Resolve synthetic compilation dir /emsdk/emscripten to obtain actual path for better source map

test/other/wasm_sourcemap_resolve_dummy_comp_dir/foo.wasm.dump

@@ -0,0 +1,113 @@
+foo.wasm:	file format WASM
+
+.debug_info contents:
+0x00000000: Compile Unit: length = 0x00000045, format = DWARF32, version = 0x0004, abbr_offset = 0x0000, addr_size = 0x04 (next unit at 0x00000049)
+
+0x0000000b: DW_TAG_compile_unit
+              DW_AT_producer	("clang version 18.0.0 (https://github.com/llvm/llvm-project 269685545e439ad050b67740533c59f965cae955)")
+              DW_AT_language	(DW_LANG_C11)
+              DW_AT_name	("no_main.c")
+              DW_AT_stmt_list	(0x00000000)
+              DW_AT_comp_dir	("/emscripten/test/other/wasm_sourcemap_resolve_dummy_comp_dir")
+              DW_AT_low_pc	(0x00000002)
+              DW_AT_high_pc	(0x0000000d)
+0x00000049: Compile Unit: length = 0x00000120, format = DWARF32, version = 0x0004, abbr_offset = 0x0034, addr_size = 0x04 (next unit at 0x0000016d)
+
+0x00000054: DW_TAG_compile_unit
+              DW_AT_stmt_list	(0x00000042)
+              DW_AT_ranges	(0x00000000
+                 [0x0000000e, 0x00000016)
+                 [0x00000017, 0x00000021)
+                 [0x00000022, 0x0000003c))
+              DW_AT_name	("system/lib/compiler-rt/stack_ops.S")
+              DW_AT_comp_dir	("/emsdk/emscripten")
+              DW_AT_producer	("clang version 18.0.0 (https://github.com/llvm/llvm-project 269685545e439ad050b67740533c59f965cae955)")
+              DW_AT_language	(DW_LANG_Mips_Assembler)
+
+.debug_line contents:
+debug_line[0x00000000]
+Line table prologue:
+    total_length: 0x0000003e
+          format: DWARF32
+         version: 4
+ prologue_length: 0x00000021
+ min_inst_length: 1
+max_ops_per_inst: 1
+ default_is_stmt: 1
+       line_base: -5
+      line_range: 14
+     opcode_base: 13
+standard_opcode_lengths[DW_LNS_copy] = 0
+standard_opcode_lengths[DW_LNS_advance_pc] = 1
+standard_opcode_lengths[DW_LNS_advance_line] = 1
+standard_opcode_lengths[DW_LNS_set_file] = 1
+standard_opcode_lengths[DW_LNS_set_column] = 1
+standard_opcode_lengths[DW_LNS_negate_stmt] = 0
+standard_opcode_lengths[DW_LNS_set_basic_block] = 0
+standard_opcode_lengths[DW_LNS_const_add_pc] = 0
+standard_opcode_lengths[DW_LNS_fixed_advance_pc] = 1
+standard_opcode_lengths[DW_LNS_set_prologue_end] = 0
+standard_opcode_lengths[DW_LNS_set_epilogue_begin] = 0
+standard_opcode_lengths[DW_LNS_set_isa] = 1
+file_names[  1]:
+           name: "no_main.c"
+      dir_index: 0
+       mod_time: 0x00000000
+         length: 0x00000000
+
+Address            Line   Column File   ISA Discriminator OpIndex Flags
+------------------ ------ ------ ------ --- ------------- ------- -------------
+0x0000000000000005     12      3      1   0             0       0  is_stmt prologue_end
+0x000000000000000d     12      3      1   0             0       0  is_stmt end_sequence
+
+debug_line[0x00000042]
+Line table prologue:
+    total_length: 0x000000ea
+          format: DWARF32
+         version: 4
+ prologue_length: 0x0000003a
+ min_inst_length: 1
+max_ops_per_inst: 1
+ default_is_stmt: 1
+       line_base: -5
+      line_range: 14
+     opcode_base: 13
+standard_opcode_lengths[DW_LNS_copy] = 0
+standard_opcode_lengths[DW_LNS_advance_pc] = 1
+standard_opcode_lengths[DW_LNS_advance_line] = 1
+standard_opcode_lengths[DW_LNS_set_file] = 1
+standard_opcode_lengths[DW_LNS_set_column] = 1
+standard_opcode_lengths[DW_LNS_negate_stmt] = 0
+standard_opcode_lengths[DW_LNS_set_basic_block] = 0
+standard_opcode_lengths[DW_LNS_const_add_pc] = 0
+standard_opcode_lengths[DW_LNS_fixed_advance_pc] = 1
+standard_opcode_lengths[DW_LNS_set_prologue_end] = 0
+standard_opcode_lengths[DW_LNS_set_epilogue_begin] = 0
+standard_opcode_lengths[DW_LNS_set_isa] = 1
+include_directories[  1] = "system/lib/compiler-rt"
+file_names[  1]:
+           name: "stack_ops.S"
+      dir_index: 1
+       mod_time: 0x00000000
+         length: 0x00000000
+
+Address            Line   Column File   ISA Discriminator OpIndex Flags
+------------------ ------ ------ ------ --- ------------- ------- -------------
+0x000000000000000e     18      0      1   0             0       0  is_stmt
+0x0000000000000011     19      0      1   0             0       0  is_stmt
+0x0000000000000012     19      0      1   0             0       0  is_stmt end_sequence
+0x0000000000000013     23      0      1   0             0       0  is_stmt
+0x0000000000000016     24      0      1   0             0       0  is_stmt
+0x0000000000000018     25      0      1   0             0       0  is_stmt
+0x0000000000000019     25      0      1   0             0       0  is_stmt end_sequence
+0x000000000000001f     30      0      1   0             0       0  is_stmt
+0x0000000000000021     32      0      1   0             0       0  is_stmt
+0x0000000000000023     34      0      1   0             0       0  is_stmt
+0x0000000000000024     36      0      1   0             0       0  is_stmt
+0x0000000000000026     37      0      1   0             0       0  is_stmt
+0x0000000000000027     38      0      1   0             0       0  is_stmt
+0x0000000000000029     39      0      1   0             0       0  is_stmt
+0x000000000000002b     40      0      1   0             0       0  is_stmt
+0x000000000000002d     41      0      1   0             0       0  is_stmt
+0x000000000000002e     41      0      1   0             0       0  is_stmt end_sequence
+

test/other/wasm_sourcemap_resolve_dummy_comp_dir/no_main.c

@@ -0,0 +1,13 @@
+/*
+ * Copyright 2023 The Emscripten Authors.  All rights reserved.
+ * Emscripten is available under two separate licenses, the MIT license and the
+ * University of Illinois/NCSA Open Source License.  Both these licenses can be
+ * found in the LICENSE file.
+ */
+
+// embuilder build libcompiler_rt --force
+// emcc -nostartfiles -sASSERTIONS=0 -sSUPPORT_ERRNO=0 \
+//   -sEXPORTED_FUNCTIONS=_foo -Wl,--no-entry -g -o foo.wasm no_main.c
+// llvm-dwarfdump -debug-info -debug-line --recurse-depth=0 foo.wasm \
+//   >foo.wasm.dump
+int foo() { return 42; }

test/test_other.py

@@ -9561,6 +9561,18 @@ def test_wasm_sourcemap_dead(self):
     # has only two entries
     self.assertRegexpMatches(output, r'"mappings":\s*"[A-Za-z0-9+/]+,[A-Za-z0-9+/]+"')
 
+  def test_wasm_sourcemap_resolve_dummy_comp_dir(self):
+    wasm_map_cmd = [PYTHON, path_from_root('tools/wasm-sourcemap.py'),
+                    '--dwarfdump-output',
+                    test_file('other/wasm_sourcemap_resolve_dummy_comp_dir/foo.wasm.dump'),
+                    '-o', 'a.out.wasm.map',
+                    test_file('other/wasm_sourcemap_resolve_dummy_comp_dir/foo.wasm'),
+                    '--basepath=' + path_from_root()]
+    self.run_process(wasm_map_cmd, stdout=PIPE, stderr=PIPE)
+    output = read_file('a.out.wasm.map')
+    # has an actual path to the source file of system library, not a dummy path (/emsdk/emscripten/**)
+    self.assertIn('"system/lib/compiler-rt/stack_ops.S"', output)
+
   def test_wasm_sourcemap_relative_paths(self):
     ensure_dir('build')
 
@@ -9803,11 +9815,11 @@ def test_dwarf_system_lib(self):
     if config.FROZEN_CACHE:
       self.skipTest("test doesn't work with frozen cache")
     self.run_process([EMBUILDER, 'build', 'libemmalloc', '--force'])
-    libc = os.path.join(config.CACHE, 'sysroot', 'lib', 'wasm32-emscripten', 'libemmalloc.a')
-    self.assertExists(libc)
+    system_lib = os.path.join(config.CACHE, 'sysroot', 'lib', 'wasm32-emscripten', 'libemmalloc.a')
+    self.assertExists(system_lib)
 
     dwdump = self.run_process(
-        [LLVM_DWARFDUMP, libc, '-debug-info', '-debug-line', '--recurse-depth=0'],
+        [LLVM_DWARFDUMP, system_lib, '-debug-info', '-debug-line', '--recurse-depth=0'],
         stdout=PIPE).stdout
     # Check that the embedded location of the source file is correct.
     self.assertIn('DW_AT_name\t("system/lib/emmalloc.c")', dwdump)

tools/system_libs.py

@@ -39,6 +39,9 @@
 # link time.
 USE_NINJA = int(os.environ.get('EMCC_USE_NINJA', '0'))
 
+# A dummy path used as a compilation directory for deterministic builds.
+DUMMY_EMSCRIPTEN_ROOT = '/emsdk/emscripten'
+
 
 def files_in_path(path, filenames):
   srcdir = utils.path_from_root(path)
@@ -472,9 +475,9 @@ def generate_ninja(self, build_dir, libname):
     if self.deterministic_paths:
       source_dir = utils.path_from_root()
       relative_source_dir = os.path.relpath(source_dir, build_dir)
-      cflags += [f'-ffile-prefix-map={source_dir}=/emsdk/emscripten',
+      cflags += [f'-ffile-prefix-map={source_dir}={DUMMY_EMSCRIPTEN_ROOT}',
                  f'-ffile-prefix-map={relative_source_dir}/=',
-                 '-fdebug-compilation-dir=/emsdk/emscripten']
+                 f'-fdebug-compilation-dir={DUMMY_EMSCRIPTEN_ROOT}']
     asflags = get_base_cflags(preprocess=False)
     input_files = self.get_files()
     ninja_file = os.path.join(build_dir, 'build.ninja')
@@ -494,9 +497,9 @@ def build_objects(self, build_dir):
     if self.deterministic_paths:
       source_dir = utils.path_from_root()
       relative_source_dir = os.path.relpath(source_dir, build_dir)
-      cflags += [f'-ffile-prefix-map={source_dir}=/emsdk/emscripten',
+      cflags += [f'-ffile-prefix-map={source_dir}={DUMMY_EMSCRIPTEN_ROOT}',
                  f'-ffile-prefix-map={relative_source_dir}/=',
-                 '-fdebug-compilation-dir=/emsdk/emscripten']
+                 f'-fdebug-compilation-dir={DUMMY_EMSCRIPTEN_ROOT}']
     case_insensitive = is_case_insensitive(build_dir)
     for src in self.get_files():
       ext = shared.suffix(src)

tools/wasm-sourcemap.py

@@ -25,6 +25,7 @@
 sys.path.insert(0, __rootdir__)
 
 from tools import utils
+from tools import system_libs
 
 logger = logging.getLogger('wasm-sourcemap')
 
@@ -238,6 +239,9 @@ def read_dwarf_entries(wasm, options):
     for file in re.finditer(r"file_names\[\s*(\d+)\]:\s+name: \"([^\"]*)\"\s+dir_index: (\d+)", line_chunk):
       dir = include_directories[file.group(3)]
       file_path = os.path.join(dir, file.group(2))
+      if file_path.startswith(f'{system_libs.DUMMY_EMSCRIPTEN_ROOT}{os.path.sep}'):
+        sub_dir = os.path.relpath(file_path, system_libs.DUMMY_EMSCRIPTEN_ROOT)
+        file_path = utils.path_from_root(sub_dir)
       files[file.group(1)] = file_path
 
     for line in re.finditer(r"\n0x([0-9a-f]+)\s+(\d+)\s+(\d+)\s+(\d+)(.*?end_sequence)?", line_chunk):