Skip to contentSkip to conversation

lock validator: add s390 to supported options

PATCH In review Heiko Carstens 7 participants 19 messages Jun 14, 2006 – Jun 23, 2006

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 8/8 lock validator: add s390 to supported options

From: Heiko Carstens <[email protected]>

Enable DEBUG_SPINLOCK_ALLOC, DEBUG_RWLOCK_ALLOC, DEBUG_MUTEX_ALLOC, DEBUG_RWSEM_ALLOC and LOCKDEP for s390.

---

 lib/Kconfig.debug |   10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)
lib/Kconfig.debug +5 -5
diff -purN a/lib/Kconfig.debug b/lib/Kconfig.debug
@@ -175,7 +175,7 @@ config DEBUG_SPINLOCK
 
 config DEBUG_SPINLOCK_ALLOC
 	bool "Spinlock debugging: detect incorrect freeing of live spinlocks"
-	depends on DEBUG_SPINLOCK && X86
+	depends on DEBUG_SPINLOCK && (X86 || S390)
 	select LOCKDEP
 	help
 	 This feature will check whether any held spinlock is incorrectly
@@ -224,7 +224,7 @@ config PROVE_SPIN_LOCKING
 
 config DEBUG_RWLOCK_ALLOC
 	bool "rw-lock debugging: detect incorrect freeing of live rwlocks"
-	depends on DEBUG_SPINLOCK && X86
+	depends on DEBUG_SPINLOCK && (X86 || S390)
 	select LOCKDEP
 	help
 	 This feature will check whether any held rwlock is incorrectly
@@ -281,7 +281,7 @@ config DEBUG_MUTEXES
 
 config DEBUG_MUTEX_ALLOC
 	bool "Mutex debugging: detect incorrect freeing of live mutexes"
-	depends on DEBUG_MUTEXES && X86
+	depends on DEBUG_MUTEXES && (X86 || S390)
 	select LOCKDEP
 	help
 	 This feature will check whether any held mutex is incorrectly
@@ -337,7 +337,7 @@ config DEBUG_RWSEMS
 
 config DEBUG_RWSEM_ALLOC
 	bool "rwsem debugging: detect incorrect freeing of live rwsems"
-	depends on DEBUG_RWSEMS && X86
+	depends on DEBUG_RWSEMS && (X86 || S390)
 	select LOCKDEP
 	help
 	 This feature will check whether any held rwsem is incorrectly
@@ -389,7 +389,7 @@ config LOCKDEP
 	select FRAME_POINTER
 	select KALLSYMS
 	select KALLSYMS_ALL
-	depends on X86
+	depends on X86 || S390
 
 config DEBUG_NON_NESTED_UNLOCKS
 	bool "Detect non-nested unlocks"
#1 lore raw

lib/Kconfig.debug config DEBUG_SPINLOCK · line 178

Quoted 5 lines · Heiko Carstens

config DEBUG_SPINLOCK_ALLOC

lib/Kconfig.debug +1 -1
 	bool "Spinlock debugging: detect incorrect freeing of live spinlocks"
-	depends on DEBUG_SPINLOCK && X86
+	depends on DEBUG_SPINLOCK && (X86 || S390)

Can we please stomp this out before it starts to look like CONFIG_FRAME_POINTER?

We should define CONFIG_ARCH_SUPPORTS_LOCKDEP down in arch/[i386|x86_64|s390]/Kconfig and use that in lib/Kconfig.debug.

#2 lore raw

RE: Wish for 2006 to Alan Cox and Jeff Garzik: A functional Driver for PDC202XX

Good thing that I googled groups this thread. I thought that I was the only one that had this problem.

I've gone from three 133TX2 controllers, 5 80-GB drives in a RAID5, and 3 drives in drive linking (all managed by EVMS) to 2 controllers with just the 3 drive linking drives, and still get the mda_expiry error messages.

Last thing that I've done is to force the data drives to use UDMA2, with the hope that it'll work around this problem. May be working, as I've not had an error in 24 hours of running the system, but then I'm not accessing the drives either. They are dismounted to protect the data. Didn't matter before, got the errors with just the smartd daemon checking up on the drive health, but so far so good.

Regardless, count me as another one of the interested parties for a cure. I've read the thread, and will prepare two current kernels, one using the PDC202XX_NEW and one using the PDC202XX_OLD configuration options. I'm hoping that the PDC202XX_OLD will also resolve this issue.

Any further advice on how to work around this would be greatly appreciated.

Erik.

#3 lore raw

RE: Wish for 2006 to Alan Cox and Jeff Garzik: A functional Driver for PDC202XX

Ar Llu, 2006-06-19 am 18:18 -0400, ysgrifennodd Erik Ohrnberger:

Quoted 4 lines

Regardless, count me as another one of the interested parties for a cure. I've read the thread, and will prepare two current kernels, one using the PDC202XX_NEW and one using the PDC202XX_OLD configuration options. I'm hoping that the PDC202XX_OLD will also resolve this issue.

Bartlomiej is the old IDE layer maintainer. I would direct any enquiries to him about those drivers.

Quoted 1 line

Any further advice on how to work around this would be greatly appreciated.

2.6.17 with the libata pata patch from
http://zeniv.linux.org.uk/~alan/IDE has a Promise driver for the PDC 20268 and higher that was written by Albert Lee. There is also a test driver for the older chips (20265 etc).

To try that build 2.6.17 with the patch and then say "N" to CONFIG_IDE, "Y" to the SATA options under SCSI and the right controller. It will move your disks to /dev/sda /dev/sdb etc as it uses the SCSI layer.

Alan

#4 lore raw

RE: Wish for 2006 to Alan Cox and Jeff Garzik: A functional Driver for PDC202XX

Alan,
I thank you persaonlly for these pointers.

Erik.

Quoted 25 lines · Alan Cox

Ar Llu, 2006-06-19 am 18:18 -0400, ysgrifennodd Erik Ohrnberger:

Quoted 6 lines

Regardless, count me as another one of the interested parties for a cure.
I've read the thread, and will prepare two current kernels, one using the
PDC202XX_NEW and one using the PDC202XX_OLD configuration options. I'm hoping that the PDC202XX_OLD will also resolve this issue.

Bartlomiej is the old IDE layer maintainer. I would direct any enquiries to him about those drivers.

Quoted 2 lines

Any further advice on how to work around this would be greatly appreciated.

2.6.17 with the libata pata patch from
http://zeniv.linux.org.uk/~alan/IDE has a Promise driver for the PDC 20268 and higher that was written by Albert Lee. There is also a test driver for the older chips (20265 etc).

To try that build 2.6.17 with the patch and then say "N" to CONFIG_IDE, "Y" to the SATA options under SCSI and the right controller. It will move your disks to /dev/sda /dev/sdb etc as it uses the SCSI layer.

Alan

#5 lore raw

RE: Wish for 2006 to Alan Cox and Jeff Garzik: A functional Driver for PDC202XX

Hate to be a bother, but I think I must have botched something.

Here is what I did:
Downloaded kernel code.
Applied the IDE patches (http://zeniv.linux.org.uk/~alan/IDE)

	(had to apply the change from ata_data_xfer calls to ops->data_xfer
	(no big deal, I speak C )

CONFIG_IDE is set to N
Device Drives\SCSI device support\Probe all LUNs on each SCSI device: Y Device Drives\SCSI device support\Serial ATA (SATA) support: Y Device Drives\SCSI device support\Serial ATA (SATA) support\NVIDIA SATA
support: Y
Device Drives\SCSI device support\Serial ATA (SATA) support\Promise SATA TX2/TX4 Support: Y
Device Drives\SCSI device support\Serial ATA (SATA) support\Promise SATA SX4
Support: Y
Device Drives\SCSI device support\Older Promise PATA controller Support (Raving Lunatic): Y
Device Drives\SCSI device support\Legacy ISA PATA support (Raving Lunatic):
Y
Device Drives\SCSI device support\Older Promise PATA controller support (Raving Lunatic): Y

Build the kernel, put it in /boot etc. and boot it. The kernel finds the motherboard IDE controllers and recognizes the two hard drives and the DVD ROM drive there, but it does not appear to find the two Promise 20269 controllers. What'd I do wrong? Should I have applied the kernel-2.6.17 patches from kernel.org before the http://zeniv.linux.org.uk/~alan/IDE
patches?

If I ATA/ATAPI/MFM/RLL support\PROMISE PDC202{68|69|70|71|75|76|77} support:
Y
Then the kernel find the controllers, but uses that failing driver rather than the SCSI/PATA driver. Yea, well duhh. So I can't turn that on.

lspci:
00:00.0 Host bridge: nVidia Corporation nForce2 AGP (different version?) (rev c1)
00:00.1 RAM memory: nVidia Corporation nForce2 Memory Controller 0 (rev c1) 00:00.2 RAM memory: nVidia Corporation nForce2 Memory Controller 4 (rev c1) 00:00.3 RAM memory: nVidia Corporation nForce2 Memory Controller 3 (rev c1) 00:00.4 RAM memory: nVidia Corporation nForce2 Memory Controller 2 (rev c1) 00:00.5 RAM memory: nVidia Corporation nForce2 Memory Controller 5 (rev c1) 00:01.0 ISA bridge: nVidia Corporation nForce2 ISA Bridge (rev a4) 00:01.1 SMBus: nVidia Corporation nForce2 SMBus (MCP) (rev a2) 00:08.0 PCI bridge: nVidia Corporation nForce2 External PCI Bridge (rev a3) 00:09.0 IDE interface: nVidia Corporation nForce2 IDE (rev a2) 00:1e.0 PCI bridge: nVidia Corporation nForce2 AGP (rev c1) 01:06.0 Mass storage controller: Promise Technology, Inc. 20269 (rev 02) 01:07.0 Mass storage controller: Promise Technology, Inc. 20269 (rev 02) 01:0a.0 Ethernet controller: Accton Technology Corporation SMC2-1211TX (rev
10)
02:00.0 VGA compatible controller: ATI Technologies Inc Radeon R200 QM [Radeon 9100]

Quoted 44 lines

-----Original Message-----
From: Erik Ohrnberger [mailto:[email protected]]
Sent: Tuesday, June 20, 2006 8:52 AM
To: Alan Cox
Cc: [email protected]
Subject: RE: Wish for 2006 to Alan Cox and Jeff Garzik: A functional Driver for PDC202XX

Alan,
I thank you persaonlly for these pointers.

Erik.

Quoted 2 lines · Alan Cox

Ar Llu, 2006-06-19 am 18:18 -0400, ysgrifennodd Erik Ohrnberger:

Quoted 1 line

Regardless, count me as another one of the interested

parties for a

Quoted 2 lines
Quoted 2 lines

cure.
I've read the thread, and will prepare two current

kernels, one using

Quoted 7 lines
Quoted 3 lines

the PDC202XX_NEW and one using the PDC202XX_OLD configuration options. I'm hoping that the PDC202XX_OLD will also resolve this issue.

Bartlomiej is the old IDE layer maintainer. I would direct any enquiries to him about those drivers.

Quoted 6 lines
Quoted 2 lines

Any further advice on how to work around this would be greatly appreciated.

2.6.17 with the libata pata patch from
http://zeniv.linux.org.uk/~alan/IDE has a Promise driver for the PDC 20268 and higher that was written by Albert Lee. There is

also a test

Quoted 5 lines

driver for the older chips (20265 etc).

To try that build 2.6.17 with the patch and then say "N" to CONFIG_IDE, "Y" to the SATA options under SCSI and the right controller. It will move your disks to /dev/sda /dev/sdb

etc as it uses the SCSI layer.

Quoted 3 lines

Alan

#6 lore raw

RE: Wish for 2006 to Alan Cox and Jeff Garzik: A functional Driver for PDC202XX

Ar Maw, 2006-06-20 am 21:19 -0400, ysgrifennodd Erik Ohrnberger:

Quoted 4 lines

Downloaded kernel code.
Applied the IDE patches (http://zeniv.linux.org.uk/~alan/IDE)

	(had to apply the change from ata_data_xfer calls to ops->data_xfer
	(no big deal, I speak C )

What tree are you working against. The patches I put up are versus 2.6.17 not -mm. Also check in include/linux/libata.h that ATA_ENABLE_PATA is defined (its down with the debug options). The patch should have enabled it however.

#7 lore raw

RE: Wish for 2006 to Alan Cox and Jeff Garzik: A functional Driver for PDC202XX

Quoted 11 lines · Alan Cox

Ar Maw, 2006-06-20 am 21:19 -0400, ysgrifennodd Erik Ohrnberger:

Quoted 4 lines

Downloaded kernel code.
Applied the IDE patches (http://zeniv.linux.org.uk/~alan/IDE)

	(had to apply the change from ata_data_xfer calls to ops->data_xfer
	(no big deal, I speak C )

What tree are you working against. The patches I put up are versus 2.6.17 not -mm. Also check in include/linux/libata.h that ATA_ENABLE_PATA is defined (its down with the debug options). The patch should have enabled it however.

I downloaded linux-2.6.17.tar.gz from kernel.org, when extraceted created linux-2.6.17 directory. I zmore'd the patch-2.6.17-rc4-ide1.gz file and noticed that it was working against linux-2.6.17-rc4-ide1, so I renamed the linux-2.6.17 to linux-2.6.17-rc4-ide1 and zcat | patch -p0. (Did I do this wrong? Seemed to apply except for the two function calls that I had to edit by hand)

Checking include/linux/libata.h reveals that: ATA_ENABLE_PATA is defined.

Thanks for all the help. I really appreciate it, but hope that it's not taking up too much of your time.

Erik.

#8 lore raw

RE: Wish for 2006 to Alan Cox and Jeff Garzik: A functional Driver for PDC202XX

Ar Mer, 2006-06-21 am 09:09 -0400, ysgrifennodd Erik Ohrnberger:

Quoted 2 lines

noticed that it was working against linux-2.6.17-rc4-ide1, so I renamed the linux-2.6.17 to linux-2.6.17-rc4-ide1 and zcat | patch -p0.

Get linux-2.6.17-ide1.gz instead of the rc4 patch

Quoted 2 lines

Thanks for all the help. I really appreciate it, but hope that it's not taking up too much of your time.

Not a problem. I'd like to get to the bottom of the problems you are seeing and this all helps.

#9 lore raw

Re: Wish for 2006 to Alan Cox and Jeff Garzik: A functional Driver for PDC202XX

Quoted 5 lines · Alan Cox

Ar Mer, 2006-06-21 am 09:09 -0400, ysgrifennodd Erik Ohrnberger:

Quoted 2 lines

noticed that it was working against linux-2.6.17-rc4-ide1, so I renamed the linux-2.6.17 to linux-2.6.17-rc4-ide1 and zcat | patch -p0.

Get linux-2.6.17-ide1.gz instead of the rc4 patch

I have a PDC20267 and tried out 2.6.17 + linux-2.6.17-ide1.gz The resulting pata_pdc202xx_old module didn't initialize the controller on boot (via initrd). Is there some other CONFIG option or linux cmdline that should be used?

thanks.

#10 lore raw

Re: Wish for 2006 to Alan Cox and Jeff Garzik: A functional Driver for PDC202XX

Ar Mer, 2006-06-21 am 09:34 -0400, ysgrifennodd Mike Snitzer:

Quoted 4 lines

I have a PDC20267 and tried out 2.6.17 + linux-2.6.17-ide1.gz The resulting pata_pdc202xx_old module didn't initialize the controller on boot (via initrd). Is there some other CONFIG option or linux cmdline that should be used?

No it should find it as soon as you load the module. If the device is not the primary IDE controller (hda-hdd) you can set CONFIG_IDE = y, include drivers for your other controller and get some debug info.

Usually "it didnt initialize" reports are from people with mixed old/new IDE in their kernel or initrd, or initrds missing the required modules.

Stick a printk in pdc_init before the pci_register_driver call and you'll know for sure if the module is getting loaded. If it is then send me an lspci and I'll have a deeper look

Alan

#11 lore raw

Re: Wish for 2006 to Alan Cox and Jeff Garzik: A functional Driver for PDC202XX

Quoted 10 lines · Alan Cox

Ar Mer, 2006-06-21 am 09:34 -0400, ysgrifennodd Mike Snitzer:

Quoted 4 lines

I have a PDC20267 and tried out 2.6.17 + linux-2.6.17-ide1.gz The resulting pata_pdc202xx_old module didn't initialize the controller on boot (via initrd). Is there some other CONFIG option or linux cmdline that should be used?

No it should find it as soon as you load the module. If the device is not the primary IDE controller (hda-hdd) you can set CONFIG_IDE = y, include drivers for your other controller and get some debug info.

Quoted 6 lines

Usually "it didnt initialize" reports are from people with mixed old/new IDE in their kernel or initrd, or initrds missing the required modules.

Stick a printk in pdc_init before the pci_register_driver call and you'll know for sure if the module is getting loaded. If it is then send me an lspci and I'll have a deeper look

The PDC20267 is the primary controller, I do not have CONFIG_IDE set. The issue with my initrd was mkinitrd --with=pata_pdc202xx_old didn't include sd_mod; setting scsi_hostadapter accordingly in modprobe.conf took care of this.

Once the kernel booted the system was seemingly starting without any problems until various services tried to write files to the root filesystem. Each write attempt resulted in a "Read-only filesystem" error yet all filesystems were mounted read-write.

The kernel had a ton of errors:

Buffer I/O error on device sda3, logical block 4869508 lost page write due to I/O error on sda3
ata1: translated ATA stat/err 0x51/84 to SCSI SK/ASC/ASCQ 0xb/47/00
ata1: status=0x51 { DriveReady SeekComplete Error }
ata1: error=0x84 { DriveStatusError BadCRC }
ata1: translated ATA stat/err 0x51/84 to SCSI SK/ASC/ASCQ 0xb/47/00
ata1: status=0x51 { DriveReady SeekComplete Error }
ata1: error=0x84 { DriveStatusError BadCRC }
ata1: translated ATA stat/err 0x51/84 to SCSI SK/ASC/ASCQ 0xb/47/00
ata1: status=0x51 { DriveReady SeekComplete Error }
ata1: error=0x84 { DriveStatusError BadCRC }
ata1: translated ATA stat/err 0x51/84 to SCSI SK/ASC/ASCQ 0xb/47/00
ata1: status=0x51 { DriveReady SeekComplete Error }
ata1: error=0x84 { DriveStatusError BadCRC }
ata1: translated ATA stat/err 0x51/84 to SCSI SK/ASC/ASCQ 0xb/47/00
ata1: status=0x51 { DriveReady SeekComplete Error }
ata1: error=0x84 { DriveStatusError BadCRC }
ata1: translated ATA stat/err 0x51/84 to SCSI SK/ASC/ASCQ 0xb/47/00
ata1: status=0x51 { DriveReady SeekComplete Error }
ata1: error=0x84 { DriveStatusError BadCRC }
sd 0:0:0:0: SCSI error: return code = 0x8000002
sda: Current: sense key: Aborted Command
Additional sense: Scsi parity error
Info fld=0x27a3f9b

lspci output is:

00:00.0 Host bridge: Intel Corporation 82865G/PE/P DRAM Controller/Host-Hub Interface (rev 02)
00:02.0 VGA compatible controller: Intel Corporation 82865G Integrated Graphics Controller (rev 02)
00:1d.0 USB Controller: Intel Corporation 82801EB/ER (ICH5/ICH5R) USB UHCI Controller #1 (rev 02)
00:1d.1 USB Controller: Intel Corporation 82801EB/ER (ICH5/ICH5R) USB UHCI Controller #2 (rev 02)
00:1d.3 USB Controller: Intel Corporation 82801EB/ER (ICH5/ICH5R) USB UHCI Controller #4 (rev 02)
00:1d.7 USB Controller: Intel Corporation 82801EB/ER (ICH5/ICH5R) USB2 EHCI Controller (rev 02)
00:1e.0 PCI bridge: Intel Corporation 82801 PCI Bridge (rev c2) 00:1f.0 ISA bridge: Intel Corporation 82801EB/ER (ICH5/ICH5R) LPC Interface Bridge (rev 02)
00:1f.1 IDE interface: Intel Corporation 82801EB/ER (ICH5/ICH5R) IDE Controller (rev 02)
00:1f.3 SMBus: Intel Corporation 82801EB/ER (ICH5/ICH5R) SMBus Controller (rev 02)
00:1f.5 Multimedia audio controller: Intel Corporation 82801EB/ER (ICH5/ICH5R) AC'97 Audio Controller (rev 02)
01:01.0 Unknown mass storage controller: Promise Technology, Inc. PDC20267 (FastTrak100/Ultra100) (rev 02)
01:08.0 Ethernet controller: Intel Corporation 82562EZ 10/100 Ethernet Controller (rev 02)

#12 lore raw

Re: Wish for 2006 to Alan Cox and Jeff Garzik: A functional Driver for PDC202XX

Ar Mer, 2006-06-21 am 11:20 -0400, ysgrifennodd Mike Snitzer:

Quoted 3 lines

ata1: translated ATA stat/err 0x51/84 to SCSI SK/ASC/ASCQ 0xb/47/00
ata1: status=0x51 { DriveReady SeekComplete Error }
ata1: error=0x84 { DriveStatusError BadCRC }

Thats a speed mistune somewhere in the code.

Can you send me an lspci -vxx and the information on the drive (or dmesg) with a "normal" kernel boot and I'll hunt it down.

Alan

#13 lore raw

Re: Wish for 2006 to Alan Cox and Jeff Garzik: A functional Driver for PDC202XX

Quoted 9 lines · Alan Cox

Ar Mer, 2006-06-21 am 11:20 -0400, ysgrifennodd Mike Snitzer:

Quoted 3 lines

ata1: translated ATA stat/err 0x51/84 to SCSI SK/ASC/ASCQ 0xb/47/00
ata1: status=0x51 { DriveReady SeekComplete Error }
ata1: error=0x84 { DriveStatusError BadCRC }

Thats a speed mistune somewhere in the code.

Can you send me an lspci -vxx and the information on the drive (or dmesg) with a "normal" kernel boot and I'll hunt it down.

Please see attached for the lspci -vxx output.

Here is the the relevant IDE info from dmesg:

ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
PDC20267: IDE controller at PCI slot 0000:01:01.0
ACPI: PCI Interrupt Link [LNKG] enabled at IRQ 10
PCI: setting IRQ 10 as level-triggered
ACPI: PCI Interrupt 0000:01:01.0[A] -> Link [LNKG] -> GSI 10 (level, low) -> IRQ 10
PDC20267: chipset revision 2
PDC20267: ROM enabled at 0x40000000
PDC20267: 100% native mode on irq 10
PDC20267: (U)DMA Burst Bit ENABLED Primary PCI Mode Secondary PCI Mode.
ide0: BM-DMA at 0xdf00-0xdf07, BIOS settings: hda:pio, hdb:pio
ide1: BM-DMA at 0xdf08-0xdf0f, BIOS settings: hdc:pio, hdd:pio
Probing IDE interface ide0...
hda: SAMSUNG SP0401N, ATA DISK drive
ide0 at 0xdef0-0xdef7,0xdeea on irq 10
Probing IDE interface ide1...
ICH5: IDE controller at PCI slot 0000:00:1f.1
ACPI: PCI Interrupt Link [LNKC] enabled at IRQ 5
PCI: setting IRQ 5 as level-triggered
ACPI: PCI Interrupt 0000:00:1f.1[A] -> Link [LNKC] -> GSI 5 (level, low) -> IRQ 5
ICH5: chipset revision 2
ICH5: not 100% native mode: will probe irqs later
ide2: BM-DMA at 0xffa8-0xffaf, BIOS settings: hde:DMA, hdf:pio
Probing IDE interface ide2...
hde: GCR-8483B, ATAPI CD/DVD-ROM drive
ide2 at 0x170-0x177,0x376 on irq 15
Probing IDE interface ide1...
Probing IDE interface ide3...
Probing IDE interface ide4...
Probing IDE interface ide5...
hda: max request size: 128KiB
hda: Host Protected Area detected.

       current capacity is 78125000 sectors (40000 MB)
       native  capacity is 78242976 sectors (40060 MB)

hda: Host Protected Area disabled.
hda: 78242976 sectors (40060 MB) w/2048KiB Cache, CHS=16383/255/63, UDMA(100)
hda: cache flushes supported
hda: hda1 hda2 hda3

also smartctl info:

=== START OF INFORMATION SECTION ===

Device Model:     SAMSUNG SP0401N
Serial Number:    S004J10Y647557
Firmware Version: TJ100-28
User Capacity:    40,060,403,712 bytes
Device is:        In smartctl database [for details use: -P show]
ATA Version is:   7
ATA Standard is:  ATA/ATAPI-7 T13 1532D revision 0
Local Time is:    Wed Jun 21 13:45:56 2006 EDT
SMART support is: Available - device has SMART capability.
SMART support is: Enabled
#14 lore raw

RE: Wish for 2006 to Alan Cox and Jeff Garzik: A functional Driverfor PDC202XX

In reply to Alan Cox

OK, Success!

I have to admit that without the hint from Mike Snitzer, I wouldn't have put it all together. The key piece of information that he provided was:

Quoted 7 lines

From: Mike Snitzer [mailto:[email protected]]
Sent: Thursday, June 22, 2006 12:40 AM
To: [email protected]
Subject: Re: Wish for 2006 to Alan Cox and Jeff Garzik: A functional Driverfor PDC202XX

Alan's initial response to you was:

<snip>

Quoted 20 lines

But looking closer it is clear that the drivers/scsi/pata_pdc2027x.c source file provides support for your 20268 card. But the Kconfig and makefile changes to expose this driver in an official way weren't included in Alan's patch. So you just need to add something like the following to drivers/scsi/Makefile:

obj-$(CONFIG_SCSI_PATA_PDC) += libata.o pata_pdc2027x.o

Similarly you need a new entry in drivers/scsi/Kconfig that resembles something like:

config SCSI_PATA_PDC
tristate "Newer Promise PATA controller support
(Raving Lunatic)"
depends on SCSI_SATA && PCI && EXPERIMENTAL
help
This option enables support for the Promise 20268 through
20277 adapters.

Quoted 7 lines

If unsure, say N.

Once you modify those 2 files and select the "Newer Promise ..." option with make menuconfig; then rebuild with 'make' you should see something like:

Quoted 1 line

make

Many thanks Mike!

#15 lore raw
In reply to Andrew Morton
Quoted 9 lines · Andrew Morton

lib/Kconfig.debug config DEBUG_SPINLOCK · line 178

Quoted 5 lines · Heiko Carstens

config DEBUG_SPINLOCK_ALLOC

lib/Kconfig.debug +1 -1
 	bool "Spinlock debugging: detect incorrect freeing of live spinlocks"
-	depends on DEBUG_SPINLOCK && X86
+	depends on DEBUG_SPINLOCK && (X86 || S390)

Can we please stomp this out before it starts to look like CONFIG_FRAME_POINTER?

why is this even an arch option ?

#16 lore raw
Quoted 11 lines · On Tue, 20 Jun 2006 09:07:47 +0200 Arjan van de Ven <[email protected]> wrote:
Quoted 9 lines · Andrew Morton

lib/Kconfig.debug config DEBUG_SPINLOCK · line 178

Quoted 5 lines · Heiko Carstens

config DEBUG_SPINLOCK_ALLOC

lib/Kconfig.debug +1 -1
 	bool "Spinlock debugging: detect incorrect freeing of live spinlocks"
-	depends on DEBUG_SPINLOCK && X86
+	depends on DEBUG_SPINLOCK && (X86 || S390)

Can we please stomp this out before it starts to look like CONFIG_FRAME_POINTER?

Quoted 1 line

why is this even an arch option ?

I made it thus, because it failed to compile on powerpc.

#17 lore raw
In reply to Andrew Morton
Quoted 10 lines · Andrew Morton

Review

Quoted 5 lines · Heiko Carstens

config DEBUG_SPINLOCK_ALLOC

(quoted diff) +1 -1
 	bool "Spinlock debugging: detect incorrect freeing of live spinlocks"
-	depends on DEBUG_SPINLOCK && X86
+	depends on DEBUG_SPINLOCK && (X86 || S390)

Can we please stomp this out before it starts to look like CONFIG_FRAME_POINTER?

Quoted 2 lines

We should define CONFIG_ARCH_SUPPORTS_LOCKDEP down in arch/[i386|x86_64|s390]/Kconfig and use that in lib/Kconfig.debug.

How about this:

From: Heiko Carstens <[email protected]>

Add LOCKDEP_SUPPORT config option per architecture.

Cc Ingo Molnar <[email protected]>
Cc Arjan van de Ven <[email protected]>

---

 arch/i386/Kconfig.debug   |    4 ++++
 arch/s390/Kconfig.debug   |    4 ++++
 arch/x86_64/Kconfig.debug |    4 ++++
 lib/Kconfig.debug         |   10 +++++-----
 4 files changed, 17 insertions(+), 5 deletions(-)
arch/i386/Kconfig.debug +4 -0
Index: linux-2.6.17-mm1/arch/i386/Kconfig.debug
===================================================================
@@ -4,6 +4,10 @@ config TRACE_IRQFLAGS_SUPPORT
 	bool
 	default y
 
+config LOCKDEP_SUPPORT
+	bool
+	default y
+
 source "lib/Kconfig.debug"
 
 config EARLY_PRINTK
arch/s390/Kconfig.debug +4 -0
Index: linux-2.6.17-mm1/arch/s390/Kconfig.debug
===================================================================
@@ -4,6 +4,10 @@ config TRACE_IRQFLAGS_SUPPORT
 	bool
 	default y
 
+config LOCKDEP_SUPPORT
+	bool
+	default y
+
 source "lib/Kconfig.debug"
 
 endmenu
arch/x86_64/Kconfig.debug +4 -0
Index: linux-2.6.17-mm1/arch/x86_64/Kconfig.debug
===================================================================
@@ -4,6 +4,10 @@ config TRACE_IRQFLAGS_SUPPORT
 	bool
 	default y
 
+config LOCKDEP_SUPPORT
+	bool
+	default y
+
 source "lib/Kconfig.debug"
 
 config DEBUG_RODATA
lib/Kconfig.debug +5 -5
Index: linux-2.6.17-mm1/lib/Kconfig.debug
===================================================================
@@ -159,7 +159,7 @@ config DEBUG_SPINLOCK
 
 config DEBUG_SPINLOCK_ALLOC
 	bool "Spinlock debugging: detect incorrect freeing of live spinlocks"
-	depends on DEBUG_SPINLOCK && (X86 || S390)
+	depends on LOCKDEP_SUPPORT && DEBUG_SPINLOCK
 	select LOCKDEP
 	help
 	 This feature will check whether any held spinlock is incorrectly
@@ -208,7 +208,7 @@ config PROVE_SPIN_LOCKING
 
 config DEBUG_RWLOCK_ALLOC
 	bool "rw-lock debugging: detect incorrect freeing of live rwlocks"
-	depends on DEBUG_SPINLOCK && (X86 || S390)
+	depends on LOCKDEP_SUPPORT && DEBUG_SPINLOCK
 	select LOCKDEP
 	help
 	 This feature will check whether any held rwlock is incorrectly
@@ -265,7 +265,7 @@ config DEBUG_MUTEXES
 
 config DEBUG_MUTEX_ALLOC
 	bool "Mutex debugging: detect incorrect freeing of live mutexes"
-	depends on DEBUG_MUTEXES && (X86 || S390)
+	depends on LOCKDEP_SUPPORT && DEBUG_MUTEXES
 	select LOCKDEP
 	help
 	 This feature will check whether any held mutex is incorrectly
@@ -321,7 +321,7 @@ config DEBUG_RWSEMS
 
 config DEBUG_RWSEM_ALLOC
 	bool "rwsem debugging: detect incorrect freeing of live rwsems"
-	depends on DEBUG_RWSEMS && (X86 || S390)
+	depends on LOCKDEP_SUPPORT && DEBUG_RWSEMS
 	select LOCKDEP
 	help
 	 This feature will check whether any held rwsem is incorrectly
@@ -373,7 +373,7 @@ config LOCKDEP
 	select FRAME_POINTER
 	select KALLSYMS
 	select KALLSYMS_ALL
-	depends on X86 || S390
+	depends on LOCKDEP_SUPPORT
 
 config DEBUG_NON_NESTED_UNLOCKS
 	bool "Detect non-nested unlocks"
#18 lore raw
Quoted 11 lines · Heiko Carstens
Quoted 10 lines · Andrew Morton

Review

Quoted 5 lines · Heiko Carstens

config DEBUG_SPINLOCK_ALLOC

(quoted diff) +1 -1
 	bool "Spinlock debugging: detect incorrect freeing of live spinlocks"
-	depends on DEBUG_SPINLOCK && X86
+	depends on DEBUG_SPINLOCK && (X86 || S390)

Can we please stomp this out before it starts to look like CONFIG_FRAME_POINTER?

Quoted 8 lines
Quoted 2 lines

We should define CONFIG_ARCH_SUPPORTS_LOCKDEP down in arch/[i386|x86_64|s390]/Kconfig and use that in lib/Kconfig.debug.

How about this:

From: Heiko Carstens <[email protected]>

Add LOCKDEP_SUPPORT config option per architecture.

fine with me. Perhaps we can thus replace IRQ_TRACE_SUPPORT with LOCKDEP_SUPPORT as well, and would codify LOCKDEP_SUPPORT to mean irqflags-trace support and stacktrace support?

Ingo