runtime/trace: fix tracing of blocking system calls

The placement and invocation of traceGoSysCall when using
entersyscallblock() instead of entersyscall() differs enough that the
TestTraceSymbolize test can fail on some platforms.

This change moves the invocation of traceGoSysCall for entersyscall() so
that the same number of "frames to skip" are present in the trace as when
entersyscallblock() is used ensuring system call traces remain identical
regardless of internal implementation choices.

Fixes golang#12056

Change-Id: I8361e91aa3708f5053f98263dfe9feb8c5d1d969
Reviewed-on: https://go-review.googlesource.com/13861
Run-TryBot: Dmitry Vyukov <[email protected]>
Reviewed-by: Dmitry Vyukov <[email protected]>

src/runtime/proc1.go

@@ -1823,10 +1823,6 @@ func reentersyscall(pc, sp uintptr) {
 	// but can have inconsistent g->sched, do not let GC observe it.
 	_g_.m.locks++
 
-	if trace.enabled {
-		systemstack(traceGoSysCall)
-	}
-
 	// Entersyscall must not call any function that might split/grow the stack.
 	// (See details in comment above.)
 	// Catch calls that might, by replacing the stack guard with something that
@@ -1846,6 +1842,14 @@ func reentersyscall(pc, sp uintptr) {
 		})
 	}
 
+	if trace.enabled {
+		systemstack(traceGoSysCall)
+		// systemstack itself clobbers g.sched.{pc,sp} and we might
+		// need them later when the G is genuinely blocked in a
+		// syscall
+		save(pc, sp)
+	}
+
 	if atomicload(&sched.sysmonwait) != 0 { // TODO: fast atomic
 		systemstack(entersyscall_sysmon)
 		save(pc, sp)

src/runtime/trace.go

@@ -828,7 +828,7 @@ func traceGoUnpark(gp *g, skip int) {
 }
 
 func traceGoSysCall() {
-	traceEvent(traceEvGoSysCall, 4)
+	traceEvent(traceEvGoSysCall, 1)
 }
 
 func traceGoSysExit(seq uint64, ts int64) {