Date: Thu, 10 Nov 2005 08:53:30 -0500 (EST) From: Keith White <Keith.White@site.uottawa.ca> To: FreeBSD-gnats-submit@FreeBSD.org Subject: bin/88788: [patch] bin/rup displays incorrect date and uptime on sparc64 platform Message-ID: <200511101353.jAADrUU1013725@grdsav.site.uottawa.ca> Resent-Message-ID: <200511101400.jAAE0OMT031760@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
>Number: 88788
>Category: bin
>Synopsis: [patch] bin/rup displays incorrect date and uptime on sparc64 platform
>Confidential: no
>Severity: non-critical
>Priority: low
>Responsible: freebsd-bugs
>State: open
>Quarter:
>Keywords:
>Date-Required:
>Class: sw-bug
>Submitter-Id: current-users
>Arrival-Date: Thu Nov 10 14:00:24 GMT 2005
>Closed-Date:
>Last-Modified:
>Originator: Keith White
>Release: 6.0-RELEASE sparc64
>Organization:
SITE, University of Ottawa
>Environment:
System: FreeBSD grdsav.site.uottawa.ca 6.0-RELEASE FreeBSD 6.0-RELEASE #1: Wed Nov 9 10:27:29 EST 2005 root@grdsav.site.uottawa.ca:/usr/obj/usr/src/sys/ULTRA5 sparc64
>Description:
"rup" displays incorrect date and uptime when run on the sparc64 platform
(and presumably on other 64-bit time_t platforms).
The RPC rstat network packet uses 32-bit time_t. src/usr.bin/rup.c
uses gmtime() and localtime() to decode the time_t values. On the
sparc64 platform these functions expect a 64-bit time_t so the
returned values displayed by rup are incorrect.
>How-To-Repeat:
$ uname -rp
6.0-RELEASE sparc64
$ rup
a.site.uottawa. 2:43pm up 0 day, 54 mins, load average: 0.00 0.09 0.16
b.site.uottawa. 5:03am up 41 days, 2:37, load average: 0.01 0.01 0.02
c.site.uottawa. 11:30pm up 117 days, 15:01, load average: 0.00 0.00 0.00
d.site.uottawa. 9:28pm up 237 days, 20:41, load average: 0.00 0.00 0.01
e.site.uottawa. 10:51am up 18 days, 20:52, load average: 0.02 0.01 0.02
f.site.uottawa. 8:11pm up 238 days, 8:43, load average: 0.90 0.41 0.12
g.site.uottawa. 1:21am up 21 days, 6:45, load average: 0.01 0.01 0.02
h.site.uottawa. 10:26pm up 274 days, 17:18, load average: 0.00 0.00 0.00
i.site.uottawa. 4:45am up 37 days,39 mins, load average: 0.02 0.00 0.00
>Fix:
The following patch fixes the problem on sparc64 platforms.
Tested on both sparc64 (6.0-RELEASE) and i386 (6.0-RELEASE).
--- rup.patch begins here ---
--- src/usr.bin/rup/rup.c.orig Sat May 21 05:55:07 2005
+++ src/usr.bin/rup/rup.c Wed Nov 9 14:16:37 2005
@@ -101,6 +101,7 @@
struct hostent *hp;
char *host;
statstime *host_stat = (statstime *)replyp;
+ time_t tmp_time_t;
if (search_host(raddrp->sin_addr))
return(0);
@@ -118,13 +119,26 @@
printf("%-*s\t", HOST_WIDTH, host);
- tmp_time = localtime((time_t *)&host_stat->curtime.tv_sec);
- host_time = *tmp_time;
-
- host_stat->curtime.tv_sec -= host_stat->boottime.tv_sec;
-
- tmp_time = gmtime((time_t *)&host_stat->curtime.tv_sec);
- host_uptime = *tmp_time;
+ if (sizeof(time_t) == sizeof(host_stat->curtime.tv_sec)) {
+ tmp_time = localtime((time_t *)&host_stat->curtime.tv_sec);
+ host_time = *tmp_time;
+
+ host_stat->curtime.tv_sec -= host_stat->boottime.tv_sec;
+
+ tmp_time = gmtime((time_t *)&host_stat->curtime.tv_sec);
+ host_uptime = *tmp_time;
+ }
+ else { /* non-32-bit time_t */
+ tmp_time_t = host_stat->curtime.tv_sec;
+ tmp_time = localtime(&tmp_time_t);
+ host_time = *tmp_time;
+
+ host_stat->curtime.tv_sec -= host_stat->boottime.tv_sec;
+
+ tmp_time_t = host_stat->curtime.tv_sec;
+ tmp_time = gmtime(&tmp_time_t);
+ host_uptime = *tmp_time;
+ }
#define updays (host_stat->curtime.tv_sec / 86400)
if (host_uptime.tm_yday != 0)
--- rup.patch ends here ---
>Release-Note:
>Audit-Trail:
>Unformatted:
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200511101353.jAADrUU1013725>