tsan: improve detection of stack/tls races

[dvyukov]

Summary

Print meaningful stack frames for stack/tls races
(instead of PC 1/2 that don't symbolize).

Imitate stack/tls writes after we create and initialize
the new thread, otherwise the races are not detected.

This is re-submit of the following reverted commits,
but without tests as they failed on a number of OSes/arches:
"tsan: fix and test detection of TLS races"
"tsan: fix tls_race3 test on darwin"
"tsan: print a meaningful frame for stack races"

compiler-rt/lib/tsan/rtl/tsan_platform_linux.cpp

static void InitializeLongjmpXorKey() {
   asm("mov  %0, sp" : "=r" (sp));
   uptr mangled_sp = ((uptr *)&env)[LONG_JMP_SP_ENV_SLOT];
 
   // 3. xor SPs to obtain key.
   longjmp_xor_key = mangled_sp ^ sp;
 }
 #endif
 
+extern "C" void __tsan_tls_initialization() {}
+
 void ImitateTlsWrite(ThreadState *thr, uptr tls_addr, uptr tls_size) {
   // Check that the thr object is in tls;
   const uptr thr_beg = (uptr)thr;
   const uptr thr_end = (uptr)thr + sizeof(*thr);
   CHECK_GE(thr_beg, tls_addr);
   CHECK_LE(thr_beg, tls_addr + tls_size);
   CHECK_GE(thr_end, tls_addr);
   CHECK_LE(thr_end, tls_addr + tls_size);
   // Since the thr object is huge, skip it.
-  MemoryRangeImitateWrite(thr, /*pc=*/2, tls_addr, thr_beg - tls_addr);
-  MemoryRangeImitateWrite(thr, /*pc=*/2, thr_end,
-                          tls_addr + tls_size - thr_end);
+  const uptr pc = StackTrace::GetNextInstructionPc(
+      reinterpret_cast<uptr>(__tsan_tls_initialization));
+  MemoryRangeImitateWrite(thr, pc, tls_addr, thr_beg - tls_addr);
+  MemoryRangeImitateWrite(thr, pc, thr_end, tls_addr + tls_size - thr_end);
 }
 
 // Note: this function runs with async signals enabled,
 // so it must not touch any tsan state.
 int call_pthread_cancel_with_cleanup(int (*fn)(void *arg),
                                      void (*cleanup)(void *arg), void *arg) {
   // pthread_cleanup_push/pop are hardcore macros mess.
   // We can't intercept nor call them w/o including pthread.h.

compiler-rt/lib/tsan/rtl/tsan_platform_mac.cpp

uptr ExtractLongJmpSp(uptr *env) {
   uptr mangled_sp = env[LONG_JMP_SP_ENV_SLOT];
   uptr sp = mangled_sp ^ longjmp_xor_key;
   sp = (uptr)ptrauth_auth_data((void *)sp, ptrauth_key_asdb,
                                ptrauth_string_discriminator("sp"));
   return sp;
 }
 
 #if !SANITIZER_GO
+extern "C" void __tsan_tls_initialization() {}
+
 void ImitateTlsWrite(ThreadState *thr, uptr tls_addr, uptr tls_size) {
   // The pointer to the ThreadState object is stored in the shadow memory
   // of the tls.
   uptr tls_end = tls_addr + tls_size;
   uptr thread_identity = (uptr)pthread_self();
+  const uptr pc = StackTrace::GetNextInstructionPc(
+      reinterpret_cast<uptr>(__tsan_tls_initialization));
   if (thread_identity == main_thread_identity) {
-    MemoryRangeImitateWrite(thr, /*pc=*/2, tls_addr, tls_size);
+    MemoryRangeImitateWrite(thr, pc, tls_addr, tls_size);
   } else {
     uptr thr_state_start = thread_identity;
     uptr thr_state_end = thr_state_start + sizeof(uptr);
     CHECK_GE(thr_state_start, tls_addr);
     CHECK_LE(thr_state_start, tls_addr + tls_size);
     CHECK_GE(thr_state_end, tls_addr);
     CHECK_LE(thr_state_end, tls_addr + tls_size);
-    MemoryRangeImitateWrite(thr, /*pc=*/2, tls_addr,
-                            thr_state_start - tls_addr);
-    MemoryRangeImitateWrite(thr, /*pc=*/2, thr_state_end,
-                            tls_end - thr_state_end);
+    MemoryRangeImitateWrite(thr, pc, tls_addr, thr_state_start - tls_addr);
+    MemoryRangeImitateWrite(thr, pc, thr_state_end, tls_end - thr_state_end);
   }
 }
 #endif
 
 #if !SANITIZER_GO
 // Note: this function runs with async signals enabled,
 // so it must not touch any tsan state.
 int call_pthread_cancel_with_cleanup(int (*fn)(void *arg),

compiler-rt/lib/tsan/rtl/tsan_rtl_thread.cpp

if (!args->thr)  // GCD workers don't have a parent thread.
     return;
   args->thr->fast_state.IncrementEpoch();
   // Can't increment epoch w/o writing to the trace as well.
   TraceAddEvent(args->thr, args->thr->fast_state, EventTypeMop, 0);
   ReleaseImpl(args->thr, 0, &sync);
   creation_stack_id = CurrentStackId(args->thr, args->pc);
 }
 
+extern "C" void __tsan_stack_initialization() {}
+
 struct OnStartedArgs {
   ThreadState *thr;
   uptr stk_addr;
   uptr stk_size;
   uptr tls_addr;
   uptr tls_size;
 };
 
 void ThreadStart(ThreadState *thr, Tid tid, tid_t os_id,
                  ThreadType thread_type) {
   uptr stk_addr = 0;
   uptr stk_size = 0;
   uptr tls_addr = 0;
   uptr tls_size = 0;
 #if !SANITIZER_GO
   if (thread_type != ThreadType::Fiber)
     GetThreadStackAndTls(tid == kMainTid, &stk_addr, &stk_size, &tls_addr,
                          &tls_size);
-
-  if (tid != kMainTid) {
-    if (stk_addr && stk_size)
-      MemoryRangeImitateWrite(thr, /*pc=*/ 1, stk_addr, stk_size);
-
-    if (tls_addr && tls_size) ImitateTlsWrite(thr, tls_addr, tls_size);
-  }
 #endif
 
   ThreadRegistry *tr = &ctx->thread_registry;
   OnStartedArgs args = { thr, stk_addr, stk_size, tls_addr, tls_size };
   tr->StartThread(tid, os_id, thread_type, &args);
 
   while (!thr->tctx->trace.parts.Empty()) thr->tctx->trace.parts.PopBack();
 
 #if !SANITIZER_GO
   if (ctx->after_multithreaded_fork) {
     thr->ignore_interceptors++;
     ThreadIgnoreBegin(thr, 0);
     ThreadIgnoreSyncBegin(thr, 0);
   }
 #endif
+
+#if !SANITIZER_GO
+  // Don't imitate stack/TLS writes for the main thread,
+  // because its initialization is synchronized with all
+  // subsequent threads anyway.
+  if (tid != kMainTid) {
+    if (stk_addr && stk_size) {
+      const uptr pc = StackTrace::GetNextInstructionPc(
+          reinterpret_cast<uptr>(__tsan_stack_initialization));
+      MemoryRangeImitateWrite(thr, pc, stk_addr, stk_size);
+    }
+
+    if (tls_addr && tls_size)
+      ImitateTlsWrite(thr, tls_addr, tls_size);
+  }
+#endif
 }
 
 void ThreadContext::OnStarted(void *arg) {
   OnStartedArgs *args = static_cast<OnStartedArgs *>(arg);
   thr = args->thr;
   // RoundUp so that one trace part does not contain events
   // from different threads.
   epoch0 = RoundUp(epoch1 + 1, kTracePartSize);