A problem was found on PC's with the SB700 chipset: The PC fails to
load BIOS after running the 3.8.x kernel until the power is completely
cut off. It occurs in all 3.8.x versions and the mainline version as of
2/4. The issue does not occur with the 3.7.x builds.
There are two methods for accessing the watchdog registers.
1. Re-programming a resource address obtained by allocate_resource()
to chipset.
2. Use the direct memory-mapped IO access.
The method 1 can be used by all the chipsets (SP5100, SB7x0, SB8x0 or
later). However, experience shows that only PC with the SB8x0 (or
later) chipsets can use the method 2.
This patch removes the method 1, because the critical problem was found.
That's why the watchdog timer was able to be used on SP5100 and SB7x0
chipsets until now.
Link: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1116835
Link: https://lkml.org/lkml/2013/2/14/271
Signed-off-by: Takahisa Tanaka <mc74hc00@gmail.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
Cc: stable <stable@vger.kernel.org>
In case of SP5100 or SB7x0 chipsets, the sp5100_tco module writes zero to
reserved bits. The module, however, shouldn't depend on specific default
value, and should perform a read-merge-write operation for the reserved
bits.
This patch makes the sp5100_tco module perform a read-merge-write operation
on all the chipset (sp5100, sb7x0, sb8x0 or later).
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=43176
Signed-off-by: Takahisa Tanaka <mc74hc00@gmail.com>
Tested-by: Paul Menzel <paulepanter@users.sourceforge.net>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
Cc: stable <stable@vger.kernel.org>
In case of SB800 or later chipset and re-programming MMIO address(*),
sp5100_tco module may read incorrect value of reserved bit, because the module
reads a value from an incorrect I/O address. However, this bug doesn't cause
a problem, because when re-programming MMIO address, by chance the module
writes zero (this is BIOS's default value) to the low three bits of register.
* In most cases, PC with SB8x0 or later chipset doesn't need to re-programming
MMIO address, because such PC can enable AcpiMmio and can use 0xfed80b00 for
watchdog register base address.
This patch fixes this bug.
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=43176
Signed-off-by: Takahisa Tanaka <mc74hc00@gmail.com>
Tested-by: Paul Menzel <paulepanter@users.sourceforge.net>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
Cc: stable <stable@vger.kernel.org>
The current sp5100_tco driver only supports SP5100/SB7x0 chipset, doesn't
support SB8x0 chipset, because current sp5100_tco driver doesn't know that the
offset address for watchdog timer was changed from SB8x0 chipset.
The offset address of SP5100 and SB7x0 chipsets are as follows, quote from the
AMD SB700/710/750 Register Reference Guide (Page 164) and the AMD SP5100
Register Reference Guide (Page 166).
WatchDogTimerControl 69h
WatchDogTimerBase0 6Ch
WatchDogTimerBase1 6Dh
WatchDogTimerBase2 6Eh
WatchDogTimerBase3 6Fh
In contrast, the offset address of SB8x0 chipset is as follows, quote from
AMD SB800-Series Southbridges Register Reference Guide (Page 147).
WatchDogTimerEn 48h
WatchDogTimerConfig 4Ch
So, In the case of SB8x0 chipset, sp5100_tco reads meaningless MMIO
address (for example, 0xbafe00) from wrong offset address, and the following
message is logged.
SP5100 TCO timer: mmio address 0xbafe00 already in use
With this patch, sp5100_tco driver supports SB8x0 chipset, and can avoid
iomem resource conflict. The processing of this patch is as follows.
Step 1) Attempt to get the watchdog base address from indirect I/O (0xCD6
and 0xCD7).
- Go to the step 7 if obtained address hasn't conflicted with other
resource. But, currently, the address (0xfec000f0) conflicts with the
IOAPIC MMIO address, and the following message is logged.
SP5100 TCO timer: mmio address 0xfec000f0 already in use
0xfec000f0 is recommended by AMD BIOS Developer's Guide. So, go to the
next step.
Step 2) Attempt to get the SBResource_MMIO base address from AcpiMmioEN (for
SB8x0, PM_Reg:24h) or SBResource_MMIO (SP5100/SB7x0, PCI_Reg:9Ch)
register.
- Go to the step 7 if these register has enabled by BIOS, and obtained
address hasn't conflicted with other resource.
- If above condition isn't true, go to the next step.
Step 3) Attempt to get the free MMIO address from allocate_resource().
- Go to the step 7 if these register has enabled by BIOS, and obtained
address hasn't conflicted with other resource.
- Driver initialization has failed if obtained address has conflicted
with other resource, and no 'force_addr' parameter is specified.
Step 4) Use the specified address If 'force_addr' parameter is specified.
- allocate_resource() function may fail, when the PCI bridge device occupies
iomem resource from 0xf0000000 to 0xffffffff. To handle such a case,
I added 'force_addr' parameter to sp5100_tco driver. With 'force_addr'
parameter, sp5100_tco driver directly can assign MMIO address for watchdog
timer from free iomem region. Note that It's dangerous to specify wrong
address in the 'force_addr' parameter.
Example of force_addr parameter use
# cat /proc/iomem
...snip...
fec00000-fec003ff : IOAPIC 0
<--- free MMIO region
fec10000-fec1001f : pnp 00:0b
fec20000-fec203ff : IOAPIC 1
...snip...
# cat /etc/modprobe.d/sp5100_tco.conf
options sp5100_tco force_addr=0xfec00800
# modprobe sp5100_tco
# cat /proc/iomem
...snip...
fec00000-fec003ff : IOAPIC 0
fec00800-fec00807 : SP5100 TCO <--- watchdog timer MMIO address
fec10000-fec1001f : pnp 00:0b
fec20000-fec203ff : IOAPIC 1
...snip...
#
- Driver initialization has failed if specified address has conflicted
with other resource.
Step 5) Disable the watchdog timer
- To rewrite the watchdog timer register of the chipset, absolutely
guarantee that the watchdog timer is disabled.
Step 6) Re-program the watchdog timer MMIO address to chipset.
- Re-program the obtained MMIO address in Step 3 or Step 4 to chipset via
indirect I/O (0xCD6 and 0xCD7).
Step 7) Enable and setup the watchdog timer
This patch has worked fine on my test environment (ASUS M4A89GTD-PRO/USB3 and
DL165G7). therefore I believe that it's no problem to re-program the MMIO
address for watchdog timer to chipset during disabled watchdog. However,
I'm not sure about it, because I don't know much about chipset programming.
So, any comments will be welcome.
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=43176
Tested-by: Arkadiusz Miskiewicz <arekm@maven.pl>
Tested-by: Paul Menzel <paulepanter@users.sourceforge.net>
Signed-off-by: Takahisa Tanaka <mc74hc00@gmail.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
CONFIG_HOTPLUG is going away as an option so __devexit is no
longer needed.
Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Wim Van Sebroeck <wim@iguana.be>
Cc: Wan ZongShun <mcuos.com@gmail.com>
Cc: Ben Dooks <ben-linux@fluff.org>
Cc: Kukjin Kim <kgene.kim@samsung.com>
Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
CONFIG_HOTPLUG is going away as an option so __devinit is no longer
needed.
Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Wim Van Sebroeck <wim@iguana.be>
Cc: Wan ZongShun <mcuos.com@gmail.com>
Cc: Ben Dooks <ben-linux@fluff.org>
Cc: Kukjin Kim <kgene.kim@samsung.com>
Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
CONFIG_HOTPLUG is going away as an option so __devexit_p is no longer
needed.
Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Wim Van Sebroeck <wim@iguana.be>
Cc: Wan ZongShun <mcuos.com@gmail.com>
Cc: Ben Dooks <ben-linux@fluff.org>
Cc: Kukjin Kim <kgene.kim@samsung.com>
Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Pointers should not be compared to plain integers.
Quiets the sparse warning:
warning: Using plain integer as NULL pointer
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
Use the current logging styles.
Make sure all output has a prefix.
Add missing newlines.
Remove now unnecessary PFX, NAME, and miscellaneous other #defines.
Coalesce formats.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
Stefano found SP5100 TCO watchdog driver using wrong address.
[ 9.148536] SP5100 TCO timer: SP5100 TCO WatchDog Timer Driver v0.01
[ 9.148628] DEBUG __ioremap_caller WARNING address=b8fe00 size=8 valid=1 reserved=1
and e820 said that range is RAM.
We should check if we can use that reading out. BIOS could just program wrong address there.
Reported-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Signed-off-by:Yinghai Lu <yinghai@kernel.org>
Acked-by: Mike Waychison <mikew@google.com>
Tested-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
Cc: stable <stable@kernel.org>
This driver adds /dev/watchdog support for the AMD sp5100 aka SB7x0 chipsets.
It follows the same conventions found in other /dev/watchdog drivers.
Signed-off-by: Priyanka Gupta <priyankag@google.com>
Signed-off-by: Mike Waychison <mikew@google.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
