public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* Re: panic during IO-APIC detection
  2005-02-16 16:26 panic during IO-APIC detection Andreas Deresch
@ 2005-02-15 20:12 ` Andi Kleen
  0 siblings, 0 replies; 2+ messages in thread
From: Andi Kleen @ 2005-02-15 20:12 UTC (permalink / raw)
  To: Andreas Deresch; +Cc: linux-kernel

Andreas Deresch <aderesch@fs.tum.de> writes:
>
> The question is, should io_apic_get_unique_id be more fault tolerant, or
> where else should this be fixed?

Patch makes sense in theory, but I suspect there is more state to fix
up to get rid of the IO-APIC than just decreasing nr_ioapics.

-Andi


^ permalink raw reply	[flat|nested] 2+ messages in thread

* panic during IO-APIC detection
@ 2005-02-16 16:26 Andreas Deresch
  2005-02-15 20:12 ` Andi Kleen
  0 siblings, 1 reply; 2+ messages in thread
From: Andreas Deresch @ 2005-02-16 16:26 UTC (permalink / raw)
  To: linux-kernel

Hi lkml,

my notebook (Acer Aspire 1691WLMi, chipset: Intel 915PM) hangs during
IO-APIC detection. I am currently using 2.6.11-rc4-bk1, but this should be
the same for other versions.

Reason:
ACPI MADT contains info on 2 IO-APICs.
io_apic_get_unique_id panics due to being unable to assign an id to the
second one (the memory location reads as all ones). The chipset spec
(specifically: ICH6-M) is only talking about one IO-APIC, so most likely the
MADT is in error.

The question is, should io_apic_get_unique_id be more fault tolerant, or
where else should this be fixed?

TIA,
ad

For the moment I did the following:

diff -rbBu linux-2.6.11.orig/arch/i386/kernel/io_apic.c linux-2.6.11/arch/i386/kernel/io_apic.c
--- linux-2.6.11.orig/arch/i386/kernel/io_apic.c	2005-02-15 20:50:11.000000000 +0100
+++ linux-2.6.11/arch/i386/kernel/io_apic.c	2005-02-14 23:49:04.000000000 +0100
@@ -2449,8 +2449,10 @@
  		spin_unlock_irqrestore(&ioapic_lock, flags);

  		/* Sanity check */
-		if (reg_00.bits.ID != apic_id)
-			panic("IOAPIC[%d]: Unable change apic_id!\n", ioapic);
+		if (reg_00.bits.ID != apic_id) {
+			printk("IOAPIC[%d]: Unable change apic_id!\n", ioapic);
+			return -1;
+		}
  	}

  	apic_printk(APIC_VERBOSE, KERN_INFO
diff -rbBu linux-2.6.11.orig/arch/i386/kernel/mpparse.c linux-2.6.11/arch/i386/kernel/mpparse.c
--- linux-2.6.11.orig/arch/i386/kernel/mpparse.c	2005-02-15 20:50:11.000000000 +0100
+++ linux-2.6.11/arch/i386/kernel/mpparse.c	2005-02-14 23:49:23.000000000 +0100
@@ -895,6 +895,7 @@
  	u32			gsi_base)
  {
  	int			idx = 0;
+	int			tmpid = 0;

  	if (nr_ioapics >= MAX_IO_APICS) {
  		printk(KERN_ERR "ERROR: Max # of I/O APICs (%d) exceeded "
@@ -914,7 +915,11 @@
  	mp_ioapics[idx].mpc_apicaddr = address;

  	set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address);
-	mp_ioapics[idx].mpc_apicid = io_apic_get_unique_id(idx, id);
+	if ((tmpid = io_apic_get_unique_id(idx, id)) == -1) {
+		nr_ioapics--;
+		return;
+	}
+	mp_ioapics[idx].mpc_apicid = (u8)(tmpid);
  	mp_ioapics[idx].mpc_apicver = io_apic_get_version(idx);

  	/*

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2005-02-15 20:21 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-02-16 16:26 panic during IO-APIC detection Andreas Deresch
2005-02-15 20:12 ` Andi Kleen

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox