Refactor bootstrap w.r.p new Fix design #8
Conversation
| (call $get_value_type (local.get $input)) | ||
| (i32.eq (i32.const 2)) | ||
| (call $is_blob (local.get $input)) | ||
| (i32.eq (i32.const 1)) |
There was a problem hiding this comment.
is this needed? I thought Wasm uses C-style truthiness already
| @@ -156,15 +159,17 @@ | |||
| (table.set $rw_table_2 (i32.const 2) (local.get $curried_cc)) | |||
| (table.set $rw_table_2 (i32.const 3) (local.get $c_files)) | |||
| (call $create_tree_rw_table_2 (i32.const 4)) | |||
| (call $create_thunk) | |||
| (call $create_application_thunk) | |||
| (call $create_strict_encode) | |||
There was a problem hiding this comment.
one of the few times I'm glad that Wasm is a stack machine
| @@ -0,0 +1,3 @@ | |||
| [submodule "fix"] | |||
There was a problem hiding this comment.
is having it as a submodule going to make releases annoying? we could also use a subtree if we don't want to pin a specific version
There was a problem hiding this comment.
But fix has submodules...We could probably make it work by copying fix/.gitmodules to bootstrap to make it work but then bootstrap still has submodules.
| @@ -4,4 +4,9 @@ SRC=`realpath ${SRC_REL}` | |||
| FIX=${SRC}/.fix | |||
|
|
|||
| mkdir -p ${FIX} | |||
| mkdir -p ${FIX}/data | |||
There was a problem hiding this comment.
the Repository class should probably be able to do this itself
There was a problem hiding this comment.
I think the Repository class is not creating {FIX}/data if it does not exist, but maybe Repository should be the one handling creating these directories?
| @@ -199,11 +199,11 @@ class InitComposer | |||
|
|
|||
| void InitComposer::write_attach_tree() | |||
| { | |||
| result_ << "extern void fixpoint_attach_tree(__m256i, wasm_rt_externref_table_t*);" << endl; | |||
| result_ << "extern void fixpoint_attach_tree(u8x32, wasm_rt_externref_table_t*);" << endl; | |||
There was a problem hiding this comment.
Where does this typedef come from?
No description provided.