After fixing the same issue in the set_caching IOCTL and Chris' request
to check out the possibilities for an improved RPM ref handling I
noticed that we have the same issue in the set_tiling IOCTL. Fix this
up.I didn't see any bug reports about this one, but the GTT unbind
operation on this path accesses the HW, which needs the ref.
Signed-off-by: NImre Deak <imre.deak@intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1447092986-11165-1-git-send-email-imre.deak@intel.com

The old style of memory interleaving swizzled upto the end of the
first even bank of memory, and then used the remainder as unswizzled on
the unpaired bank - i.e. swizzling is not constant for all memory. This
causes problems when we try to migrate memory and so the kernel prevents
migration at all when we detect L-shaped inconsistent swizzling.
However, this issue also extends to userspace who try to manually detile
into memory as the swizzling for an individual page is unknown (it
depends on its physical address only known to the kernel), userspace
cannot correctly swizzle.

Note that this is a new attempt for the previously merged one,
reverted in

commit d82c0ba6
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Tue Jul 14 12:29:27 2015 +0200

    Revert "drm/i915: Declare the swizzling unknown for L-shaped configurations"

This is cc: stable since we need it to fix up troubles with wc cpu
mmaps that userspace recently started to use widely.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91105Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: stable@vger.kernel.org
[danvet: Add note about previous (failed attempt).]
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Chris rightfully suggested that documenting fences without documenting
the BO tiling tracking doesn't make much sense, so fix that.

The important bit to stress here (since it lead to some confusion) is
the GEM doesn't really care about tiling. Except for a few select cases
where the kernel needs to manage something that userspace can't take
care of: Namely the limited number of fences and fixing up swizzling,
although we still fail at the later.

v2: Move the low-level tiling/swizzling functions and kerneldoc to
i915_gem_fence.c and leave only the userspace interface here.
Suggested by Chris.

Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>

It fits more with the low-level fence code, and this move leaves only
the userspace tiling ioctl handling in i915_gem_tiling.c.
Suggested-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>

This reverts commit 19ee835c.

It breaks existing old userspace which doesn't handle UNKNOWN
swizzling correct. Yes UNKNOWN was a thing back in 2009 and probably
still is on some other platforms, but it still pretty clearly broke
the testers machine. If we want this we need to extend the ioctl with
new paramters that only new userspace looks at.

Cc: Harald Arnesen <harald@skogtun.org>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reported-by: NHarald Arnesen <harald@skogtun.org>
Cc: stable@vger.kernel.org
Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>

The old style of memory interleaving swizzled upto the end of the
first even bank of memory, and then used the remainder as unswizzled on
the unpaired bank - i.e. swizzling is not constant for all memory. This
causes problems when we try to migrate memory and so the kernel prevents
migration at all when we detect L-shaped inconsistent swizzling.
However, this issue also extends to userspace who try to manually detile
into memory as the swizzling for an individual page is unknown (it
depends on its physical address only known to the kernel), userspace
cannot correctly swizzle objects.

v2: Mark the global swizzling as unknown rather than adjust the value
reported to userspace.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91105Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: stable@vger.kernel.org
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Since the removal of the user pin_ioctl, the only means for pinning an
object is either through binding to the scanout or during execbuf
reservation. As the later prevents a call to set-tiling, we need only
check if the obj is pinned into the display plane to see if we need
reject the set-tiling ioctl.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

When we walk the list of vma, or even for protecting against concurrent
framebuffer creation, we must hold the struct_mutex or else a second
thread can corrupt the list as we walk it.

Fixes regression from
commit d7f46fc4
Author: Ben Widawsky <benjamin.widawsky@intel.com>
Date:   Fri Dec 6 14:10:55 2013 -0800

    drm/i915: Make pin count per VMA

References: https://bugs.freedesktop.org/show_bug.cgi?id=89085Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Cc: stable@vger.kernel.org
Signed-off-by: NJani Nikula <jani.nikula@intel.com>

The object structure contains the last read, write and fenced seqno values for
use in syncrhonisation operations. These have now been replaced with their
request structure counterparts.

Note that to ensure that objects do not end up with dangling pointers, the
assignments of last_*_req include reference count updates. Thus a request cannot
be freed if an object is still hanging on to it for any reason.

v2: Corrected 'last_rendering_' to 'last_read_' in a number of comments that did
not get updated when 'last_rendering_seqno' became 'last_read|write_seqno'
several millenia ago.

For: VIZ-4377
Signed-off-by: NJohn Harrison <John.C.Harrison@Intel.com>
Reviewed-by: NThomas Daniel <Thomas.Daniel@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Let's just throw in the towel on this one and take the cheap way out.

Based on a patch from Chris Wilson, but checking for a different bit.
Chris' patch checked for even bank layout, this one here for a magic
bit. Given the evidence we've gathered (not much) both work I think,
but checking for the magic bit might be more accurate.

Anyway, works on my gm45 here.

For paranoi restrict to gen4 (and mobile), since we've only ever seen
this on gm45 and i965gm.

Also add some debugfs output so that we can skip the tiled swapping
tests properly in these cases.

v2: Clean up the quirk'ed pin count in free_object to avoid upsetting
the WARN_ON. Spotted by Chris.

Cc: Chris Wilson <chris@chris-wilson.co.uk>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=28813
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=45092Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

As obj->map_and_fenceable computation has changed to only be set when
the object is bound inside the global GTT (and is suitable aligned to a
fence region) we need to accommodate those changes when the tiling is
adjusted. The easiest solution is to unbind from the global GTT if we
are currently fenceable, but will not be after the tiling change.

The bug has been exposed by

commit f8fcadba218fe6d23b2e353fea1cf0a4be4c9454
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Oct 31 13:53:52 2014 +0000

    drm/i915: Only mark as map-and-fenceable when bound into the GGTT

which tried to fix an oversight from

commit e6a84468
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Mon Aug 11 12:00:12 2014 +0200

    drm/i915: Force CPU relocations if not GTT mapped

which changed the handling of obj->map_and_fenceable.

Note that the alignment check is a vestige from our attempts to reduce
the alignment requirements of tiled but unfenced buffers on
gen2/3. Also, that was when unbinding from the GTT meant UC writes and
clflushing, so we went to great pains to avoid such.

That leaves the actual bug of setting map_and_fenceable to true if we're
not bound to ggtt, which violates the change introduced in the above
patch. Unbinding in that case really looks like the simplest and safest
option, we have to do it anyway.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=85896
Testcase: igt/gem_concurrent_blit/gttX*
Tested-by: huax.lu@intel.com
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Tested-by: NValtteri Rantala <valtteri.rantala@intel.com>
[Jani: amend commit message per input from Daniel and bisect result from
Valtteri]
Signed-off-by: NJani Nikula <jani.nikula@intel.com>

Userspace cares about whether or not swizzling depends on the page
address for its direct access into bound objects. Extend the get_tiling
ioctl to report the physical swizzling value in addition to the logical
swizzling value so that userspace can accurately determine when it is
possible for manual detiling.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Akash Goel <akash.goel@intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Testcase: igt/gem_tiled_wc
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Some machines (like MBAs) might use a tiled framebuffer but not enable
display swizzling at boot time.  We want to preserve that configuration
if possible to prevent a boot time mode set.  On IVB+ it shouldn't
affect performance anyway since the memory controller does internal
swizzling anyway.

For most other configs we'll be able to enable swizzling at boot time,
since the initial framebuffer won't be tiled, thus we won't see any
corruption when we enable it.

v2: preserve swizzling if BIOS had it set (Daniel)
v3: preserve swizzling only if we inherited a tiled framebuffer (Daniel)
    check display swizzle setting in detect_bit_6_swizzle (Daniel)
    use gen6 as cutoff point (Daniel)
v4: fixup swizzle preserve again, had wrong init order (Daniel)
Reported-by: NKristian Høgsberg <hoegsberg@gmail.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Previously, it was possible for the GPU memory accesses to be swizzled
to try to optimize the fetches for tiled buffers. This swizzling was on
top of what the memory controller in the uncore already does.

With broadwell, we drop that GPU side swizzling, and the corresponding
initialization in 3 units (GAM, GT, DE). All those bits are reserved, as
specs put it:

  Before Gen8, there was a historical configuration control field to
  swizzle address bit[6] for in X/Y tiling modes. This was set in three
  different places: TILECTL[1:0], ARB_MODE[5:4], and
  DISP_ARB_CTL[14:13]"

  For Gen8 the swizzle fields are all reserved, and the CPU's memory
  controller performs all address swizzling modifications.

This also means that user space doesn't have to manually swizzle when
accessing tiled buffers from the CPU, and so we always return
I915_BIT_6_SWIZZLE_NONE from i915_gem_detect_bit_6_swizzle(), which
short-circuits the initialization of the registers mentionned above in
i915_gem_init_swizzling().

v2: Refine the explanation a bit more (Daniel)
v3: Make it BDW+ specific (Steve)

Cc: Steve Aarnio <steve.j.aarnio@linux.intel.com>
Signed-off-by: NDamien Lespiau <damien.lespiau@intel.com>
[danvet: Keep the actual code to set the tiling bits for now, in case
some bios escaped to the wild that uses this - we'd need it for
fastboot.]
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

This migrates the fence tracking onto the existing seqno
infrastructure so that the later conversion to tracking via requests is
simplified.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Remove the rest of the references to drm_i915_private_t. No functional
changes.
Signed-off-by: NJani Nikula <jani.nikula@intel.com>
[danvet: Drop hunk in i915_cmd_parser.c]
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Signed-off-by: NBen Widawsky <ben@bwidawsk.net>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Assuming that all framebuffer related metadata is invariant simplifies
our userspace input data checking. And current userspace always first
updates the tiling of an object before creating a framebuffer with it.

This allows us to upconvert a check in pin_and_fence to a WARN.

In the future it should also be helpful to know which buffer objects
are potential scanout targets for e.g. frontbuffer rendering tracking
and similar things.

Note that SNA shipped for one prerelease with code which will be
broken through this patch. But users shouldn't notice since it's
purely an optimization and will transparently fall back to allocating
a new fb. i-g-t also had offending code (now fixed), but we don't
really care about breaking the test-suite.

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Grumpily-reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

No buffer overflows here, but better safe than sorry.

v2:
- Fixup the sizeof conversion, I've missed the pointer deref (Jani).
- Drop the redundant GFP_ZERO, kcalloc alreads memsets (Jani).
- Use kmalloc_array for the execbuf fastpath to avoid the memset
  (Chris). I've opted to leave all other conversions as-is since they
  aren't in a fastpath and dealing with cleared memory instead of
  random garbage is just generally nicer.

Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJani Nikula <jani.nikula@intel.com>
[danvet: Drop the contentious kmalloc_array hunk in execbuf.]
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

As alluded to in several patches, and it will be reiterated later... A
VMA is an abstraction for a GEM BO bound into an address space.
Therefore it stands to reason, that the existing bind, and unbind are
the ones which will be the most impacted. This patch implements this,
and updates all callers which weren't already updated in the series
(because it was too messy).

This patch represents the bulk of an earlier, larger patch. I've pulled
out a bunch of things by the request of Daniel. The history is preserved
for posterity with the email convention of ">" One big change from the
original patch aside from a bunch of cropping is I've created an
i915_vma_unbind() function. That is because we always have the VMA
anyway, and doing an extra lookup is useful. There is a caveat, we
retain an i915_gem_object_ggtt_unbind, for the global cases which might
not talk in VMAs.

> drm/i915: plumb VM into object operations
>
> This patch was formerly known as:
> "drm/i915: Create VMAs (part 3) - plumbing"
>
> This patch adds a VM argument, bind/unbind, and the object
> offset/size/color getters/setters. It preserves the old ggtt helper
> functions because things still need, and will continue to need them.
>
> Some code will still need to be ported over after this.
>
> v2: Fix purge to pick an object and unbind all vmas
> This was doable because of the global bound list change.
>
> v3: With the commit to actually pin/unpin pages in place, there is no
> longer a need to check if unbind succeeded before calling put_pages().
> Make put_pages only BUG() after checking pin count.
>
> v4: Rebased on top of the new hangcheck work by Mika
> plumbed eb_destroy also
> Many checkpatch related fixes
>
> v5: Very large rebase
>
> v6:
> Change BUG_ON to WARN_ON (Daniel)
> Rename vm to ggtt in preallocate stolen, since it is always ggtt when
> dealing with stolen memory. (Daniel)
> list_for_each will short-circuit already (Daniel)
> remove superflous space (Daniel)
> Use per object list of vmas (Daniel)
> Make obj_bound_any() use obj_bound for each vm (Ben)
> s/bind_to_gtt/bind_to_vm/ (Ben)
>
> Fixed up the inactive shrinker. As Daniel noticed the code could
> potentially count the same object multiple times. While it's not
> possible in the current case, since 1 object can only ever be bound into
> 1 address space thus far - we may as well try to get something more
> future proof in place now. With a prep patch before this to switch over
> to using the bound list + inactive check, we're now able to carry that
> forward for every address space an object is bound into.
Signed-off-by: NBen Widawsky <ben@bwidawsk.net>
[danvet: Rebase on top of the loss of "drm/i915: Cleanup more of VMA
in destroy".]
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Soon we want to gut a lot of our existing assumptions how many address
spaces an object can live in, and in doing so, embed the drm_mm_node in
the object (and later the VMA).

It's possible in the future we'll want to add more getter/setter
methods, but for now this is enough to enable the VMAs.

v2: Reworked commit message (Ben)
Added comments to the main functions (Ben)
sed -i "s/i915_gem_obj_set_color/i915_gem_obj_ggtt_set_color/" drivers/gpu/drm/i915/*.[ch]
sed -i "s/i915_gem_obj_bound/i915_gem_obj_ggtt_bound/" drivers/gpu/drm/i915/*.[ch]
sed -i "s/i915_gem_obj_size/i915_gem_obj_ggtt_size/" drivers/gpu/drm/i915/*.[ch]
sed -i "s/i915_gem_obj_offset/i915_gem_obj_ggtt_offset/" drivers/gpu/drm/i915/*.[ch]
(Daniel)

v3: Rebased on new reserve_node patch
Changed DRM_DEBUG_KMS to actually work (will need fixing later)
Signed-off-by: NBen Widawsky <ben@bwidawsk.net>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

BSpec contains several scattered notes which state that the maximum
fence stride was increased to 256KB on IVB.

Testing on real hardware agrees.
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Our checks for an invalid fence stride forgot to guard against
zero stride on gen4+. Fix it.

v2: Avoid duplicated code (danvet)
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

The i915 driver uses sg lists for memory without backing 'struct page'
pages, similarly to other IO memory regions, setting only the DMA
address for these. It does this, so that it can program the HW MMU
tables in a uniform way both for sg lists with and without backing pages.

Without a valid page pointer we can't call nth_page to get the current
page in __sg_page_iter_next, so add a helper that relevant users can
call separately. Also add a helper to get the DMA address of the current
page (idea from Daniel).

Convert all places in i915, to use the new API.
Signed-off-by: NImre Deak <imre.deak@intel.com>
Reviewed-by: NDamien Lespiau <damien.lespiau@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

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>