freenode/#sicl - IRC Chatlog
Search
3:50:33
beach
Contrary to what I said yesterday, I won't "tie the knot" today. Things are slightly more complicated than I thought yesterday. Not a lot, but a little. And I want to explore what I have now and clean up a few rough edges.
3:50:36
beach
I might remove the (old) Boot directory and put the one I have been working in (New-boot) in its place. I fear that the old boot is broken anyway. I just want to make sure that the "quick start" instructions in the README are still valid (or update them).
3:54:35
beach
As far as the host is concerned, everything is now either a HEADER object, or a simple vector. Not very user friendly to look at.
6:00:37
beach
I have a favor to ask (no rush): Could someone verify that the following works: (asdf:load-system #:sicl-new-boot), then (in-package #:sicl-new-boot), then (defparameter *b* (boot)), then [after around 15 minutes] (repl ee4) [you should see a SICL: prompt], then a few interactions like (defgeneric foo (x)) (defmethod foo (x) (list 234 x)) (defmethod foo ((x cons)) (list 345 (call-next-method))) (foo 234) (foo '(a))
6:01:35
beach
I don't know what size heap you need. In the end, (room) says it uses around 1GB, but I am sure it requires more to do the work. I use 10GB myself.
6:03:29
beach
And if you are interested, you can use the utility functions (at the host prompt) FC (for find-class) FBP (for fboundp) FD (for fdefinition). They all take a name and an environment. Environments can be accessed using EE1 to EE5.
6:04:18
beach
I was trying to come up with cute names like Anemic, Basic, Comfortable, Durable, Essential, Fake, etc.
6:06:34
beach
Anyway, if this works for everyone, I will remove the existing Boot directory and all its code, and rename the New-boot to Boot. Also update the README.
6:16:52
no-defun-allowed
the heck, asdf says "Component #:SICL-PACKAGE-SUPPORT not found, required by sicl-new-boot-base" and ql says "system sicl-package-support not found"
6:19:02
no-defun-allowed
well, i'm not sure what i told git to do but github says i'm even with robert-strandh/SICL now
6:20:59
no-defun-allowed
well, i'm very sorry to have warned you, i forgot to do QL:REGISTER-LOCAL-PROJECTS
6:26:13
no-defun-allowed
that or i'm procrastinating too hard while it compiles and i have lost track of time.
6:43:13
beach
That seems to be the difference. It takes a bit longer here (I don't know why) but it is comparable to what you get.
7:00:48
no-defun-allowed
with (declaim (optimize (speed 3))) it's fumbling about doing the usual SBCL note stuff now, and is still in stage 2 afaict
7:01:21
no-defun-allowed
it's probably that, don't stress yourself about it being slow, sbcl's just trying to help
7:03:26
beach
Thanks to both of you. I will eliminate the (old) Boot and turn the New boot into just Boot.
7:05:30
fiddlerwoaroof
Incidentally, doing (sb-ext:save-lisp-and-die "/Users/elangley/sicl" :executable t :toplevel (lambda () (repl ee4))) worked and gave me an executable that starts sicl directly
9:40:53
beach
I removed all occurrences of "new-", thinking I had only used that string with new-boot, but I had used it with NEW-DIRECT-SLOTS, resulting in the wrong value of a lexical variable. *blush*.
9:41:20
beach
Anyway, if someone wants to check the new instructions in the README, I would appreciate that.
9:42:08
beach
That mysterious bug made me quite agitated this morning. I think I should do some simpler stuff the rest of the day.
9:43:48
beach
Now that I discovered the factor 10 difference in boot times depending on SBCL OPTIMIZE quality values, I should be more productive.
9:57:32
beach
Perhaps it is not optimizing generic dispatch as much with a low SPEED and high DEBUG setting. And I have a lot of that.