Commit Graph

142 Commits

Author SHA1 Message Date
Alexandre Courbot af1f85ddec drm/ttm: remove cpu_address member from ttm_tt
Patch 3d50d4dcb0 exposed the CPU address of DMA-allocated pages as
returned by dma_alloc_coherent because Nouveau on Tegra needed it.

This is not required anymore - as there were no other users for it,
remove it and save some memory for everyone.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-09-16 15:53:05 -04:00
Christian König 7c732ea649 drm/ttm: move placement structures into ttm_placement.h
Makes more sense to keep that together.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-09-14 15:10:43 -04:00
Christian König 08614da704 drm/ttm: remove unused placement flags
Either never used or not used in quite a while.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-09-14 15:10:42 -04:00
Christian König 283cde69aa drm/ttm: rework handling of private mem types
Instead of keeping a bunch of potentially unused flags, just define
the start for private memory types and remove the rest.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-09-14 15:10:42 -04:00
Michel Dänzer 4499f2acd5 drm/ttm: Remove unused parameter evict from ttm_bo_move_memcpy
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-08-08 11:33:33 -04:00
Michel Dänzer 4e2f0caa39 drm/ttm: Remove unused parameter evict from ttm_bo_move_ttm
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-08-08 11:33:32 -04:00
Michel Dänzer 7b8082bc07 drm/ttm: Wait for a BO to become idle before unbinding it from GTT
Fixes hangs under memory pressure, e.g. running the piglit test
tex3d-maxsize concurrently with other tests.

Fixes: 17d33bc9d6 ("drm/ttm: drop waiting for idle in ttm_bo_evict.")
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-08-08 11:33:31 -04:00
Ken Wang ecf6a63709 drm/ttm: Add interface to export kernel_zone max memory size in ttm
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-08-08 11:33:08 -04:00
Christian König 2ff2bf1e5c drm/ttm: partial revert "cleanup ttm_tt_(unbind|destroy)" v3
We still need to unbind explicitly during a move.

This partial reverts commit ff20caa0bcbfef9f7686f8d1868a3b990921afd6.

v2: remove unnecessary check and unused variable
v3: fix typo in commit message

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-07-29 14:37:04 -04:00
Dave Airlie 5e580523d9 Linux 4.7
-----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1
 
 iQEcBAABAgAGBQJXlRXSAAoJEHm+PkMAQRiGG/gH/0Z8O4zWOsrwO+X1mRToRDBH
 joFOjAmCVe83T1VpF5LYNB+9+owL/dEDt6+ZIswnhH7AfQPjs4RqwS4PcuMbCDVO
 +mDm0PmfcKaYcQZrB2Z2OwIzRNnfCTVcsDPhIHwuIHk0m4z/xuGZonD8KoAj0+tO
 3yJF6sbE1KubDVjOb+lmZZSP3cXA0pDXrNhkYhE4Tsr8fiihGjeXSNJ8t2zPLjxo
 W3MPqo0rzDvQsOwoF4TWHHagVaFSJlhLBBgqu33fI7uO3jtfQD2G8wG68JCND1j3
 qbMoBfTLFV/yQmSIJUt0Wv1axaCcwnjpweEB35A/GEeZ0mNB1rDdoBeI1eKEQkc=
 =DGFC
 -----END PGP SIGNATURE-----

Backmerge tag 'v4.7' into drm-next

Linux 4.7

As requested by Daniel Vetter as the conflicts were getting messy.
2016-07-26 17:26:29 +10:00
Christian König 3ddf4ad917 drm/ttm: add the infrastructure for pipelined evictions
Free up the memory immediately, remember the last eviction for each domain and
make new allocations depend on the last eviction to be completed.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-07-07 14:54:41 -04:00
Christian König 5bc730677b drm/ttm: remove TTM_BO_PRIV_FLAG_MOVING
Instead of using the flag just remember the fence of the last move operation.

This avoids waiting for command submissions pipelined after the move, but
before accessing the BO with the CPU again.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-07-07 14:54:40 -04:00
Christian König 74561cd4f1 drm/ttm: remove no_gpu_wait param from ttm_bo_move_accel_cleanup
It isn't used and not waiting for the GPU after scheduling a move is
actually quite dangerous.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-07-07 14:54:39 -04:00
Christian König 77dfc28bad drm/ttm: wait for BO idle in ttm_bo_move_memcpy
When we want to pipeline accelerated moves we need to wait in the fallback path.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-07-07 14:54:35 -04:00
Christian König 089f16c55b drm/ttm: cleanup ttm_tt_(unbind|destroy)
ttm_tt_destroy should be the only one unbinding the object.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-07-07 14:54:33 -04:00
Sinclair Yeh 94477bff39 drm/ttm: Make ttm_bo_mem_compat available
There are cases where it is desired to see if a proposed placement
is compatible with a buffer object before calling ttm_bo_validate().

Signed-off-by: Sinclair Yeh <syeh@vmware.com>
Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>
Cc: <stable@vger.kernel.org>
---
This is the first of a 3-patch series to fix a black screen
issue observed on Ubuntu 16.04 server.
2016-07-01 10:47:49 -07:00
Christian König 98c2872ae9 drm/ttm: implement LRU add callbacks v2
This allows fine grained control for the driver where to add a BO into the LRU.

v2: fix typo in comment

Reviewed-by: Sinclair Yeh <syeh@vmware.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-04 20:21:38 -04:00
Christian König c3ea576e05 drm/ttm: add optional LRU removal callback v2
Useful for driver specific LRU handling.

v2: fix typo in comment

Reviewed-by: Sinclair Yeh <syeh@vmware.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-04 20:21:34 -04:00
Christian König 00cfd664f6 drm/ttm: remove unused validation sequence
Not used any more.

Reviewed-by: Sinclair Yeh <syeh@vmware.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-04 20:21:30 -04:00
Christian König 8aa6d4fc5f drm/ttm: remove lazy parameter from ttm_bo_wait
Not used any more.

Reviewed-by: Sinclair Yeh <syeh@vmware.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-04 20:21:26 -04:00
Christian König dfd5e50ea4 drm/ttm: remove use_ticket parameter from ttm_bo_reserve
Not used any more.

Reviewed-by: Sinclair Yeh <syeh@vmware.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-04 20:21:21 -04:00
Dave Airlie 49047962ec Linux 4.6-rc3
-----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1
 
 iQEcBAABAgAGBQJXCva8AAoJEHm+PkMAQRiGXBoIAIkrjxdbuT2nS9A3tHwkiFXa
 6/Th1UjbNaoLuZ+MckQHayAD9NcWY9lVjOUmFsSiSWMCQK/rTWDl8x5ITputrY2V
 VuhrJCwI7huEtu6GpRaJaUgwtdOjhIHz1Ue2MCdNIbKX3l+LjVyyJ9Vo8rruvZcR
 fC7kiivH04fYX58oQ+SHymCg54ny3qJEPT8i4+g26686m11hvZLI3UAs2PAn6ut+
 atCjxdQ4yLN3DWsbjuA7wYGWhTgFloxL4TIoisuOUc3FXnSi/ivIbXZvu4lUfisz
 LA2JBhfII3AEMBWG9xfGbXPijJTT4q7yNlTD0oYcnMtAt/Roh2F04asqB1LetEY=
 =bri6
 -----END PGP SIGNATURE-----

Merge tag 'v4.6-rc3' into drm-next

Backmerge 4.6-rc3 for i915.

Linux 4.6-rc3
2016-04-22 08:32:51 +10:00
Alex Deucher 749b48faaf drm/ttm: use phys_addr_t for ttm_bus_placement
Fixes ttm on platforms like PPC460 where the CPU
is in 32-bit mode, but the physical addresses are
>32 bits.

Extracted from a patch by Hans Verkuil.

Tested-by: Julian Margetson <runaway@candw.ms>
Acked-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Cc: Thomas Hellstrom <thellstrom@vmware.com>
Cc: Julian Margetson <runaway@candw.ms>
Cc: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-04-04 17:00:01 -04:00
Daniel Vetter e6bf6e5799 drm/ttm: Remove TTM_HAS_AGP
It tries to do fancy things with excluding agp support if ttm is
built-in, but agp isn't. Instead just express this depency like drm
does and use CONFIG_AGP everywhere.

Also use the neat Makefile magic to make the entire ttm_agp_backend
file optional.

v2: Use IS_ENABLED(CONFIG_AGP) as suggested by Ville

v3: Review from Emil.

v4: Actually get it right as spotted by 0-day.

Cc: Emil Velikov <emil.l.velikov@gmail.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/1459337046-25882-1-git-send-email-daniel.vetter@ffwll.ch
2016-03-30 17:20:43 +02:00
Christian König ab74961810 drm/ttm: add ttm_bo_move_to_lru_tail function v2
This allows the drivers to move a BO to the end of the LRU
without removing and adding it again.

v2: Make it more robust, handle pinned and swapable BOs as well.

Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-01-13 12:21:21 -05:00
Nicolai Hähnle eb227c554e drm/ttm: fix documentation of ttm_bo_reserve
Previously, the comment was inconsistent. EDEADLK is what the ww_mutex
mechanism really returns.

Signed-off-by: Nicolai Hähnle <Nicolai.Haehnle@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2015-12-15 10:58:39 +10:00
Alex Deucher 54c4cd68ed drm/ttm: device address space != CPU address space
We need to store device offsets in 64 bit as the device
address space may be larger than the CPU's.

Fixes GPU init failures on radeons with 4GB or more of
vram on 32 bit kernels.  We put vram at the start of the
GPU's address space so the gart aperture starts at 4 GB
causing all GPU addresses in the gart aperture to get
truncated.

bug:
https://bugs.freedesktop.org/show_bug.cgi?id=89072

[airlied: fix warning on nouveau build]

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: thellstrom@vmware.com
Acked-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2015-03-05 09:04:39 +10:00
Christian König aa35071c59 drm/ttm: optionally move duplicates to a separate list
This patch adds an optional list_head parameter to ttm_eu_reserve_buffers.
If specified duplicates in the execbuf list are no longer reported as errors,
but moved to this list instead.

Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-12-03 18:26:52 -05:00
Maarten Lankhorst f4f4e3e3e9 drm/ttm: add reservation_object as argument to ttm_bo_init
This allows importing reservation objects from dma-bufs.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
2014-09-30 14:04:00 +02:00
Christian König ae9c0af2c0 drm/ttm: allow fence to be added as shared
This patch adds a new flag to the ttm_validate_buffer list to
add the fence as shared to the reservation object.

Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-09-11 10:46:00 -04:00
Maarten Lankhorst f2c24b83ae drm/ttm: flip the switch, and convert to dma_fence
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
2014-09-02 16:41:50 +02:00
Maarten Lankhorst 1f0dc9a59a drm/ttm: kill off some members to ttm_validate_buffer
This reorders the list to keep track of what buffers are reserved,
so previous members are always unreserved.

This gets rid of some bookkeeping that's no longer needed,
while simplifying the code some.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
2014-09-01 10:18:03 +02:00
Maarten Lankhorst 58b4d720c1 drm/ttm: add interruptible parameter to ttm_eu_reserve_buffers
It seems some drivers really want this as a parameter,
like vmwgfx.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
2014-09-01 10:16:43 +02:00
Maarten Lankhorst dd7cfd6412 drm/ttm: kill fence_lock
No users are left, kill it off! :D
Conversion to the reservation api is next on the list, after
that the functionality can be restored with rcu.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
2014-09-01 10:16:43 +02:00
Christian König f1217ed09f drm/ttm: move fpfn and lpfn into each placement v2
This allows us to more fine grained specify where to place the buffer object.

v2: rebased on drm-next, add bochs changes as well

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-27 13:16:04 +02:00
Alexandre Courbot 3d50d4dcb0 drm/ttm: expose CPU address of DMA-allocated pages
Pages allocated using the DMA API have a coherent memory mapping. Make
this mapping visible to drivers so they can decide to use it instead of
creating their own redundant one.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Acked-by: David Airlie <airlied@linux.ie>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-08-10 01:08:03 +10:00
Christian König e3f202798a drm/ttm: fix handling of TTM_PL_FLAG_TOPDOWN v2
bo->mem.placement is not initialized when ttm_bo_man_get_node is called,
so the flag had no effect at all.

v2: change nouveau and vmwgfx as well

Signed-off-by: Christian König <christian.koenig@amd.com>
Cc: stable@vger.kernel.org
Signed-off-by: Dave Airlie <airlied@redhat.com>
2014-07-08 11:15:58 +10:00
Alexandre Courbot 15f8f65751 drm/ttm: remove declaration of ttm_tt_cache_flush
ttm_tt_cache_flush's implementation was removed in 2009 by commit
c9c97b8c, but its declaration has been hiding in ttm_bo_driver.h since
then.

It has been surviving in the dark for too long now ; give it the mercy
blow.

Reviewed-by: Thierry Reding <treding@nvidia.com>
Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
2014-07-08 00:29:57 +02:00
Alexandre Courbot 59201052df drm/ttm: fix kerneldoc of ttm_bo_create
The kerneldoc header of ttm_bo_create() was referring to another
(nonexisting) function and had a few obsolete or incorrect arguments.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2014-05-26 12:54:09 +10:00
Dave Airlie 8d51a977a4 Merge tag 'ttm-next-2014-04-04' of git://people.freedesktop.org/~thomash/linux into drm-next
Pull request of 2014-04-04

Currently only a single patch fixing up mixed use of the ttm_bo_reserve and
ww_mutex APIs

* tag 'ttm-next-2014-04-04' of git://people.freedesktop.org/~thomash/linux:
  drm/ttm: Hide the implementation details of reservation
2014-04-05 16:07:39 +10:00
Thomas Hellstrom c75230833c drm/ttm: Hide the implementation details of reservation
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
2014-04-04 08:00:59 +02:00
Lauri Kasanen 62347f9e0f drm: Add support for two-ended allocation, v3
Clients like i915 need to segregate cache domains within the GTT which
can lead to small amounts of fragmentation. By allocating the uncached
buffers from the bottom and the cacheable buffers from the top, we can
reduce the amount of wasted space and also optimize allocation of the
mappable portion of the GTT to only those buffers that require CPU
access through the GTT.

For other drivers, allocating small bos from one end and large ones
from the other helps improve the quality of fragmentation.

Based on drm_mm work by Chris Wilson.

v3: Changed to use a TTM placement flag
v2: Updated kerneldoc

Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Ben Widawsky <ben@bwidawsk.net>
Cc: Christian König <deathsimple@vodafone.de>
Signed-off-by: Lauri Kasanen <cand@gmx.com>
Signed-off-by: David Airlie <airlied@redhat.com>
2014-04-04 09:28:14 +10:00
Thomas Hellstrom 0d3215e385 drm/ttm: Add a ttm_ref_object_exists function
A function to be used to check whether a caller has put a ref object
(opened) a struct ttm_base_object

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
2014-03-28 14:19:03 +01:00
David Herrmann 44d847b743 drm: init TTM dev_mapping in ttm_bo_device_init()
With dev->anon_inode we have a global address_space ready for operation
right from the beginning. Therefore, there is no need to do a delayed
setup with TTM. Instead, set dev_mapping during initialization in
ttm_bo_device_init() and remove any "if (dev_mapping)" conditions.

Cc: Dave Airlie <airlied@redhat.com>
Cc: Ben Skeggs <bskeggs@redhat.com>
Cc: Maarten Lankhorst <maarten.lankhorst@canonical.com>
Cc: Alex Deucher <alexdeucher@gmail.com>
Cc: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
2014-03-16 12:23:42 +01:00
Alexandre Courbot 728a0cdf06 drm/ttm: declare 'struct device' in ttm_page_alloc.h
Declare 'struct device' explicitly in ttm_page_alloc.h as this file
does not include any file declaring it. This removes the following
warning:

	warning: 'struct device' declared inside parameter list

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Thierry Reding <treding@nvidia.com>
2014-02-18 14:01:48 +01:00
Dave Airlie faf096ffba Merge tag 'vmwgfx-next-2014-01-13' of git://people.freedesktop.org/~thomash/linux into drm-next
Anyway, nothing big here, Three more code cleanup patches from Rashika
Kheria, and one TTM/vmwgfx patch from me that tightens security around TTM
objects enough for them to opened using prime objects from render nodes:

Previously any client could access a shared buffer using the "name", also
without actually opening it. Now a reference is required, and for render nodes
such a reference is intended to only be obtainable using a prime fd.

vmwgfx-next 2014-01-13 pull request

* tag 'vmwgfx-next-2014-01-13' of git://people.freedesktop.org/~thomash/linux:
  drivers: gpu: Mark functions as static in vmwgfx_fence.c
  drivers: gpu: Mark functions as static in vmwgfx_buffer.c
  drivers: gpu: Mark functions as static in vmwgfx_kms.c
  drm/ttm: ttm object security fixes for render nodes
2014-01-14 10:55:36 +10:00
Thomas Hellstrom 05efb1abec drm/ttm: ttm object security fixes for render nodes
When a client looks up a ttm object, don't look it up through the device hash
table, but rather from the file hash table. That makes sure that the client
has indeed put a reference on the object, or in gem terms, has opened
the object; either using prime or using the global "name".

To avoid a performance loss, make sure the file hash table entries can be
looked up from under an RCU lock, and as a consequence, replace the rwlock
with a spinlock, since we never need to take it in read mode only anymore.

Finally add a ttm object lookup function for the device hash table, that is
intended to be used when we put a ref object on a base object or, in  gem terms,
when we open the object.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
2014-01-08 10:11:57 +01:00
Thomas Hellstrom 58aa6622d3 drm/ttm: Correctly set page mapping and -index members
Needed for some vm operations; most notably unmap_mapping_range() with
even_cows = 0.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
2014-01-08 10:08:28 +01:00
Dave Airlie 27946e9710 Merge branch 'ttm-fixes-3.13' of git://people.freedesktop.org/~thomash/linux into drm-fixes
The set_need_resched() removal fix and yet another fix in
ttm_bo_move_memcpy().

* 'ttm-fixes-3.13' of git://people.freedesktop.org/~thomash/linux:
  drm/ttm: Remove set_need_resched from the ttm fault handler
  drm/ttm: Don't move non-existing data
2013-11-21 18:46:56 +10:00
Thomas Hellstrom c58f009e01 drm/ttm: Remove set_need_resched from the ttm fault handler
Addresses
"[BUG] completely bonkers use of set_need_resched + VM_FAULT_NOPAGE".

In the first occurence it was used to try to be nice while releasing the
mmap_sem and retrying the fault to work around a locking inversion.
The second occurence was never used.

There has been some discussion whether we should change the locking order to
mmap_sem -> bo_reserve. This patch doesn't address that issue, and leaves
that locking order undefined. The solution that we release the mmap_sem if
tryreserve fails and wait for the buffer to become unreserved is something
we want in any case, and follows how the core vm system waits for pages
to be come unlocked while releasing the mmap_sem.

The code also outlines what needs to be changed if we want to establish the
locking order as mmap_sem -> bo::reserve.

One slight issue that remains with this code is that the fault handler might
be prone to starvation if another thread countinously reserves the buffer.
IMO that usage pattern is highly unlikely.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2013-11-20 03:46:54 -08:00