Profile: make heap snapshots viewable in vscode viewer (#53833)

Author: IanButterworth

src/gc-heap-snapshot.cpp

@@ -606,7 +606,9 @@ void _record_gc_just_edge(const char *edge_type, size_t from_idx, size_t to_idx,
 void final_serialize_heap_snapshot(ios_t *json, ios_t *strings, HeapSnapshot &snapshot, char all_one)
 {
     // mimicking https://github.com/nodejs/node/blob/5fd7a72e1c4fbaf37d3723c4c81dce35c149dc84/deps/v8/src/profiler/heap-snapshot-generator.cc#L2567-L2567
+    // also https://github.com/microsoft/vscode-v8-heap-tools/blob/c5b34396392397925ecbb4ecb904a27a2754f2c1/v8-heap-parser/src/decoder.rs#L43-L51
     ios_printf(json, "{\"snapshot\":{");
+
     ios_printf(json, "\"meta\":{");
     ios_printf(json, "\"node_fields\":[\"type\",\"name\",\"id\",\"self_size\",\"edge_count\",\"trace_node_id\",\"detachedness\"],");
     ios_printf(json, "\"node_types\":[");
@@ -617,10 +619,26 @@ void final_serialize_heap_snapshot(ios_t *json, ios_t *strings, HeapSnapshot &sn
     ios_printf(json, "\"edge_types\":[");
     snapshot.edge_types.print_json_array(json, false);
     ios_printf(json, ",");
-    ios_printf(json, "\"string_or_number\",\"from_node\"]");
+    ios_printf(json, "\"string_or_number\",\"from_node\"],");
+    // not used. Required by microsoft/vscode-v8-heap-tools
+    ios_printf(json, "\"trace_function_info_fields\":[\"function_id\",\"name\",\"script_name\",\"script_id\",\"line\",\"column\"],");
+    ios_printf(json, "\"trace_node_fields\":[\"id\",\"function_info_index\",\"count\",\"size\",\"children\"],");
+    ios_printf(json, "\"sample_fields\":[\"timestamp_us\",\"last_assigned_id\"],");
+    ios_printf(json, "\"location_fields\":[\"object_index\",\"script_id\",\"line\",\"column\"]");
+    // end not used
     ios_printf(json, "},\n"); // end "meta"
+
     ios_printf(json, "\"node_count\":%zu,", snapshot.num_nodes);
-    ios_printf(json, "\"edge_count\":%zu", snapshot.num_edges);
-    ios_printf(json, "}\n"); // end "snapshot"
+    ios_printf(json, "\"edge_count\":%zu,", snapshot.num_edges);
+    ios_printf(json, "\"trace_function_count\":0"); // not used. Required by microsoft/vscode-v8-heap-tools
+    ios_printf(json, "},\n"); // end "snapshot"
+
+    // not used. Required by microsoft/vscode-v8-heap-tools
+    ios_printf(json, "\"trace_function_infos\":[],");
+    ios_printf(json, "\"trace_tree\":[],");
+    ios_printf(json, "\"samples\":[],");
+    ios_printf(json, "\"locations\":[]");
+    // end not used
+
     ios_printf(json, "}");
 }

stdlib/Manifest.toml

@@ -2,7 +2,7 @@
 
 julia_version = "1.12.0-DEV"
 manifest_format = "2.0"
-project_hash = "13f2dd600364a1e8b659dc5796bf185b37d1c95d"
+project_hash = "d3a1f6b706609fe0c59521e1d770be6e2b8c489d"
 
 [[deps.ArgTools]]
 uuid = "0dad84c5-d112-42e6-8d28-ef12dabb789f"
@@ -23,7 +23,7 @@ version = "1.11.0"
 [[deps.CompilerSupportLibraries_jll]]
 deps = ["Artifacts", "Libdl"]
 uuid = "e66e0078-7015-5450-92f7-15fbd957f2ae"
-version = "1.1.0+0"
+version = "1.1.1+0"
 
 [[deps.Dates]]
 deps = ["Printf"]
@@ -113,7 +113,7 @@ version = "1.11.0+1"
 [[deps.LibUV_jll]]
 deps = ["Artifacts", "Libdl"]
 uuid = "183b4373-6708-53ba-ad28-60e28bb38547"
-version = "2.0.1+15"
+version = "2.0.1+16"
 
 [[deps.LibUnwind_jll]]
 deps = ["Artifacts", "Libdl"]
@@ -155,7 +155,7 @@ version = "1.11.0"
 
 [[deps.MozillaCACerts_jll]]
 uuid = "14a3606d-f60d-562e-9121-12d972cd8159"
-version = "2023.12.12"
+version = "2024.3.11"
 
 [[deps.NetworkOptions]]
 uuid = "ca575930-c2e3-43a9-ace4-1e988b2c1908"
@@ -177,9 +177,9 @@ uuid = "efcefdf7-47ab-520b-bdef-62a2eaa19f15"
 version = "10.43.0+0"
 
 [[deps.Pkg]]
-deps = ["Artifacts", "Dates", "Downloads", "FileWatching", "LibGit2", "Libdl", "Logging", "Markdown", "Printf", "Random", "SHA", "Serialization", "TOML", "Tar", "UUIDs", "p7zip_jll"]
+deps = ["Artifacts", "Dates", "Downloads", "FileWatching", "LibGit2", "Libdl", "Logging", "Markdown", "Printf", "Random", "SHA", "TOML", "Tar", "UUIDs", "p7zip_jll"]
 uuid = "44cfe95a-1eb2-52ea-b672-e2afdf69b78f"
-version = "1.11.0"
+version = "1.12.0"
 weakdeps = ["REPL"]
 
     [deps.Pkg.extensions]
@@ -191,6 +191,7 @@ uuid = "de0858da-6303-5e67-8744-51eddeeeb8d7"
 version = "1.11.0"
 
 [[deps.Profile]]
+deps = ["Unicode"]
 uuid = "9abbd945-dff8-562f-b5e8-e1ebf5ef1b79"
 version = "1.11.0"
 

stdlib/Profile/Project.toml

@@ -2,6 +2,12 @@ name = "Profile"
 uuid = "9abbd945-dff8-562f-b5e8-e1ebf5ef1b79"
 version = "1.11.0"
 
+[deps]
+Unicode = "4ec0a83e-493e-50e2-b9ac-8f72acf5a8f5"
+
+[compat]
+Unicode = "1.11.0"
+
 [extras]
 Base64 = "2a0f44e3-6c83-55bd-87e4-b1978d98bd5f"
 Logging = "56ddb016-857b-54e1-b83d-db4d58db5568"

stdlib/Profile/src/Profile.jl

@@ -1286,6 +1286,7 @@ function take_heap_snapshot(filepath::AbstractString, all_one::Bool=false; strea
         prefix = filepath
         _stream_heap_snapshot(prefix, all_one)
         Profile.HeapSnapshot.assemble_snapshot(prefix, filepath)
+        Profile.HeapSnapshot.cleanup_streamed_files(prefix)
     end
     return filepath
 end

stdlib/Profile/src/heapsnapshot_reassemble.jl

@@ -2,6 +2,8 @@
 
 module HeapSnapshot
 
+using Unicode
+
 """
     assemble_snapshot(filepath::AbstractString, out_file::AbstractString)
 
@@ -92,7 +94,7 @@ function assemble_snapshot(in_prefix, io::IO)
     node_count = parse(Int, String(@view preamble[pos:endpos]))
 
     pos = last(findnext("edge_count\":", preamble, endpos)) + 1
-    endpos = findnext(==('}'), preamble, pos) - 1
+    endpos = findnext(==(','), preamble, pos) - 1
     edge_count = parse(Int, String(@view preamble[pos:endpos]))
 
     nodes = Nodes(node_count, edge_count)
@@ -137,7 +139,8 @@ function assemble_snapshot(in_prefix, io::IO)
     end
 
     _digits_buf = zeros(UInt8, ndigits(typemax(UInt)))
-    println(io, @view(preamble[1:end-2]), ",") # remove trailing "}\n", we don't end the snapshot here
+    println(io, @view(preamble[1:end-1]), ",") # remove trailing "}" to reopen the object
+
     println(io, "\"nodes\":[")
     for i in 1:length(nodes)
         i > 1 && println(io, ",")
@@ -182,12 +185,11 @@ function assemble_snapshot(in_prefix, io::IO)
             str_bytes = read(strings_io, str_size)
             str = String(str_bytes)
             if first
-                print_str_escape_json(io, str)
                 first = false
             else
                 print(io, ",\n")
-                print_str_escape_json(io, str)
             end
+            print_str_escape_json(io, str)
         end
     end
     print(io, "]}")
@@ -202,6 +204,19 @@ function assemble_snapshot(in_prefix, io::IO)
     return nothing
 end
 
+"""
+    cleanup_streamed_files(prefix::AbstractString)
+
+Remove files streamed during `take_heap_snapshot` in streaming mode.
+"""
+function cleanup_streamed_files(prefix::AbstractString)
+    rm(string(prefix, ".metadata.json"))
+    rm(string(prefix, ".nodes"))
+    rm(string(prefix, ".edges"))
+    rm(string(prefix, ".strings"))
+    return nothing
+end
+
 function print_str_escape_json(stream::IO, s::AbstractString)
     print(stream, '"')
     for c in s
@@ -221,6 +236,9 @@ function print_str_escape_json(stream::IO, s::AbstractString)
             print(stream, "\\t")
         elseif '\x00' <= c <= '\x1f'
             print(stream, "\\u", lpad(string(UInt16(c), base=16), 4, '0'))
+        elseif !Unicode.isassigned(c)
+            # we have to do this because vscode's viewer doesn't like the replace character
+            print(stream, "[invalid unicode character]")
         else
             print(stream, c)
         end