Currently all devices are resumed at system resume time, including any
that were individually powered off ("at runtime") prior to the system
suspend.  In certain cases it can be nice to force back on individually
suspended devices, such as the display, but hopefully this policy can be
left up to userspace power managers; the kernel should probably honor
the settings previously made by userspace/drivers.  This seems
preferable to requiring a power-conscious system to re-suspend devices
after a system resume; furthermore, for certain platforms (such as
XScale PXA27X) there can be disastrous consequences of powering up
devices when the system is in a state incompatible with operation of the
device.

Suggested patch does this:

(1) At system resume, checks power_state to see if the device was
suspended prior to system suspend, and skips powering on the device if
so.

(2) Does not re-suspend an already-suspended device at system suspend
(using a different method than is currently employed, which reorders the
list, see #3).

(3) Preserves the active/off device list order despite the above changes
to suspend/resume behavior, to avoid dependency problems that tend to
occur when the list is reordered.
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

(1) Set device power state at runtime resume (as is done for runtime
suspend) so that a later suspend does not think the device is still
suspended (refusing to suspend it again).

(2) Move devices from the active list to the off list only when
suspending all devices as part of a system suspend, not for runtime
suspend.  This matches the resume code, which only moves devices from
off to active during system resume, such that runtime resume currently
doesn't move the suspended device back to the active list.  (This also
avoids reordering the device list for runtime suspends; the list is in
order of registration and suspend/resume works best that way -- granted,
more sweeping improvements in how device dependencies are accounted for
in the suspend/resume order are also needed someday.)

Runtime device suspend/resume is in some cases used frequently on
battery-powered embedded devices, to save additional power and to handle
device power state interactions with overall system power state on
certain platforms.
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

You can do the same from userspace, and the author requests that the
driver be deleted from the kernel tree before people start using it.

- Rediffed the patch for 2.6.6-mm3 to fix rejects in the latest changes
  in sysfs code.

o The symlinks code in sysfs doesnot point to the correct target kobject
  whenever target kobject is renamed and suffers from dangling symlinks
  if target kobject is removed.

o The following patch implements ->readlink and ->follow_link operations
  for sysfs instead of using the page_symlink_inode_operations.
  The pointer to target kobject is saved in the link dentry's d_fsdata field.
  The target path is generated everytime we do ->readlink and ->follow_link.
  This results in generating the correct target path during readlink and
  follow_link operations inspite of renamed target kobject.

o This also pins the target kobject during link creation and the ref. is
  released when the link is removed.

o Apart from being correct this patch also saves some memory by not pinning
  a whole page for saving the target information.

o Used a rw_semaphor sysfs_rename_sem to avoid clashing with renaming of
  ancestors while the target path is generated.

o Used dcache_lock in fs/sysfs/sysfs.h:sysfs_get_kobject() because of using
  d_drop() while removing dentries.

into ppc970.osdl.org:/home/torvalds/v2.6/linux

ata_scsi_translate() sets this flag for all ATA->SCSI translated
commands, so it need not be done in atapi_xlat().  The now-removed
use in atapi_xlat() was also inconsistent WRT PIO versus DMA.

DMADIR bit is necessary for some PATA->SATA bridges.  These
bridges require the OS driver to specify the data xfer direction,
for PACKET (a.k.a. scsi) commands.

A reliable DMADIR detection method hasn't yet been developed,
and ATAPI is still a WIP, so DMADIR is enabled with an ifdef for
now.

into ppc970.osdl.org:/home/torvalds/v2.6/linux

Since all boards specify base_baud, the code to default base_baud to
the architecture-defined BASE_BAUD is redundant.  Also, defaulting
to the architecture-defined BASE_BAUD is wrong since the UARTs on a
serial PCI card will be clocked at the same rate no matter what the
architecture of the host machine.

into ppc970.osdl.org:/home/torvalds/v2.6/linux

This eliminates the last SWI user from the kernel - now all SWI calls
will only come from userspace.  More importantly, this also allows us
to empty the kernel stack when starting userspace programs from
kernelspace, thereby ensuring that the user registers always appear
at the top of the kernel stack.

This moves a bunch of EXPORT_SYMBOL() statements from armksyms.c
into the file which defines the function/variable such that the
exports are localised.

This also means we can get rid of the ugly __xxx_to_xxx__is_a_macro
in include/asm-arm/arch-*/memory.h

into ppc970.osdl.org:/home/torvalds/v2.6/linux

into redhat.com:/spare/repo/net-drivers-2.6

No more users of the autoirq_xxx() API existed, so this file is not
only unused, it isn't even listed in any makefiles.

This fixes the gfp_mask setting on the quota inode.

into redhat.com:/spare/repo/net-drivers-2.6

This patch makes the b44-after-bcm4400 scenario work for
me. What was happening is that the broadcom driver sets a "power off MAC"
bit, and we didn't remove that when initializing the chip.
Also added some (a bit ugly, I know  ) logic to clear up the address
filter stuff, which is what recent broadcom drivers do...

It's easier to do it this way, than polling, at the moment.

Also, fix a test in ata_scsi_translate that was incorrectly
erroring-out non-data commands.

* now that ATAPI is close to working, making ATAPI DMA interrupts
  in ata_host_intr
* remove unnecessary space character in printk() output (oh, the horror)

From: Ivan Kokshaysky <ink@jurassic.park.msu.ru>

This skips the GP-loading function prologue (two instructions: 8 bytes)
on BRSGP linkage correctly, fixing an oops on alpha while loading the
aic7xxx driver.

mapped memory.

A regular "memset()" may be using cache control
instructions etc, which is not appropriate for
memory-mapped IO.

This also fixes a warning.

into ppc970.osdl.org:/home/torvalds/v2.6/linux

Patch from Robin Farine

Fix the xscale cache handling routines that were invalidating a D 
cache line instead of draining the write & fill buffer as intended.

Patch from Catalin Marinas

Unlike the v5 architecture, the ARM1136 requires that BIT4 is 0 in
the first level page descriptor (ARM1136 TRM, page 6-39). It works
at the moment but it might break future v6 cores.

Patch from Catalin Marinas

The v6_dma_invalidate_range - the "mcr" instruction for draining

the write buffer requires r0 == 0. A "cmp" instruction for

testing the end address is missing in the v6_dma_flush_range

function.

Patch from Catalin Marinas

The _raw_write_(try)lock functions in include/asm-arm/spinlock.h

should use the early clobber modifier (&) for the "tmp" register.

A newer compiler (gcc-3.4.0) generates an "strexeq %0, %1, [%2]"

instruction where %0 is the same as %2, which is illegal.

Patch from Catalin Marinas

In the blockops_check() function, cache_type is uninitialised

because an "mcr" instruction is used instead of "mrc".

into ppc970.osdl.org:/home/torvalds/v2.6/linux

into ppc970.osdl.org:/home/torvalds/v2.6/linux

This fixes the problem with recursion into filesystem when inode of
quota file needs a page + some other allocation problems.  I hope I got
the GFP mask setting right..

into ppc970.osdl.org:/home/torvalds/v2.6/linux

Remove unused 'done_late' arg to ata_qc_complete(), which was never
useful in 2.4, and never used at all in 2.6.

This allows us to eliminate the same arg from ata_dma_complete(),
and also make it more correct by passing the command rather than
the ATA port structure as arg0.

into redhat.com:/spare/repo/libata-2.6

We need to always allocate at least one indirect block
pointer, since we always fill out blocks[0] even if
we don't have any groups.

Trivia.

CMD640 driver doesn't use generic IDE PCI code (it doesn't even include
this header).

- many Maxtor disks incorrectly claim CACHE FLUSH EXT command support,
  fix it by checking both CACHE FLUSH EXT command and LBA48 support
  (thanks to Eric D. Mudama for help in fixing this)

- write_cache() was called with 'drive->id->cfs_enable_2 & 0x3000' as 'int arg'
  argument which was always truncated to zero due to 'u8 drive->wcache = arg'
  assignment so write cache was indeed enabled but drive->wcache was zero
  (thanks to Rene Herman for help in debugging this)

- flush cache in idedisk_start_power_step() only if ATA-6 CACHE FLUSH (EXT)
  bits are present in disk's identify data (prevents sending unknown commands)

- set drive->wcache in idedisk_setup() not idedisk_attach() (no need to check
  id->command_set_2 - we check id->cfs_enable_2 instead in write_cache() call)

- use ide_cacheflush_p() in idedisk_setup()

- minor cleanups

This code is shared between i386 and x86-64, and x86-64 needs to check
for empty nodes here.  Otherwise you can get oopses at boot in some
circumstances. 

This handles empty nodes != 0; empty node zero are still broken in other
ways.

This changes the /dev/mem caching behaviour on x86-64 to be compatible
with i386.

By default everything is set cached.

This actually makes WC MTRRs on AMD systems work, which would get
overriden by the UC PAT bits that were set earlier.  This can make DVD
decoding with hardware support a lot faster. 

It also supports O_SYNC now, like i386, although that is not really
safe, because it allows the user to create undefined cache attribute
conflicts that can corrupt caches in some circumstances.  I kept it for
now.  Better would to disallow it, until Terrence Ripperda's PAT
framework is getting merged, that can avoid these problems. 

Actually it would be probably a good idea to add a printk here to catch
broken programs for i386 and x86-64, but that is for another patch.