This is a first step in allowing people to by-pass udev for loading
device firmware.  Current versions of udev will deadlock (causing us to
block for the 30 second timeout) under some circumstances if the
firmware is loaded as part of the module initialization path, and this
is causing problems for media drivers in particular.

The current patch hardcodes the firmware path that udev uses by default,
and will fall back to the legacy udev mode if the firmware cannot be
found there.  We'd like to add support for both configuring the paths
and the fallback behaviour, but in the meantime this hopefully fixes the
immediate problem, while also giving us a way forward.

[ v2: Some VFS layer interface cleanups suggested by Al Viro ]
[ v3: use the default udev paths suggested by Kay Sievers ]
Suggested-by: NIvan Kalvachev <ikalvachev@gmail.com>
Acked-by: NGreg KH <gregkh@linuxfoundation.org>
Acked-by: NAl Viro <viro@zeniv.linux.org.uk>
Cc: Mauro Carvalho Chehab <mchehab@redhat.com>
Cc: Kay Sievers <kay@redhat.com>
Cc: Ming Lei <ming.lei@canonical.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Explicitly test for GLOBAL_ROOT_UID and GLOBAL_ROOT_GID
instead of using 0, allowing dgrp to compile and work
properly when user namespace support is enabled.

Cc: Bill Pemberton <wfp5p@virginia.edu>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: N"Eric W. Biederman" <ebiederm@xmission.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Details of the problem, and solution, are in comments in the commit
proper.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Hopefully fixed the tlb flush timeout issue.  Was able to observe this
condition occur occasionally, and it appears the binary driver doesn't
wait on the old condition either..

Should give 39-bit DMA addressing on the relevant chipsets.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Something seems to be missing in regards to flushing specific ranges of
the TLB.  For the moment, flushing the entire thing seems to make it
work alright.

Should give 39-bit DMA addressing on the relevant chipsets.

v2: allocate contig 16KiB for dummy pages, reported by mwk on irc
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

We don't need to pull the page address out of the page tables on nv4x
chips that have a real GART.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

For now, only 2 control modes are available:
- NONE: The fan is never touched (default)
- MANUAL: The fan is set to the user-defined fan speed (pwm1)

This patch introduces a distinction between ptherm internal fan management
and external fan management. The latter is bound to respect the fan mode
while the first can still select the speed it wants unless the NONE mode
is selected. This is important for automatic fan management.
Signed-off-by: NMartin Peres <martin.peres@labri.fr>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

This was reported by tizbac on IRC.
Signed-off-by: NMartin Peres <martin.peres@labri.fr>
Reviewed-by: NBen Skeggs <bskeggs@redhat.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

v2: Martin Peres <martin.peres@labri.fr>
- fixed unintentional use of floating point
Signed-off-by: NMartin Peres <martin.peres@labri.fr>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

The previous driver waited for 250ms to accumulate data. This version times a
complete fan rotation and extrapolates to RPM.

The fan rotational speed should now be read in less than 250ms (worst case)
and usually in less 50ms.
Signed-off-by: NMartin Peres <martin.peres@labri.fr>
Reviewed-by: NBen Skeggs <bskeggs@redhat.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

It looks scary because of the size, but I tried to keep the differences minimal.
Further patches will fix the actual "driver" code and add new features.

v2: change filenames, split to submodules

v3: add a missing include

v4: Ben Skeggs <bskeggs@redhat.com>
- fixed set_defaults() to allow min_duty < 30 (thermal table will
  override this if it's actually necessary)
- fixed set_defaults() to not provide pwm_freq so nv4x (which only has
  pwm_div) can actually work.  the boards using pwm_freq will have a
  thermal table entry to provide us the value.
- removed unused files
Signed-off-by: NMartin Peres <martin.peres@labri.fr>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

v2: perf_table now is more in line with the other functions
Signed-off-by: NMartin Peres <martin.peres@labri.fr>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

This commit also adds a static list of all known devices and their possible
i2c addresses.

v2: use the common table parsing technique as suggested by darktama
Signed-off-by: NMartin Peres <martin.peres@labri.fr>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

As an accident, it should also fix temperature reading on nv4x.

v2: introduce nvbios_therm_entry as advised by darktama
Signed-off-by: NMartin Peres <martin.peres@labri.fr>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NMartin Peres <martin.peres@labri.fr>
Reviewed-by: NBen Skeggs <bskeggs@redhat.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

If nouveau_pm_perflvl_get() fails, pm->profiles list will be left
uninitialized, which causes oops during nouveau_pm_fini().

Move INIT_LIST_HEAD before call to nouveau_pm_perflvl_get().
Signed-off-by: NDmitry Eremin-Solenikov <dmitry_eremin@mentor.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

This will make it more obvious which application caused particular messages.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

May kill the DRM version completely at some point, undecided..
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Can be somewhat more informative that way...
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Seems to be required to "re-arm" the engines after a vm fault.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Not really sure how to confirm this 100%, but, the numbers match on all the
traces I have for NVCx (2 LTS), NVD9 (1LTS) and NVEx (4LTS).
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Graphics acceleration is still disabled by default due to lingering issues
that need to be solved.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Triggers PIBUS interrupts due to register not existing anymore, and as
a result HUB_SET_CHAN times out.

After this commit, our fuc loads and can accelerate at least fbcon, X,
glxgears and OA on NVE4.  NVE7 not tested as of yet.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Currently identical except the available chipset register lists.  This will
*not* currently work and is disabled by default because of this.

May get merged again later, remains to be seen what further changes will be
required.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Came in useful for debugging another issue earlier, so keep it around.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NMarcin Slusarz <marcin.slusarz@gmail.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NMarcin Slusarz <marcin.slusarz@gmail.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Will probably flesh the documentation of the classes out a bit too at some
later point.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>