From: Mike Waychison <mikew-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
To: acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
Subject: [PATCH] IO APIC existence check (2.6.11.10)
Date: Mon, 23 May 2005 15:23:12 -0700 [thread overview]
Message-ID: <429257D0.3020703@google.com> (raw)
[-- Attachment #1: Type: text/plain, Size: 587 bytes --]
Hi,
We've come across some boards where the MADT will list IO APICs that
don't really exist in hardware.
The current 2.6, the kernel will panic in io_apic_get_unique_id because
it will fail to force set the APIC id.
This following patch adds a quick check when registering IO APICs to
make sure they exist. We do this by actually checking to see if the data
register loads the version properly and by examining the index register
to see that it was actually modified from 0xff after the io_apic_read.
Signed-off-by: Mike Waychison <mikew-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
[-- Attachment #2: check_io_apic_existance.patch --]
[-- Type: text/plain, Size: 2316 bytes --]
--- 2.6/include/asm-i386/io_apic.h 2005-05-23 12:45:51.000000000 -0700
+++ 2.6/include/asm-i386/io_apic.h 2005-05-23 12:45:51.000000000 -0700
@@ -196,6 +196,7 @@ extern int skip_ioapic_setup;
#define io_apic_assign_pci_irqs (mp_irq_entries && !skip_ioapic_setup && io_apic_irqs)
#ifdef CONFIG_ACPI_BOOT
+extern unsigned char io_apic_read_index (int ioapic);
extern int io_apic_get_unique_id (int ioapic, int apic_id);
extern int io_apic_get_version (int ioapic);
extern int io_apic_get_redir_entries (int ioapic);
--- 2.6/arch/i386/kernel/mpparse.c 2005-05-23 13:12:35.000000000 -0700
+++ 2.6/arch/i386/kernel/mpparse.c 2005-05-23 13:12:35.000000000 -0700
@@ -907,16 +907,34 @@ void __init mp_register_ioapic (
return;
}
- idx = nr_ioapics++;
+ /*
+ * Probe the IO APIC. Set up the ioapic structures so we can do reads,
+ * but delay incrementing nr_ioapics in case the probe fails
+ */
+ idx = nr_ioapics;
mp_ioapics[idx].mpc_type = MP_IOAPIC;
mp_ioapics[idx].mpc_flags = MPC_APIC_USABLE;
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);
+
mp_ioapics[idx].mpc_apicver = io_apic_get_version(idx);
-
+ /*
+ * Probe to see if the io_apic looks good.
+ * index register should be set as well.
+ */
+ if (mp_ioapics[idx].mpc_apicver == 0xff) {
+ if (io_apic_read_index(idx) == 0xff) {
+ printk(KERN_ERR "WARNING: Bogus (0x%08x) I/O APIC address"
+ " found in MADT table, skipping!\n", address);
+ return;
+ }
+ }
+ nr_ioapics++;
+
+ mp_ioapics[idx].mpc_apicid = io_apic_get_unique_id(idx, id);
+
/*
* Build basic GSI lookup table to facilitate gsi->io_apic lookups
* and to prevent reprogramming of IOAPIC pins (PCI GSIs).
--- 2.6/arch/i386/kernel/io_apic.c 2005-05-23 12:48:32.000000000 -0700
+++ 2.6/arch/i386/kernel/io_apic.c 2005-05-23 12:48:32.000000000 -0700
@@ -2459,6 +2459,15 @@ int __init io_apic_get_unique_id (int io
return apic_id;
}
+unsigned char __init io_apic_read_index (int ioapic)
+{
+ unsigned long flags;
+ unsigned char index_reg;
+ spin_lock_irqsave(&ioapic_lock, flags);
+ index_reg = *IO_APIC_BASE(ioapic);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+ return index_reg;
+}
int __init io_apic_get_version (int ioapic)
{
reply other threads:[~2005-05-23 22:23 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=429257D0.3020703@google.com \
--to=mikew-hpiqsd4aklfqt0dzr+alfa@public.gmane.org \
--cc=acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox