commit baec802b21387d04aebb10ac29e719a1800c5aa0
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Sun Dec 21 14:41:12 2014 +0000

    2.99.917 snapshot

commit 01ce7efe73538047abd38bbbb95fc4012ebeb9b4
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Sat Dec 20 13:36:50 2014 +0000

    sna: Fix Drawable offsets prior to performing shadow discard
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit c27e6306e86797553351a71e2917a9b1158ec388
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Sat Dec 20 12:46:52 2014 +0000

    sna/dri2: Check for implicit flushing after updating a Pixmap buffer
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 5c16a4ec9b59123336483238ee6017db5491e255
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Sat Dec 20 09:38:41 2014 +0000

    sna: Skip allocation of new TearFree buffer if not actually flipping
    
    After showing the new front buffer, we have to avoid writing to the back
    buffer whilst the flip is still pending (as it may still be being shown
    by the display). To do this we check if there are any outstanding flips,
    and reallocate if so. However, with a CRTC override, we may just be
    flipping that one CRTC and not the current TearFree back buffer and in
    that case we can avoid the reallocation.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 9b3b591476cc52cc285a8a1f02462db37d5e10c2
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Dec 19 12:25:30 2014 +0000

    sna/dri2: Preserve the shadow CRTCs when copying to other areas
    
    In the Composite setup, if we are doing a DRI2 copy onto the front
    buffer, we are fully cognizant that the copy will not go onto the
    unredirected Windows of another Client. So we can preserve the shadow
    CRTC mapping for those Clients, and prevent ping-ponging between CRTC
    modes.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 391b50c81decd208c47c109217b2c882c064ac69
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Dec 19 11:11:38 2014 +0000

    sna/dri2: Report changes in pitch when swapping bo
    
    As we have relaxed the pitch restriction between front/back buffers when
    swapping, we need to make sure that any change is also reported back
    along with the change in front/back names.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit ada97d27250aa09b1a27d5f6640fea011fade997
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Dec 19 09:43:33 2014 +0000

    sna/dri2: Look for preferred CRTC for visible portion of the window
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 6f99bbe67e6e70d5b24611c7afa0dffc5632aaab
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Dec 19 08:37:07 2014 +0000

    sna/dri2: Prevent NULL pointer dereference of Window private
    
    In case we receive quick successive calls to DRI2GetBuffers on the same
    Window, we need to check that the private exists before dereferencing
    it.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit fdd7508e53043f82f9cf422fc968c6574334a667
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Dec 19 07:59:24 2014 +0000

    sna/dri2: Decouple Window cache on Pixmap changes
    
    If the Pixmap for a Window is changed (i.e. Composite
    redirection/unredirection), we also need to decouple any associated DRI2
    front buffer for the Pixmap (e.g. used for single CRTC flipping).
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 82e480267079f8678245e37724f95ae87f55b582
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Dec 17 08:01:00 2014 +0000

    intel-virtual-output: Do not attempt to tidy up the local display if it fails
    
    Just fix the segfault dereferencing the primary display if we bail
    during its setup.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit c211b3bc61efa7d13a5f5f1276649d90bf4ef854
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Dec 17 07:51:31 2014 +0000

    intel-virtual-output: Add version number to verbose output
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 123754de41380b94922768d021987b1dea3b39c4
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Tue Dec 16 10:31:22 2014 +0000

    sna: Make error debug tracing available at normal debug levels
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 99537089346ed7a1bc6b02349fad93845e865bb9
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Dec 12 15:36:25 2014 +0000

    sna: Tune handling of kernel error detection during intial setup
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 01692b8cf229f45813cf3a1059577a779d71dfa8
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Dec 12 15:13:04 2014 +0000

    sna: Fix TearFree multi-monitor operation whilst wedged
    
    The crtc flip (for when the kernel can't do a pageflip) made the mistake
    of setting the shadow buffer to the scanout - nullifying the effect of
    the TearFree.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 343299c78bfc1044da0b965e99de9ad312b05cca
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Dec 12 09:40:51 2014 +0000

    sna/trapezoids: Check the pixmap is wrapped before deferencing it
    
    When we AddTraps to a low resolution bitmap, we need to fallback as we
    don't wrap it with a GPU pixmap.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 60ebac591596ccb768b684b1909064570dd1cf47
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Dec 12 07:55:51 2014 +0000

    sna: Add another Xinerama vs RandR guard
    
    Another place used rrScrPriv without checking for its validity.
    
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=87257
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit f587d71f3c91f9ddcc4cdbe8a0f9ce59590d9db2
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Thu Dec 11 08:53:54 2014 +0000

    sna: Catch NULL RandR screen private
    
    If Xinerama is enabled, than RandR12 will be silently disabled. Be
    careful not to dereference the rrScrPiv when it doesn't exist.
    
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=87207
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 04a09d353fc5ad8a55eb2457dc8bb43638ba879e
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Tue Dec 9 20:43:52 2014 +0000

    uxa/dri: Remove defunct intel->back_pixmap
    
    This was only required for glamor support.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 08ee9a75df16ee276835df5fc613648db2566154
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Tue Dec 9 20:40:20 2014 +0000

    uxa/dri: Fix triple buffer swapping
    
    Fixes regression from
    
    commit 0aa2edbd29c8dd26a5f3748e3875c445ea358a6d
    Author: Chris Wilson <chris@chris-wilson.co.uk>
    Date:   Wed Nov 5 11:56:20 2014 +0000
    
        Remove defunct glamor support
    
    where the wrong branch of pixmap exchange upon SwapBuffers was kept when
    removing the glamor paths.
    
    Reported-by: Rui Matos
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit fc984e8953d61901b255422c8f56eb79a2dd2a28
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Tue Jul 22 08:38:42 2014 +0100

    sna/dri3: Mesa relies upon implicit fences for X/Compositor synchronisation
    
    The decision has been made that DRI3/intel shall continue with DRI2-style
    implicit fencing for synchronisation between X and clients using pixmaps
    as texture sources. (The other way around uses explicit fencing!)
    
    References: https://bugs.freedesktop.org/show_bug.cgi?id=81551
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit d247cb7d0cdb73736f31612157e47f166af68ba0
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Mon Dec 8 10:07:25 2014 +0000

    sna/gen6: Poke PSMI control around WAIT_FOR_EVENT to prevent idling
    
    The bspec recommends preventing the hardware from going to sleep around
    a WAIT_FOR_EVENT, and tells us to use disable sleep bit in PSMI control
    to accomplish this.
    
    References: https://bugs.freedesktop.org/show_bug.cgi?id=62373
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit d39dc73a80161d290555877942e5990daf0c0411
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Mon Dec 8 08:33:11 2014 +0000

    sna: Inline check for pending events before reading from drm fd
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit f8aecdffada438306dff8923cdfc0df527509672
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Thu Dec 4 11:23:11 2014 +0000

    sna: Clear the read flags for the shared drm fd after use
    
    In ZaphodHeads, we may reuse the same select read flags and attempt to
    read from a blocking drm fd multiple times. However, if we clear the
    read flags after first exhausting the fd, we shouldn't then block on
    subsequent heads.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 594470c809627feab4891bade5ec6d80e1d0a919
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Thu Dec 4 11:16:58 2014 +0000

    sna: Update the fence marker after submitting a new batch
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 9c2c485df9fd39ae36779f765a892e36835a8001
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Mon Dec 1 14:26:46 2014 +0000

    sna: Validate that the reg/insn size is well defined
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 263d10998617ba935179942d2440a1607cd3c21c
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Mon Dec 1 13:26:46 2014 +0000

    sna: Clamp downsampling tile size
    
    For very large scale factors, it is possible for the current calculation
    to underflow and try negative tile sizes.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 88e9bb760fbd1249bb364a6b68aceee0b442738c
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Mon Dec 1 10:47:32 2014 +0000

    sna: Explicitly disable unused pipes after an output reconfiguration
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit ddf5feba786cf0147c4db9720c96e9a5636573b0
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Mon Dec 1 10:00:32 2014 +0000

    sna: Generate a backtrace on assertion failures
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit e12ec452a99396f76138a5961c8195e22ca12ac1
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Mon Dec 1 09:46:20 2014 +0000

    sna: Cancel pending mode change notification
    
    Along a couple of paths, we either do not care about the notification
    (i.e. during suspend) or do it explicitly. There we should mark the work
    as done.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 700c036d180a09e4f7ec30466f7f6dda1c1d0066
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Mon Dec 1 09:31:41 2014 +0000

    sna: Reject modesets with 0 outputs
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit dd421005cac981e75f734d50263675521943ed7b
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Mon Dec 1 09:29:07 2014 +0000

    sna: Add a few more asserts to sna_crtc_flip()
    
    Check that we are updating an active CRTC with a valid mode and
    associated outputs.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 104517507f267aca7433f8b85dec89e9013a640e
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Mon Dec 1 09:24:50 2014 +0000

    sna: Do not post shadow updates in the middle of output reconfiguration
    
    Wait until the screens and framebuffers settle before we start updating.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit faef6e5e1ab52819d4e9a19738ec7d8419841613
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Mon Dec 1 08:29:47 2014 +0000

    sna: Clean up DBG compile warnings
    
    Including a potential crash due to using a unspecified string, oops.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 0bec625cc45eb8de85a5013506cce5f789dc8e45
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Mon Dec 1 08:11:18 2014 +0000

    sna: Report when we disable TearFree due to an error in operation
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 6059b51b59537682b3d9c64c69ea72616462c41c
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Nov 28 07:33:42 2014 +0000

    sna: Reject attempts to set an invalid mode
    
    Let's be sure the mode has been established before we attempt to apply
    it to a CRTC - just in case the kernel tries to use the invalid mode and
    blows up.
    
    References: https://bugs.freedesktop.org/show_bug.cgi?id=86679
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 0f15b8b45bb6de10ce1926db303247a5bd3c1c08
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Nov 19 18:42:32 2014 +0000

    sna/transform: Correctly check for imprecise fractional translations
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 7725c423cda217e0dbd95003bb7a80b4b3a73bb4
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Nov 19 18:30:20 2014 +0000

    sna/dri2: Improve precision of completion event for no-op swaps
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit f367f3dee5e2ecba265a8ff7e98e196189793613
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Nov 19 18:19:13 2014 +0000

    sna/dri2: Add more DBG to explain invalid frames
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 05a1aba1ab5d288918d6689080170fc83a80db61
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Nov 19 18:12:10 2014 +0000

    sna: DBG compile fix
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit a90cc3b3889fafbd91c11c42d068a9d6474e218b
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Tue Nov 18 08:37:25 2014 +0000

    sna: Tweak alignment constraints on gen8 to allow BLT
    
    The previous commits prevent us from using the BLT if the destination
    address is misaligned. Honour that restriction when creating buffers as
    well, so that they are always usuable by the BLT.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 8dee52997891108eec8e4df12dd02f3a060d9cb8
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Nov 19 13:38:20 2014 +0000

    sna: Add more checks and asserts for BLT capable bo
    
    Before we use the BLT for core acceleration, double check that we can.
    This should catch the case where we attempt to operate on SHM pixmaps
    which do not meet the restrictions.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 3a22b6f6d55a5b1e0a1c0a3d597996268ed439ad
Author: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Date:   Wed Nov 19 15:10:05 2014 +0200

    sna: gen8 BLT broken when address has bit 4 set
    
    With bit 4 set in address, the gen8 blitter fails and blits errorneously
    into the cacheline preceeding the destination and similarly when reading from
    the source, corrupting memory.
    
    v2: Update the destination base offset pattern as revealed
        by igt/tests/gem_userptr_blits/destination-bo-align
    
    v3: Check base address as well
    
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=79053
    Cc: Chris Wilson <chris@chris-wilson.co.uk>
    Tested-by: xunx.fang@intel.com [v2]
    Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>

commit 29aab766f4b59c71181875880c8c943b93008bdf
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Tue Nov 18 20:54:39 2014 +0000

    sna/dri2: Remove unused function
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit b6eeb7a1f7efa591504070b606be655e27e6e9c2
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Nov 5 13:03:41 2014 +0000

    Disable DRI3 by default
    
    The external libraries, both in git, and especially shipping already
    enabled in distributions, are buggy and lead to server crashes and
    lockups. Caveat emptor.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 6c2707d7bbc0ebb422be66618b6f78887c46446e
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Thu Nov 13 08:37:13 2014 +0000

    sna: Unroll pwrite using our cached mmappings
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit f9f85b88032b2458d4d24ceca450fdaa65e6b94d
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Nov 12 11:50:33 2014 +0000

    sna: Reorder PARAMs based on upstream inclusion
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 13fa1f1a26aed83cd8fccafb9a0bdb096f69534b
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Nov 12 10:54:36 2014 +0000

    sna/gen7: Be paranoid and force a pipecontrol stall at least every 4 flushes
    
    The documentation warns of potential GPU hangs if we emit more than 4
    pipecontrol flushes without a CS stall. This is highly unlikely given
    how frequently we must inject stalls into the pipeline, but force a
    stall just in case!
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit b2492fa45beaf3c676eea2d5d75e35d60b91d769
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Mon Nov 10 13:30:15 2014 +0000

    sna: Only reshow unhidden cursors after modesetting
    
    Reported-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 0267f5902ddaceced870dcd8451ef8729e33e5c2
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Mon Nov 10 07:42:42 2014 +0000

    sna: Check after filtering points before submitting boxes to be drawn
    
    As we clip the points when converting them into GPU boxes, check that we
    have something to draw before submitting the commands.
    
    Reported-by: Ian Gay <gay@sfu.ca>
    References: https://bugs.freedesktop.org/show_bug.cgi?id=86075
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 9ea242c275a70351abcfbc0f8f3233c122abecbe
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Sun Nov 9 08:59:08 2014 +0000

    sna: Check picture format on destination before beginning a GPU operation
    
    This is more important in the multiple stage operations like glyph
    rendering where we do not want to initiate an operation on the GPU only
    then to fallback due to an incompatible destination.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit e96520327bd2ef4fbc7b7b5169a17b966d9f85f3
Author: Sedat Dilek <sedat.dilek@gmail.com>
Date:   Fri Nov 7 10:54:18 2014 +0100

    sna: wc-mmap: Fix warning in configure
    
    This fixes the following warning due to a typo:
    
    configure: WARNING: unrecognized options: --enable-wc-map
    Signed-off-by: Sedat Dilek <sedat.dilek@gmail.com>

commit ab6680375cc6d5efebe8e40816d1906b15678d6e
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Nov 7 09:54:35 2014 +0000

    sna/xv: Give the illusion of hardware progress
    
    The number of ports we provide for Xv textured video is arbitrary. The
    main cost is reservation of a number of XIDs and preallocation of a
    block of memory. Whatever value we pick, someone will always want
    more...
    
    References: https://bugs.freedesktop.org/show_bug.cgi?id=85974
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 0aa2edbd29c8dd26a5f3748e3875c445ea358a6d
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Nov 5 11:56:20 2014 +0000

    Remove defunct glamor support
    
    It never was a stable or complete replacement, and now it is
    incorporated in Xorg itself!
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 4281bbdd5be1f846c217cdf6d09536779cd324ce
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Oct 24 08:50:55 2014 +0100

    sna: Fix debug mmapping of active + released aux buffers
    
    The debug code wants to print the batch and the aux buffers. To do so,
    it needs to bypass the assertions on the lifetime of the buffers.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 87802b3402ec4a829c6507b4b78fcd03e30b6fe1
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Thu Oct 23 10:34:44 2014 +0100

    sna: Experimental support for write-combining mmaps
    
    If we have a linear buffer, we can request the kernel mmap it directly
    with write-combining without having to pin it into the GTT. This allows
    us to efficiently upload very large buffers, and can avoid the dreaded
    aperture thrashing.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 33764685cbcd01548b6a23c217c7b511c1639694
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Thu Nov 6 12:40:20 2014 +0000

    sna: Wrap prefaulting with sigtrap()
    
    Handle a potential SIGBUS due to kernel bugs when prefaulting the
    scanout surface.
    
    References: https://bugs.freedesktop.org/show_bug.cgi?id=85959
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 97fe3c1c860978c7a649cba93a55fa497010ccc1
Author: Rodrigo Vivi <rodrigo.vivi@intel.com>
Date:   Wed Nov 5 15:48:14 2014 -0800

    sna: Use VMask in 3DSTATE_PS
    
    Using dispatch mask cause hangs waiting PS Done on some cases like bug #83207,
    with larger screen or when scaling it.
    
    Also mesa uses VMask instead of Dmask for 3DSTATE_PS because in some cases
    they were getting incorrect derivatives for subspans.
    
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=83207
    Cc: Timo Aaltonen <tjaalton@ubuntu.com>
    Cc: Gary Wang <gary.c.wang@intel.com>
    Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
    Tested-by: Timo Aaltonen <tjaalton@ubuntu.com>

commit 0532a3313ad9c76a6e1d28e8a1c2ea495583fead
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Nov 5 20:11:54 2014 +0000

    sna/gen8: Clear instancing enabled bit between batches
    
    gen8 sets the instancing bit relative to the vertex element, but we were
    clearing it for the vertex buffer. As the maximum number of vertex
    elements is fixed, just clear them all when emitting our header. Note
    that VF_SGVS is not sufficient by itself to disable all side-effects of
    instancing.
    
    Thanks to Kenneth Graunke for pointing out the change from vertex buffer
    to vertex element of the instancing enable bit.
    
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=84958
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit ba408bf21c4b65f19c7b581e4c88c92805184334
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Tue Nov 4 13:39:52 2014 +0000

    sna: Correct units for videoRam
    
    The mappable aperture size is stored as a page count, videoRam expects
    size in KiB.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 7ed5b1156cbf173a68f4fe007c51b911e7c5a1a2
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Tue Nov 4 12:54:00 2014 +0000

    sna: Reuse the queried aperture size for videoRam determination
    
    Rather than repeat our derivation of how much RAM we can map, just reuse
    the already determined value.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit a972c9886edc1141ad3f8a0a33d2f7f18d0008bc
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Oct 24 12:16:11 2014 +0100

    sna: Use v2 of get_tiling ioctl to check for bit17 swizzling
    
    With the new version of the get_tiling ioctl we can reliably detect if
    we can use manual detiling (i.e. CPU mmaps) on older gen.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 7ecc778691c452285f754743a93a46fa1d3da52f
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Tue Nov 4 10:36:26 2014 +0000

    backlight: bl_power uses 0 for on and 4 for off
    
    The backlight/bl_power file didn't obey the obvious semantics, but
    instead operates using some interesting framebuffer API. Oh well.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit cc3b8a542ecb1ba873efefaeab630fa8f69b5b96
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Oct 31 20:38:40 2014 +0000

    sna: Treat external bo carefully when marking as idle
    
    When we flag a bo as being externally modified, we set a special flag on
    its request pointer. This forces us to inspect whether the bo is busy
    before accessing it the next, and if it is idle, we must be careful that
    we then do not dereference the special request.
    
    Fixes regression from
    commit 9115406b651eb11779f37c4b048d55188b9d6e02
    Author: Chris Wilson <chris@chris-wilson.co.uk>
    Date:   Thu Oct 30 19:21:02 2014 +0000
    
        sna: Retire all requests when a bo is found to be not busy
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 122f6065604480a57b86f04d5d09011ee6e8829b
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Oct 31 16:50:04 2014 +0000

    sna: Fix forced fallbacks from init
    
    As we now override a few hooks when we mark the GPU as wedged, we also
    need to preserve those overrides during initialisation. Fixes DBG_NO_HW.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit d959014c10e2408ed2f073bd5fcb8c558a17da07
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Oct 31 16:12:37 2014 +0000

    sna: Modicum of extra DBG for transformed redisplay
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 1040d92e31eea4556c2605c809d18ec4c864424a
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Oct 31 16:11:54 2014 +0000

    sna/gen6+: Fix projective sample coordinates
    
    Reported-by: "Ding, Chi" <chix.ding@intel.com>
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 30fded8dfa77c74de165ad53f82412a2e7fa88ae
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Oct 31 16:11:28 2014 +0000

    sna/brw: Fix EU tests to select the right gen
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 71d79daf458c05f771441eb268693c34be0a97c3
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Thu Oct 30 19:20:13 2014 +0000

    sna: Remove extraneous tab whitespace before parameter name
    
    Reported-by: Arthur Koziol
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 9115406b651eb11779f37c4b048d55188b9d6e02
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Thu Oct 30 19:21:02 2014 +0000

    sna: Retire all requests when a bo is found to be not busy
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 9d48d2817927bc2532953f09b6be2002c2086e1f
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Oct 24 09:12:49 2014 +0100

    sna: Reuse the same fence until retired
    
    In order to reduce the number of breadcrumbs the kernel must emit to
    track our batches, reuse the last query until it has retired.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 61436c2fabe117b85404eecb06158ba0a63a7741
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Oct 29 16:52:44 2014 +0000

    sna: Fix build for GC API changes in Xorg-1.16.99.901
    
    The GC struct was changed in
    
    commit 73e2383b7350723256894c5076d0c731aec5cd1f
    Author: Adam Jackson <ajax@redhat.com>
    Date:   Fri Jun 20 13:31:20 2014 -0400
    
        dix: Always store GC client clip as a region (v2)
    
    update to match.
    
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=85581
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit d08a5f555a0c47ae23c0f9a890b512cb23e74feb
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Oct 24 09:53:29 2014 +0100

    sna/trapezoids: Prevent overflow of edge gradient in mono rasteriser
    
    References: https://bugs.freedesktop.org/show_bug.cgi?id=70461#c76
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit f611f9580469661e585f419a7dd033ddffd7e20d
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Oct 24 09:41:47 2014 +0100

    sna/trapezoids: Difference between two 32-bit quantities is 33-bits in size
    
    When computing the edge distance, we subtract on 32-bit quantity from
    another. This requires 33-bits to store the full result so promote the
    subtraction to 64-bits (rather than the result of that subtraction as
    done currently).
    
    Reported-by: Jiri Slaby <jirislaby@gmail.com>
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=70461#c76
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 4df0052a21efd744c4b8cb2409139ded6e45f5c8
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Oct 24 08:08:23 2014 +0100

    sna: Update the reloc->delta for the shrunk batch
    
    As well as actually updating the value in the batch for the relocation
    address, we need to tell the kernel in case it has to update the
    address.
    
    Another regression from
    commit b4872f74c05a92c91d7b57146eb35ed2ecc86d97
    Author: Chris Wilson <chris@chris-wilson.co.uk>
    Date:   Tue Oct 21 10:38:22 2014 +0100
    
        sna: Avoid pwriting large batches
    
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=85389
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 8df8618c62c0918416ca3e21b081299eacb9b054
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Oct 24 08:02:50 2014 +0100

    sna: Adapt to changes in internal render API in Xorg-1.17
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit a8d1eaf6ee1fed0aeff681d33864e17cb904ddd3
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Oct 22 21:12:56 2014 +0100

    sna: Fix use of uninitialized value when handling migration for PRIME
    
    sna_accel.c: In function 'sna_pixmap_undo_cow':
    sna_accel.c:1933:11: warning: 'create' may be used uninitialized in this function [-Wuninitialized]
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit e46f26b9e2671468e9accd2f69ffc22fb96b5849
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Oct 22 19:37:42 2014 +0100

    sna/gen8: Load the sampler result directly into the fb write message
    
    With the unified register file (from gen6+), we no longer need to copy
    the sample results into the final write message.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit dee9ea29917dd5ef4a7de6570cd091f08d334df9
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Oct 22 19:40:14 2014 +0100

    sna/gen8: Override the URB output from the VS
    
    Using the override seems to be marginally faster.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 7a9bdadd71730adc5266bc6758982abec5917b93
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Oct 22 19:31:10 2014 +0100

    sna/gen8: Clamp URB allocations for GT3
    
    GT3 requires some reserved space in the URB allocation and so we must
    reduce the amount we allocate to our vertices.
    
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=81583
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit de54a93217cc550c44ee138f0511ede6925d84e0
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Oct 22 19:30:21 2014 +0100

    sna/gen8: Annotate more 64bit pointer locations
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 2114b54a0b8a561dd7db1a38a87654571b4aa853
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Oct 22 18:51:39 2014 +0100

    sna: Fix double application of batch shrinking to relocation targets
    
    Fixes regression from
    
    commit b4872f74c05a92c91d7b57146eb35ed2ecc86d97
    Author: Chris Wilson <chris@chris-wilson.co.uk>
    Date:   Tue Oct 21 10:38:22 2014 +0100
    
        sna: Avoid pwriting large batches
    
    NB: Remember to check your tree before pushing a simple bug fix.
    
    Signed-off-by: root <root@huronriver.alporthouse.com>

commit 6e486e11a53f93b628a3a82d5b086e63b7dc6c1e
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Oct 22 07:24:10 2014 +0100

    sna/dri2: Fix interoperation with PRIME and older Xorg
    
    When using the fake triple buffer implementation with recent enough Xorg
    to support both real triple buffering and PRIME, we have to disable use
    of SwapLimit as that freezes the client when we are instead faking the
    triple buffer. (We have to use the fake support on those Xservers to
    workaround crashes inside the Xserver.)
    
    Reported-by: Paulo Dias <paulo.miguel.dias@gmail.com>
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=85144#c12
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit de96fbf66cf9a0df85cfd1dad8cf786a1d6f8325
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Tue Oct 21 12:20:29 2014 +0100

    sna/trapezoids: Prefer GPU for AddTraps()
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit b4872f74c05a92c91d7b57146eb35ed2ecc86d97
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Tue Oct 21 10:38:22 2014 +0100

    sna: Avoid pwriting large batches
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 83a3d9147308f4777324abdea859ac0c108f03c6
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Sun Oct 19 09:11:33 2014 +0100

    sna: Prefer the GPU for wide lines and arcs
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 316155db98aac4d5d0a7077e86453e4d41a3029d
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Sat Oct 18 10:58:03 2014 +0100

    sna/gen6: Apply gen7 flushing
    
    Clemens Eisserer noticed that glyphs would randomly disapper whilst
    being rendered on his Sandybridge, a sign that the GPU is not flushing
    its internal state on pipeline changes. As a precaution, adopt the
    Ivybridge flush semantics (whilst preserving the required Sandybridge
    workarounds).
    
    Reported-and-tested-by: Clemens Eisserer <linuxhippy@gmail.com>
    Reported-and-tested-by: Ilia Mirkin <imirkin@alum.mit.edu>
    References: https://bugs.freedesktop.org/show_bug.cgi?id=85158
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit bf7b5a24bbfc8a3553be6baa457ec92559dd3d65
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Sat Oct 18 10:13:31 2014 +0100

    sna/dri2: Exchange pitches as well
    
    When swapping over the names, we also need to swap the pitches exposed
    through the protocol as well.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 6b98f16241c2a4788f3b5fe4c0d956a849d2ac05
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Thu Oct 16 08:26:03 2014 +0100

    sna: Allow TearFree updates to continue even when the GPU is wedged
    
    Even if we cannot render using the GPU, we should still be able to
    request that the outputs be flipped. So try, and only if that fails,
    resort to writing directly into the scanout.
    
    References: https://bugs.freedesktop.org/show_bug.cgi?id=85058
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit c896f249c7b64683839665ff528920382f4fc58e
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Thu Oct 16 08:17:09 2014 +0100

    sna: Fixup a failed per-CRTC TearFree flip
    
    If we fail to flip onto a new per-CRTC bo, first try a normal modeset.
    This catches instances where the pitch changes for example.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 3812dfce175eeaef7f41632d2df6a0d88c0ae19b
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Thu Oct 16 07:46:26 2014 +0100

