[release-branch.go1.22] runtime: traceAcquire and traceRelease across all P steals

Currently there are a few places where a P can get stolen where the
runtime doesn't traceAcquire and traceRelease across the steal itself.
What can happen then is the following scenario:
- Thread 1 enters a syscall and writes an event about it.
- Thread 2 steals Thread 1's P.
- Thread 1 exits the syscall and writes one or more events about it.
- Tracing ends (trace.gen is set to 0).
- Thread 2 checks to see if it should write an event for the P it just
  stole, sees that tracing is disabled, and doesn't.

This results in broken traces, because there's a missing ProcSteal
event. The parser always waits for a ProcSteal to advance a
GoSyscallEndBlocked event, and in this case, it never comes.

Fixes #65181.

Change-Id: I437629499bb7669bf7fe2fc6fc4f64c53002916b
Reviewed-on: https://go-review.googlesource.com/c/go/+/560235
Reviewed-by: Michael Pratt <mpratt@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
(cherry picked from commit c9d88ea2aa628cae224335c49f256e13adfce337)
Reviewed-on: https://go-review.googlesource.com/c/go/+/559958
Auto-Submit: Michael Knyszek <mknyszek@google.com>

1 files changed, 15 insertions, 6 deletions

diff --git a/src/runtime/proc.go b/src/runtime/proc.go
index c2676c43b2..061673150f 100644
--- a/src/runtime/proc.go
+++ b/src/runtime/proc.go
@@ -4404,8 +4404,8 @@ func entersyscall_gcwait() {
 	pp := gp.m.oldp.ptr()
 
 	lock(&sched.lock)
+	trace := traceAcquire()
 	if sched.stopwait > 0 && atomic.Cas(&pp.status, _Psyscall, _Pgcstop) {
-		trace := traceAcquire()
 		if trace.ok() {
 			if goexperiment.ExecTracer2 {
 				// This is a steal in the new tracer. While it's very likely
@@ -4428,6 +4428,8 @@ func entersyscall_gcwait() {
 		if sched.stopwait--; sched.stopwait == 0 {
 			notewakeup(&sched.stopnote)
 		}
+	} else if trace.ok() {
+		traceRelease(trace)
 	}
 	unlock(&sched.lock)
 }
@@ -4605,12 +4607,19 @@ func exitsyscallfast(oldp *p) bool {
 	}
 
 	// Try to re-acquire the last P.
+	trace := traceAcquire()
 	if oldp != nil && oldp.status == _Psyscall && atomic.Cas(&oldp.status, _Psyscall, _Pidle) {
 		// There's a cpu for us, so we can run.
 		wirep(oldp)
-		exitsyscallfast_reacquired()
+		exitsyscallfast_reacquired(trace)
+		if trace.ok() {
+			traceRelease(trace)
+		}
 		return true
 	}
+	if trace.ok() {
+		traceRelease(trace)
+	}
 
 	// Try to get any other idle P.
 	if sched.pidle != 0 {
@@ -4646,10 +4655,9 @@ func exitsyscallfast(oldp *p) bool {
 // syscall.
 //
 //go:nosplit
-func exitsyscallfast_reacquired() {
+func exitsyscallfast_reacquired(trace traceLocker) {
 	gp := getg()
 	if gp.m.syscalltick != gp.m.p.ptr().syscalltick {
-		trace := traceAcquire()
 		if trace.ok() {
 			// The p was retaken and then enter into syscall again (since gp.m.syscalltick has changed).
 			// traceGoSysBlock for this syscall was already emitted,
@@ -4666,7 +4674,6 @@ func exitsyscallfast_reacquired() {
 					// Denote completion of the current syscall.
 					trace.GoSysExit(true)
 				}
-				traceRelease(trace)
 			})
 		}
 		gp.m.p.ptr().syscalltick++
@@ -6146,8 +6153,8 @@ func retake(now int64) uint32 {
 			// Otherwise the M from which we retake can exit the syscall,
 			// increment nmidle and report deadlock.
 			incidlelocked(-1)
+			trace := traceAcquire()
 			if atomic.Cas(&pp.status, s, _Pidle) {
-				trace := traceAcquire()
 				if trace.ok() {
 					trace.GoSysBlock(pp)
 					trace.ProcSteal(pp, false)
@@ -6156,6 +6163,8 @@ func retake(now int64) uint32 {
 				n++
 				pp.syscalltick++
 				handoffp(pp)
+			} else if trace.ok() {
+				traceRelease(trace)
 			}
 			incidlelocked(1)
 			lock(&allpLock)