This was the competing idea long ago, but it was only with the rewrite
of the idr as an radixtree and using the radixtree directly ourselves,
along with the realisation that we can store the vma directly in the
radixtree and only need a list for the reverse mapping, that made the
patch performant enough to displace using a hashtable. Though the vma ht
is fast and doesn't require any extra allocation (as we can embed the node
inside the vma), it does require a thread for resizing and serialization
and will have the occasional slow lookup. That is hairy enough to
investigate alternatives and favour them if equivalent in peak performance.
One advantage of allocating an indirection entry is that we can support a
single shared bo between many clients, something that was done on a
first-come first-serve basis for shared GGTT vma previously. To offset
the extra allocations, we create yet another kmem_cache for them.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20170816085210.4199-5-chris@chris-wilson.co.uk
Another month, another story in the cache coherency saga. This time, we
come to the realisation that i915_gem_object_is_coherent() has been
reporting whether we can read from the target without requiring a cache
invalidate; but we were using it in places for testing whether we could
write into the object without requiring a cache flush. So split the
tracking into two, one to decide before reads, one after writes.
See commit e27ab73d17 ("drm/i915: Mark CPU cache as dirty on every
transition for CPU writes") for the previous entry in this saga.
v2: Be verbose
v3: Remove unused function (i915_gem_object_is_coherent)
v4: Fix inverted coherency check prior to execbuf (from v2)
v5: Add comment for nasty code where we are optimising on gcc's behalf.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101109
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101555
Testcase: igt/kms_mmap_write_crc
Testcase: igt/kms_pwrite_crc
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Dongwon Kim <dongwon.kim@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Tested-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Acked-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20170811111116.10373-1-chris@chris-wilson.co.uk
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
The advent of full-ppgtt lead to an extra indirection between the object
and its binding. That extra indirection has a noticeable impact on how
fast we can convert from the user handles to our internal vma for
execbuffer. In order to bypass the extra indirection, we use a
resizable hashtable to jump from the object to the per-ctx vma.
rhashtable was considered but we don't need the online resizing feature
and the extra complexity proved to undermine its usefulness. Instead, we
simply reallocate the hastable on demand in a background task and
serialize it before iterating.
In non-full-ppgtt modes, multiple files and multiple contexts can share
the same vma. This leads to having multiple possible handle->vma links,
so we only use the first to establish the fast path. The majority of
buffers are not shared and so we should still be able to realise
speedups with multiple clients.
v2: Prettier names, more magic.
v3: Many style tweaks, most notably hiding the misuse of execobj[].rsvd2
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
For ease of use (i.e. avoiding a few checks and function calls), store
the object's cache coherency next to the cache is dirty bit.
Specifically this patch aims to reduce the frequency of no-op calls to
i915_gem_object_clflush() to counter-act the increase of such calls for
GPU only objects in the previous patch.
v2: Replace cache_dirty & ~cache_coherent with cache_dirty &&
!cache_coherent as gcc generates much better code for the latter
(Tvrtko)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Dongwon Kim <dongwon.kim@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Tested-by: Dongwon Kim <dongwon.kim@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170616105455.16977-1-chris@chris-wilson.co.uk
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Add kerneldoc for the vma_list stored on the i915_gem_object, in
particular, documenting the expected ordering of elements -- i.e. that
we do expect GGTT VMA first followed by the ppGTT VMA.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170525204818.12044-1-chris@chris-wilson.co.uk
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Having just watched someone add a new value, 0x3, without realising that
the flags were bit values, I have come to appreciate the value in using
BIT.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170523103116.32239-1-chris@chris-wilson.co.uk
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
i915_gem_object_is_dead() was a temporary lockdep aide whilst
transitioning to a new locking structure for obj->mm. Since commit
1233e2db19 ("drm/i915: Move object backing storage manipulation to its
own locking") it is now unused and should be removed.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170308132629.7987-2-chris@chris-wilson.co.uk
Backmerge drm-next to get at all the good stuff in drm-misc. We need
that because:
- drm_connector_list_iter conversion for i915 needs the core patches.
- Maarten's patches to use the new atomic state iterators also need
the core patches.
- We need the new link status property to complete the DP retraining
work, merging through 2 branches wasn't a good idea and we had to
partially backtrack.
- Chris needs reservation_object_trylock and we want to roll out
kref_read everywhere.
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
4 weeks worth of stuff since I was traveling&lazy:
- lspcon improvements (Imre)
- proper atomic state for cdclk handling (Ville)
- gpu reset improvements (Chris)
- lots and lots of polish around fences, requests, waiting and
everything related all over (both gem and modeset code), from Chris
- atomic by default on gen5+ minus byt/bsw (Maarten did the patch to
flip the default, really this is a massive joint team effort)
- moar power domains, now 64bit (Ander)
- big pile of in-kernel unit tests for various gem subsystems (Chris),
including simple mock objects for i915 device and and the ggtt
manager.
- i915_gpu_info in debugfs, for taking a snapshot of the current gpu
state. Same thing as i915_error_state, but useful if the kernel didn't
notice something is stick. From Chris.
- bxt dsi fixes (Umar Shankar)
- bxt w/a updates (Jani)
- no more struct_mutex for gem object unreference (Chris)
- some execlist refactoring (Tvrtko)
- color manager support for glk (Ander)
- improve the power-well sync code to better take over from the
firmware (Imre)
- gem tracepoint polish (Tvrtko)
- lots of glk fixes all around (Ander)
- ctx switch improvements (Chris)
- glk dsi support&fixes (Deepak M)
- dsi fixes for vlv and clanups, lots of them (Hans de Goede)
- switch to i915.ko types in lots of our internal modeset code (Ander)
- byt/bsw atomic wm update code, yay (Ville)
* tag 'drm-intel-next-2017-03-06' of git://anongit.freedesktop.org/git/drm-intel: (432 commits)
drm/i915: Update DRIVER_DATE to 20170306
drm/i915: Don't use enums for hardware engine id
drm/i915: Split breadcrumbs spinlock into two
drm/i915: Refactor wakeup of the next breadcrumb waiter
drm/i915: Take reference for signaling the request from hardirq
drm/i915: Add FIFO underrun tracepoints
drm/i915: Add cxsr toggle tracepoint
drm/i915: Add VLV/CHV watermark/FIFO programming tracepoints
drm/i915: Add plane update/disable tracepoints
drm/i915: Kill level 0 wm hack for VLV/CHV
drm/i915: Workaround VLV/CHV sprite1->sprite0 enable underrun
drm/i915: Sanitize VLV/CHV watermarks properly
drm/i915: Only use update_wm_{pre,post} for pre-ilk platforms
drm/i915: Nuke crtc->wm.cxsr_allowed
drm/i915: Compute proper intermediate wms for vlv/cvh
drm/i915: Skip useless watermark/FIFO related work on VLV/CHV when not needed
drm/i915: Compute vlv/chv wms the atomic way
drm/i915: Compute VLV/CHV FIFO sizes based on the PM2 watermarks
drm/i915: Plop vlv/chv fifo sizes into crtc state
drm/i915: Plop vlv wm state into crtc_state
...
Before we instantiate/pin the backing store for our use, we
can prepopulate the shmemfs filp efficiently using a write into the
pagecache. We avoid the penalty of instantiating all the pages, important
if the user is just writing to a few and never uses the object on the GPU,
and using a direct write into shmemfs allows it to avoid the cost of
retrieving a page (mostly the clear-before-use, but in theory we could
curtail swapin) before it is overwritten.
This can be extended later to provide additional specialisation for
other backends (other than shmemfs). For now it provides a defense
against very large write-only allocations from exhausting all of system
memory.
v2: Smelling fixes.
Fixes: fe115628d5 ("drm/i915: Implement pwrite without struct-mutex")
References: https://bugs.freedesktop.org/show_bug.cgi?id=99107
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Matthew Auld <matthew.william.auld@gmail.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: <stable@vger.kernel.org> # v4.10+
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170307120338.7277-2-chris@chris-wilson.co.uk
Reintroduce a lock around tiling vs framebuffer creation to prevent
modification of the obj->tiling_and_stride whilst the framebuffer is
being created. Rather than use struct_mutex once again, use the
per-object lock - this will also be required in future to prevent
changing the tiling whilst submitting rendering.
Reported-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Fixes: 24dbf51a55 ("drm/i915: struct_mutex is not required for allocating the framebuffer")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170301154128.2841-2-chris@chris-wilson.co.uk
-----BEGIN PGP SIGNATURE-----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=Yjfg
-----END PGP SIGNATURE-----
Merge tag 'drm-for-v4.11-less-shouty' of git://people.freedesktop.org/~airlied/linux
Pull drm updates from Dave Airlie:
"This is the main drm pull request for v4.11.
Nothing too major, the tinydrm and mmu-less support should make
writing smaller drivers easier for some of the simpler platforms, and
there are a bunch of documentation updates.
Intel grew displayport MST audio support which is hopefully useful to
people, and FBC is on by default for GEN9+ (so people know where to
look for regressions). AMDGPU has a lot of fixes that would like new
firmware files installed for some GPUs.
Other than that it's pretty scattered all over.
I may have a follow up pull request as I know BenH has a bunch of AST
rework and fixes and I'd like to get those in once they've been tested
by AST, and I've got at least one pull request I'm just trying to get
the author to fix up.
Core:
- drm_mm reworked
- Connector list locking and iterators
- Documentation updates
- Format handling rework
- MMU-less support for fbdev helpers
- drm_crtc_from_index helper
- Core CRC API
- Remove drm_framebuffer_unregister_private
- Debugfs cleanup
- EDID/Infoframe fixes
- Release callback
- Tinydrm support (smaller drivers for simple hw)
panel:
- Add support for some new simple panels
i915:
- FBC by default for gen9+
- Shared dpll cleanups and docs
- GEN8 powerdomain cleanup
- DMC support on GLK
- DP MST audio support
- HuC loading support
- GVT init ordering fixes
- GVT IOMMU workaround fix
amdgpu/radeon:
- Power/clockgating improvements
- Preliminary SR-IOV support
- TTM buffer priority and eviction fixes
- SI DPM quirks removed due to firmware fixes
- Powerplay improvements
- VCE/UVD powergating fixes
- Cleanup SI GFX code to match CI/VI
- Support for > 2 displays on 3/5 crtc asics
- SI headless fixes
nouveau:
- Rework securre boot code in prep for GP10x secure boot
- Channel recovery improvements
- Initial power budget code
- MMU rework preperation
vmwgfx:
- Bunch of fixes and cleanups
exynos:
- Runtime PM support for MIC driver
- Cleanups to use atomic helpers
- UHD Support for TM2/TM2E boards
- Trigger mode fix for Rinato board
etnaviv:
- Shader performance fix
- Command stream validator fixes
- Command buffer suballocator
rockchip:
- CDN DisplayPort support
- IOMMU support for arm64 platform
imx-drm:
- Fix i.MX5 TV encoder probing
- Remove lower fb size limits
msm:
- Support for HW cursor on MDP5 devices
- DSI encoder cleanup
- GPU DT bindings cleanup
sti:
- stih410 cleanups
- Create fbdev at binding
- HQVDP fixes
- Remove stih416 chip functionality
- DVI/HDMI mode selection fixes
- FPS statistic reporting
omapdrm:
- IRQ code cleanup
dwi-hdmi bridge:
- Cleanups and fixes
adv-bridge:
- Updates for nexus
sii8520 bridge:
- Add interlace mode support
- Rework HDMI and lots of fixes
qxl:
- probing/teardown cleanups
ZTE drm:
- HDMI audio via SPDIF interface
- Video Layer overlay plane support
- Add TV encoder output device
atmel-hlcdc:
- Rework fbdev creation logic
tegra:
- OF node fix
fsl-dcu:
- Minor fixes
mali-dp:
- Assorted fixes
sunxi:
- Minor fix"
[ This was the "fixed" pull, that still had build warnings due to people
not even having build tested the result. I'm not a happy camper
I've fixed the things I noticed up in this merge. - Linus ]
* tag 'drm-for-v4.11-less-shouty' of git://people.freedesktop.org/~airlied/linux: (1177 commits)
lib/Kconfig: make PRIME_NUMBERS not user selectable
drm/tinydrm: helpers: Properly fix backlight dependency
drm/tinydrm: mipi-dbi: Fix field width specifier warning
drm/tinydrm: mipi-dbi: Silence: ‘cmd’ may be used uninitialized
drm/sti: fix build warnings in sti_drv.c and sti_vtg.c files
drm/amd/powerplay: fix PSI feature on Polars12
drm/amdgpu: refuse to reserve io mem for split VRAM buffers
drm/ttm: fix use-after-free races in vm fault handling
drm/tinydrm: Add support for Multi-Inno MI0283QT display
dt-bindings: Add Multi-Inno MI0283QT binding
dt-bindings: display/panel: Add common rotation property
of: Add vendor prefix for Multi-Inno
drm/tinydrm: Add MIPI DBI support
drm/tinydrm: Add helper functions
drm: Add DRM support for tiny LCD displays
drm/amd/amdgpu: post card if there is real hw resetting performed
drm/nouveau/tmr: provide backtrace when a timeout is hit
drm/nouveau/pci/g92: Fix rearm
drm/nouveau/drm/therm/fan: add a fallback if no fan control is specified in the vbios
drm/nouveau/hwmon: expose power_max and power_crit
..
We have three different paths by which userspace wants to flush the
display plane (i.e. objects with obj->pin_display). Use a common helper
to identify those paths and to simplify a later change.
v2: Include the conditional in the name, i915_gem_object_flush_if_display
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170222114049.28456-3-chris@chris-wilson.co.uk
We do not need to hold struct_mutex for destroying drm_i915_gem_objects
any longer, and with a little care taken over tracking
obj->framebuffer_references, we can relinquish BKL locking around the
destroy of intel_framebuffer.
v2: Use atomic check for WARN_ON framebuffer miscounting
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170216094621.3426-1-chris@chris-wilson.co.uk
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Allocate objects with varying number of pages (which should hopefully
consist of a mixture of contiguous page chunks and so coalesced sg
lists) and check that the sg walkers in insert_pages cope.
v2: Check both small <-> large
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170213171558.20942-28-chris@chris-wilson.co.uk
We would like to be able to exercise huge allocations even on memory
constrained devices. To do this we create an object that allocates only
a few pages and remaps them across its whole range - each page is reused
multiple times. We can therefore pretend we are rendering into a much
larger object.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170213171558.20942-9-chris@chris-wilson.co.uk
Since we need to change the implementation, stop exposing internals.
Provide kref_read() to read the current reference count; typically
used for debug messages.
Kills two anti-patterns:
atomic_read(&kref->refcount)
kref->refcount.counter
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
I tried to avoid having to track the write for every VMA by only
tracking writes to the ggtt. However, for the purposes of frontbuffer
tracking this is insufficient as we need to invalidate around writes not
just to the the ggtt but all aliased ppgtt views of the framebuffer. By
moving the critical section to the object and only doing so for
framebuffer writes we can reduce the tracking even further by only
watching framebuffers and not vma.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Paulo Zanoni <paulo.r.zanoni@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161116190704.5293-1-chris@chris-wilson.co.uk
Tested-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
As a side product, had to split two other files;
- i915_gem_fence_reg.h
- i915_gem_object.h (only parts that needed immediate untanglement)
I tried to move code in as big chunks as possible, to make review
easier. i915_vma_compare was moved to a header temporarily.
v2:
- Use i915_gem_fence_reg.{c,h}
v3:
- Rebased
v4:
- Fix building when DEBUG_GEM is enabled by reordering a bit.
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1478861034-30643-1-git-send-email-joonas.lahtinen@linux.intel.com
