So far the assumption was that each dma scatter list entry contains only
a single page. This might not hold in the future, when we'll introduce
compact scatter lists, so prepare for this everywhere in the i915 code
where we walk such a list.

We'll fix the place _creating_ these lists separately in the next patch
to help the reviewing/bisectability.

Reference: http://www.spinics.net/lists/dri-devel/msg33917.htmlSigned-off-by: NImre Deak <imre.deak@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

The purpose of the gtt structure is to help isolate our gtt specific
properties from the rest of the code (in doing so it help us finish the
isolation from the AGP connection).

The following members are pulled out (and renamed):
gtt_start
gtt_total
gtt_mappable_end
gtt_mappable
gtt_base_addr
gsm

The gtt structure will serve as a nice place to put gen specific gtt
routines in upcoming patches. As far as what else I feel belongs in this
structure: it is meant to encapsulate the GTT's physical properties.
This is why I've not added fields which track various drm_mm properties,
or things like gtt_mtrr (which is itself a pretty transient field).
Reviewed-by: NRodrigo Vivi <rodrigo.vivi@gmail.com>
[Ben modified commit messages]
Signed-off-by: NBen Widawsky <ben@bwidawsk.net>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Signed-off-by: NImre Deak <imre.deak@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

The two functions are rather similar, so merge them.
Signed-off-by: NImre Deak <imre.deak@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

On a machine with bit17 swizzling, we need to store the bit17 of the
physical page address in put-pages. This requires a memory allocation,
on average less than a page, which may be difficult to satisfy is the
request to put-pages is on behalf of the shrinker. We could allow that
allocation to pull from the reserved memory pools, but it seems much
safer to preallocate the array for tiled objects on affected machines.

v2: Export i915_gem_object_needs_bit17_swizzle() for reuse.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

We don't have bit 6 swizzling on VLV, so this function is easy.
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

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>

Remove redundant DRM UAPI header #inclusions from drivers/gpu/.

Remove redundant #inclusions of core DRM UAPI headers (drm.h, drm_mode.h and
drm_sarea.h).  They are now #included via drmP.h and drm_crtc.h via a preceding
patch.

Without this patch and the patch to make include the UAPI headers from the core
headers, after the UAPI split, the DRM C sources cannot find these UAPI headers
because the DRM code relies on specific -I flags to make #include "..."  work
on headers in include/drm/ - but that does not work after the UAPI split without
adding more -I flags.
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>

Rather than have multiple data structures for describing our page layout
in conjunction with the array of pages, we can migrate all users over to
a scatterlist.

One major advantage, other than unifying the page tracking structures,
this offers is that we replace the vmalloc'ed array (which can be up to
a megabyte in size) with a chain of individual pages which helps reduce
memory pressure.

The disadvantage is that we then do not have a simple array to iterate,
or to access randomly. The common case for this is in the relocation
processing, which will typically fit within a single scatterlist page
and so be almost the same cost as the simple array. For iterating over
the array, the extra function call could be optimised away, but in
reality is an insignificant cost of either binding the pages, or
performing the pwrite/pread.

v2: Fix drm_clflush_sg() to not invoke wbinvd as well! And fix the
trivial compile error from rebasing.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

If we fail to unbind and so abort the change in tiling, we will have
removed the VMA for the object for no reason. The likelihood of unbind
failing is slim (other than ERESTARTSYS which will cause userspace to
try again), so the change is mostly for the principle.

Also improve the slightly stale comment.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Rename obj->tiling_changed to obj->fence_dirty so that it is clear that
it flags when the parameters for an active fence (including the
no-fence) register are changed.

Also, do not set this flag when the object does not have a fence
register allocated currently and the gpu does not depend upon the
unfence. This case works exactly like when a tiled object lost its
fence and hence does not need additional handling for the tiling
change in the code.

v2: Use fence_dirty to better express what the flag tracks and add a few
more details to the comments to serve as a reminder of how the GPU also
uses the unfenced register slot.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
[danvet: Add some bikeshed to the commit message about the stricter
use of fence_dirty.]
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

We have to do this manually. Somebody had a Great Idea.

I've measured speed-ups just a few percent above the noise level
(below 5% for the best case), but no slowdows. Chris Wilson measured
quite a bit more (10-20% above the usual snb variance) on a more
recent and better tuned version of sna, but also recorded a few
slow-downs on benchmarks know for uglier amounts of snb-induced
variance.

v2: Incorporate Ben Widawsky's preliminary review comments and
elaborate a bit about the performance impact in the changelog.

v3: Add a comment as to why we don't need to check the 3rd memory
channel.

v4: Fixup whitespace.
Acked-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NBen Widawsky <ben@bwidawsk.net>
Reviewed-by: NEric Anholt <eric@anholt.net>
Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

It looks like the desktop variants of i915 and i945 also have the DCC
register to control dram channel interleave and cpu side bit6
swizzling.

Unfortunately internal Cspec/ConfigDB documentation for these ancient chips
have already been dropped and there seem to be no archives. Also
somebody thought the swizzling behaviour is surely a worthy secret to
keep and redacted any mention of these fields from the published Intel
datasheets.

I suspect the hw engineers were really proud of the page coloring
they've achieved in their first dual channel dram controller with
bit17 - after all Bspec explains in great length the optimal layout of
page frame numbers modulo 4 for the color and depth buffers, too.
Later on when they've started to work on VT-d they shamefully
discoverd their stupidity and tried to cover the tracks ...

Tested-by: Daniel Vetter <daniel.vetter@ffwll.ch> (i915g)
Tested-by: Pavel Ondračka <pavel.ondracka@email.cz> (i945g)
Tested-by: NChris Wilson <chris@chris-wilson.co.uk>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=42625Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Use the helper function already employed by the pwrite/pread
functions.
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NKeith Packard <keithp@keithp.com>

Fixes tests/gem_tiled_pread on my snb. I know, mesa doesn't use this
on gen6+, but I also hate failing testcases.
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: NBen Widawsky <ben@bwidawsk.net>
Signed-off-by: NKeith Packard <keithp@keithp.com>

Align unfenced buffers on older hardware to the power-of-two object
size.  The docs suggest that it should be possible to align only to a
power-of-two tile height, but using the already computed fence size is
easier and always correct. We also have to make sure that we unbind
misaligned buffers upon tiling changes.

In order to prevent a repetition of this bug, we change the interface
to the alignment computation routines to force the caller to provide
the requested alignment and size of the GTT binding rather than assume
the current values on the object.
Reported-and-tested-by: NSitosfe Wheeler <sitsofe@yahoo.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=36326Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: stable@kernel.org
Reviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: NKeith Packard <keithp@keithp.com>

Treat it like Ironlake and Sandy Bridge.
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>
Reviewed-by: NKeith Packard <keithp@keithp.com>
Signed-off-by: NKeith Packard <keithp@keithp.com>

Early gen3 and gen2 chipset do not have the relaxed per-surface tiling
constraints of the later chipsets, so we need to check that the GTT
alignment is correct for the new tiling. If it is not, we need to
rebind.
Reported-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>

This reverts commit c2e0eb16.

As it turns out, userspace already depends upon being able to enable
tiling on existing bo which it promises to be large enough for its
purposes i.e. it will not access beyond the end of the last full-tile
row.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=35016Reported-and-tested-by: NKamal Mostafa <kamal@canonical.com>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>

It looks like gen2 has a peculiar interleaved 2-row inter-tile
layout. Probably inherited from i81x which had 2kb tiles (which
naturally fit an even-number-of-tile-rows scheme to fit onto 4kb
pages). There is no other mention of this in any docs (also not
in the Intel internal documention according to Chris Wilson).

Problem manifests itself in corruptions in the second half of the
last tile row (if the bo has an odd number of tiles). Which can
only happen with relaxed tiling (introduced in a00b10c3).

So reject set_tiling calls that don't satisfy this constrain to
prevent broken userspace from causing havoc. While at it, also
check the size for newer chipsets.

LKML: https://lkml.org/lkml/2011/2/19/5Reported-by: NIndan Zupancic <indan@nul.nu>
Tested-by: NIndan Zupancic <indan@nul.nu>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>

Dave Airlie spotted that we had a potential bug should we ever rearrange
the drm_i915_gem_object so not the base drm_gem_object was not its first
member. He noticed that we often convert the return of
drm_gem_object_lookup() immediately into drm_i915_gem_object and then
check the result for nullity. This is only valid when the base object is
the first member and so the superobject has the same address. Play safe
instead and use the compiler to convert back to the original return
address for sanity testing.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>

Instead of reporting EIO upfront in the entrance of an ioctl that may or
may not attempt to use the GPU, defer the actual detection of an invalid
ioctl to when we issue a GPU instruction. This allows us to continue to
use bo in video memory (via pread/pwrite and mmap) after the GPU has hung.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>

With this change, every batchbuffer can use all available fences (save
pinned and scanout, of course) without ever stalling the gpu!

In theory. Currently the actual pipelined update of the register is
disabled due to some stability issues. However, just the deferred update
is a significant win.

Based on a series of patches by Daniel Vetter.

The premise is that before every access to a buffer through the GTT we
have to declare whether we need a register or not. If the access is by
the GPU, a pipelined update to the register is made via the ringbuffer,
and we track the last seqno of the batches that access it. If by the
CPU we wait for the last GPU access and update the register (either
to clear or to set it for the current buffer).

One advantage of being able to pipeline changes is that we can defer the
actual updating of the fence register until we first need to access the
object through the GTT, i.e. we can eliminate the stall on set_tiling.
This is important as the userspace bo cache does not track the tiling
status of active buffers which generate frequent stalls on gen3 when
enabling tiling for an already bound buffer.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

A glorified s/obj_priv/obj/ with a net reduction of over a 100 lines and
many characters!
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>

As we may bind an object with the correct alignment, but with an invalid
size, it may pass the current checks on whether the object may be reused
with a fence.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>

So long as we adhere to the fence registers rules for alignment and no
overlaps (including with unfenced accesses to linear memory) and account
for the tiled access in our size allocation, we do not have to allocate
the full fenced region for the object. This allows us to fight the bloat
tiling imposed on pre-i965 chipsets and frees up RAM for real use. [Inside
the GTT we still suffer the additional alignment constraints, so it doesn't
magic allow us to render larger scenes without stalls -- we need the
expanded GTT and fence pipelining to overcome those...]
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>

So remove the redundant bit in the capabilities block and
s/IS_IRONLAKE/IS_GEN5/.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>

Owain Ainsworth reported an issue between the interaction of the
hangcheck and userspace immediately (and permanently) falling back to
s/w rasterisation. In order to break the mutex and begin resetting the
GPU, we must abort the current operation (usually within the wait) and
climb sufficiently far back up the call chain to drop the mutex. In his
implementation, Owain has a loop within the ioctl handler to detect the
hang and then sleep until the error handler has run. I've chosen to
return to userspace and report an EAGAIN which should trigger the
userspace ioctl handler to repeat the call (simply because it felt less
invasive...). Before hitting a wedged GPU, we then wait upon completion
of the error handler.
Reported-by: NOwain G. Ainsworth <zerooa@googlemail.com>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>

Avoid confusion between i965g meaning broadwater and the gen4+ chipset
families.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>

As we currently may need to acquire a fence register during a modeset,
we need to be able to do so in an uninterruptible manner. So expose that
parameter to the callers of the fence management code.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>

Our usage of kmap() cannot return NULL here, so remove the unnecessary
error handling.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>

This is consistent with trying to access a filename that not exist
within a directory which is a good analogy here. The main reason for the
change is that it is easy to confuse the error code of EBADF as an
performing an ioctl on an invalid file descriptor (rather than an
unknown object).
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDave Airlie <airlied@redhat.com>

We generally issue an error message at the point of failure, and so this
warning with a fairly pointless stacktrace is superfluous and ugly.
Needless to say, the common trigger for this WARN happens to be EIO
where this is pure noise.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NEric Anholt <eric@anholt.net>

Makes no sense and complicates matters for pipelined tiling changes.
So don't allow it and return -EBUSY.

v2: Fix reference leak. Thanks to Owain Ainsworth for spotting this.
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: NEric Anholt <eric@anholt.net>

Current code is definitely crap: Largest pitch allowed spills into
the TILING_Y bit of the fence registers ... :(

I've rewritten the limits check under the assumption that 3rd gen hw
has a 3d pitch limit of 8kb (like 2nd gen). This is supported by an
otherwise totally misleading XXX comment.

This bug mostly resulted in tiling-corrupted pixmaps because the kernel
allowed too wide buffers to be tiled. Bug brought to the light by the
xf86-video-intel 2.11 release because that unconditionally enabled
tiling for pixmaps, relying on the kernel to check things. Tiling for
the framebuffer was not affected because the ddx does some additional
checks there ensure the buffer is within hw-limits.

v2: Instead of computing the value that would be written into the
hw fence registers and then checking the limits simply check whether
the stride is above the 8kb limit. To better document the hw, add
some WARN_ONs in i915_write_fence_reg like I've done for the i830
case (using the right limits).
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=27449Tested-by: NAlexander Lam <lambchop468@gmail.com>
Cc: stable@kernel.org
Signed-off-by: NEric Anholt <eric@anholt.net>

This is a purely cosmetic change to make changes in this area easier.
And hey, it's not only clearer and typechecked, but actually shorter,
too!

[anholt: To clarify, this is a change to let us later make
drm_i915_gem_object subclass drm_gem_object, instead of having
drm_gem_object have a pointer to i915's private data]
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Acked-by: NDave Airlie <airlied@gmail.com>
Signed-off-by: NEric Anholt <eric@anholt.net>

Bug introduced in

commit 10ae9bd2
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Mon Feb 1 13:59:17 2010 +0100

    drm/i915: blow away userspace mappings before fence change

The problem is that when there's no fence reg assigned and the object
is mapped at a fenceable offset in the gtt, the userspace mappings won't
be torn down. Which happens on untiled->tiled transition quite often
on 4th gen and later because there fencing does not have any special
alignment constraints (as opposed to 2nd and 3rd gen on which I've tested
the original commit).

Bugzilla: http://bugs.freedesktop.org/show_bug.cgi?id=26993Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Tested-by: Eric Anholt <eric@anholt.net> (fixes OpenArena)
Signed-off-by: NEric Anholt <eric@anholt.net>

Signed-off-by: NEric Anholt <eric@anholt.net>

i915_gem_object_fenceable was mostly just a repeat of the
i915_gem_object_fence_offset_ok, but also checking the size (which was
checkecd when we allowed that BO to be tiled in the first place). So
instead, export the latter function and use it in place.
Signed-Off-By: NOwain G. Ainsworth <oga@openbsd.org>
Signed-off-by: NEric Anholt <eric@anholt.net>

This aligns it with the other user of i915_gem_clear_fence_reg,
which blows away the mapping before changing the fence reg.

Only affects userspace if it races against itself when changing
tiling parameters, i.e. behaviour is undefined, anyway.
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NEric Anholt <eric@anholt.net>