aboutsummaryrefslogtreecommitdiff
path: root/README.md
diff options
context:
space:
mode:
Diffstat (limited to 'README.md')
-rw-r--r--README.md3
1 files changed, 1 insertions, 2 deletions
diff --git a/README.md b/README.md
index 0bdef99..00eafdd 100644
--- a/README.md
+++ b/README.md
@@ -5,8 +5,7 @@ It does almost nothing, and what it does, it does badly and is not useful.
- There is no parser, you have to write your program in the internal Clean
representation (see `Start` in [`sjit.icl`](/sjit.icl)).
-- There is no type checker, you have to guess the implicit rules (such as that
- a `fun_expr` must always be an `Abstr`).
+- There is no type checker, you have to guess the implicit rules.
- There is no register allocation, everything is done on the stack.
- There is no code optimisation, not even to eliminate `push rbx` followed by
`pop rbx`.