diff --git a/src/runtime/os_solaris.go b/src/runtime/os_solaris.go
index 703a2e5430..4575b5e641 100644
--- a/src/runtime/os_solaris.go
+++ b/src/runtime/os_solaris.go
@@ -37,12 +37,14 @@ func sysvicall0(fn *libcFunc) uintptr {
 	if gp != nil {
 		mp = gp.m
 	}
-	if mp != nil {
+	if mp != nil && mp.libcallsp == 0 {
 		mp.libcallg.set(gp)
 		mp.libcallpc = getcallerpc()
 		// sp must be the last, because once async cpu profiler finds
 		// all three values to be non-zero, it will use them
 		mp.libcallsp = getcallersp()
+	} else {
+		mp = nil // See comment in sys_darwin.go:libcCall
 	}
 
 	var libcall libcall
@@ -64,12 +66,14 @@ func sysvicall1(fn *libcFunc, a1 uintptr) uintptr {
 	if gp != nil {
 		mp = gp.m
 	}
-	if mp != nil {
+	if mp != nil && mp.libcallsp == 0 {
 		mp.libcallg.set(gp)
 		mp.libcallpc = getcallerpc()
 		// sp must be the last, because once async cpu profiler finds
 		// all three values to be non-zero, it will use them
 		mp.libcallsp = getcallersp()
+	} else {
+		mp = nil
 	}
 
 	var libcall libcall
@@ -92,12 +96,14 @@ func sysvicall2(fn *libcFunc, a1, a2 uintptr) uintptr {
 	if gp != nil {
 		mp = gp.m
 	}
-	if mp != nil {
+	if mp != nil && mp.libcallsp == 0 {
 		mp.libcallg.set(gp)
 		mp.libcallpc = getcallerpc()
 		// sp must be the last, because once async cpu profiler finds
 		// all three values to be non-zero, it will use them
 		mp.libcallsp = getcallersp()
+	} else {
+		mp = nil
 	}
 
 	var libcall libcall
@@ -119,12 +125,14 @@ func sysvicall3(fn *libcFunc, a1, a2, a3 uintptr) uintptr {
 	if gp != nil {
 		mp = gp.m
 	}
-	if mp != nil {
+	if mp != nil && mp.libcallsp == 0 {
 		mp.libcallg.set(gp)
 		mp.libcallpc = getcallerpc()
 		// sp must be the last, because once async cpu profiler finds
 		// all three values to be non-zero, it will use them
 		mp.libcallsp = getcallersp()
+	} else {
+		mp = nil
 	}
 
 	var libcall libcall
@@ -146,12 +154,14 @@ func sysvicall4(fn *libcFunc, a1, a2, a3, a4 uintptr) uintptr {
 	if gp != nil {
 		mp = gp.m
 	}
-	if mp != nil {
+	if mp != nil && mp.libcallsp == 0 {
 		mp.libcallg.set(gp)
 		mp.libcallpc = getcallerpc()
 		// sp must be the last, because once async cpu profiler finds
 		// all three values to be non-zero, it will use them
 		mp.libcallsp = getcallersp()
+	} else {
+		mp = nil
 	}
 
 	var libcall libcall
@@ -173,12 +183,14 @@ func sysvicall5(fn *libcFunc, a1, a2, a3, a4, a5 uintptr) uintptr {
 	if gp != nil {
 		mp = gp.m
 	}
-	if mp != nil {
+	if mp != nil && mp.libcallsp == 0 {
 		mp.libcallg.set(gp)
 		mp.libcallpc = getcallerpc()
 		// sp must be the last, because once async cpu profiler finds
 		// all three values to be non-zero, it will use them
 		mp.libcallsp = getcallersp()
+	} else {
+		mp = nil
 	}
 
 	var libcall libcall
@@ -200,12 +212,14 @@ func sysvicall6(fn *libcFunc, a1, a2, a3, a4, a5, a6 uintptr) uintptr {
 	if gp != nil {
 		mp = gp.m
 	}
-	if mp != nil {
+	if mp != nil && mp.libcallsp == 0 {
 		mp.libcallg.set(gp)
 		mp.libcallpc = getcallerpc()
 		// sp must be the last, because once async cpu profiler finds
 		// all three values to be non-zero, it will use them
 		mp.libcallsp = getcallersp()
+	} else {
+		mp = nil
 	}
 
 	var libcall libcall
diff --git a/src/runtime/os_windows.go b/src/runtime/os_windows.go
index 6180dd3a60..5607bf95c1 100644
--- a/src/runtime/os_windows.go
+++ b/src/runtime/os_windows.go
@@ -734,17 +734,20 @@ func stdcall(fn stdFunction) uintptr {
 	gp := getg()
 	mp := gp.m
 	mp.libcall.fn = uintptr(unsafe.Pointer(fn))
-
-	if mp.profilehz != 0 {
+	resetLibcall := false
+	if mp.profilehz != 0 && mp.libcallsp == 0 {
 		// leave pc/sp for cpu profiler
 		mp.libcallg.set(gp)
 		mp.libcallpc = getcallerpc()
 		// sp must be the last, because once async cpu profiler finds
 		// all three values to be non-zero, it will use them
 		mp.libcallsp = getcallersp()
+		resetLibcall = true // See comment in sys_darwin.go:libcCall
 	}
 	asmcgocall(asmstdcallAddr, unsafe.Pointer(&mp.libcall))
-	mp.libcallsp = 0
+	if resetLibcall {
+		mp.libcallsp = 0
+	}
 	return mp.libcall.r1
 }
 
diff --git a/src/runtime/sys_darwin.go b/src/runtime/sys_darwin.go
index f0d0815903..7efbef746c 100644
--- a/src/runtime/sys_darwin.go
+++ b/src/runtime/sys_darwin.go
@@ -18,12 +18,30 @@ func libcCall(fn, arg unsafe.Pointer) int32 {
 	if gp != nil {
 		mp = gp.m
 	}
-	if mp != nil {
+	if mp != nil && mp.libcallsp == 0 {
 		mp.libcallg.set(gp)
 		mp.libcallpc = getcallerpc()
 		// sp must be the last, because once async cpu profiler finds
 		// all three values to be non-zero, it will use them
 		mp.libcallsp = getcallersp()
+	} else {
+		// Make sure we don't reset libcallsp. This makes
+		// libcCall reentrant; We remember the g/pc/sp for the
+		// first call on an M, until that libcCall instance
+		// returns.  Reentrance only matters for signals, as
+		// libc never calls back into Go.  The tricky case is
+		// where we call libcX from an M and record g/pc/sp.
+		// Before that call returns, a signal arrives on the
+		// same M and the signal handling code calls another
+		// libc function.  We don't want that second libcCall
+		// from within the handler to be recorded, and we
+		// don't want that call's completion to zero
+		// libcallsp.
+		// We don't need to set libcall* while we're in a sighandler
+		// (even if we're not currently in libc) because we block all
+		// signals while we're handling a signal. That includes the
+		// profile signal, which is the one that uses the libcall* info.
+		mp = nil
 	}
 	res := asmcgocall(fn, arg)
 	if mp != nil {
diff --git a/src/runtime/traceback.go b/src/runtime/traceback.go
index 4953653900..d8c225d975 100644
--- a/src/runtime/traceback.go
+++ b/src/runtime/traceback.go
@@ -679,7 +679,14 @@ func traceback(pc, sp, lr uintptr, gp *g) {
 // the initial PC must not be rewound to the previous instruction.
 // (All the saved pairs record a PC that is a return address, so we
 // rewind it into the CALL instruction.)
+// If gp.m.libcall{g,pc,sp} information is available, it uses that information in preference to
+// the pc/sp/lr passed in.
 func tracebacktrap(pc, sp, lr uintptr, gp *g) {
+	if gp.m.libcallsp != 0 {
+		// We're in C code somewhere, traceback from the saved position.
+		traceback1(gp.m.libcallpc, gp.m.libcallsp, 0, gp.m.libcallg.ptr(), 0)
+		return
+	}
 	traceback1(pc, sp, lr, gp, _TraceTrap)
 }