unwind the latest continuation-space fix, because it breaks something else

svn: r6213
This commit is contained in:
Matthew Flatt 2007-05-12 22:23:27 +00:00
parent ef645e466c
commit 0420f9c814

View File

@ -509,6 +509,8 @@ scheme_init_fun (Scheme_Env *env)
original_default_prompt = MALLOC_ONE_TAGGED(Scheme_Prompt);
original_default_prompt->so.type = scheme_prompt_type;
original_default_prompt->tag = scheme_default_prompt_tag;
REGISTER_SO(scheme_overflow_reply);
}
Scheme_Object *
@ -1653,9 +1655,9 @@ void scheme_really_create_overflow(void *stack_base)
scheme_init_jmpup_buf(&jmp->cont);
if (scheme_setjmpup(&jmp->cont, jmp, stack_base)) {
/* A jump into here is a rquest to handle overflow.
/* A jump into here is a request to handle overflow.
The way to continue is in scheme_overflow_k.
Whe we get back, put the result into
When we get back, put the result into
scheme_overflow_reply. The route to return is
in the thread's `overflow' field. */
Scheme_Thread * volatile p;
@ -3621,7 +3623,7 @@ static Scheme_Saved_Stack *copy_out_runstack(Scheme_Thread *p,
}
isaved->runstack_size = size;
start = MALLOC_N(Scheme_Object*, size);
isaved->runstack_start = start;
memcpy(isaved->runstack_start,
@ -3897,11 +3899,11 @@ static Scheme_Meta_Continuation *clone_meta_cont(Scheme_Meta_Continuation *mc,
break;
if (!mc->pseudo && SAME_OBJ(mc->prompt_tag, limit_tag))
break;
if (for_composable && mc->pseudo && mc->empty_to_next && mc->next
if (0 /* FIXME! */ && for_composable && mc->pseudo && mc->empty_to_next && mc->next
&& SAME_OBJ(mc->next->prompt_tag, limit_tag)) {
/* We don't need to keep the compose-introduced meta-continuation,
because it represents an empty continuation relative to the
prompt. */
/* We don't need to keep the compose-introduced
meta-continuation, because it represents an empty
continuation relative to the prompt. */
break;
}
naya = MALLOC_ONE_RT(Scheme_Meta_Continuation);