The moves themselves were generally async to graphics previously, with
the exception that if the "main" channel is used to synchronise a
page flip at the same time, it can end up blocked for a noticable amount
of time for large buffer moves.

Not really critical, and there's better ways of handling this, but they
are all rather invasive, so this is fine for now.

Based on a patch by Maarten Lankhorst addressing the same issue.
Reported-by: NMaarten Lankhorst <maarten.lankhorst@canonical.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
Acked-by: NMaarten Lankhorst <maarten.lankhorst@canonical.com>

This should no longer be required, and is harmful for framebuffer pinning.
Also add a warning if unpin causes the pin count to drop below 0.
Signed-off-by: NMaarten Lankhorst <maarten.lankhorst@canonical.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

b580c9e2 introduced additional problems
while trying to solve issues that became apparent while porting to the
new reservation stuff.

The major problem was that the the previously mentioned patch took the
client mutex earlier than previously, but the pinning of new_bo can
can potentially cause a buffer move, which would result in attempting to
acquire the same mutex again.

This commit attempts to fix that "fix".

Thanks to Maarten for the tips on keeping lockdep happy and cooking :)
Reported-by: NMaarten Lankhorst <maarten.lankhorst@canonical.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
Acked-by: NMaarten Lankhorst <maarten.lankhorst@canonical.com>

Reported-by: NMaarten Lankhorst <maarten.lankhorst@canonical.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

As opposed to an explicit wait.  Allows userspace to not stall waiting
on buffer deletion.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

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

Currently used only on original nv50, nvaa and nvac.
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>

On tile architecture (with "make allyesconfig") including
<linux/swiotlb.h> is required to call swiotlb_nr_tbl().
Signed-off-by: NChris Metcalf <cmetcalf@tilera.com>
Acked-by: NMaarten Lankhorst <maarten.lankhorst@canonical.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

On tile architecture (with "make allyesconfig") including
<linux/swiotlb.h> is required to call swiotlb_nr_tbl().
Signed-off-by: NChris Metcalf <cmetcalf@tilera.com>
Acked-by: NMaarten Lankhorst <maarten.lankhorst@canonical.com>

... by moving the bo_pin/bo_unpin manipulation of the pin_refcount
under the protection of the ttm reservation lock. pin/unpin seems
to get called from all over the place, so atm this is completely racy.

After this patch there are only a few places in cleanup functions
left which access ->pin_refcount without locking. But I'm hoping that
those are safe and some other code invariant guarantees that this
won't blow up.

In any case, I only need to fix up pin/unpin to make ->pageflip work
safely, so let's keep it at that.

Add a comment to the header to explain the new locking rule.
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Check that the AGP aperture can be mapped. This follows a similar change
done for Radeon (commit 365048ff, drm/radeon: AGP memory is only I/O if
the aperture can be mapped by the CPU.).

The patch fixes the following error seen on G5 iMac:

	nouveau E[     DRM] failed to create kernel channel, -12

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=58806
Cc: stable@vger.kernel.org
Reviewed-by: NMichel Dänzer <michel@daenzer.net>
Signed-off-by: NAaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: NDave Airlie <airlied@redhat.com>

All items on the lru list are always reservable, so this is a stupid
thing to keep. Not only that, it is used in a way which would
guarantee deadlocks if it were ever to be set to block on reserve.

This is a lot of churn, but mostly because of the removal of the
argument which can be nested arbitrarily deeply in many places.

No change of code in this patch except removal of the no_wait_reserve
argument, the previous patch removed the use of no_wait_reserve.

v2:
 - Warn if -EBUSY is returned on reservation, all objects on the list
   should be reservable. Adjusted patch slightly due to conflicts.
v3:
 - Focus on no_wait_reserve removal only.
Signed-off-by: NMaarten Lankhorst <maarten.lankhorst@canonical.com>
Reviewed-by: NThomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

Signed-off-by: NMaarten Lankhorst <maarten.lankhorst@canonical.com>
Reviewed-By: NThomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

Signed-off-by: NMaarten Lankhorst <maarten.lankhorst@canonical.com>
Reviewed-By: NThomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

All drivers set it to 0 and nothing uses it.
Signed-off-by: NMarcin Slusarz <marcin.slusarz@gmail.com>
Reviewed-by: NThomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

ttm_agp_tt_create is itself defined under CONFIG_AGP, so there's no
point calling it otherwise.
Signed-off-by: NMax Filippov <jcmvbkbc@gmail.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Kepler PFIFO lost the ability to address multiple engines from a single
channel, so we need a separate one for the copy engine.

v2: Marcin Slusarz <marcin.slusarz@gmail.com>
- regression fix: restore hw accelerated buffer copies
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

v2: Ben Skeggs <bskeggs@redhat.com>
- fill in nouveau_pm.dev to prevent oops
- fix ppc issues (build + OF shadow)
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

This is a HUGE commit, but it's not nearly as bad as it looks - any problems
can be isolated to a particular chipset and engine combination.  It was
simply too difficult to port each one at a time, the compat layers are
*already* ridiculous.

Most of the changes here are simply to the glue, the process for each of the
engine modules was to start with a standard skeleton and copy+paste the old
code into the appropriate places, fixing up variable names etc as needed.

v2: Marcin Slusarz <marcin.slusarz@gmail.com>
- fix find/replace bug in license header

v3: Ben Skeggs <bskeggs@redhat.com>
- bump indirect pushbuf size to 8KiB, 4KiB barely enough for userspace and
  left no space for kernel's requirements during GEM pushbuf submission.
- fix duplicate assignments noticed by clang

v4: Marcin Slusarz <marcin.slusarz@gmail.com>
- add sparse annotations to nv04_fifo_pause/nv04_fifo_start
- use ioread32_native/iowrite32_native for fifo control registers

v5: Ben Skeggs <bskeggs@redhat.com>
- rebase on v3.6-rc4, modified to keep copy engine fix intact
- nv10/fence: unmap fence bo before destroying
- fixed fermi regression when using nvidia gr fuc
- fixed typo in supported dma_mask checking
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

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

v2 (Ben Skeggs):
- some fixes for 64KiB PAGE_SIZE
- fix porting issues in (currently unused) nv41/nv44 pciegart code
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

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

These currently just call the existing ones in nouveau_drv.c, but will be
extended in upcoming commits.  This needed to be separated from the current
code as there will be some header clashes until things are ported.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Future work will be headed in the way of separating the policy supplied by
the nouveau drm module from the mechanisms provided by the driver core.

There will be a couple of major classes (subdev, engine) of driver modules
that have clearly defined tasks, and the further directory structure change
is to reflect this.

No code changes here whatsoever, aside from fixing up a couple of include
file pathnames.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Convert #include "..." to #include <path/...> in drivers/gpu/.
Signed-off-by: NDavid Howells <dhowells@redhat.com>
Acked-by: NDave Airlie <airlied@redhat.com>
Acked-by: NArnd Bergmann <arnd@arndb.de>
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Acked-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: NDave Jones <davej@redhat.com>

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

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

Disabled for the moment until some performance issues are sorted out, code
committed as a reference 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>

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

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

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

This adds prime->fd and fd->prime support to nouveau,
it passes the SG object to TTM, and then populates the
GART entries using it.

v2: add stubbed kmap + use new function to fill out pages array
for faulting + add reimport test.
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

This adds the ability for ttm common code to take an SG table
and use it as the backing for a slave TTM object.

The drivers can then populate their GTT tables using the SG object.

v2: make sure to setup VM for sg bos as well.
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NJerome Glisse <jglisse@redhat.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

This seems to be wrong to me, spotted while thinking about dma-buf.
Reviewed-by: NBen Skeggs <bskeggs@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: NDave Airlie <airlied@redhat.com>

There was once good reasons for wanting the drm to be able to use M2MF etc
on user channels, but they're not relevant anymore.  For the general
buffer move case, we've already lost by transferring between vram/sysmem
already so the context switching overhead is minimal in comparison.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Both changes in dc97b340 cause serious
regressions in the nouveau driver.

move_notify() was originally able to presume that bo->mem is the old node,
and new_mem is the new node.  The above commit moves the call to
move_notify() to after move() has been done, which means that now, sometimes,
new_mem isn't the new node at all, bo->mem is, and new_mem points at a
stale, possibly-just-been-killed-by-move node.

This is clearly not a good situation.  This patch reverts this change, and
replaces it with a cleanup in the move() failure path instead.

The second issue is that the call to move_notify() from cleanup_memtype_use()
causes the TTM ghost objects to get passed into the driver.  This is clearly
bad as the driver knows nothing about these "fake" TTM BOs, and ends up
accessing uninitialised memory.

I worked around this in nouveau's move_notify() hook by ensuring the BO
destructor was nouveau's.  I don't particularly like this solution, and
would rather TTM never pass the driver these objects.  However, I don't
clearly understand the reason why we're calling move_notify() here anyway
and am happy to work around the problem in nouveau instead of breaking the
behaviour expected by other drivers.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
Reviewed-by: NThomas Hellstrom <thellstrom@vmware.com>
Cc: Jerome Glisse <j.glisse@gmail.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>