From 1adc1af6d9b52ad33f0b773cfdcd64573a2b7b9e Mon Sep 17 00:00:00 2001
From: Peter Bex <peter@more-magic.net>
Date: Sat, 18 Nov 2017 13:58:13 +0100
Subject: Run signal handler in a separate thread

This seems to be necessary so that condition-variable-broadcast! works
when the signal handler is running in the same thread as the waiting
thread.
---
 scsh-process.scm | 35 ++++++++++++++++++++---------------
 tests/run.scm    |  9 ++++++---
 2 files changed, 26 insertions(+), 18 deletions(-)

diff --git a/scsh-process.scm b/scsh-process.scm
index 1a6c22b..fd7ff3f 100644
--- a/scsh-process.scm
+++ b/scsh-process.scm
@@ -177,21 +177,26 @@
                   (else (syntax-rules () ((_ val) val))))))
      (let ((old-handler (workaround (signal-handler signal/chld))))
        (lambda (signal)
-         (for-each (lambda (pid)
-                     (handle-exceptions exn
-                       ;; User might have waited manually
-                         (begin (remove-scsh-pending-process! pid) (void))
-                       (receive (pid ok? status)
-                         (posix-process-wait pid #t)
-                         (unless (zero? pid)
-                           (let ((p (hash-table-ref *scsh-pending-processes* pid)))
-                             (scsh-process-exit-status-set! p status)
-                             (scsh-process-ok?-set! p ok?)
-			     (condition-variable-broadcast!
-                              (scsh-process-child-condition p))
-                             ;; The GC can clean it up
-                             (remove-scsh-pending-process! pid))))))
-                   (hash-table-keys *scsh-pending-processes*))
+         ;; Run the signal handler in another thread. This is needed
+         ;; because the current thread may be waiting on a condition
+         ;; variable, and we can't wake ourselves up.
+         (thread-start!
+          (lambda ()
+            (for-each (lambda (pid)
+                        (handle-exceptions exn
+                            ;; User might have waited manually
+                            (begin (remove-scsh-pending-process! pid) (void))
+                          (receive (pid ok? status)
+                              (posix-process-wait pid #t)
+                            (unless (zero? pid)
+                              (let ((p (hash-table-ref *scsh-pending-processes* pid)))
+                                (scsh-process-exit-status-set! p status)
+                                (scsh-process-ok?-set! p ok?)
+                                (condition-variable-broadcast!
+                                 (scsh-process-child-condition p))
+                                ;; The GC can clean it up
+                                (remove-scsh-pending-process! pid))))))
+                      (hash-table-keys *scsh-pending-processes*))))
          (when old-handler (old-handler signal)))))))
 
 (define (signal-process proc sig)
diff --git a/tests/run.scm b/tests/run.scm
index f05d643..7690752 100644
--- a/tests/run.scm
+++ b/tests/run.scm
@@ -18,9 +18,12 @@
         "This is a test"
         (run/string* (lambda () (display "This is a test"))))
 
-  ;; We must mask sigchld, because otherwise our (wait #f) will fail
-  ;; due to scsh-process' signal handler possibly reaping the child
-  ;; before our wait is able to do so.
+  ;; Ensure all processes up to here have been reaped
+  (handle-exceptions exn (void) (let lp () (when (wait #f) (lp))))
+
+  ;; We must mask sigchld, because otherwise our next (wait #f) will
+  ;; fail due to scsh-process' signal handler possibly reaping the
+  ;; child before our wait is able to do so.
   (signal-mask! signal/chld)
 
   (test "wait for next process to exit"
-- 
cgit v1.2.3