Switch LLVM reflection utilities to be CodeInstance based
#56876
Conversation
It is in this PR. |
|
Then this PR cannot be merged, since it cannot be optional, as codegen has no other reliable way to get that |
|
It just picks out the |
vtjnash
dismissed
their stale review
Can proceed with other changes here now, if still relevant
|
I rebased this because I thought I needed it, but turns out I don't immediately, but i think it might be useful for @topolarity shortly, so leaving this up anyway. |
|
|
||
| jl_printf(stream, "---- dumping IR for ----\n"); | ||
| jl_static_show(stream, (jl_value_t*)mi); | ||
| jl_printf(stream, "\n----\n"); | ||
|
|
||
| jl_printf(stream, "\n---- unoptimized IR ----\n"); | ||
| jl_get_llvmf_defn(&llvmf_dump, mi, src, 0, false, jl_default_cgparams); | ||
| jl_get_llvmf_defn(&llvmf_dump, ci, NULL, 0, false, jl_default_cgparams); |
There was a problem hiding this comment.
It is not legal now to pass NULL here as the second argument, as codegen (intentionally) lacks the ability now to do inference itself
This is a natural continuation of the stream of work begun in #53219. In particular, with #56555, it is important to be able to pass a CodeInstance, since the ABI may be overwritten and without knowing that, codegen will generated incorrect code for the CodeInstance.