Jason Ekstrand [Thu, 8 Jul 2021 15:48:11 +0000 (10:48 -0500)]
drm/i915: Drop the CONTEXT_CLONE API (v2)
This API allows one context to grab bits out of another context upon
creation. It can be used as a short-cut for setparam(getparam()) for
things like I915_CONTEXT_PARAM_VM. However, it's never been used by any
real userspace. It's used by a few IGT tests and that's it. Since it
doesn't add any real value (most of the stuff you can CLONE you can copy
in other ways), drop it.
There is one thing that this API allows you to clone which you cannot
clone via getparam/setparam: timelines. However, timelines are an
implementation detail of i915 and not really something that needs to be
exposed to userspace. Also, sharing timelines between contexts isn't
obviously useful and supporting it has the potential to complicate i915
internally. It also doesn't add any functionality that the client can't
get in other ways. If a client really wants a shared timeline, they can
use a syncobj and set it as an in and out fence on every submit.
v2 (Jason Ekstrand):
- More detailed commit message
Jason Ekstrand [Thu, 8 Jul 2021 15:48:09 +0000 (10:48 -0500)]
drm/i915/gem: Set the watchdog timeout directly in intel_context_set_gem (v2)
Instead of handling it like a context param, unconditionally set it when
intel_contexts are created. For years we've had the idea of a watchdog
uAPI floating about. The aim was for media, so that they could set very
tight deadlines for their transcodes jobs, so that if you have a corrupt
bitstream (especially for decoding) you don't hang your desktop too
hard. But it's been stuck in limbo since forever, and this simplifies
things a bit in preparation for the proto-context work. If we decide to
actually make said uAPI a reality, we can do it through the proto-
context easily enough.
This does mean that we move from reading the request_timeout_ms param
once per engine when engines are created instead of once at context
creation. If someone changes request_timeout_ms between creating a
context and setting engines, it will mean that they get the new timeout.
If someone races setting request_timeout_ms and context creation, they
can theoretically end up with different timeouts. However, since both
of these are fairly harmless and require changing kernel params, we
don't care.
v2 (Tvrtko Ursulin):
- Add a comment about races with request_timeout_ms
Jason Ekstrand [Thu, 8 Jul 2021 15:48:08 +0000 (10:48 -0500)]
drm/i915: Drop I915_CONTEXT_PARAM_NO_ZEROMAP
The idea behind this param is to support OpenCL drivers with relocations
because OpenCL reserves 0x0 for NULL and, if we placed memory there, it
would confuse CL kernels. It was originally sent out as part of a patch
series including libdrm [1] and Beignet [2] support. However, the
libdrm and Beignet patches never landed in their respective upstream
projects so this API has never been used. It's never been used in Mesa
or any other driver, either.
Dropping this API allows us to delete a small bit of code.
Jason Ekstrand [Thu, 8 Jul 2021 15:48:07 +0000 (10:48 -0500)]
drm/i915: Stop storing the ring size in the ring pointer (v3)
Previously, we were storing the ring size in the ring pointer before it
was actually allocated. We would then guard setting the ring size on
checking for CONTEXT_ALLOC_BIT. This is error-prone at best and really
only saves us a few bytes on something that already burns at least 4K.
Instead, this patch adds a new ring_size field and makes everything use
that.
v2 (Daniel Vetter):
- Replace 512 * SZ_4K with SZ_2M
v2 (Jason Ekstrand):
- Rebase on top of page migration code
Jason Ekstrand [Thu, 8 Jul 2021 15:48:06 +0000 (10:48 -0500)]
drm/i915: Drop I915_CONTEXT_PARAM_RINGSIZE
This reverts commit c96dc848a38b ("drm/i915: Allow userspace to specify
ringsize on construction"). This API was originally added for OpenCL
but the compute-runtime PR has sat open for a year without action so we
can still pull it out if we want. I argue we should drop it for three
reasons:
1. If the compute-runtime PR has sat open for a year, this clearly
isn't that important.
2. It's a very leaky API. Ring size is an implementation detail of the
current execlist scheduler and really only makes sense there. It
can't apply to the older ring-buffer scheduler on pre-execlist
hardware because that's shared across all contexts and it won't
apply to the GuC scheduler that's in the pipeline.
3. Having userspace set a ring size in bytes is a bad solution to the
problem of having too small a ring. There is no way that userspace
has the information to know how to properly set the ring size so
it's just going to detect the feature and always set it to the
maximum of 512K. This is what the compute-runtime PR does. The
scheduler in i915, on the other hand, does have the information to
make an informed choice. It could detect if the ring size is a
problem and grow it itself. Or, if that's too hard, we could just
increase the default size from 16K to 32K or even 64K instead of
relying on userspace to do it.
Let's drop this API for now and, if someone decides they really care
about solving this problem, they can do it properly.
John Harrison [Sat, 26 Jun 2021 00:45:22 +0000 (17:45 -0700)]
drm/i915/adlp: Add ADL-P GuC/HuC firmware files
Add ADL-P to the list of supported GuC and HuC firmware versions. For
HuC, it reuses the existing TGL firmware file. For GuC, there is a
dedicated firmware release.
46 bit addressing enables you to use 4 bits to support some
MKTME features, and 3 more bits for Optane support that uses
a subset of MTKME for persistent memory.
But GTT addressing sticking to 39 bit addressing, thus setting
dma_mask_size to 39 fixes below tests :
igt@i915_selftest@live@mman
igt@kms_big_fb@linear-32bpp-rotate-0
igt@gem_create@create-clear
igt@gem_mmap_offset@clear
igt@gem_mmap_gtt@cpuset-big-copy
In a way solves Gitlab#3142
https://gitlab.freedesktop.org/drm/intel/-/issues/3142, which had
following errors :
DMAR: DRHD: handling fault status reg 2
DMAR: [DMA Write] Request device [00:02.0] PASID ffffffff fault addr 7effff9000 [fault reason 05] PTE Write access is not set
0x7effff9000 is suspiciously exactly 39 bits, so it seems likely that
the HW just ends up masking off those extra bits hence DMA errors.
Changes since V2 :
- dim checkpatch error solved
Changes since V1 :
- Added more details to commit message - Matthew Auld
Lucas De Marchi [Wed, 7 Jul 2021 18:13:23 +0000 (11:13 -0700)]
drm/i915/gt: finish INTEL_GEN and friends conversion
Commit 63aa5d1b965e ("drm/i915/gt: replace IS_GEN and friends with
GRAPHICS_VER") converted INTEL_GEN and friends to the new version check
macros. Meanwhile, some changes sneaked in to use INTEL_GEN. Remove the
last users so we can remove the macros.
drm/i915/tracepoints: Add request submit and execute tracepoints
to "alleviate the performance impact concerns."
Which is nonsense.
Tvrtko and Joonas pointed out on irc that the real (but undocumented
reason) was stable abi concerns for tracepoints, see
https://lwn.net/Articles/705270/
and the specific change that was blocked around tracepoints:
https://lwn.net/Articles/442113/
Anyway to make it a notch clearer why we have this Kconfig option
consistly add the "Recommended for driver developers only." to it and
all the other debug options we have.
The code in xcs_resume() probably didn't work as intended. It uses
struct drm_device.irq, which is allocated to 0, but never initialized
by i915 to the device's interrupt number.
Change all calls to synchronize_hardirq() to intel_synchronize_irq(),
which uses the correct interrupt. _hardirq() functions are not needed
in this context.
v5:
* go back to _hardirq() after PCI probe reported wrong
context; add rsp comment
v4:
* switch everything to intel_synchronize_irq() (Daniel)
v3:
* also use intel_synchronize_hardirq() at another callsite
v2:
* wrap irq code in intel_synchronize_hardirq() (Ville)
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de> Fixes: 0209dc948f6e ("drm/i915/gt: Call stop_ring() from ring resume, again") Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: https://patchwork.freedesktop.org/patch/msgid/20210701173618.10718-2-tzimmermann@suse.de
Matthew Auld [Fri, 25 Jun 2021 10:38:24 +0000 (11:38 +0100)]
drm/i915/gtt: ignore min_page_size for paging structures
The min_page_size is only needed for pages inserted into the GTT, and
for our paging structures we only need at most 4K bytes, so simply
ignore the min_page_size restrictions here, otherwise we might see some
severe overallocation on some devices.
Matthew Auld [Fri, 25 Jun 2021 10:38:23 +0000 (11:38 +0100)]
drm/i915: support forcing the page size with lmem
For some specialised objects we might need something larger than the
regions min_page_size due to some hw restriction, and slightly more
hairy is needing something smaller with the guarantee that such objects
will never be inserted into any GTT, which is the case for the paging
structures.
This also fixes how we setup the BO page_alignment, if we later migrate
the object somewhere else. For example if the placements are {SMEM,
LMEM}, then we might get this wrong. Pushing the min_page_size behaviour
into the manager should fix this.
v2(Thomas): push the default page size behaviour into buddy_man, and let
the user override it with the page-alignment, which looks cleaner
Thomas Hellström [Tue, 29 Jun 2021 15:12:03 +0000 (17:12 +0200)]
drm/i915/display: Migrate objects to LMEM if possible for display
Objects intended to be used as display framebuffers must reside in
LMEM for discrete. If they happen to not do that, migrate them to
LMEM before pinning.
Matthew Auld [Tue, 29 Jun 2021 15:12:02 +0000 (17:12 +0200)]
drm/i915/gem: Introduce a selftest for the gem object migrate functionality
A selftest for the gem object migrate functionality. Slightly adapted
from the original by Matthew to the new interface and new fill blit
code.
v4:
- Initialize buffers and check contents after migration
(Suggested by Matthew Auld)
- Perform async migration (if implemented) in the igt_lmem_pages_migrate
test
- Test also migration to the current region.
Thomas Hellström [Tue, 29 Jun 2021 15:12:01 +0000 (17:12 +0200)]
drm/i915/gem: Implement object migration
Introduce an interface to migrate objects between regions.
This is primarily intended to migrate objects to LMEM for display and
to SYSTEM for dma-buf, but might be reused in one form or another for
performance-based migration.
v2:
- Verify that the memory region given as an id really exists.
(Reported by Matthew Auld)
- Call i915_gem_object_{init,release}_memory_region() when switching region
to handle also switching region lists. (Reported by Matthew Auld)
v3:
- Fix i915_gem_object_can_migrate() to return true if object is already in
the correct region, even if the object ops doesn't have a migrate()
callback.
- Update typo in commit message.
- Fix kerneldoc of i915_gem_object_wait_migration().
v4:
- Improve documentation (Suggested by Mattew Auld and Michael Ruhl)
- Always assume TTM migration hits a TTM move and unsets the pages through
move_notify. (Reported by Matthew Auld)
- Add a dma_fence_might_wait() annotation to
i915_gem_object_wait_migration() (Suggested by Daniel Vetter)
v5:
- Re-add might_sleep() instead of __dma_fence_might_wait(), Sent
v4 with the wrong version, didn't compile and __dma_fence_might_wait()
is not exported.
- Added an R-B.
Matthew Brost [Tue, 29 Jun 2021 19:35:11 +0000 (12:35 -0700)]
drm/doc/rfc: i915 new parallel submission uAPI plan
Add entry for i915 new parallel submission uAPI plan.
v2:
(Daniel Vetter):
- Expand logical order explaination
- Add dummy header
- Only allow N BBs in execbuf IOCTL
- Configure parallel submission per slot not per gem context
v3:
(Marcin Ślusarz):
- Lot's of typos / bad english fixed
(Tvrtko Ursulin):
- Consistent pseudo code, clean up wording in descriptions
v4:
(Daniel Vetter)
- Drop flags
- Add kernel doc
- Reword a few things / fix typos
(Tvrtko)
- Reword a few things / fix typos
v5:
(Checkpatch)
- Fix typos
(Docs)
- Fix warning
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Tony Ye <tony.ye@intel.com> CC: Carl Zhang <carl.zhang@intel.com> Cc: Daniel Vetter <daniel.vetter@intel.com> Cc: Jason Ekstrand <jason@jlekstrand.net> Signed-off-by: Matthew Brost <matthew.brost@intel.com> Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch> Acked-by: Tony Ye <tony.ye@intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: https://patchwork.freedesktop.org/patch/msgid/20210629193511.124099-3-matthew.brost@intel.com
Matthew Brost [Tue, 29 Jun 2021 19:35:10 +0000 (12:35 -0700)]
drm/doc/rfc: i915 GuC submission / DRM scheduler
Add entry for i915 GuC submission / DRM scheduler integration plan.
Follow up patch with details of new parallel submission uAPI to come.
v2:
(Daniel Vetter)
- Expand explaination of why bonding isn't supported for GuC
submission
- CC some of the DRM scheduler maintainers
- Add priority inheritance / boosting use case
- Add reasoning for removing in order assumptions
(Daniel Stone)
- Add links to priority spec
v4:
(Tvrtko)
- Add TODOs section
(Daniel Vetter)
- Pull in 1 line from following patch
v5:
(Checkpatch)
- Fix typos
Cc: Christian König <christian.koenig@amd.com> Cc: Luben Tuikov <luben.tuikov@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Steven Price <steven.price@arm.com> Cc: Jon Bloomfield <jon.bloomfield@intel.com> Cc: Dave Airlie <airlied@gmail.com> Cc: Daniel Vetter <daniel.vetter@intel.com> Cc: Jason Ekstrand <jason@jlekstrand.net> Cc: dri-devel@lists.freedesktop.org Signed-off-by: Matthew Brost <matthew.brost@intel.com> Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> Acked-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: https://patchwork.freedesktop.org/patch/msgid/20210629193511.124099-2-matthew.brost@intel.com
Thomas Hellström [Thu, 24 Jun 2021 11:29:14 +0000 (13:29 +0200)]
drm/i915: Reinstate the mmap ioctl for some platforms
Reinstate the mmap ioctl for all current integrated platforms.
The intention was really to have it disabled for discrete graphics
where we enforce a single mmap mode.
This was reported to break ADL-P with the media stack, which was not the
intention. Although longer term we do still plan to sunset this ioctl
even for integrated, in favour of using mmap_offset instead.
Thomas Hellström [Thu, 24 Jun 2021 08:42:40 +0000 (10:42 +0200)]
drm/i915/ttm: Use TTM for system memory
For discrete, use TTM for both cached and WC system memory. That means
we currently rely on the TTM memory accounting / shrinker. For cached
system memory we should consider remaining shmem-backed, which can be
implemented from our ttm_tt_populate callback. We can then also reuse our
own very elaborate shrinker for that memory.
If an object is evicted to a gem allowable region, we will now consider
the object migrated, and we flip the gem region and move the object to a
different region list. Since we are now changing gem regions, we can't
any longer rely on the CONTIGUOUS flag being set based on the region
min page size, so remove that flag update. If we want to reintroduce it,
we need to put it in the mutable flags.
Thomas Hellström [Thu, 24 Jun 2021 08:42:39 +0000 (10:42 +0200)]
drm/i915/ttm: Adjust gem flags and caching settings after a move
After a TTM move or object init we need to update the i915 gem flags and
caching settings to reflect the new placement. Currently caching settings
are not changed during the lifetime of an object, although that might
change moving forward if we run into performance issues or issues with
WC system page allocations.
Also introduce gpu_binds_iomem() and cpu_maps_iomem() to clean up the
various ways we previously used to detect this.
Finally, initialize the TTM object reserved to be able to update
flags and caching before anyone else gets hold of the object.
Thomas Hellström [Thu, 24 Jun 2021 08:42:38 +0000 (10:42 +0200)]
drm/i915: Update object placement flags to be mutable
The object ops i915_GEM_OBJECT_HAS_IOMEM and the object
I915_BO_ALLOC_STRUCT_PAGE flags are considered immutable by
much of our code. Introduce a new mem_flags member to hold these
and make sure checks for these flags being set are either done
under the object lock or with pages properly pinned. The flags
will change during migration under the object lock.
drm/i915: Restrict pagefault disabling to just around copy_from_user()
we entirely missed that there's a slow path call to eb_relocate_entry
(or i915_gem_execbuffer_relocate_entry as it was called back then)
which was left fully wrapped by pagefault_disable/enable() calls.
Previously any issues with blocking calls where handled by the
following code:
/* we can't wait for rendering with pagefaults disabled */
if (pagefault_disabled() && !object_is_idle(obj))
return -EFAULT;
Now at this point the prefaulting was still around, which means in
normal applications it was very hard to hit this bug. No idea why the
regressions in igts weren't caught.
Now this all changed big time with 2 patches merged closely together.
drm/i915: Eliminate lots of iterations over the execobjects array
removes the prefaulting from the first relocation path, pushing it into
the first slowpath (of which this patch added a total of 3 escalation
levels). This would have really quickly uncovered the above bug, were
it not for immediate adding a duct-tape on top with
Since pagefault_disable/enable is an atomic section, any sleeping in
there is prohibited, and we definitely do that without gpu relocations
since we have to wait for the gpu usage to finish before we can patch
up the relocations.
Reviewed-by: Matthew Auld <matthew.auld@intel.com> Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@intel.com> Cc: Jon Bloomfield <jon.bloomfield@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: "Thomas Hellström" <thomas.hellstrom@linux.intel.com> Cc: Matthew Auld <matthew.auld@intel.com> Cc: Lionel Landwerlin <lionel.g.landwerlin@intel.com> Cc: Dave Airlie <airlied@redhat.com> Cc: Jason Ekstrand <jason@jlekstrand.net> Link: https://patchwork.freedesktop.org/patch/msgid/20210618214503.1773805-1-daniel.vetter@ffwll.ch
Tvrtko Ursulin [Fri, 18 Jun 2021 15:00:36 +0000 (16:00 +0100)]
drm/i915: Document the Virtual Engine uAPI
A little bit of documentation covering the topics of engine discovery,
context engine maps and virtual engines. It is not very detailed but
supposed to be a starting point of giving a brief high level overview of
general principles and intended use cases.
v2:
* Have the text in uapi header and link from there.
Michal Wajdeczko [Wed, 16 Jun 2021 00:13:02 +0000 (17:13 -0700)]
drm/i915/doc: Include GuC ABI documentation
GuC ABI documentation is now ready to be included in i915.rst
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com> Signed-off-by: Matthew Brost <matthew.brost@intel.com> Cc: Piotr Piórkowski <piotr.piorkowski@intel.com> Reviewed-by: Matthew Brost <matthew.brost@intel.com> Signed-off-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20210616001302.84233-4-matthew.brost@intel.com
Michal Wajdeczko [Wed, 16 Jun 2021 00:13:01 +0000 (17:13 -0700)]
drm/i915/guc: Update firmware to v62.0.0
Most of the changes to the 62.0.0 firmware revolved around CTB
communication channel. Conform to the new (stable) CTB protocol.
v2:
(Michal)
Add values back to kernel DOC for actions
(Docs)
Add 'CT buffer' back in to fix warning
Signed-off-by: John Harrison <John.C.Harrison@Intel.com> Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com> Signed-off-by: Matthew Brost <matthew.brost@intel.com> Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
[mattrope: Tweaked kerneldoc while pushing as suggested by Daniele/Michal] Signed-off-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20210616001302.84233-3-matthew.brost@intel.com
Michal Wajdeczko [Wed, 16 Jun 2021 00:13:00 +0000 (17:13 -0700)]
drm/i915/guc: Introduce unified HXG messages
New GuC firmware will unify format of MMIO and CTB H2G messages.
Introduce their definitions now to allow gradual transition of
our code to match new changes.
Signed-off-by: Matthew Brost <matthew.brost@intel.com> Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Michał Winiarski <michal.winiarski@intel.com> Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Signed-off-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20210616001302.84233-2-matthew.brost@intel.com
Matthew Brost [Fri, 18 Jun 2021 01:06:38 +0000 (18:06 -0700)]
drm/i915: Move submission tasklet to i915_sched_engine
The submission tasklet operates on i915_sched_engine, thus it is the
correct place for it.
v3:
(Jason Ekstrand)
Change sched_engine->engine to a void* private data pointer
Add kernel doc
v4:
(Daniele)
Update private_data comment
Set queue_priority_hint in kick_execlists
v5:
(CI)
Rebase and fix build error
Matthew Brost [Fri, 18 Jun 2021 01:06:36 +0000 (18:06 -0700)]
drm/i915: Add kick_backend function to i915_sched_engine
Not all back-ends require a kick after a scheduling update, so make the
kick a call-back function that the back-end can opt-in to. Also move
the current kick function from the scheduler to the execlists file as it
is specific to that back-end.
Matthew Brost [Fri, 18 Jun 2021 01:06:34 +0000 (18:06 -0700)]
drm/i915: Move active tracking to i915_sched_engine
Move active request tracking and its lock to i915_sched_engine. This
lock is also the submission lock so having it in the i915_sched_engine
is the correct place.
Matthew Brost [Fri, 18 Jun 2021 01:06:33 +0000 (18:06 -0700)]
drm/i915: Reset sched_engine.no_priolist immediately after dequeue
Rather than touching schedule state in the generic PM code, reset the
priolist allocation when empty in the submission code. Add a wrapper
function to do this and update the backends to call it in the correct
place.
v3:
(Jason Ekstrand)
Update patch commit message with a better description
Matthew Brost [Fri, 18 Jun 2021 01:06:31 +0000 (18:06 -0700)]
drm/i915: Move priolist to new i915_sched_engine object
Introduce i915_sched_engine object which is lower level data structure
that i915_scheduler / generic code can operate on without touching
execlist specific structures. This allows additional submission backends
to be added without breaking the layering. Currently the execlists
backend uses 1 of these object per each engine (physical or virtual) but
future backends like the GuC will point to less instances utilizing the
reference counting.
This is a bit of detour to integrating the i915 with the DRM scheduler
but this object will still exist when the DRM scheduler lands in the
i915. It will however look a bit different. It will encapsulate the
drm_gpu_scheduler object plus and common variables (to the backends)
related to scheduling. Regardless this is a step in the right direction.
This patch starts the aforementioned transition by moving the priolist
into the i915_sched_engine object.
v3:
(Jason Ekstrand)
Update comment next to intel_engine_cs.virtual
Add kernel doc
(Checkpatch)
Fix double the in commit message
v4:
(Daniele)
Update comment message.
Add comment about subclass field
Thomas Hellström [Fri, 18 Jun 2021 13:25:15 +0000 (15:25 +0200)]
drm/i915/ttm: Fix incorrect assumptions about ttm_bo_validate() semantics
We have assumed that if the current placement was not the requested
placement, but instead one of the busy placements, a TTM move would have
been triggered. That is not the case.
So when we initially place LMEM objects in "Limbo", (that is system
placement without any pages allocated), to be able to defer clearing
objects until first get_pages(), the first get_pages() would happily keep
objects in system memory if that is one of the allowed placements. And
since we don't yet support i915 GEM system memory from TTM, everything
breaks apart.
So make sure we try the requested placement first, if no eviction is
needed. If that fails, retry with all allowed placements also allowing
evictions. Also make sure we handle TTM failure codes correctly.
Also temporarily (until we support i915 GEM system on TTM), restrict
allowed placements to the requested placement to avoid things falling
apart should LMEM be full.
Matt Roper [Thu, 17 Jun 2021 21:14:25 +0000 (14:14 -0700)]
drm/i915: Add support for explicit L3BANK steering
Because Render Power Gating restricts us to just a single subslice as a
valid steering target for reads of multicast registers in a SUBSLICE
range, the default steering we setup at init may not lead to a suitable
target for L3BANK multicast register. In cases where it does not, use
explicit runtime steering whenever an L3BANK multicast register is read.
While we're at it, let's simplify the function a little bit and drop its
support for gen10/CNL since no such platforms ever materialized for real
use. Multicast register steering is already an area that causes enough
confusion; no need to complicate it with what's effectively dead code.
v2:
- Use gt->uncore instead of gt->i915->uncore. (Tvrtko)
- Use {} as table terminator. (Rodrigo)
v3:
- L3bank fuse register is a disable mask rather than an enable mask.
We need to invert it before use. (CI)
v4:
- L3bank ID goes in the subslice field, not the slice field. (CI)
Matt Roper [Thu, 17 Jun 2021 21:14:24 +0000 (14:14 -0700)]
drm/i915: Add GT support for multiple types of multicast steering
Although most of our multicast registers are replicated per-subslice, we
also have a small number of multicast registers that are replicated
per-l3 bank instead. For both types of multicast registers we need to
make sure we steer reads of these registers to a valid instance.
Ideally we'd like to find a specific instance ID that would steer reads
of either type of multicast register to a valid instance (i.e., not
fused off and not powered down), but sometimes the combination of
part-specific fusing and the additional restrictions imposed by Render
Power Gating make it impossible to find any overlap between the set of
valid subslices and valid l3 banks. This problem will become even more
noticeable on our upcoming platforms since they will be adding
additional types of multicast registers with new types of replication
and rules for finding valid instances for reads.
To handle this we'll continue to pick a suitable subslice instance at
driver startup and program this as the default (sliceid,subsliceid)
setting in the steering control register (0xFDC). In cases where we
need to read another type of multicast GT register, but the default
subslice steering would not correspond to a valid instance, we'll
explicitly re-steer the single read to a valid value, perform the read,
and then reset the steering to it's "subslice" default.
This patch adds the general functionality to prepare for this explicit
steering of other multicast register types. We'll plug L3 bank steering
into this in the next patch, and then add additional types of multicast
registers when the support for our next upcoming platform arrives.
v2:
- Use entry->end==0 as table terminator. (Rodrigo)
- Grab forcewake in wa_list_verify() now that we're using accessors
that assume forcewake is already held.
v3:
- Fix loop condition when iterating over steering range tables.
(Rodrigo)
Thomas Hellström [Thu, 17 Jun 2021 06:30:18 +0000 (08:30 +0200)]
drm/i915/gem: Zap the i915_gem_object_blt code
It's unused with the exception of selftest. Replace a call in the
memory_region live selftest with a call into a corresponding
function in the new migrate code.
Chris Wilson [Thu, 17 Jun 2021 06:30:13 +0000 (08:30 +0200)]
drm/i915/gt: Pipelined page migration
If we pipeline the PTE updates and then do the copy of those pages
within a single unpreemptible command packet, we can submit the copies
and leave them to be scheduled without having to synchronously wait
under a global lock. In order to manage migration, we need to
preallocate the page tables (and keep them pinned and available for use
at any time), causing a bottleneck for migrations as all clients must
contend on the limited resources. By inlining the ppGTT updates and
performing the blit atomically, each client only owns the PTE while in
use, and so we can reschedule individual operations however we see fit.
And most importantly, we do not need to take a global lock on the shared
vm, and wait until the operation is complete before releasing the lock
for others to claim the PTE for themselves.
Thomas Hellström [Thu, 17 Jun 2021 06:30:07 +0000 (08:30 +0200)]
drm/i915: Reference objects on the ww object list
Since the ww transaction endpoint easily end up far out-of-scope of
the objects on the ww object list, particularly for contending lock
objects, make sure we reference objects on the list so they don't
disappear under us.
This comes with a performance penalty so it's been debated whether this
is really needed. But I think this is motivated by the fact that locking
is typically difficult to get right, and whatever we can do to make it
simpler for developers moving forward should be done, unless the
performance impact is far too high.
drm/i915/gem: Allow backends to override pread implementation
this accidentally landed twice.
Reviewed-by: Matthew Auld <matthew.auld@intel.com> Cc: Matthew Auld <matthew.auld@intel.com Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com> Cc: Jason Ekstrand <jason@jlekstrand.net> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Daniel Vetter <daniel.vetter@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20210616090350.828696-1-daniel.vetter@ffwll.ch
Matthew Auld [Wed, 16 Jun 2021 15:25:01 +0000 (16:25 +0100)]
drm/i915/ttm: restore min_page_size behaviour
We now have bo->page_alignment which perfectly describes what we need if
we have min page size restrictions for lmem. We can also drop the flag
here, since this is the default behaviour for all objects.
Matthew Auld [Wed, 16 Jun 2021 15:25:00 +0000 (16:25 +0100)]
drm/i915/ttm: switch over to ttm_buddy_man
Move back to the buddy allocator for managing device local memory, and
restore the lost mock selftests. Keep around the range manager related
bits, since we likely need this for managing stolen at some point. For
stolen we also don't need to reserve anything so no need to support a
generic reserve interface.
v2(Thomas):
- bo->page_alignment is in page units, not bytes
Matthew Auld [Wed, 16 Jun 2021 15:24:59 +0000 (16:24 +0100)]
drm/i915/ttm: remove node usage in our naming
Now that ttm_resource_manager just returns a generic ttm_resource we
don't need to reference the mm_node stuff anymore which mostly only
makes sense for drm_mm_node. In the next few patches we want switch over
to the ttm_buddy_man which is just another type of ttm_resource so
reflect that in the naming.
Matthew Auld [Wed, 16 Jun 2021 15:24:58 +0000 (16:24 +0100)]
drm/i915/ttm: pass along the I915_BO_ALLOC_CONTIGUOUS
Currently we just ignore the I915_BO_ALLOC_CONTIGUOUS flag, which is
fine since everything is already contiguous with the ttm range manager.
However in the next patch we want to switch over to the ttm buddy
manager, where allocations are by default not contiguous.
v2(Thomas):
- Forward ALLOC_CONTIG for all regions
Thomas Hellström [Wed, 16 Jun 2021 15:24:57 +0000 (16:24 +0100)]
drm/i915/ttm: Calculate the object placement at get_pages time
Instead of relying on a static placement, calculate at get_pages() time.
This should work for LMEM regions and system for now. For stolen we need
to take preallocated range into account. That will if needed be added
later.
Matthew Auld [Wed, 16 Jun 2021 15:24:56 +0000 (16:24 +0100)]
drm/i915/ttm: add i915_sg_from_buddy_resource
We need to be able to build an sg table from our list of buddy blocks,
so that we can later plug this into our ttm backend, and replace our use
of the range manager.
Matthew Auld [Wed, 16 Jun 2021 15:24:55 +0000 (16:24 +0100)]
drm/i915/ttm: add ttm_buddy_man
Add back our standalone i915_buddy allocator and integrate it into a
ttm_resource_manager. This will plug into our ttm backend for managing
device local-memory in the next couple of patches.
v2(Thomas):
- Return -ENOSPC from the buddy; ttm expects this in order to
trigger eviction
- Drop the unnecessary inline
- bo->page_alignment is in page units
Thomas Hellström [Thu, 10 Jun 2021 14:35:25 +0000 (16:35 +0200)]
drm/i915: Simplify userptr locking
Use an rwlock instead of spinlock for the global notifier lock
to reduce risk of contention in execbuf.
Protect object state with the object lock whenever possible rather
than with the global notifier lock
Don't take an explicit page_ref in userptr_submit_init() but rather
call get_pages() after obtaining the page list so that
get_pages() holds the page_ref. This means we don't need to call
userptr_submit_fini(), which is needed to avoid awkward locking
in our upcoming VM_BIND code.
Joonas Lahtinen [Mon, 14 Jun 2021 09:24:24 +0000 (12:24 +0300)]
Merge tag 'topic/i915-ttm-2021-06-11' of git://anongit.freedesktop.org/drm/drm-misc into drm-intel-gt-next
drm-misc and drm-intel pull request for topic/i915-ttm:
- Convert i915 lmem handling to ttm.
- Add a patch to temporarily add a driver_private member to vma_node.
- Use this to allow mixed object mmap handling for i915.
Use the ttm handlers for servicing page faults, and vm_access.
We do our own validation of read-only access, otherwise use the
ttm handlers as much as possible.
Because the ttm handlers expect the vma_node at vma->base, we slightly
need to massage the mmap handlers to look at vma_node->driver_private
to fetch the bo, if it's NULL, we assume i915's normal mmap_offset uapi
is used.
This is the easiest way to achieve compatibility without changing ttm's
semantics.
Thomas Hellström [Thu, 10 Jun 2021 07:01:50 +0000 (09:01 +0200)]
drm/i915/lmem: Verify checks for lmem residency
Since objects can be migrated or evicted when not pinned or locked,
update the checks for lmem residency or future residency so that
the value returned is not immediately stale.
Thomas Hellström [Thu, 10 Jun 2021 07:01:49 +0000 (09:01 +0200)]
drm/i915/ttm: Introduce a TTM i915 gem object backend
Most logical place to introduce TTM buffer objects is as an i915
gem object backend. We need to add some ops to account for added
functionality like delayed delete and LRU list manipulation.
Initially we support only LMEM and SYSTEM memory, but SYSTEM
(which in this case means evicted LMEM objects) is not
visible to i915 GEM yet. The plan is to move the i915 gem system region
over to the TTM system memory type in upcoming patches.
We set up GPU bindings directly both from LMEM and from the system region,
as there is no need to use the legacy TTM_TT memory type. We reserve
that for future porting of GGTT bindings to TTM.
Dave Airlie [Fri, 11 Jun 2021 04:19:06 +0000 (14:19 +1000)]
Merge tag 'exynos-drm-next-for-v5.14' of git://git.kernel.org/pub/scm/linux/kernel/git/daeinki/drm-exynos into drm-next
Two cleanups
- These patches make Exynos DRM driver to use pm_runtime_resume_and_get()
function instead of m_runtime_get_sync() to deal with usage counter.
pm_runtime_get_sync() increases the usage counter even when it failed,
which could make callers to forget to decrease the usage counter.
pm_runtime_resume_and_get() decreases the usage counter regardless of
whether it failed or not.
Dave Airlie [Fri, 11 Jun 2021 03:34:42 +0000 (13:34 +1000)]
Merge tag 'drm-intel-gt-next-2021-06-10' of git://anongit.freedesktop.org/drm/drm-intel into drm-next
UAPI Changes:
- Disable mmap ioctl for gen12+ (excl. TGL-LP)
- Start enabling HuC loading by default for upcoming Gen12+
platforms (excludes TGL and RKL)
Core Changes:
- Backmerge of drm-next
Driver Changes:
- Revert "i915: use io_mapping_map_user" (Eero, Matt A)
- Initialize the TTM device and memory managers (Thomas)
- Major rework to the GuC submission backend to prepare
for enabling on new platforms (Michal Wa., Daniele,
Matt B, Rodrigo)
- Fix i915_sg_page_sizes to record dma segments rather
than physical pages (Thomas)
- Locking rework to prep for TTM conversion (Thomas)
- Replace IS_GEN and friends with GRAPHICS_VER (Lucas)
- Use DEVICE_ATTR_RO macro (Yue)
- Static code checker fixes (Zhihao)
Dave Airlie [Fri, 11 Jun 2021 02:50:47 +0000 (12:50 +1000)]
Merge branch 'etnaviv/next' of https://git.pengutronix.de/git/lst/linux into drm-next
- remove redundant NULL checks by various people
- fix sparse checker warnings from Marc
- expose more GPU ID values to userspace from Christian
- add HWDB entry for GPU found on i.MX8MP from Sascha
- rework of the linear window calculation to better deal with
systems with large regions of reserved RAM
Inki Dae [Tue, 25 May 2021 10:51:39 +0000 (19:51 +0900)]
drm/exynos: use pm_runtime_resume_and_get()
Use pm_runtime_resume_and_get() instead of pm_runtime_get_sync()
to deal with usage counter. pm_runtime_get_sync() increases the
usage counter even when it failed, which makes callers to forget
to decrease the usage counter and resulted in reference leak.
pm_runtime_resume_and_get() function decreases the usage counter
when it failed internally so it can avoid the reference leak.
Changelog v1:
- Fix an build error reported by kernel test robot of Intel.
Signed-off-by: Inki Dae <inki.dae@samsung.com> Reported-by: kernel test robot <lkp@intel.com>
Tian Tao [Mon, 24 May 2021 12:07:57 +0000 (20:07 +0800)]
drm/exynos: Use pm_runtime_resume_and_get() to replace open coding
use pm_runtime_resume_and_get() to replace pm_runtime_get_sync and
pm_runtime_put_noidle to avoid continuing to increase the refcount
when pm_runtime_get_sync fails.
Signed-off-by: Tian Tao <tiantao6@hisilicon.com> Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Inki Dae <inki.dae@samsung.com>
Alex Deucher [Mon, 7 Jun 2021 20:19:46 +0000 (16:19 -0400)]
drm/amdgpu/vcn: drop gfxoff control for VCN2+
Drop disabling of gfxoff during VCN use. This allows gfxoff
to kick in and potentially save power if the user is not using
gfx for color space conversion or scaling.
VCN1.0 had a bug which prevented it from working properly with
gfxoff, so we disabled it while using VCN. That said, most apps
today use gfx for scaling and color space conversion rather than
overlay planes so it was generally in use anyway. This was fixed
on VCN2+, but since we mostly use gfx for color space conversion
and scaling and rapidly powering up/down gfx can negate the
advantages of gfxoff, we left gfxoff disabled. As more
applications use overlay planes for color space conversion
and scaling, this starts to be a win, so go ahead and leave
gfxoff enabled.
Note that VCN1.0 uses vcn_v1_0_idle_work_handler() and
vcn_v1_0_ring_begin_use() so they are not affected by this
patch.
Reviewed-by: James Zhu <James.Zhu@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Reviewed-by: Boyuan Zhang <Boyuan.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
* ttm: Make struct ttm_resource the base of all managers + changes
in all users of TTM; Add a generic memcpy for page-based iomem; Remove
use of VM_MIXEDMAP; Cleanups
Driver Changes:
* drm/bridge: Add TI SN65DSI83 and SN65DSI84 + DT bindings
* drm/hyperv: Add DRM driver for HyperV graphics output
* drm/msm: Fix module dependencies
* drm/panel: KD53T133: Support rotation
* drm/pl111: Fix module dependencies
* drm/qxl: Fixes
* drm/stm: Cleanups
* drm/sun4i: Be explicit about format modifiers
* drm/vc4: Use struct gpio_desc; Cleanups
* drm/vgem: Cleanups
* drm/vmwgfx: Use ttm_bo_move_null() if there's nothing to copy
John Harrison [Thu, 3 Jun 2021 16:48:12 +0000 (09:48 -0700)]
drm/i915/uc: Use platform specific defaults for GuC/HuC enabling
The meaning of 'default' for the enable_guc module parameter has been
updated to accurately reflect what is supported on current platforms.
So start using the defaults instead of forcing everything off.
Although, note that right now, the default is for everything to be off
anyway. So this is not a change for current platforms.
drm/i915/adl_p: Same slices mask is not same Dbuf state
We currently treat same slice mask as a same DBuf state and skip
updating the Dbuf slices, if we detect that.
This is wrong as if we have a multi to single pipe change or
vice versa, that would be treated as a same Dbuf state and thus
no changes required, so we don't get Mbus updated, causing issues.
Solution: check also mbus_join, in addition to slices mask.
Colin Ian King [Tue, 8 Jun 2021 16:13:13 +0000 (17:13 +0100)]
drm: qxl: ensure surf.data is ininitialized
The object surf is not fully initialized and the uninitialized
field surf.data is being copied by the call to qxl_bo_create
via the call to qxl_gem_object_create. Set surf.data to zero
to ensure garbage data from the stack is not being copied.
Ville Syrjälä [Tue, 8 Jun 2021 08:54:15 +0000 (11:54 +0300)]
drm/i915: Disable PSR around cdclk changes
AUX logic is often clocked from cdclk. Disable PSR to make sure
there are no hw initiated AUX transactions in flight while we
change the cdclk frequency.
Gwan-gyeong Mun [Tue, 8 Jun 2021 08:54:14 +0000 (11:54 +0300)]
drm/i915/display: Introduce new intel_psr_pause/resume function
This introduces the following function that can exit and activate a psr
source when intel_psr is already enabled.
- intel_psr_pause(): Pause current PSR. It deactivates current psr state.
- intel_psr_resume(): Resume paused PSR. It activates paused psr state.
v2: Address Jose's review comment.
- Remove unneeded changes around the intel_psr_enable().
- Add intel_psr_post_exit() which processes waiting until PSR is idle
and WA for SelectiveFetch.
v3: Address Jose's review comment.
- Rename intel_psr_post_exit() to intel_psr_wait_exit_locked().
- Move WA_1408330847 to intel_psr_disable_locked()
- If the PSR is paused by an explicit intel_psr_paused() call, make the
intel_psr_flush() not to activate PSR.
v4: Address Jose's review comment.
- In order to avoid the scenario of PSR is not active but there is a
scheduled psr->work, it changes the check routine of intel_psr_pause()
for PSR's enablement from "psr->active" to "psr->enable".
Cc: José Roberto de Souza <jose.souza@intel.com> Cc: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Signed-off-by: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com> Signed-off-by: Matt Roper <matthew.d.roper@intel.com> Reviewed-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20210608085415.515342-1-gwan-gyeong.mun@intel.com
Marek Vasut [Mon, 7 Jun 2021 17:24:57 +0000 (19:24 +0200)]
drm/stm: Remove usage of drm_display_mode_to_videomode()
There is not much value in the extra conversion step, the calculations
required for the LTDC IP are different than what is used in the
drm_display_mode_to_videomode(), so just do the right ones in the LTDC
driver right away.
Signed-off-by: Marek Vasut <marex@denx.de> Cc: Alexandre Torgue <alexandre.torgue@foss.st.com> Cc: Antonio Borneo <antonio.borneo@foss.st.com> Cc: Benjamin Gaignard <benjamin.gaignard@foss.st.com> Cc: Maxime Coquelin <mcoquelin.stm32@gmail.com> Cc: Philippe Cornu <philippe.cornu@foss.st.com> Cc: Sam Ravnborg <sam@ravnborg.org> Cc: Vincent Abriou <vincent.abriou@foss.st.com> Cc: Yannick Fertre <yannick.fertre@foss.st.com> Cc: linux-arm-kernel@lists.infradead.org Cc: linux-stm32@st-md-mailman.stormreply.com
To: dri-devel@lists.freedesktop.org Tested-by: Yannick Fertré <yannick.fertre@st.com> Signed-off-by: Philippe Cornu <philippe.cornu@foss.st.com> Link: https://patchwork.freedesktop.org/patch/msgid/20210607172457.14471-1-marex@denx.de
projects / kernel.git / log