freenode/#sicl - IRC Chatlog
Search
7:17:43
heisig
beach: Did I read that correctly, you are looking for volunteers to refine your bootstrapping code?
7:19:59
beach
Yes, I think so. And also the rest of the SICL code based on fewer restrictions of the new bootstrapping procedure. But I haven't committed the new bootstrapping code yet, so it is not ready yet.
7:25:11
beach
It is messy because it is not a modification of the existing code. I didn't want to break the existing code.
7:26:42
heisig
There is no hurry because I am still busy with other projects. But once you upload something, I'd like to contribute.
7:28:15
heisig
The SICL way of bootstrapping is remarkable. It bothers me that we don't yet have the full image.
7:35:31
heisig
I cannot state how eager I am to lay my hands on SICL. Persistent memory is also around the corner, so we will need something to replace Unix :)
7:43:14
ck_
That wasn't meant to be offending, sorry. You said the words "end of the year" once, but as you said new work can't really be put on a schedule. It's better not to rush, and enjoy the finish line when it happens.
8:24:52
no-defun-allowed
I am trying to write a trivial .class file to test my exporting code for HIR-to-JVM. My challenge today is exporting the "constant pool", and whoever wrote the specification agrees with me that it's a bit messy at times: "In retrospect, making 8-byte constants take two constant pool entries was a poor choice."
9:08:41
no-defun-allowed
Oh, and you report the number of elements in the constant pool as one more than there actually is.
9:11:59
no-defun-allowed
It works, and javap and a small Java program I wrote to load a class accept my class file: https://plaster.tymoon.eu/view/2100#2100