Bug 242817 - [regression] kdump initialization always fails on x86_64
[regression] kdump initialization always fails on x86_64
Status: CLOSED DUPLICATE of bug 242648
Product: Red Hat Enterprise Linux 5
Classification: Red Hat
Component: kernel (Show other bugs)
5.1
x86_64 Linux
high Severity high
: ---
: ---
Assigned To: Red Hat Kernel Manager
Martin Jenner
:
Depends On:
Blocks:
  Show dependency treegraph
 
Reported: 2007-06-05 18:56 EDT by Jun'ichi Nomura (Red Hat)
Modified: 2013-04-02 19:51 EDT (History)
3 users (show)

See Also:
Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Environment:
Last Closed: 2007-06-07 10:27:28 EDT
Type: ---
Regression: ---
Mount Type: ---
Documentation: ---
CRM:
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---


Attachments (Terms of Use)
use end_pfn instead of uninitialized max_low_pfn (1.04 KB, patch)
2007-06-05 19:01 EDT, Jun'ichi Nomura (Red Hat)
no flags Details | Diff

  None (edit)
Description Jun'ichi Nomura (Red Hat) 2007-06-05 18:56:34 EDT
Description of problem:
  kdump initialization always fails on x86_64

Version-Release number of selected component:
  2.6.18-15.el5 or later

How reproducible:
  Always on x86_64

Steps to Reproduce:
  0. setup kdump
  1. pass "crashkernel=128M@16M" boot parameter to kernel
  2. do "service kdump start"
  3. grep "Crash" /proc/iomem

Actual results:
  The step 2 fails.
  The step 3 shows "00000000-00000000 : Crash kernel".
  dmesg shows "Memory for crash kernel (0x1000000 to 0x8ffffff) notwithin
  permissible range" "disabling kdump".

Expected results:
  The step 2 succeeds.
  The step 3 shows "01000000-08ffffff : Crash kernel".

Hardware info:

Additional Info:
  This is a regression introduced since 2.6.18-15.el5.

  The following patch is a suspect.
    linux-2.6-kdump-bounds-checking-for-crashkernel-args.patch

  The patch adds a code to check the boundary based on max_low_pfn.
  The code for x86_64 is implemented in setup_arch(), which is called
  earlier than mem_init(), which sets max_low_pfn.
  So the check always fails.
Comment 1 Jun'ichi Nomura (Red Hat) 2007-06-05 19:01:39 EDT
Created attachment 156293 [details]
use end_pfn instead of uninitialized max_low_pfn

max_low_pfn is not yet initialized in setup_arch().
Using end_pfn would solve the issue.
Tested on x86_64.
Comment 2 Jun'ichi Nomura (Red Hat) 2007-06-07 10:27:28 EDT

*** This bug has been marked as a duplicate of 242648 ***

Note You need to log in before you can comment on or make changes to this bug.