Skip to content
  1. Oct 06, 2021
    • Tvrtko Ursulin's avatar
      drm/i915: Handle Intel igfx + Intel dgfx hybrid graphics setup · 07f82a47
      Tvrtko Ursulin authored 
      

In short this makes i915 work for hybrid setups (DRI_PRIME=1 with Mesa)
when rendering is done on Intel dgfx and scanout/composition on Intel
igfx.

Before this patch the driver was not quite ready for that setup, mainly
because it was able to emit a semaphore wait between the two GPUs, which
results in deadlocks because semaphore target location in HWSP is neither
shared between the two, nor mapped in both GGTT spaces.

To fix it the patch adds an additional check to a couple of relevant code
paths in order to prevent using semaphores for inter-engine
synchronisation when relevant objects are not in the same GGTT space.

v2:
 * Avoid adding rq->i915. (Chris)

v3:
 * Use GGTT which describes the limit more precisely.

Signed-off-by: default avatarTvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: default avatarThomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211005113135.768295-1-tvrtko.ursulin@linux.intel.com
      07f82a47
  3. Oct 05, 2021
  5. Oct 02, 2021
  7. Oct 01, 2021
    • Thomas Hellström's avatar
      drm/i915/ttm: Rework object initialization slightly · 068396bb
      Thomas Hellström authored 
      We may end up in i915_ttm_bo_destroy() in an error path before the
object is fully initialized. In that case it's not correct to call
__i915_gem_free_object(), because that function
a) Assumes the gem object refcount is 0, which it isn't.
b) frees the placements which are owned by the caller until the
init_object() region ops returns successfully. Fix this by providing
a lightweight cleanup function __i915_gem_object_fini() which is also
called by __i915_gem_free_object().

While doing this, also make sure we call dma_resv_fini() as part of
ordinary object destruction and not from the RCU callback that frees
the object. This will help track down bugs where the object is incorrectly
locked from an RCU lookup.

Finally, make sure the object isn't put on the region list until it's
either locked or fully initialized in order to block list processing of
partially initialized objects.

v2:
- The TTM object backend memory was freed before the gem pages were
  put. Separate this functionality into __i915_gem_object_pages_fini()
  and call it from the TTM delete_mem_notify() callback.
v3:
- Include i915_gem_object_free_mmaps() in __i915_gem_object_pages_fini()
  to make sure we don't inadvertedly introduce a race.

Fixes: 48b09612

 ("drm/i915: Move __i915_gem_free_object to ttm_bo_destroy")
Signed-off-by: default avatarThomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com> #v1
Link: https://patchwork.freedesktop.org/patch/msgid/20210930113236.583531-1-thomas.hellstrom@linux.intel.com
      068396bb
  9. Sep 30, 2021
  11. Sep 29, 2021
  13. Sep 27, 2021
  15. Sep 25, 2021
    • Janusz Krzysztofik's avatar
      drm/i915: Flush buffer pools on driver remove · 74af1e2c
      Janusz Krzysztofik authored 
      

We currently do an explicit flush of the buffer pools within the call path
of drm_driver.release(); this removes all buffers, regardless of their age,
freeing the buffers' associated resources (objects, address space areas).
However there is other code that runs within the drm_driver.release() call
chain that expects objects and their associated address space areas have
already been flushed.

Since buffer pools auto-flush old buffers once per second in a worker
thread, there's a small window where if we remove the driver while there
are still objects in buffers with an age of less than one second, the
assumptions of the other release code may be violated.

By moving the flush to driver remove (which executes earlier via the
pci_driver.remove() flow) we're ensuring that all buffers are flushed and
their associated objects freed before some other code in
pci_driver.remove() flushes those objects so they are released before
_any_ code in drm_driver.release() that check completness of those
flushes executes.

v2: Reword commit description as suggested by Matt.

Signed-off-by: default avatarJanusz Krzysztofik <janusz.krzysztofik@linux.intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: default avatarMatt Roper <matthew.d.roper@intel.com>
Reviewed-by: default avatarMatt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210924163825.634606-1-janusz.krzysztofik@linux.intel.com
      74af1e2c
  17. Sep 24, 2021
    • Tejas Upadhyay's avatar
      drm/i915: Remove warning from the rps worker · a837a068
      Tejas Upadhyay authored 
      In commit 4e5c8a99 ("drm/i915: Drop i915_request.lock requirement
for intel_rps_boost()"), we decoupled the rps worker from the pm so
that we could avoid the synchronization penalty which makes the
assertion liable to run too early. Which makes warning invalid hence
removed.

Fixes: 4e5c8a99

 ("drm/i915: Drop i915_request.lock requirement for intel_rps_boost()")

Reviewed-by: default avatarChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: default avatarTejas Upadhyay <tejaskumarx.surendrakumar.upadhyay@intel.com>
Signed-off-by: default avatarMatt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210914090412.1393498-1-tejaskumarx.surendrakumar.upadhyay@intel.com
      a837a068
    • Matthew Auld's avatar
      drm/i915/selftests: exercise shmem_writeback with THP · 6341eb6f
      Matthew Auld authored 
      In commit:

commit 1e6decf3


Author: Hugh Dickins <hughd@google.com>
Date:   Thu Sep 2 14:54:43 2021 -0700

    shmem: shmem_writepage() split unlikely i915 THP

it looks THP + shmem_writeback was an unexpected combination, and ends up
hitting some BUG_ON, but it also looks like that is now fixed.

While the IGTs did eventually hit this(although not during pre-merge it
seems), it's likely worthwhile adding some explicit coverage for this
scenario in the shrink_thp selftest.

References: https://gitlab.freedesktop.org/drm/intel/-/issues/4166
Signed-off-by: default avatarMatthew Auld <matthew.auld@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Reviewed-by: default avatarTvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210921142116.3807946-1-matthew.auld@intel.com
      6341eb6f
    • Matthew Auld's avatar
      drm/i915/request: fix early tracepoints · be988eae
      Matthew Auld authored 
      Currently we blow up in trace_dma_fence_init, when calling into
get_driver_name or get_timeline_name, since both the engine and context
might be NULL(or contain some garbage address) in the case of newly
allocated slab objects via the request ctor. Note that we also use
SLAB_TYPESAFE_BY_RCU here, which allows requests to be immediately
freed, but delay freeing the underlying page by an RCU grace period.
With this scheme requests can be re-allocated, at the same time as they
are also being read by some lockless RCU lookup mechanism.

In the ctor case, which is only called for new slab objects(i.e allocate
new page and call the ctor for each object) it's safe to reset the
context/engine prior to calling into dma_fence_init, since we can be
certain that no one is doing an RCU lookup which might depend on peeking
at the engine/context, like in active_engine(), since the object can't
yet be externally visible.

In the recycled case(which might also be externally visible) the request
refcount always transitions from 0->1 after we set the context/engine
etc, which should ensure it's valid to dereference the engine for
example, when doing an RCU list-walk, so long as we can also increment
the refcount first. If the refcount is already zero, then the request is
considered complete/released.  If it's non-zero, then the request might
be in the process of being re-allocated, or potentially still in flight,
however after successfully incrementing the refcount, it's possible to
carefully inspect the request state, to determine if the request is
still what we were looking for. Note that all externally visible
requests returned to the cache must have zero refcount.

One possible fix then is to move dma_fence_init out from the request
ctor. Originally this was how it was done, but it was moved in:

commit 855e39e6
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Mon Feb 3 09:41:48 2020 +0000

    drm/i915: Initialise basic fence before acquiring seqno

where it looks like intel_timeline_get_seqno() relied on some of the
rq->fence state, but that is no longer the case since:

commit 12ca695d
Author: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date:   Tue Mar 23 16:49:50 2021 +0100

    drm/i915: Do not share hwsp across contexts any more, v8.

intel_timeline_get_seqno() could also be cleaned up slightly by dropping
the request argument.

Moving dma_fence_init back out of the ctor, should ensure we have enough
of the request initialised in case of trace_dma_fence_init.
Functionally this should be the same, and is effectively what we were
already open coding before, except now we also assign the fence->lock
and fence->ops, but since these are invariant for recycled
requests(which might be externally visible), and will therefore already
hold the same value, it shouldn't matter.

An alternative fix, since we don't yet have a fully initialised request
when in the ctor, is just setting the context/engine as NULL, but this
does require adding some extra handling in get_driver_name etc.

v2(Daniel):
  - Try to make the commit message less confusing

Fixes: 855e39e6

 ("drm/i915: Initialise basic fence before acquiring seqno")
Signed-off-by: default avatarMatthew Auld <matthew.auld@intel.com>
Cc: Michael Mason <michael.w.mason@intel.com>
Cc: Daniel Vetter <daniel@ffwll.ch>
Reviewed-by: default avatarDaniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20210921134202.3803151-1-matthew.auld@intel.com
      be988eae
    • Thomas Hellström's avatar
      drm/i915: Reduce the number of objects subject to memcpy recover · a259cc14
      Thomas Hellström authored 
      

We really only need memcpy restore for objects that affect the
operability of the migrate context. That is, primarily the page-table
objects of the migrate VM.

Add an object flag, I915_BO_ALLOC_PM_EARLY for objects that need early
restores using memcpy and a way to assign LMEM page-table object flags
to be used by the vms.

Restore objects without this flag with the gpu blitter and only objects
carrying the flag using TTM memcpy.

Initially mark the migrate, gt, gtt and vgpu vms to use this flag, and
defer for a later audit which vms actually need it. Most importantly, user-
allocated vms with pinned page-table objects can be restored using the
blitter.

Performance-wise memcpy restore is probably as fast as gpu restore if not
faster, but using gpu restore will help tackling future restrictions in
mappable LMEM size.

v4:
- Don't mark the aliasing ppgtt page table flags for early resume, but
  rather the ggtt page table flags as intended. (Matthew Auld)
- The check for user buffer objects during early resume is pointless, since
  they are never marked I915_BO_ALLOC_PM_EARLY. (Matthew Auld)
v5:
- Mark GuC LMEM objects with I915_BO_ALLOC_PM_EARLY to have them restored
  before we fire up the migrate context.

Cc: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: default avatarThomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: default avatarMatthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210922062527.865433-8-thomas.hellstrom@linux.intel.com
      a259cc14
    • Thomas Hellström's avatar
      drm/i915: Don't back up pinned LMEM context images and rings during suspend · 0d8ee5ba
      Thomas Hellström authored 
      

Pinned context images are now reset during resume. Don't back them up,
and assuming that rings can be assumed empty at suspend, don't back them
up either.

Introduce a new object flag, I915_BO_ALLOC_PM_VOLATILE meaning that an
object is allowed to lose its content on suspend.

v3:
- Slight documentation clarification (Matthew Auld)

Signed-off-by: default avatarThomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: default avatarMatthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210922062527.865433-7-thomas.hellstrom@linux.intel.com
      0d8ee5ba
    • Thomas Hellström's avatar
      drm/i915/gt: Register the migrate contexts with their engines · 3e42cc61
      Thomas Hellström authored 
      

Pinned contexts, like the migrate contexts need reset after resume
since their context image may have been lost. Also the GuC needs to
register pinned contexts.

Add a list to struct intel_engine_cs where we add all pinned contexts on
creation, and traverse that list at resume time to reset the pinned
contexts.

This fixes the kms_pipe_crc_basic@suspend-read-crc-pipe-a selftest for now,
but proper LMEM backup / restore is needed for full suspend functionality.
However, note that even with full LMEM backup / restore it may be
desirable to keep the reset since backing up the migrate context images
must happen using memcpy() after the migrate context has become inactive,
and for performance- and other reasons we want to avoid memcpy() from
LMEM.

Also traverse the list at guc_init_lrc_mapping() calling
guc_kernel_context_pin() for the pinned contexts, like is already done
for the kernel context.

v2:
- Don't reset the contexts on each __engine_unpark() but rather at
  resume time (Chris Wilson).
v3:
- Reset contexts in the engine sanitize callback. (Chris Wilson)

Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Brost Matthew <matthew.brost@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: default avatarThomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: default avatarMatthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210922062527.865433-6-thomas.hellstrom@linux.intel.com
      3e42cc61
    • Thomas Hellström's avatar
      drm/i915 Implement LMEM backup and restore for suspend / resume · c56ce956
      Thomas Hellström authored 
      

Just evict unpinned objects to system. For pinned LMEM objects,
make a backup system object and blit the contents to that.

Backup is performed in three steps,
1: Opportunistically evict evictable objects using the gpu blitter.
2: After gt idle, evict evictable objects using the gpu blitter. This will
be modified in an upcoming patch to backup pinned objects that are not used
by the blitter itself.
3: Backup remaining pinned objects using memcpy.

Also move uC suspend to after 2) to make sure we have a functional GuC
during 2) if using GuC submission.

v2:
- Major refactor to make sure gem_exec_suspend@hang-SX subtests work, and
  suspend / resume works with a slightly modified GuC submission enabling
  patch series.

v3:
- Fix a potential use-after-free (Matthew Auld)
- Use i915_gem_object_create_shmem() instead of
  i915_gem_object_create_region (Matthew Auld)
- Minor simplifications (Matthew Auld)
- Fix up kerneldoc for i195_ttm_restore_region().
- Final lmem_suspend() call moved to i915_gem_backup_suspend from
  i915_gem_suspend_late, since the latter gets called at driver unload
  and we don't unnecessarily want to run it at that time.

v4:
- Interface change of ttm- & lmem suspend / resume functions to use
  flags rather than bools. (Matthew Auld)
- Completely drop the i915_gem_backup_suspend change (Matthew Auld)

Signed-off-by: default avatarThomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: default avatarMatthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210922062527.865433-5-thomas.hellstrom@linux.intel.com
      c56ce956
    • Thomas Hellström's avatar
      drm/i915/gt: Increase suspend timeout · 81387fc4
      Thomas Hellström authored 
      

With GuC submission on DG1, the execution of the requests times out
for the gem_exec_suspend igt test case after executing around 800-900
of 1000 submitted requests.

Given the time we allow elsewhere for fences to signal (in the order of
seconds), increase the timeout before we mark the gt wedged and proceed.

Signed-off-by: default avatarThomas Hellström <thomas.hellstrom@linux.intel.com>
Acked-by: default avatarMatthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210922062527.865433-4-thomas.hellstrom@linux.intel.com
      81387fc4
    • Thomas Hellström's avatar
      drm/i915/gem: Implement a function to process all gem objects of a region · d80ee88e
      Thomas Hellström authored 
      

An upcoming common pattern is to traverse the region object list and
perform certain actions on all objects in a region. It's a little tricky
to get the list locking right, in particular since a gem object may
change region unless it's pinned or the object lock is held.

Define a function that does this for us and that takes an argument that
defines the action to be performed on each object.

v3:
- Improve structure documentation a bit (Matthew Auld)

Signed-off-by: default avatarThomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: default avatarMatthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210922062527.865433-3-thomas.hellstrom@linux.intel.com
      d80ee88e
    • Thomas Hellström's avatar
      drm/i915/ttm: Implement a function to copy the contents of two TTM-based objects · 0d938863
      Thomas Hellström authored 
      

When backing up or restoring contents of pinned objects at suspend /
resume time we need to allocate a new object as the backup. Add a function
to facilitate copies between the two. Some data needs to be copied before
the migration context is ready for operation, so make sure we can
disable accelerated copies.

v2:
- Fix a missing return value check (Matthew Auld)

Signed-off-by: default avatarThomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: default avatarMatthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210922062527.865433-2-thomas.hellstrom@linux.intel.com
      0d938863