proposal: stop_at_err_count patch

[Martin d Anjou]

Hello GNU make gurus,

Instead of the patch proposed here:
http://lists.gnu.org/archive/html/bug-make/2009-01/msg00007.html

I would like to propose the patch below instead. The difference is that I 
renamed keep_going_count to stop_at_err_count which I think is more 
descriptive of the intent. I have also added a hunk in job.c which forces 
GNU Make to return immediately when the number of errors is reached.

However, there is one problem which manifest itself when using parallel 
make, and I need help to solve it (I am not familiar with GNU Make
code). Here is a description of the problem (the patch is further down
below).

Given this makefile:

all: t1 t2
t1:
        sleep 30 && echo done || echo failed
t2:
        sleep 10 && exit 1

and this command: ../make-3.81/make -j 2 --stop-at-err-count=1

The problem is that --stop-at-err-count=1 forces make to return but
unfinished children are not killed by make, as you can see in this
sample run:

$ ../make-3.81/make -j 2 --stop-at-err-count=1
sleep 30 && echo done || echo failed
sleep 10 && exit 1
make: *** [t2] Error 1
make: *** Waiting for unfinished jobs....
make: *** Aborting after 1 failures
$ done

$ echo $?
2
$

I examined the processes and yes in fact make launches both in parallel:

boo 22255  2955  0 08:42 pts/38   00:00:00 ../make-3.81/make -j 2 
--stop-at-err-count=1
boo 22256 22255  0 08:42 pts/38   00:00:00   /bin/sh -c sleep 30 && echo done 
|| echo failed
boo 22257 22256  0 08:42 pts/38   00:00:00     sleep 30
boo 22258 22255  0 08:42 pts/38   00:00:00   /bin/sh -c sleep 10 && exit 1
boo 22259 22258  0 08:42 pts/38   00:00:00     sleep 10

and when 22258 exits, GNU Make also exits, but 22256 keeps running. I
want all children to be killed before GNU Make returns. How do I modify
the proposed patch to achieve that?

I would really appreciate feedback on this proposal and on how to fix the 
aforementionned problem.

Regards,
Martin d'Anjou

Here is the said patch, it should apply cleanly to 3.81.

----------------------------

diff -u ./job.c /home/boo/local/SRC/make-3.81a/job.c
--- ./job.c     2006-03-19 22:03:04.000000000 -0500
+++ /home/boo/local/SRC/make-3.81a/job.c        2009-01-09 17:31:09.384897000 
-0500
@@ -379,9 +379,19 @@
 child_error (char *target_name, int exit_code, int exit_sig, int coredump,
              int ignored)
 {
+  if (max_stop_at_err_count <= stop_at_err_count)
+    {
+      error (NILF, _("*** [%s] Error (aborted after %d failures) %d"),
+            target_name, stop_at_err_count, exit_code);
+      die (2);
+    }
+
   if (ignored && silent_flag)
     return;

+  /* Only increase STOP_AT_ERR_COUNT if the job _really_ failed.  */
+  stop_at_err_count++;
+
 #ifdef VMS
   if (!(exit_code & 1))
       error (NILF,
@@ -485,7 +495,14 @@
              Only print this message once no matter how many jobs are left.  */
          fflush (stdout);
           if (!printed)
+          {
             error (NILF, _("*** Waiting for unfinished jobs...."));
+            if (max_stop_at_err_count <= stop_at_err_count)
+            {
+              error (NILF, _("*** Aborting after %d 
failures"),stop_at_err_count);
+              die (2);
+            }
+          }
           printed = 1;
        }

diff -u ./main.c /home/boo/local/SRC/make-3.81a/main.c
--- ./main.c    2006-03-19 21:36:37.000000000 -0500
+++ /home/boo/local/SRC/make-3.81a/main.c       2009-01-09 17:31:02.674041000 
-0500
@@ -189,9 +189,13 @@

 /* Nonzero means keep going even if remaking some file fails (-k).  */

-int keep_going_flag;
+int keep_going_flag = 0;
 int default_keep_going_flag = 0;

+int stop_at_err_count = 0;
+unsigned int max_stop_at_err_count = -1;
+static unsigned int inf_stop_at_err_count = 0;
+
 /* Nonzero means check symlink mtimes.  */

 int check_symlink_flag = 0;
@@ -342,6 +346,8 @@
   -S, --no-keep-going, --stop\n\
                               Turns off -k.\n"),
     N_("\
+      --stop-at-err-count[=N] Stop when N errors are reached. Default 0 
(infinite).\n"),
+    N_("\
   -t, --touch                 Touch targets instead of remaking them.\n"),
     N_("\
   -v, --version               Print the version number of make and exit.\n"),
@@ -416,6 +422,14 @@
     { 'W', string, (char *) &new_files, 0, 0, 0, 0, 0, "what-if" },
     { CHAR_MAX+4, flag, (char *) &warn_undefined_variables_flag, 1, 1, 0, 0, 0,
       "warn-undefined-variables" },
+    /* BUG: There is a bug in the argument parser where if you have a
+       long-optin, but no corresponding short option in the ASCII
+       range, the output will produce a charachter that can corrupt
+       the terminal.
+    */
+    { CHAR_MAX+5, positive_int, (char *) &max_stop_at_err_count, 1, 1, 0,
+      (char *) &inf_stop_at_err_count,
+      (char *) &max_stop_at_err_count, "stop-at-err-count" },
     { 0, 0, 0, 0, 0, 0, 0, 0, 0 }
   };

diff -u ./make.h /home/boo/local/SRC/make-3.81a/make.h
--- ./make.h    2006-02-15 18:54:43.000000000 -0500
+++ /home/boo/local/SRC/make-3.81a/make.h       2009-01-09 17:36:55.992563000 
-0500
@@ -499,6 +499,9 @@
 extern const struct floc *reading_file;
 extern const struct floc **expanding_var;

+extern int stop_at_err_count;
+extern unsigned int max_stop_at_err_count;
+
 extern char **environ;

 extern int just_print_flag, silent_flag, ignore_errors_flag, keep_going_flag;
diff -u ./remake.c /home/boo/local/SRC/make-3.81a/remake.c
--- ./remake.c  2006-03-19 21:36:37.000000000 -0500
+++ /home/boo/local/SRC/make-3.81a/remake.c     2008-09-04 08:12:17.000000000 
-0400
@@ -311,7 +311,7 @@
       alloca (0);

       /* If we got an error, don't bother with double_colon etc.  */
-      if (status != 0 && !keep_going_flag)
+      if (status != 0 && keep_going_flag == 1)
        return status;

       if (f->command_state == cs_running
@@ -350,7 +350,7 @@
   const char *msg_parent
     = _("%sNo rule to make target `%s', needed by `%s'%s");

-  if (!keep_going_flag)
+  if (keep_going_flag == 0)
     {
       if (file->parent == 0)
         fatal (NILF, msg_noparent, "", file->name, "");
@@ -529,7 +529,7 @@
        while (f != 0);
       }

-      if (dep_status != 0 && !keep_going_flag)
+      if (dep_status != 0 && keep_going_flag == 0)
        break;

       if (!running)
@@ -587,7 +587,7 @@
              while (f != 0);
            }

-           if (dep_status != 0 && !keep_going_flag)
+           if (dep_status != 0 && keep_going_flag == 0)
              break;

            if (!running)
@@ -619,7 +619,7 @@

       DBF (DB_VERBOSE, _("Giving up on target file `%s'.\n"));

-      if (depth == 0 && keep_going_flag
+      if (depth == 0 && keep_going_flag == 1
          && !just_print_flag && !question_flag)
        error (NILF,
                _("Target `%s' not remade because of errors."), file->name);
@@ -1017,7 +1017,7 @@
               if (! d->ignore_mtime)
                 *must_make_ptr = maybe_make;
              check_renamed (d->file);
-             if (dep_status != 0 && !keep_going_flag)
+             if (dep_status != 0 && keep_going_flag == 0)
                break;

              if (d->file->command_state == cs_running