Move rewriting of bs_match from beam_clean to beam_z

The actual bs_match_string instruction has four operands:

   bs_match_string {f,Lbl} Ctxt NumBits {string,ListOfBytes}

However, v3_codegen emits a more compact representation where
the bits to match are packaged in a bitstring:

   bs_match_string {f,Lbl} Ctxt Bitstring

Currently, beam_clean:clean_labels/1 will rewrite the compact
representation to the final representation. That is unfortunate
since clean_labels/1 is called by beam_dead, which means that
the less compact representation will be introduced long before
it is actually needed by beam_asm. It will also complicate any
optimizations that we might want to do.

Move the rewriting of bs_match_string from beam_clean:clean_labels/1
to the beam_z pass, which is the last pass executed before
beam_validator and beam_asm.

4 files changed, 13 insertions, 10 deletions

diff --git a/lib/compiler/src/beam_a.erl b/lib/compiler/src/beam_a.erl
index dd7e03dd28..410f598665 100644
--- a/lib/compiler/src/beam_a.erl
+++ b/lib/compiler/src/beam_a.erl
@@ -91,6 +91,10 @@ rename_instr({bs_private_append=I,F,Sz,U,Src,Flags,Dst}) ->
     {bs_init,F,{I,U,Flags},none,[Sz,Src],Dst};
 rename_instr(bs_init_writable=I) ->
     {bs_init,{f,0},I,1,[{x,0}],{x,0}};
+rename_instr({test,Op,F,[Ctx,Bits,{string,Str}]}) ->
+    %% When compiling from a .S file.
+    <<Bs:Bits/bits,_/bits>> = list_to_binary(Str),
+    {test,Op,F,[Ctx,Bs]};
 rename_instr({put_map_assoc,Fail,S,D,R,L}) ->
     {put_map,Fail,assoc,S,D,R,L};
 rename_instr({put_map_exact,Fail,S,D,R,L}) ->
diff --git a/lib/compiler/src/beam_clean.erl b/lib/compiler/src/beam_clean.erl
index b68b8702e0..1d26993103 100644
--- a/lib/compiler/src/beam_clean.erl
+++ b/lib/compiler/src/beam_clean.erl
@@ -184,14 +184,6 @@ function_replace([{function,Name,Arity,Entry,Asm0}|Fs], Dict, Acc) ->
     function_replace(Fs, Dict, [{function,Name,Arity,Entry,Asm}|Acc]);
 function_replace([], _, Acc) -> Acc.
 
-replace([{test,bs_match_string=Op,{f,Lbl},[Ctx,Bin0]}|Is], Acc, D) ->
-    Bits = bit_size(Bin0),
-    Bin = case Bits rem 8 of
-	      0 -> Bin0;
-	      Rem -> <<Bin0/bitstring,0:(8-Rem)>>
-	  end,
-    I = {test,Op,{f,label(Lbl, D)},[Ctx,Bits,{string,binary_to_list(Bin)}]},
-    replace(Is, [I|Acc], D);
 replace([{test,Test,{f,Lbl},Ops}|Is], Acc, D) ->
     replace(Is, [{test,Test,{f,label(Lbl, D)},Ops}|Acc], D);
 replace([{test,Test,{f,Lbl},Live,Ops,Dst}|Is], Acc, D) ->
diff --git a/lib/compiler/src/beam_dead.erl b/lib/compiler/src/beam_dead.erl
index f4515ba2a7..adc3cebc62 100644
--- a/lib/compiler/src/beam_dead.erl
+++ b/lib/compiler/src/beam_dead.erl
@@ -459,8 +459,8 @@ count_bits_matched([{test,_,_,_,[_,Sz,U,{field_flags,_}],_}|Is], SavePoint, Bits
 	{integer,N} -> count_bits_matched(Is, SavePoint, Bits+N*U);
 	_ -> count_bits_matched(Is, SavePoint, Bits)
     end;
-count_bits_matched([{test,bs_match_string,_,[_,Bits,_]}|Is], SavePoint, Bits0) ->
-    count_bits_matched(Is, SavePoint, Bits0+Bits);
+count_bits_matched([{test,bs_match_string,_,[_,Bs]}|Is], SavePoint, Bits) ->
+    count_bits_matched(Is, SavePoint, Bits+bit_size(Bs));
 count_bits_matched([{test,_,_,_}|Is], SavePoint, Bits) ->
     count_bits_matched(Is, SavePoint, Bits);
 count_bits_matched([{bs_save2,Reg,SavePoint}|_], {Reg,SavePoint}, Bits) ->
diff --git a/lib/compiler/src/beam_z.erl b/lib/compiler/src/beam_z.erl
index 0c7bef9183..47e786034d 100644
--- a/lib/compiler/src/beam_z.erl
+++ b/lib/compiler/src/beam_z.erl
@@ -74,6 +74,13 @@ undo_rename({bs_init,F,{I,Extra,U,Flags},Live,[Sz,Src],Dst}) ->
     {I,F,Sz,Extra,Live,U,Src,Flags,Dst};
 undo_rename({bs_init,_,bs_init_writable=I,_,_,_}) ->
     I;
+undo_rename({test,bs_match_string=Op,F,[Ctx,Bin0]}) ->
+    Bits = bit_size(Bin0),
+    Bin = case Bits rem 8 of
+	      0 -> Bin0;
+	      Rem -> <<Bin0/bitstring,0:(8-Rem)>>
+	  end,
+    {test,Op,F,[Ctx,Bits,{string,binary_to_list(Bin)}]};
 undo_rename({put_map,Fail,assoc,S,D,R,L}) ->
     {put_map_assoc,Fail,S,D,R,L};
 undo_rename({put_map,Fail,exact,S,D,R,L}) ->