On Fri, Mar 16, 2012 at 10:33:15AM +0100, Igor Mammedov wrote:
Reduce amount of consumed cpu time (i.e. don't spin forever) if number of started cpus are more then expected. And print a bug message into debug port.
And it is not safe to continue to boot if CountCPUs is more than cmos_smp_count and might be still racing, due to mp/acpi could be build with still changing CountCPUs. It is safer to stop and allow user to notice and restart system rather than debug possible inxplicable bugs later.
I don't understand the gain in this. The CPU is spinning on the count, so (unless I missed something) there's a very good chance that this check wont trigger even if the emulator sets the counts wrong. It seems like we're just adding a random halt. Why bother?
[...]
--- a/src/smp.c +++ b/src/smp.c @@ -115,8 +115,14 @@ smp_probe(void) msleep(10); } else { u8 cmos_smp_count = inb_cmos(CMOS_BIOS_SMP_COUNT);
while (cmos_smp_count + 1 != readl(&CountCPUs))
while (cmos_smp_count + 1 != readl(&CountCPUs)) {if (cmos_smp_count + 1 < readl(&CountCPUs)) {dprintf(1, "BUG: Expected %d cpu(s) but %d cpus started\n",cmos_smp_count + 1, readl(&CountCPUs));hlt();
The panic() call should be used instead of dprintf()/hlt().
-Kevin