Allow interval trees to quickly check for overlaps to avoid unnecesary
tree lookups in interval_tree_iter_first().
As of this patch, all interval tree flavors will require using a
'rb_root_cached' such that we can have the leftmost node easily
available. While most users will make use of this feature, those with
special functions (in addition to the generic insert, delete, search
calls) will avoid using the cached option as they can do funky things
with insertions -- for example, vma_interval_tree_insert_after().
[jglisse@redhat.com: fix deadlock from typo vm_lock_anon_vma()]
Link: http://lkml.kernel.org/r/20170808225719.20723-1-jglisse@redhat.com
Link: http://lkml.kernel.org/r/20170719014603.19029-12-dave@stgolabs.net
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
Acked-by: Christian König <christian.koenig@amd.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Doug Ledford <dledford@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Cc: David Airlie <airlied@linux.ie>
Cc: Jason Wang <jasowang@redhat.com>
Cc: Christian Benvenuti <benve@cisco.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The drm_mm range manager claimed to support top-down insertion, but it
was neither searching for the top-most hole that could fit the
allocation request nor fitting the request to the hole correctly.
In order to search the range efficiently, we create a secondary index
for the holes using either their size or their address. This index
allows us to find the smallest hole or the hole at the bottom or top of
the range efficiently, whilst keeping the hole stack to rapidly service
evictions.
v2: Search for holes both high and low. Rename flags to mode.
v3: Discover rb_entry_safe() and use it!
v4: Kerneldoc for enum drm_mm_insert_mode.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: "Christian König" <christian.koenig@amd.com>
Cc: David Airlie <airlied@linux.ie>
Cc: Russell King <rmk+kernel@armlinux.org.uk>
Cc: Daniel Vetter <daniel.vetter@intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Sean Paul <seanpaul@chromium.org>
Cc: Lucas Stach <l.stach@pengutronix.de>
Cc: Christian Gmeiner <christian.gmeiner@gmail.com>
Cc: Rob Clark <robdclark@gmail.com>
Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: Stephen Warren <swarren@wwwdotorg.org>
Cc: Alexandre Courbot <gnurou@gmail.com>
Cc: Eric Anholt <eric@anholt.net>
Cc: Sinclair Yeh <syeh@vmware.com>
Cc: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Sinclair Yeh <syeh@vmware.com> # vmwgfx
Reviewed-by: Lucas Stach <l.stach@pengutronix.de> #etnaviv
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20170202210438.28702-1-chris@chris-wilson.co.uk
Rather than using "struct file*", use "struct drm_file*" as tag VM tag for
BOs. This will pave the way for "struct drm_file*" without any "struct
file*" back-pointer.
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20160901124837.680-3-dh.herrmann@gmail.com
Just fallout from switching from asciidoc to sphinx/rst.
v2: Found more. Also s/\//#/ in the vgpu ascii-art - sphinx treats
those as comments and switch to variable-width, which wreaks the
layout.
v3: Undo some of the hacks, rebasing onto latest version of Jani's
series fixed it.
Acked-by: Liviu Dudau <Liviu.Dudau@arm.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Compared to wrapping the final kref_put with dev->struct_mutex this
allows us to only acquire the offset manager look both in the final
cleanup and in the lookup. Which has the upside that no locks leak out
of the core abstractions. But it means that we need to hold a
temporary reference to the object while checking mmap constraints, to
make sure the object doesn't disappear. Extended the critical region
would have worked too, but would result in more leaky locking.
Also, this is the final bit which required dev->struct_mutex in gem
core, now modern drivers can be completely struct_mutex free!
This needs a new drm_vma_offset_exact_lookup_locked and makes both
drm_vma_offset_exact_lookup and drm_vma_offset_lookup unused.
v2: Don't leak object references in failure paths (David).
v3: Add a comment from Chris explaining how the ordering works, with
the slight adjustment that I dropped any mention of struct_mutex since
with this patch it's now immaterial ot core gem.
Cc: David Herrmann <dh.herrmann@gmail.com>
Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://mid.gmane.org/1444901623-18918-1-git-send-email-daniel.vetter@ffwll.ch
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
One bit in ->vm_flags is unused now!
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Dan Carpenter <dan.carpenter@oracle.com>
Cc: Michal Hocko <mhocko@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The VMA offset manager uses a device-global address-space. Hence, any
user can currently map any offset-node they want. They only need to guess
the right offset. If we wanted per open-file offset spaces, we'd either
need VM_NONLINEAR mappings or multiple "struct address_space" trees. As
both doesn't really scale, we implement access management in the VMA
manager itself.
We use an rb-tree to store open-files for each VMA node. On each mmap
call, GEM, TTM or the drivers must check whether the current user is
allowed to map this file.
We add a separate lock for each node as there is no generic lock available
for the caller to protect the node easily.
As we currently don't know whether an object may be used for mmap(), we
have to do access management for all objects. If it turns out to slow down
handle creation/deletion significantly, we can optimize it in several
ways:
- Most times only a single filp is added per bo so we could use a static
"struct file *main_filp" which is checked/added/removed first before we
fall back to the rbtree+drm_vma_offset_file.
This could be even done lockless with rcu.
- Let user-space pass a hint whether mmap() should be supported on the
bo and avoid access-management if not.
- .. there are probably more ideas once we have benchmarks ..
v2: add drm_vma_node_verify_access() helper
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Add a "best_match" flag similar to the drm_mm_search_*() helpers so we
can convert TTM to use them in follow up patches. We can also inline the
non-generic helpers and move them into the header to allow compile-time
optimizations.
To make calls to drm_mm_{search,insert}_node() more readable, this
converts the boolean argument to a flagset. There are pending patches that
add additional flags for top-down allocators and more.
v2:
- use flag parameter instead of boolean "best_match"
- convert *_search_free() helpers to also use flags argument
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Dave Airlie <airlied@redhat.com>
If we want to map GPU memory into user-space, we need to linearize the
addresses to not confuse mm-core. Currently, GEM and TTM both implement
their own offset-managers to assign a pgoff to each object for user-space
CPU access. GEM uses a hash-table, TTM uses an rbtree.
This patch provides a unified implementation that can be used to replace
both. TTM allows partial mmaps with a given offset, so we cannot use
hashtables as the start address may not be known at mmap time. Hence, we
use the rbtree-implementation of TTM.
We could easily update drm_mm to use an rbtree instead of a linked list
for it's object list and thus drop the rbtree from the vma-manager.
However, this would slow down drm_mm object allocation for all other
use-cases (rbtree insertion) and add another 4-8 bytes to each mm node.
Hence, use the separate tree but allow for later migration.
This is a rewrite of the 2012-proposal by David Airlie <airlied@linux.ie>
v2:
- fix Docbook integration
- drop drm_mm_node_linked() and use drm_mm_node_allocated()
- remove unjustified likely/unlikely usage (but keep for rbtree paths)
- remove BUG_ON() as drm_mm already does that
- clarify page-based vs. byte-based addresses
- use drm_vma_node_reset() for initialization, too
v4:
- allow external locking via drm_vma_offset_un/lock_lookup()
- add locked lookup helper drm_vma_offset_lookup_locked()
v5:
- fix drm_vma_offset_lookup() to correctly validate range-mismatches
(fix (offset > start + pages))
- fix drm_vma_offset_exact_lookup() to actually do what it says
- remove redundant vm_pages member (add drm_vma_node_size() helper)
- remove unneeded goto
- fix documentation
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Dave Airlie <airlied@gmail.com>
