On Fri, Aug 01, 2025 at 01:09:07AM -0600, Theo de Raadt wrote:
> There are two issues being solved. One is blocking kernel behaviors
> during the time when the kernel is fully in control. The second issue
> is ignoring the user's requests. If the timer is simply extended
> once the kernel-resume parts have finished, we get a nice 10 second
> block on user interaction on all machines.
I see. It becomes a much simpler diff:
M sys/kern/subr_suspend.c | 2+ 1-
1 file changed, 2 insertions(+), 1 deletion(-)
commit - 3b565b6513502b4d00eb2ea3c0e8d1aa648f5db2
commit + 10825f0f1315cbb61913a0003512928e8419ae8a
blob - 3c83a049cb97e4b3026fd9c3b75d3b1eff2d3caa
blob + 08dd1e46c3fab68bdb8759bc6704af153d45ccf0
--- sys/kern/subr_suspend.c
+++ sys/kern/subr_suspend.c
@@ -218,11 +218,12 @@ fail_hiballoc:
cpu_setperf(perflevel); /* Restore hw.setperf */
if (suspend_finish(v) == EAGAIN)
goto top;
+ resume_time = getuptime();
return (error);
}
int
resuming(void)
{
- return (getuptime() < resume_time + 15);
+ return (getuptime() < resume_time + 10);
}
