On Sat, Mar 20, 2021 at 07:46:38PM +0100, Klemens Nanni wrote:
> Otherwise there is no way other than reading driver code to determine
> why running zzz(8) for example does not do anything on certain machines.
>
> apm(4/arm64) for one currently does not implement suspend and resume,
> i.e. it yields ENOSUPP which gets lost in userland.
>
> This still does not make `zzz' or `apm -z' report the informative
> warning, but syslog now has it:
>
> $ zzz
> Suspending system...
> $ tail -n3 /var/log/messages
> Mar 20 19:16:57 pine64 apmd: system suspending
> Mar 20 19:16:57 pine64 apmd: battery status: unknown. external power
> status: not known. estimated battery life 0%
> Mar 20 19:16:58 pine64 apmd: suspend: Operation not supported
>
>
> Feedback? OK?
Anyone?
This is a trivial but helpful change, just like the ones in
these other mails on tech@:
"apm/macppc: Mark unsupported ioctl as such, merge cases"
"apm/arm64: fix errno, merge ioctl cases"
Index: apmd.c
===================================================================
RCS file: /cvs/src/usr.sbin/apmd/apmd.c,v
retrieving revision 1.101
diff -u -p -r1.101 apmd.c
--- apmd.c 16 Mar 2021 09:00:43 -0000 1.101
+++ apmd.c 20 Mar 2021 18:47:13 -0000
@@ -329,7 +329,8 @@ suspend(int ctl_fd)
do_etc_file(_PATH_APM_ETC_SUSPEND);
sync();
sleep(1);
- ioctl(ctl_fd, APM_IOC_SUSPEND, 0);
+ if (ioctl(ctl_fd, APM_IOC_SUSPEND, 0) == -1)
+ logmsg(LOG_WARNING, "%s: %s", __func__, strerror(errno));
}
void
@@ -340,7 +341,8 @@ stand_by(int ctl_fd)
do_etc_file(_PATH_APM_ETC_STANDBY);
sync();
sleep(1);
- ioctl(ctl_fd, APM_IOC_STANDBY, 0);
+ if (ioctl(ctl_fd, APM_IOC_STANDBY, 0) == -1)
+ logmsg(LOG_WARNING, "%s: %s", __func__, strerror(errno));
}
void
@@ -351,7 +353,8 @@ hibernate(int ctl_fd)
do_etc_file(_PATH_APM_ETC_HIBERNATE);
sync();
sleep(1);
- ioctl(ctl_fd, APM_IOC_HIBERNATE, 0);
+ if (ioctl(ctl_fd, APM_IOC_HIBERNATE, 0) == -1)
+ logmsg(LOG_WARNING, "%s: %s", __func__, strerror(errno));
}
void
