[release-branch.go1.17] runtime: don't block preemption signal in new M's or ensureSigM

No test because we already have a test in the syscall package.
The issue reports 1 failure per 100,000 iterations, which is rare enough
that our builders won't catch the problem.

For #52226
Fixes #52374

Change-Id: I17633ff6cf676b6d575356186dce42cdacad0746
Reviewed-on: https://go-review.googlesource.com/c/go/+/400315
Run-TryBot: Ian Lance Taylor <[email protected]>
TryBot-Result: Gopher Robot <[email protected]>
Reviewed-by: Michael Knyszek <[email protected]>
Reviewed-by: Ian Lance Taylor <[email protected]>
(cherry picked from commit e398266)
Reviewed-on: https://go-review.googlesource.com/c/go/+/400317
Reviewed-by: Austin Clements <[email protected]>

Author: ianlancetaylor

src/runtime/signal_unix.go

@@ -1151,6 +1151,7 @@ func unminitSignals() {
 // blockableSig reports whether sig may be blocked by the signal mask.
 // We never want to block the signals marked _SigUnblock;
 // these are the synchronous signals that turn into a Go panic.
+// We never want to block the preemption signal if it is being used.
 // In a Go program--not a c-archive/c-shared--we never want to block
 // the signals marked _SigKill or _SigThrow, as otherwise it's possible
 // for all running threads to block them and delay their delivery until
@@ -1161,6 +1162,9 @@ func blockableSig(sig uint32) bool {
 	if flags&_SigUnblock != 0 {
 		return false
 	}
+	if sig == sigPreempt && preemptMSupported && debug.asyncpreemptoff == 0 {
+		return false
+	}
 	if isarchive || islibrary {
 		return true
 	}