Bug 163419

Summary: hard disk host protect area handling
Product: [Fedora] Fedora Reporter: Parameshwara Bhat <peebhat>
Component: kernelAssignee: Kernel Maintainer List <kernel-maint>
Status: CLOSED CURRENTRELEASE QA Contact: Brian Brock <bbrock>
Severity: medium Docs Contact:
Priority: medium    
Version: 5CC: alan, davej, ncunning, pp, wtogami, zing
Target Milestone: ---   
Target Release: ---   
Hardware: i686   
OS: Linux   
URL: http://bugme.osdl.org/show_bug.cgi?id=6840
Whiteboard:
Fixed In Version: fc7 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2007-07-23 23:15:55 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:

Description Parameshwara Bhat 2005-07-16 02:46:59 UTC
From Bugzilla Helper:
User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; X11; Linux i686; en) Opera 8.01

Description of problem:
The recent kernel releases from fedora fail to handle host protect area of the 
hard disk as they should. Host protect area is the hidden remaining capacity of 
the hard disk set by a jumper to overcome BIOS limitations of hard disk sizes.

Linux kernel's behaviour with respect to this was set by IDE_DISK_STROKE 
parameter in kernel configuration. Somewhere along 2.6.X kernels,this parameter 
became boot time diskwise kernel option. I learned this option also went and 
kernel handles this intelligently now. ( OR DISK_STROKE became default ).Example 
is kernel-2.6.11-1.1226_FC4 which handles this well to disable host protect area 
and restore full capacity.

Come later kernels, kernel-2.6.11-1.1369_FC4,kernel-2.6.12-1.1390_FC4, this 
ability of the kernel is inhibited by Fedora developer's configuration of 
kernel. Contemporary kernels from other distributions do not exhibit this 
problem.

Though this doen't do any more harm than allowing just BIOS limited size, it 
cripples kernel's ability of restoring full capacity of hard disk. Also, now I 
am unable to look into my data partitions beyond 33.8 GB on the hard disk 
created during kernel-2.6.11-1.1226_FC4. Also I had to install FC afresh on a 
second hard disk.



Version-Release number of selected component (if applicable):
kernel-2.6.11-1.1369_FC4, kernel-2.6.12-1.1390_FC4

How reproducible:
Always

Steps to Reproduce:
1.just boot with mentioned kernels
2.
3.
  

Actual Results:  Fc4 not able to boot off this disk with mentioned kernels

Size fo the hard disk reported wrongly

Expected Results:  Kernel should have automatically read the native capacity of the hard disk.

Additional info:

Output of 2.6.12-1.1390_FC4 kernel

Jul 15 06:49:22 sidharth kernel: hdb: max request size: 1024KiB
Jul 15 06:49:22 sidharth kernel: hdb: Host Protected Area detected.
Jul 15 06:49:22 sidharth kernel: 	current capacity is 66055248 sectors (33820 
MB)
Jul 15 06:49:22 sidharth kernel: 	native  capacity is 78165360 sectors (40020 
MB)
Jul 15 06:49:22 sidharth kernel: hdb: task_no_data_intr: status=0x51 { 
DriveReady SeekComplete Error }
Jul 15 06:49:22 sidharth kernel: hdb: task_no_data_intr: error=0x04 { 
DriveStatusError }
Jul 15 06:49:22 sidharth kernel: ide: failed opcode was: 0xf9
Jul 15 06:49:22 sidharth kernel: hdb: 66055248 sectors (33820 MB) w/2048KiB 
Cache, CHS=16383/255/63, UDMA(33)


Output of kernel 2.6.11-1.1226_FC4

hdb: max request size: 1024KiB
Jul 15 08:16:53 sidharth kernel: hdb: Host Protected Area detected.
Jul 15 08:16:53 sidharth kernel: 	current capacity is 66055248 sectors (33820 
MB)
Jul 15 08:16:53 sidharth kernel: 	native  capacity is 78165360 sectors (40020 
MB)
Jul 15 08:16:53 sidharth kernel: hdb: Host Protected Area disabled.
Jul 15 08:16:53 sidharth kernel: hdb: 78165360 sectors (40020 MB) w/2048KiB 
Cache, CHS=16383/255/63, UDMA(33)
Jul 15 08:16:53 sidharth kernel: hdb: cache flushes supported
Jul 15 08:16:53 sidharth kernel:  hdb: hdb1 hdb3 < hdb5 hdb6 hdb7 hdb8 >

Comment 1 Parameshwara Bhat 2005-07-16 02:57:46 UTC
Also, this choice is historical with redhat, as all kernels from earlier feodra 
Cores had IDE_DISK_STROKE not set in kernel configuration.This was inconsistent 
with the minimum requirement claims of P II / III processors, as BIOSes from 
that era all had this limitation for hard Disk size.

Also, this is denying the kernel's potential to users.

Allowing  this ability in kernel doesn't do any harm to those who do not need 
it.

Parameshwara Bhat, peebhat 

Comment 2 Alan Cox 2005-07-20 14:23:28 UTC
That bug should be fixed in the newest FC3/FC4 kernels. We issued a 48bit LBA
HPA request for 28bit only aware devices. Please test with a current kernel


Comment 3 Parameshwara Bhat 2005-07-23 04:06:00 UTC
As of kernel-2.6.12-1.1398_FC4(Latest) the issue is not resolved. Relevant 
portion of boot log (dmesg) follows:
hda: max request size: 128KiB
hda: 16841664 sectors (8622 MB) w/512KiB Cache, CHS=16708/16/63, UDMA(33)
hda: cache flushes not supported
 hda: hda1 hda2 hda3
hdb: max request size: 1024KiB
hdb: Host Protected Area detected.
        current capacity is 66055248 sectors (33820 MB)
        native  capacity is 78165360 sectors (40020 MB)
hdb: task_no_data_intr: status=0x51 { DriveReady SeekComplete Error }
hdb: task_no_data_intr: error=0x04 { DriveStatusError }
ide: failed opcode was: 0xf9
hdb: 66055248 sectors (33820 MB) w/2048KiB Cache, CHS=16383/255/63, UDMA(33)
hdb: cache flushes supported
 hdb: hdb1 hdb3 < hdb5 hdb6 hdb7 hdb8 >

Parameshwara Bhat



Comment 4 Pekka Pietikäinen 2005-07-23 14:08:58 UTC
There's a nasty FC3 <-> FC4 regression too related to host protected areas.

IBM thinkpads have a few GB recovery area in the host protected area that FC3
happily ignored. FC4 overrides the BIOS/harddisk and thinks it can use the
entire disk. Which means anaconda makes partitions that fill the entire disk.
Everything is fine until you come back from suspend -> that part of the disk is
unaccessible since the bios has locked it -> your system gets very confused.

So there's two scenarios, some people really want to use that space and others
absolutely require that the kernel doesn't touch it. Lovely :)

Comment 5 Alan Cox 2005-07-25 10:44:35 UTC
Please file seperate bugs for seperate problems. The HPA/BIOS/anaconda item is
unrelated to the bug report. I'd suggest Anaconda is probably the best place to
file your bug.

Thanks


Comment 6 Alan Cox 2005-08-08 16:13:28 UTC
I talked with Matthew Garrett about the HPA on resume problem. It looks like
upstream needs to specifically handle this case. Fortunately it seems Matthew
has an afflicted machine


Comment 7 Dave Jones 2005-08-26 21:41:24 UTC
*** Bug 164139 has been marked as a duplicate of this bug. ***

Comment 8 Parameshwara Bhat 2005-10-23 01:44:40 UTC
In Fedora-2.6.13-1.1532_FC4 kernel update, the behaviour of the kernel is 
restored.Kernel automatically disables host protect area.From dmesg:
hdb: max request size: 1024KiB
hdb: Host Protected Area detected.
	current capacity is 66055248 sectors (33820 MB)
	native  capacity is 78165360 sectors (40020 MB)
hdb: Host Protected Area disabled.
hdb: 78165360 sectors (40020 MB) w/2048KiB Cache, CHS=16383/255/63, UDMA(33)
hdb: cache flushes supported
 hdb: hdb1 hdb3 < hdb5 hdb6 hdb7 hdb8 >


Comment 9 Dave Jones 2005-11-10 20:15:29 UTC
2.6.14-1.1637_FC4 has been released as an update for FC4.
Please retest with this update, as a large amount of code has been changed in
this release, which may have fixed your problem.

Thank you.


Comment 10 Parameshwara Bhat 2005-11-14 07:52:14 UTC
kernel behaviour wrt HPA has not changed in kernel-2.6.13-1.1532_FC4 from 
release kernel-2.6.13-1.1532_FC4.But shows something of the behaviour of bugs #  
163423, 162347. A few lines of I/O error messages and then goes on.

Parameshwara Bhat

Comment 11 John Thacker 2006-04-22 04:30:36 UTC
Just changing component to kernel, since it's clearly not a 4Suite issue.  Does
this behavior still occur in FC5?

Comment 12 Parameshwara Bhat 2006-04-22 08:42:23 UTC
It was never filed as 4suite issue. It is a kernel issue.

For FC5 I haven't tested as yet.As my bandwideth is consumed for April, I do not 
hink I can test it before that either.

Parameshwara bhat

Comment 13 Dave Jones 2006-09-17 03:02:27 UTC
[This comment added as part of a mass-update to all open FC4 kernel bugs]

FC4 has now transitioned to the Fedora legacy project, which will continue to
release security related updates for the kernel.  As this bug is not security
related, it is unlikely to be fixed in an update for FC4, and has been migrated
to FC5.

Please retest with Fedora Core 5.

Thank you.


Comment 14 Dave Jones 2006-10-17 00:10:46 UTC
A new kernel update has been released (Version: 2.6.18-1.2200.fc5)
based upon a new upstream kernel release.

Please retest against this new kernel, as a large number of patches
go into each upstream release, possibly including changes that
may address this problem.

This bug has been placed in NEEDINFO state.
Due to the large volume of inactive bugs in bugzilla, if this bug is
still in this state in two weeks time, it will be closed.

Should this bug still be relevant after this period, the reporter
can reopen the bug at any time. Any other users on the Cc: list
of this bug can request that the bug be reopened by adding a
comment to the bug.

In the last few updates, some users upgrading from FC4->FC5
have reported that installing a kernel update has left their
systems unbootable. If you have been affected by this problem
please check you only have one version of device-mapper & lvm2
installed.  See bug 207474 for further details.

If this bug is a problem preventing you from installing the
release this version is filed against, please see bug 169613.

If this bug has been fixed, but you are now experiencing a different
problem, please file a separate bug for the new problem.

Thank you.

Comment 15 Pekka Pietikäinen 2006-10-18 16:08:16 UTC
Still no change (well, FC6 not FC5 + 2.6.18 update kernel). Patch for resetting
the HPA after resume seemed to be in 2.6.18-rcX-mm for a while, but it got
reverted, so it's not there anymore either.


Comment 20 Pekka Pietikäinen 2006-12-04 23:33:12 UTC
My case was:

Have Linux installed on a Thinkpad X31 with funny harddrive that has HPA that
can't be removed permanently (it can on several identical machines, just not
this one harddrive. I've tried. Several times. )

After powerup machine works fine, kernel disables HPA and the whole harddrive is
available.  Suspend, resume, and the last 6 GB are gone -> much unhappyness if
you have data there.

I can imagine use cases that don't depend on a weird harddrive that would show
problems ;) I ended up repartitioning so that the last 6GB doesn't get used,
ages ago (FC2/FC3-era) Linux didn't see it and that worked just fine. Probably
should ask for a non-broken drive without HPA, that would be the easy fix ;)

Comment 28 Nigel Cunningham 2007-07-23 23:15:55 UTC
Marking as WONTFIX, sorry. I'm not sure whether newer releases would have 
better driver level support for HPA. It might be worth giving them a try.

Comment 29 Pekka Pietikäinen 2007-07-24 00:07:20 UTC
Actually the switch to libata for PATA in F7 made things work, it honors it by
default and can be overridden by a commandline option (would make things work
for original submitter) + even the suspend/resume paths probably are fixed ->
changing to CURRENTRELEASE ;)

Laptop with funny harddrive with impossible-to-remove-HPA is now
someone-elses-problem anyway. Woo! An entire 6 GB of extra space on my laptop!


Comment 30 Nigel Cunningham 2007-07-24 08:49:23 UTC
The original bug was against FC5, which we won't (AFAIK) be fixing, so WONTFIX 
is actually the right closure. Nevertheless, great to hear it's working now.