Bug 191847

Summary: REGRESSION: kernel-2.6.9.36 does not boot on ALTIX systems
Product: Red Hat Enterprise Linux 4 Reporter: Prarit Bhargava <prarit>
Component: kernelAssignee: Prarit Bhargava <prarit>
Status: CLOSED ERRATA QA Contact: Brian Brock <bbrock>
Severity: high Docs Contact:
Priority: urgent    
Version: 4.0CC: akeshava, jbaron
Target Milestone: ---Keywords: Regression
Target Release: ---   
Hardware: ia64   
OS: Linux   
Whiteboard:
Fixed In Version: RHSA-2006-0575 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2006-08-10 23:18:47 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 181409    
Attachments:
Description Flags
Patch to replace panic with printk none

Description Prarit Bhargava 2006-05-16 01:58:02 UTC 
Description of problem:

kernel-2.6.9.36 fails due to panic in MADT ACPI table code.

Version-Release number of selected component (if applicable):
kernel-2.6.9.36

How reproducible: 100%


Steps to Reproduce:
1. Boot kernel 2.6.9-36
2.
3.
  
Actual results:

Loading initrd initrd-2.6.9-36.EL.img...done
Linux version 2.6.9-36.EL (bhcompile.redhat.com) (gcc version
3.4.5 20051201 (Red Hat 3.4.5-2)) #1 SMP Tue May 9 09:06:25 EDT 2006
EFI v1.10 by INTEL: SALsystab=0x3002814b60 ACPI 2.0=0x300281b8e0
booting generic kernel on platform sn2
Number of logical nodes in system = 32
Number of memory chunks in system = 32
Initial ramdisk at: 0xe0000fb07a0ba000 (3331564 bytes)
SAL 2.9: SGI SN2 version 4.43
SAL Platform features: ITC_Drift
SAL: AP wakeup using external interrupt vector 0x12
No logical to physical processor mapping available
ACPI: Local APIC address c0000000fee00000
ACPI: Error parsing MADT - no IOSAPIC entries
register_intr: No IOSAPIC for GSI 52
Kernel panic - not syncing: acpi_register_gsi: out of interrupt vectors!

Expected results:

Boot should succeed.
Additional info: Binary search through patches indicates Anil's
acpi_register_gsi patch is the culprit.  Anil modified the code such that
acpi_register_gsi panic's if iosapic_register_intr returns < 0.  However, the
MADT code that registers the SCI interrupt can (and does handle) a return < 0.

Anil's patch should not panic if < 0 -- just output a warning?
Comment 2 Prarit Bhargava 2006-05-16 13:33:17 UTC 
Anil,

This statement in your patch is wrong:

+       /* Caller of acpi_register_gsi() assumes this function always succeeds,
+        * hence calling panic() for backward compatibility

IMO, calling panic() to maintain backward compatibility is incorrect here as this
will break all systems with bad ACPI tables (of which there are many).
Comment 4 Anil S Keshavamurthy 2006-05-16 17:16:59 UTC 
Ha..good that I had a panic() call which caught the buggy ACPI table bug:) 
I am okay with you replacing panic() call with printk.
Comment 5 Prarit Bhargava 2006-05-16 17:18:02 UTC 
Created attachment 129237 [details]
Patch to replace panic with printk

Patch to replace panic with printk
Comment 6 Jay Turner 2006-05-16 19:15:02 UTC 
Regression in behavior introduced in U4 code development.  Not a beta blocker
and would agree to taking during beta since it's just a behavior change.
Comment 7 Daniel Riek 2006-05-17 21:05:16 UTC 
PM ACK for U4
Comment 9 Jason Baron 2006-05-22 19:21:54 UTC 
committed in stream U4 build 36.1. A test kernel with this patch is available
from http://people.redhat.com/~jbaron/rhel4/
Comment 13 Red Hat Bugzilla 2006-08-10 23:18:50 UTC 
An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on the solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.

http://rhn.redhat.com/errata/RHSA-2006-0575.html