On 8/10/22 00:42, Jonathan Gray wrote:
On Tue, Aug 09, 2022 at 10:29:41AM -0400, Nick Holland wrote:
On 8/9/22 09:36, Jonathan Gray wrote:
> On Sat, Aug 06, 2022 at 10:28:06AM -0400, Nick Holland wrote:
> > > When looking at top what does Xorg have for WAIT?
> >
> > PID USERNAME PRI NICE SIZE RES STATE WAIT TIME CPU COMMAND
> > 41914 _x11 -22 -1 15M 26M idle schto 0:01 0.00% Xorg
> >
> > > this will run X with the older Mesa driver
> > > MESA_LOADER_DRIVER_OVERRIDE=i965 startx
> > > does that change anything?
> >
> > Changes a little.
> > X doesn't work, same WAIT, same blank screen.
> > BUT... a "reboot" (via ssh) is actually successful(!).
>
> Do you still see that when disabling acceleration?
>
> before starting X create a /etc/X11/xorg.conf with:
>
> Section "Device"
> Identifier "Intel device"
> Driver "modesetting"
> Option "AccelMethod" "none"
> EndSection
WORKS!
Also works running xenodm. Probably not a surprise.
a long shot but does this change anything if you move away xorg.conf
No change from before -- black screen, system stays responsive remotely until
attempting to reboot it, at which point it hangs.
Nick.
errata described in
https://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/pentium-celeron-n-series-j-series-datasheet-spec-update.pdf
Index: sys/arch/amd64/amd64/cpu.c
===================================================================
RCS file: /cvs/src/sys/arch/amd64/amd64/cpu.c,v
retrieving revision 1.157
diff -u -p -r1.157 cpu.c
--- sys/arch/amd64/amd64/cpu.c 7 Aug 2022 23:56:06 -0000 1.157
+++ sys/arch/amd64/amd64/cpu.c 10 Aug 2022 04:04:14 -0000
@@ -465,6 +465,11 @@ cpu_init_mwait(struct cpu_softc *sc)
if ((cpu_ecxfeature & CPUIDECX_MWAIT) == 0 || cpuid_level < 0x5)
return;
+ /* APL30 A Store Instruction May Not Wake up MWAIT */
+ if (strcmp(cpu_vendor, "GenuineIntel") == 0 &&
+ curcpu()->ci_family == 6 && curcpu()->ci_model == 0x5c)
+ return;
+
/* get the monitor granularity */
CPUID(0x5, smallest, largest, extensions, cpu_mwait_states);
smallest &= 0xffff;
