profilers: print user-friendly errors

mkokryashkin · mkokryashkin · commit c9e7600b213f · 2023-12-12T15:38:03.000+03:00
Prior to this patch, there was no error-checking in profilers, which resulted in raw Lua errors being reported in cases of non-existing paths or corrupt file structures. This patch adds error handling, so all parsing errors are now reported in a user-friendly manner. Event parsing is now moved into a separate profiler-agnostic module. Tool CLI flag tests are adapted correspondingly to error message changes. Resolves tarantool/tarantool#9217 Part of tarantool/tarantool#5994
diff --git a/Makefile.original b/Makefile.original
@@ -97,7 +97,7 @@ FILES_JITLIB= bc.lua bcsave.lua dump.lua p.lua v.lua zone.lua \
 	      dis_x86.lua dis_x64.lua dis_arm.lua dis_arm64.lua \
 	      dis_arm64be.lua dis_ppc.lua dis_mips.lua dis_mipsel.lua \
 	      dis_mips64.lua dis_mips64el.lua vmdef.lua
-FILES_UTILSLIB= avl.lua bufread.lua symtab.lua
+FILES_UTILSLIB= avl.lua bufread.lua evread.lua symtab.lua
 FILES_MEMPROFLIB= humanize.lua parse.lua process.lua
 FILES_SYSPROFLIB= parse.lua
 FILES_TOOLSLIB= memprof.lua sysprof.lua
diff --git a/test/tarantool-tests/gh-5688-tool-cli-flag.test.lua b/test/tarantool-tests/gh-5688-tool-cli-flag.test.lua
@@ -42,7 +42,7 @@ local SMOKE_CMD_SET = {
 local MEMPROF_CMD_SET = {
   {
     cmd = MEMPROF_PARSER .. BAD_PATH,
-    like = 'fopen, errno: 2',
+    like = 'Failed to open.*fopen, errno: 2',
   },
   {
     cmd = MEMPROF_PARSER .. TMP_BINFILE_MEMPROF,
@@ -61,7 +61,7 @@ local MEMPROF_CMD_SET = {
 local SYSPROF_CMD_SET = {
   {
     cmd = SYSPROF_PARSER .. BAD_PATH,
-    like = 'fopen, errno: 2',
+    like = 'Failed to open.*fopen, errno: 2',
   },
   {
     cmd = SYSPROF_PARSER .. TMP_BINFILE_SYSPROF,
diff --git a/test/tarantool-tests/gh-9217-profile-parsers-error-handling.test.lua b/test/tarantool-tests/gh-9217-profile-parsers-error-handling.test.lua
@@ -0,0 +1,90 @@
+local tap = require('tap')
+local test = tap.test('gh-9217-profile-parsers-error-handling'):skipcond({
+  ['Profile tools are implemented for x86_64 only'] = jit.arch ~= 'x86' and
+                                                      jit.arch ~= 'x64',
+  ['Profile tools are implemented for Linux only'] = jit.os ~= 'Linux',
+  -- XXX: Tarantool integration is required to run this test properly.
+  -- luacheck: no global
+  ['No profile tools CLI option integration'] = _TARANTOOL,
+})
+
+jit.off()
+jit.flush()
+
+local table_new = require('table.new')
+local utils = require('utils')
+
+local BAD_PATH = utils.tools.profilename('bad-path-tmp.bin')
+local NON_PROFILE_DATA = utils.tools.profilename('not-profile-data.tmp.bin')
+local CORRUPT_PROFILE = utils.tools.profilename('profdata.tmp.bin')
+
+local EXECUTABLE = utils.exec.luacmd(arg)
+local PARSERS = {
+  memprof = EXECUTABLE .. ' -tm ',
+  sysprof = EXECUTABLE .. ' -ts ',
+}
+local REDIRECT_OUTPUT = ' 2>&1'
+
+local TABLE_SIZE = 20
+
+local TEST_CASES = {
+  {
+    path = BAD_PATH,
+    err_msg = 'Failed to open'
+  },
+  {
+    path = NON_PROFILE_DATA,
+    err_msg = 'Failed to parse symtab from'
+  },
+  {
+    path = CORRUPT_PROFILE,
+    err_msg = 'Failed to parse profile data from'
+  },
+}
+
+test:plan(2 * #TEST_CASES)
+
+local function generate_non_profile_data(path)
+  local file = io.open(path, 'w')
+  file:write('data')
+  file:close()
+end
+
+local function generate_corrupt_profile(path)
+  local res, err = misc.memprof.start(path)
+  -- Should start successfully.
+  assert(res, err)
+
+  local _ = table_new(TABLE_SIZE, 0)
+   _ = nil
+  collectgarbage()
+
+  res, err = misc.memprof.stop()
+  -- Should stop successfully.
+  assert(res, err)
+
+  local file = io.open(path, 'r')
+  local content = file:read('*all')
+  file:close()
+  local index = string.find(content, 'ljm')
+
+  file = io.open(path, 'w')
+  file:write(string.sub(content, 1, index - 1))
+  file:close()
+end
+
+generate_non_profile_data(NON_PROFILE_DATA)
+generate_corrupt_profile(CORRUPT_PROFILE)
+
+for _, case in ipairs(TEST_CASES) do
+  for profiler, parser in pairs(PARSERS) do
+    local path = case.path
+    local err_msg = case.err_msg
+    local output = io.popen(parser .. path .. REDIRECT_OUTPUT):read('*all')
+    test:like(output, err_msg, string.format('%s: %s', profiler, err_msg))
+  end
+end
+
+os.remove(NON_PROFILE_DATA)
+os.remove(CORRUPT_PROFILE)
+test:done(true)
diff --git a/tools/CMakeLists.txt b/tools/CMakeLists.txt
@@ -18,6 +18,7 @@ else()
     memprof.lua
     utils/avl.lua
     utils/bufread.lua
+    utils/evread.lua
     utils/symtab.lua
   )
   list(APPEND LUAJIT_TOOLS_DEPS tools-parse-memprof)
@@ -36,6 +37,7 @@ else()
   install(FILES
       ${CMAKE_CURRENT_SOURCE_DIR}/utils/avl.lua
       ${CMAKE_CURRENT_SOURCE_DIR}/utils/bufread.lua
+      ${CMAKE_CURRENT_SOURCE_DIR}/utils/evread.lua
       ${CMAKE_CURRENT_SOURCE_DIR}/utils/symtab.lua
     DESTINATION ${LUAJIT_DATAROOTDIR}/utils
     PERMISSIONS
@@ -65,6 +67,7 @@ else()
     sysprof.lua
     utils/avl.lua
     utils/bufread.lua
+    utils/evread.lua
     utils/symtab.lua
   )
   list(APPEND LUAJIT_TOOLS_DEPS tools-parse-sysprof)
@@ -81,6 +84,7 @@ else()
   install(FILES
       ${CMAKE_CURRENT_SOURCE_DIR}/utils/avl.lua
       ${CMAKE_CURRENT_SOURCE_DIR}/utils/bufread.lua
+      ${CMAKE_CURRENT_SOURCE_DIR}/utils/evread.lua
       ${CMAKE_CURRENT_SOURCE_DIR}/utils/symtab.lua
     DESTINATION ${LUAJIT_DATAROOTDIR}/utils
     PERMISSIONS
diff --git a/tools/memprof.lua b/tools/memprof.lua
@@ -10,10 +10,9 @@
 -- Major portions taken verbatim or adapted from the LuaVela.
 -- Copyright (C) 2015-2019 IPONWEB Ltd.
 
-local bufread = require "utils.bufread"
 local memprof = require "memprof.parse"
 local process = require "memprof.process"
-local symtab = require "utils.symtab"
+local evread = require "utils.evread"
 local view = require "memprof.humanize"
 
 local stdout, stderr = io.stdout, io.stderr
@@ -108,9 +107,11 @@ local function parseargs(args)
 end
 
 local function dump(inputfile)
-  local reader = bufread.new(inputfile)
-  local symbols = symtab.parse(reader)
-  local events = memprof.parse(reader, symbols)
+  -- XXX: This function exits with a non-zero exit code and
+  -- prints an error message if it encounters any failure during
+  -- the process of parsing.
+  local events, symbols = evread(memprof.parse, inputfile)
+
   if not config.leak_only then
     view.profile_info(events, config)
   end
diff --git a/tools/sysprof.lua b/tools/sysprof.lua
@@ -1,6 +1,5 @@
-local bufread = require "utils.bufread"
+local evread = require "utils.evread"
 local sysprof = require "sysprof.parse"
-local symtab = require "utils.symtab"
 
 local stdout, stderr = io.stdout, io.stderr
 local match, gmatch = string.match, string.gmatch
@@ -78,9 +77,10 @@ local function parseargs(args)
 end
 
 local function dump(inputfile)
-  local reader = bufread.new(inputfile)
-  local symbols = symtab.parse(reader)
-  local events = sysprof.parse(reader, symbols)
+  -- XXX: This function exits with a non-zero exit code and
+  -- prints an error message if it encounters any failure during
+  -- the process of parsing.
+  local events = evread(sysprof.parse, inputfile)
 
   for stack, count in pairs(events) do
     print(stack, count)
diff --git a/tools/sysprof/parse.lua b/tools/sysprof/parse.lua
@@ -237,7 +237,7 @@ function M.parse(reader, symbols)
   local _ = reader:read_octets(3)
 
   if magic ~= LJP_MAGIC then
-    error("Bad LJP format prologue: "..magic)
+    error("Bad LJP format prologue: " .. tostring(magic))
   end
 
   if string.byte(version) ~= LJP_CURRENT_VERSION then
diff --git a/tools/utils/evread.lua b/tools/utils/evread.lua
@@ -0,0 +1,32 @@
+local bufread = require('utils.bufread')
+local symtab = require('utils.symtab')
+
+local function make_error_handler(fmt, inputfile)
+  return function(err)
+    io.stderr:write(string.format(fmt, inputfile))
+    io.stderr:write(string.format('\n\t%s\n', err))
+    os.exit(1, true)
+  end
+end
+
+return function(parser, inputfile)
+  local _, reader = xpcall(
+    bufread.new,
+    make_error_handler('Failed to open %s.', inputfile),
+    inputfile
+  )
+
+  local _, symbols = xpcall(
+    symtab.parse,
+    make_error_handler('Failed to parse symtab from %s.', inputfile),
+    reader
+  )
+
+  local _, events = xpcall(
+    parser,
+    make_error_handler('Failed to parse profile data from %s.', inputfile),
+    reader,
+    symbols
+  )
+  return events, symbols
+end
diff --git a/tools/utils/symtab.lua b/tools/utils/symtab.lua
@@ -95,7 +95,7 @@ function M.parse(reader)
   local _ = reader:read_octets(3)
 
   if magic ~= LJS_MAGIC then
-    error("Bad LJS format prologue: "..magic)
+    error("Bad LJS format prologue: " .. tostring(magic))
   end
 
   if string.byte(version) ~= LJS_CURRENT_VERSION then

Original file line number	Diff line number	Diff line change
`@@ -42,7 +42,7 @@ local SMOKE_CMD_SET = {`
`42`	`42`	`local MEMPROF_CMD_SET = {`
`43`	`43`	`{`
`44`	`44`	`cmd = MEMPROF_PARSER .. BAD_PATH,`
`45`		`- like = 'fopen, errno: 2',`
	`45`	`+ like = 'Failed to open.*fopen, errno: 2',`
`46`	`46`	`},`
`47`	`47`	`{`
`48`	`48`	`cmd = MEMPROF_PARSER .. TMP_BINFILE_MEMPROF,`
`@@ -61,7 +61,7 @@ local MEMPROF_CMD_SET = {`
`61`	`61`	`local SYSPROF_CMD_SET = {`
`62`	`62`	`{`
`63`	`63`	`cmd = SYSPROF_PARSER .. BAD_PATH,`
`64`		`- like = 'fopen, errno: 2',`
	`64`	`+ like = 'Failed to open.*fopen, errno: 2',`
`65`	`65`	`},`
`66`	`66`	`{`
`67`	`67`	`cmd = SYSPROF_PARSER .. TMP_BINFILE_SYSPROF,`