With multi-socket systems becoming more prevalent, and the continued
increase in cores per processors, I thought it would be nice for
'make -j' to gain some automation.
Attached is a patch that makes "-j-" be the same as
"-j `sysctl -n kern.smp.cpus`" and "-j=" be twice that.
I've also thought that maybe just supporting "-j-" would be better - with
a definition of
num_core = `sysctl -n kern.smp.cpus`
-j => MAX(num_core * 5 / 4, num_core + 1)
the idea being one would want a few more jobs than cores, but not a whole
lot more.
comments? (redirected back to list)
--
-- David ([EMAIL PROTECTED])
Index: main.c
===================================================================
RCS file: /home/ncvs/src/usr.bin/make/main.c,v
retrieving revision 1.160
diff -u -p -r1.160 main.c
--- main.c 17 Jul 2006 19:16:12 -0000 1.160
+++ main.c 14 Dec 2006 02:26:15 -0000
@@ -456,11 +456,20 @@ rearg:
char *endptr;
forceJobs = TRUE;
+ size_t jLlen = sizeof(jobLimit);
jobLimit = strtol(optarg, &endptr, 10);
- if (jobLimit <= 0 || *endptr != '\0') {
- warnx("illegal number, -j argument -- %s",
- optarg);
- usage();
+ if ((*optarg == '-' || *optarg == '=') &&
+ *endptr != '\0') {
+ sysctlbyname("kern.smp.cpus", &jobLimit, &jLlen,
+ NULL, 0);
+ if (*optarg == '=')
+ jobLimit *= 2;
+ } else {
+ if (jobLimit <= 0 || *endptr != '\0') {
+ warnx("illegal number, -j argument -- %s",
+ optarg);
+ usage();
+ }
}
MFLAGS_append("-j", optarg);
break;
Index: make.1
===================================================================
RCS file: /home/ncvs/src/usr.bin/make/make.1,v
retrieving revision 1.99
diff -u -p -r1.99 make.1
--- make.1 29 Sep 2006 21:17:10 -0000 1.99
+++ make.1 14 Dec 2006 04:19:22 -0000
@@ -218,6 +218,14 @@ may have running at any one time.
Turns compatibility mode off, unless the
.Fl B
flag is also specified.
+The special values
+.It Ar -
+and
+.It Ar =
+causes
+.It Ar max_jobs
+to be set to the value returned from the kern.smp.cpus sysctl and twice
+kern.smp.cpus respectively.
.It Fl k
Continue processing after errors are encountered, but only on those targets
that do not depend on the target whose creation caused the error.
_______________________________________________
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
