The following patch fixes two things: First, DEVICE_SUSPEND errors are
no longer ignored. Since we have defaults for methods we should no
longer ignore these errors. Also, DEVICE_RESUME wasn't done when the
apm_suspend_system event failed.
Second, in the ordering of DEVICE_* and apm_hook_* is not symmetric for
suspend and resume. The patch reverses the resume case as well to be
apm_hook_resume(...)
DEVICE_RESUME(root)
Send me some feedback on this. I've mailed the people mentioned in the
head of the file, but so far got no response. I would like to make sure
we get this right when USB devices are used. Some of them don't suspend
and the suspend should be refused in that case.
Nick
--
[EMAIL PROTECTED]
[EMAIL PROTECTED] USB project
http://www.etla.net/~n_hibma/
--- /usr/src/sys/i386/apm/apm.c Thu Dec 2 03:13:00 1999
+++ /heidi/usr/current/src/sys/i386/apm/apm.c Sun Jan 23 14:07:49 2000
@@ -473,19 +473,19 @@
if (sc->initialized) {
error = DEVICE_SUSPEND(root_bus);
- /*
- * XXX Shouldn't ignore the error like this, but should
- * instead fix the newbus code. Until that happens,
- * I'm doing this to get suspend working again.
- */
- if (error)
- printf("DEVICE_SUSPEND error %d, ignored\n", error);
- apm_execute_hook(hook[APM_HOOK_SUSPEND]);
- if (apm_suspend_system(PMST_SUSPEND) == 0)
- apm_processevent();
- else
- /* Failure, 'resume' the system again */
- apm_execute_hook(hook[APM_HOOK_RESUME]);
+ if (error) {
+ printf("Suspend failed, error %d\n", error);
+ DEVICE_RESUME(root_bus);
+ } else {
+ apm_execute_hook(hook[APM_HOOK_SUSPEND]);
+ if (apm_suspend_system(PMST_SUSPEND) == 0)
+ apm_processevent();
+ else {
+ /* Failure, 'resume' the system again */
+ apm_execute_hook(hook[APM_HOOK_RESUME]);
+ DEVICE_RESUME(root_bus);
+ }
+ }
}
}
@@ -593,8 +593,8 @@
return;
if (sc->initialized) {
- DEVICE_RESUME(root_bus);
apm_execute_hook(hook[APM_HOOK_RESUME]);
+ DEVICE_RESUME(root_bus);
}
}
