joined #commonlisp
occ has joined #commonlisp
igemnace has joined #commonlisp
a51 has joined #commonlisp
mm007emko has quit [Ping timeout: 268 seconds]
mm007emko has joined #commonlisp
danse-nr3 has quit [Ping timeout: 256 seconds]
igemnace has quit [Quit: WeeChat 4.2.1]
OlCe has joined #commonlisp
piglet has quit [Quit: Konversation terminated!]
tyson2 has quit [Remote host closed the connection]
mgl has quit [Ping timeout: 255 seconds]
ebrasca has quit [Remote host closed the connection]
ebrasca has joined #commonlisp
danse-nr3 has joined #commonlisp
chomwitt has quit [Ping timeout: 268 seconds]
awlygj has quit [Quit: leaving]
dra has joined #commonlisp
dra has quit [Changing host]
dra has joined #commonlisp
mm007emko has quit [Read error: Connection reset by peer]
mm007emko has joined #commonlisp
eddof13 has joined #commonlisp
thuna` has joined #commonlisp
eddof13 has quit [Quit: eddof13]
ym has joined #commonlisp
Pixel_Outlaw has joined #commonlisp
eddof13 has joined #commonlisp
danse-nr3 has quit [Ping timeout: 255 seconds]
pfdietz has quit [Quit: Client closed]
attila_lendvai has joined #commonlisp
varjag has joined #commonlisp
amb007 has joined #commonlisp
King_julian has joined #commonlisp
a51 has quit [Ping timeout: 260 seconds]
a51 has joined #commonlisp
ec has quit [Remote host closed the connection]
ec has joined #commonlisp
a51 has quit [Quit: WeeChat 4.2.1]
eddof13_ has joined #commonlisp
ec has quit [Remote host closed the connection]
ec has joined #commonlisp
eddof13_ has quit [Quit: My MacBook Air has gone to sleep. ZZZzzz…]
skeemer has joined #commonlisp
cage has joined #commonlisp
attila_lendvai has quit [Remote host closed the connection]
attila_lendvai has joined #commonlisp
King_julian has quit [Ping timeout: 272 seconds]
jonatack has joined #commonlisp
jon_atack has quit [Ping timeout: 268 seconds]
jon_atack has joined #commonlisp
nij- has quit [Ping timeout: 256 seconds]
jonatack has quit [Ping timeout: 256 seconds]
eddof13_ has joined #commonlisp
tedwing has joined #commonlisp
rgherdt has quit [Quit: Leaving]
rgherdt has joined #commonlisp
pfdietz has joined #commonlisp
mgl has joined #commonlisp
varjag has quit [Remote host closed the connection]
pve has quit [Quit: leaving]
waleee has joined #commonlisp
eddof13_ has quit [Ping timeout: 268 seconds]
eddof13 has quit [Quit: eddof13]
eddof13 has joined #commonlisp
eddof13 has quit [Client Quit]
varjag has joined #commonlisp
kevingal has joined #commonlisp
cage has quit [Quit: rcirc on GNU Emacs 29.2]
robin has quit [Remote host closed the connection]
robin has joined #commonlisp
eddof13 has joined #commonlisp
eddof13 has quit [Ping timeout: 255 seconds]
X-Scale has quit [Quit: Client closed]
ebrasca has quit [Remote host closed the connection]
wacki has quit [Quit: My iMac has gone to sleep. ZZZzzz…]
tedwing has quit [Ping timeout: 260 seconds]
josrr has quit [Remote host closed the connection]
alcor has quit [Ping timeout: 268 seconds]
varjag has quit [Ping timeout: 260 seconds]
nij- has joined #commonlisp
amb007 has quit [Ping timeout: 260 seconds]
nij- has quit [Ping timeout: 260 seconds]
reb has joined #commonlisp
Inline has quit [Quit: Leaving]
Inline has joined #commonlisp
tyson2 has joined #commonlisp
amb007 has joined #commonlisp
tedwing has joined #commonlisp
amb007 has quit [Ping timeout: 255 seconds]
ronald has quit [Ping timeout: 256 seconds]
attila_lendvai has quit [Ping timeout: 260 seconds]
ronald has joined #commonlisp
nij- has joined #commonlisp
amb007 has joined #commonlisp
dino_tutter has quit [Ping timeout: 255 seconds]
amb007 has quit [Read error: Connection reset by peer]
amb007 has joined #commonlisp
donleo has quit [Ping timeout: 268 seconds]
rgherdt has quit [Quit: Leaving]
dra has quit [Ping timeout: 268 seconds]
ronald has quit [Read error: Connection reset by peer]