Looks like someone got this backwards, highlighting the perils of the
? : !!! :)
Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Handle recursive oopses, like on x86. We had a few cases recently where
we locked up in oops printing and didnt make it into crashdump.
Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Move pmac_backlight_unblank into its own function and only take the
pmac_backlight_mutex when we are on a pmac for that added bit of
paranoia.
Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Add missing oops_enter/oops_exit, makes pause_on_oops boot parameter work.
Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Remove last_syscall from 32bit powerpc, its been gone in 64bit for years.
Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

We already have an inline __get_SP, no need for yet another one.
Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Rework how multi-function PCI devices are identified and traversed.
This fixes a bug with multi-function recovery on Power4 that was
introduced by a recent Power4 EEH patch.
Signed-off-by: NLinas Vepstas <linas@austin.ibm.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

After requesting a state change, verify that the state change
actually ocurred, and the system ends up in the expected state.
Signed-off-by: NLinas Vepstas <linas@austin.ibm.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

The EEH event notification system passes around data that is
not needed or at least, not used properly. Stop passing this
data; get it in a more reliable fashion.
Signed-off-by: NLinas Vepstas <linas@austin.ibm.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Modify routine that returns PCI slot status to wait for slot status
to become available. This is needed, as slots that are in some remote
card cage may go offline for extended periods of time. New users for
this routine in following patches.
Signed-off-by: NLinas Vepstas <linas@austin.ibm.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Some firmware versions will return a slot reset state of "1"
when a slot is EEH frozen. Recognize this as a state that can be
handled.
Signed-off-by: NLinas Vepstas <linas@austin.ibm.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

If the second or higher function of a multi-function device fails
to recover, this failure is not reported upwards. Fix this.
Signed-off-by: NLinas Vepstas <linas@austin.ibm.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

If a device driver does not have native PCI error recovery,
a hotplug error recovery will be attemped. In this case,
the device driver will not report back whether its healthy
or not; simply assume that it is.
Signed-off-by: NLinas Vepstas <linas@austin.ibm.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Provide support for the new ibm,get-config-addr-info2 RTAS token,
whenever it is actually available.
Signed-off-by: NLinas Vepstas <linas@austin.ibm.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Some drivers will attempt to perform a lot of mmio even after
an EEH event was detected. This is especially the case for fast cpu's
and PCI-E slots. Be a bit more lenient in allowing this.
Signed-off-by: NLinas Vepstas <linas@austin.ibm.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

There are multiple code patchs tht resuls in a "permanent
failure"; when examining rare events, it can be hard to see
which was taken. This patch adds printk's to assist.
Signed-off-by: NLinas Vepstas <linas@austin.ibm.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Change the order in which pci error state is examined;
the "capabilites" is not valid if "reset state" is 5.
Signed-off-by: NLinas Vepstas <linas@austin.ibm.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Move the slb_shadow_ptr field into the first cache line since it is
(like everything there) read-only after boot.  It is in fact statically
initialised and thereafter only read.
Signed-off-by: NStephen Rothwell <sfr@canb.auug.org.au>
Acked-by: NMichael Neuling <mikey@neuling.org>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Move the declaration of flush_cache to ops.h for use by platform code.
Signed-off-by: NMilton Miller <miltonm@bga.com>
Acked-by: NDavid Gibson <david@gibson.dropbear.id.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Since there is magic defined per platform in the wrapper script, the
zImage targets should depend on it.
Signed-off-by: NMilton Miller <miltonm@bga.com>
Acked-by: NDavid Gibson <david@gibson.dropbear.id.au>
Acked-by: NSegher Boessenkool <segher@kernel.crashing.org>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Signed-off-by: NScott Wood <scottwood@freescale.com>
Acked-by: NDavid Gibson <david@gibson.dropbear.id.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

To allow more robust association of each network device node with an
index (such as is used by the firmware or an EEPROM to indicate MAC
addresses), a network device's node may specify the index explicitly.
Signed-off-by: NScott Wood <scottwood@freescale.com>
Acked-by: David Gibson david@gibson.dropbear.id.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

The --no-gzip option can be passed to the wrapper so that the kernel
image is included uncompressed into the zImage.  This is intended for
bootloaders where the zImage itself can be compressed, or where boot time
is considered more important than kernel image size.
Signed-off-by: NScott Wood <scottwood@freescale.com>
Acked-by: NDavid Gibson <david@gibson.dropbear.id.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Add get_parent, create_node, and find_node_by_prop_value to dt_ops.
Currently only implemented by flatdevtree_misc.

Also, add a _str convenience wrapper for setprop.
Signed-off-by: NScott Wood <scottwood@freescale.com>
Acked-by: NDavid Gibson <david@gibson.dropbear.id.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Correct the apparent misspelling of "XMON" to "CONFIG_XMON".
Signed-off-by: NRobert P. J. Day <rpjday@mindspring.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

This patch adds documenting comments to the gunzip convenience
functions added in commit ad9d2716.
It also removes a stray newline, and an unused global variable.
Signed-off-by: NDavid Gibson <david@gibson.dropbear.id.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

This patch updates booting-without-of.txt to describe version 17 of
the flattened device tree format.  Version 17 is a small, backwards
compatible change from version 16, adding an extra field giving the
size of the device tree's structure block.  At this time, the kernel
has no use for the extra information, however its presence can make
life easier for bootloaders or other software manipulating the tree.

In addition this patch adds information on the size_dt_strings field
of the device tree header, present since version 3 of the flattened
tree format, but omitted from the documentation.  It also makes
changes to consistently refer to versions 16 and 17 as versions 16 and
17 in decimal, rather than version 0x10 which was occasionally used
for version 16 previously.

Finally, we also add the new field to the definition of the device
tree header structure in prom.h
Signed-off-by: NDavid Gibson <david@gibson.dropbear.id.au>
Acked-by: NJon Loeliger <jdl@freescale.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

This balances parenthesis in powerpc 8xx header files.
Signed-off-by: NMariusz Kozlowski <m.kozlowski@tuxland.pl>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

This adds two sysfs attributes to /sys/bus/ibmebus which can be used to
notify the ebus driver of added / removed ebus devices in the OF device
tree.

Echoing the device's location code (as found in the OFDT "ibm,loc-code"
property) into the "probe" attribute will notify ebus of addition of the
device and cause the appropriate device driver's probe function to be called
on the device.

Likewise, echoing the location code into the "remove" attribute will cause
the device to be removed from the system.

The writes will block until the respective operation has finished and return
an error code if the operation failed.

In addition, two minor tidbits are fixed:

- The fake root device used to provide a common parent for all ebus devices
  is now based on device instead of of_device - it had no associated devtree
  node. This saves several checks throughout the ebus driver.

- The sysfs attributes are now generated automagically by device_register()
  instead of by the ibmebus code, which saves a few compiler warnings about
  unused return codes.
Signed-off-by: NJoachim Fenkes <fenkes@de.ibm.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

This fixes a lot of whitespace in ibmebus.[ch]
Signed-off-by: NJoachim Fenkes <fenkes@de.ibm.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Currently, early_init() in setup_32.c zeroes from '_bss_start' to '_end'.
It should only zero from '__bss_start' to '__bss_stop'.  This patch does that.
Signed-off-by: NMark A. Greer <mgreer@mvista.com>
Acked-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Signed-off-by: NScott Wood <scottwood@freescale.com>
Acked-by: NDavid Gibson <david@gibson.dropbear.id.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

ft_find_node_by_prop_value() finds nodes with the specified
property/value pair.
Signed-off-by: NScott Wood <scottwood@freescale.com>
Acked-by: NMark A. Greer <mgreer@mvista.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Most of ft_get_parent() is factored out into __ft_get_parent(), which
deals only in internal node pointers.  The ft_get_parent() wrapper
handles phandle conversion in both directions (previously,
ft_get_parent() did not convert its return value).

It also now returns NULL as the parent of the toplevel node, rather than
just returning the toplevel node again (which made it rather useless in
loops).
Signed-off-by: NScott Wood <scottwood@freescale.com>
Acked-by: NMark A. Greer <mgreer@mvista.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

The property searching part of ft_get_prop is factored out into an
internal __ft_get_prop() which does not deal with phandles and does not
copy the property data.  ft_get_prop() is then a wrapper that does the
phandle translation and copying.
Signed-off-by: NScott Wood <scottwood@freescale.com>
Acked-by: NMark A. Greer <mgreer@mvista.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Add a function to look up a relative, rather than absolute, path name.
Signed-off-by: NScott Wood <scottwood@freescale.com>
Acked-by: NMark A. Greer <mgreer@mvista.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

When adding a property, the property name should be added to the string
table if it doesn't already exist.  map_string() does that;
lookup_string() will fail instead.
Signed-off-by: NScott Wood <scottwood@freescale.com>
Acked-by: NDavid Gibson <david@gibson.dropbear.id.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Move the caller's pointer back to match the change in the region's start,
rather than alter a byte of the device tree's content.
Signed-off-by: NScott Wood <scottwood@freescale.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

The ft_reorder() function may change the start of the region of interest,
so the pointer provided by the caller into that region must be fixed up
to still point to the same datum.
Signed-off-by: NScott Wood <scottwood@freescale.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Currently, if ft_get_phandle() is passed NULL it will allocate an entry
for it and return a non-NULL phandle.  This patch makes it simply pass
the NULL through.
Signed-off-by: NScott Wood <scottwood@freescale.com>
Acked-by: NMark A. Greer <mgreer@mvista.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>