Add a cancel request selftest that results in an engine reset to cancel
the request as it is non-preemptable. Also insert a NOP request after
the cancelled request and confirm that it completes successfully.
v2:
(Tvrtko)
- Skip test if preemption timeout compiled out
- Skip test if engine reset isn't supported
- Update debug prints to be more descriptive
v3:
- Add comment explaining test
v4:
(John Harrison)
- Fix typos in comment explaining test
- goto out_rq is NOP creation fails
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220113181351.21296-2-matthew.brost@intel.com
Add a flag PIN_VALIDATE, to indicate we don't need to pin and only
protected by the object lock.
This removes the need to unpin, which is done by just releasing the
lock.
eb_reserve is slightly reworked for readability, but the same steps
are still done:
- First pass pins with NONBLOCK.
- Second pass unbinds all objects first, then pins.
- Third pass is only called when not all objects are softpinned, and
unbinds all objects, then calls i915_gem_evict_vm(), then pins.
Changes since v1:
- Split out eb_reserve() into separate functions for readability.
Changes since v2:
- Make batch buffer mappable on platforms where only GGTT is available,
to prevent moving the batch buffer during relocations.
Changes since v3:
- Preserve current behavior for batch buffer, instead be cautious when
calling i915_gem_object_ggtt_pin_ww, and re-use the current batch vma
if it's inside ggtt and map-and-fenceable.
- Remove impossible condition check from eb_reserve. (Matt)
Changes since v5:
- Do not even temporarily pin, just call i915_gem_evict_vm() and mark
all vma's as unpinned.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220114132320.109030-7-maarten.lankhorst@linux.intel.com
We want to remove more members of i915_vma, which requires the locking to
be held more often.
Start requiring gem object lock for i915_vma_unbind, as it's one of the
callers that may unpin pages.
Some special care is needed when evicting, because the last reference to
the object may be held by the VMA, so after __i915_vma_unbind, vma may be
garbage, and we need to cache vma->obj before unlocking.
Changes since v1:
- Make trylock failing a WARN. (Matt)
- Remove double i915_vma_wait_for_bind() (Matt)
- Move atomic_set to right before mutex_unlock(), to make it more clear
they belong together. (Matt)
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.william.auld@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220114132320.109030-5-maarten.lankhorst@linux.intel.com
i915_gem_evict_vm will need to be able to evict objects that are
locked by the current ctx. By testing if the current context already
locked the object, we can do this correctly. This allows us to
evict the entire vm even if we already hold some objects' locks.
Previously, this was spread over several commits, but it makes
more sense to commit the changes to i915_gem_evict_vm separately
from the changes to i915_gem_evict_something() and
i915_gem_evict_for_node().
Changes since v1:
- Handle evicting dead objects better.
Changes since v2:
- Use for_i915_gem_ww in igt_evict_vm. (Thomas)
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
[mlankhorst: Fix up doc warning.]
Link: https://patchwork.freedesktop.org/patch/msgid/20220117075604.131477-1-maarten.lankhorst@linux.intel.com
Maarten needs backmerge to account for header file renames/changes which
landed via drm-intel-next and are interfering with his pinning work.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Two DT bindings fixes for meson, a device refcounting fix for sun4i, a
probe fix for vga16fb, a locking fix for the CMA dma-buf heap and a
compilation fix for ttm.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
[danvet: I made sure I have exactly the same conflict resolution as
Linus in 8d0749b4f8 ("Merge tag 'drm-next-2022-01-07' of
git://anongit.freedesktop.org/drm/drm") to avoid further conflict fun.
From: Maxime Ripard <maxime@cerno.tech>
Link: https://patchwork.freedesktop.org/patch/msgid/20220114125454.zs46ny52lrxk3ljz@houat
Even if it's probably not really useful, it can get selected
by e.g. randconfig builds, and then failing to compile is an
annoyance. Unfortunately, it's hard to fix in Kconfig, since
DRM_TTM is selected by many things that don't really depend
on any specific architecture, and just depend on PCI (which
is indeed now available in ARCH=um via simulation/emulation).
Fix this in the code instead by just ifdef'ing the relevant
two lines that depend on "real X86".
Reported-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20211220111519.a4c8c6eff702.Ie4cf4e68698f6a9f546b83379bc52c266504424f@changeid
For some reason this file isn't using the appropriate register
headers for DCN headers, which means that on DCN2 we're getting
the VIEWPORT_DIMENSION offset wrong.
This means that we're not correctly carving out the framebuffer
memory correctly for a framebuffer allocated by EFI and
therefore see corruption when loading amdgpu before the display
driver takes over control of the framebuffer scanout.
Fix this by checking the DCE_HWIP and picking the correct offset
accordingly.
Long-term we should expose this info from DC as GMC shouldn't
need to know about DCN registers.
Cc: stable@vger.kernel.org
Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
PMFW only returns 0 on master die and sends NACK back on other dies for
the message.
v2: only send GmiPwrDnControl msg on master die instead of all
dies.
v3: remove the pointer check for get_socket_id and get_die_id as they
should be present on Aldebaran.
Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The remove_list head was only used for keeping track of existing ranges
that are to be removed from the svms->list. The update_list was used for
new or existing ranges that need updated attributes. These two cases are
mutually exclusive (i.e. the same range will never be on both lists).
Therefore we can use the update_list head to track the remove_list and
save another 16 bytes in the svm_range struct.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Philip Yang <Philip.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
There are seven list_heads in struct svm_range: list, update_list,
remove_list, insert_list, svm_bo_list, deferred_list, child_list. This
patch and the next one remove two of them that are redundant.
The insert_list head was only used for new ranges that are not on the
svms->list yet. So we can use that list head for keeping track of
new ranges before they get added, and use list_move_tail to move them
to the svms->list when ready.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Philip Yang <Philip.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
As the possible failure of the allocation, kmemdup() may return NULL
pointer.
Therefore, it should be better to check the 'props2' in order to prevent
the dereference of NULL pointer.
Fixes: 3a87177eb1 ("drm/amdkfd: Add topology support for dGPUs")
Signed-off-by: Jiasheng Jiang <jiasheng@iscas.ac.cn>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The function performs a check on the "adev" input parameter, however, it
is used before the check.
Initialize the "dev" variable after the sanity check to avoid a possible
NULL pointer dereference.
Fixes: e27c41d5b0 ("drm/amd/display: Support for DMUB HPD interrupt handling")
Addresses-Coverity-ID: 1493909 ("Null pointer dereference")
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: José Expósito <jose.exposito89@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Otherwise future commands may fail as well leading to downstream
problems that look like they stemmed from a timeout the first time
but really didn't.
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The function amdgpu_mm_wreg_mmio_rlc() is used by debugfs to write to
MMIO registers. It didn't support registers beyond the BAR mapped MMIO
space. This adds pcie indirect write support.
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Get rid off pin/unpin of gart BO at resume/suspend and
instead pin only once and try to recover gart content
at resume time. This is much more stable in case there
is OOM situation at 2nd call to amdgpu_device_evict_resources()
while evicting GART table.
v3: remove gart recovery from other places
v2: pin gart at amdgpu_gart_table_vram_alloc()
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Do not allow exported amdgpu_vram_mgr_*() to accept
any ttm_resource_manager pointer. Also there is no need
to force other module to call a ttm function just to
eventually call vram_mgr functions.
v2: pass adev's vram_mgr instead of adev
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Do not allow exported amdgpu_gtt_mgr_*() to accept
any ttm_resource_manager pointer. Also there is no need
to force other module to call a ttm function just to
eventually call gtt_mgr functions.
v4: remove unused adev.
v3: upcast mgr from ttm resopurce manager instead of
getting it from adev.
v2: pass adev's gtt_mgr instead of adev.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Patch: 3efb17ae7e92 ("drm/amdgpu: Call amdgpu_device_unmap_mmio() if device
is unplugged to prevent crash in GPU initialization failure") makes call to
amdgpu_device_unmap_mmio() conditioned on device unplugged. This patch unmaps
MMIO mappings even when device is not unplugged.
v2: Add condition of drm_dev_enter() to deleted unmaps in patch
"drm/amdgpu: Unmap all MMIO mappings"
Signed-off-by: Leslie Shi <Yuliang.Shi@amd.com>
Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Memory of err_data should be cleaned before usage
when there're multiple entry in ras ih.
Otherwise garbage data from last loop will be used.
Signed-off-by: Jiawei Gu <Jiawei.Gu@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
If the GuC fails to load, it is useful to know what firmware file /
version was attempted. So move the version info report to before the
load attempt rather than only after a successful load.
If the GuC does fail to load, then make the error messages visible
rather than being 'debug' prints that do not appears in dmesg output
by default.
When waiting for the GuC to load, it used to be necessary to check for
two different states - READY and (LAPIC_DONE | MIA_CORE). Apparently
the second signified init complete on RC6 exit. However, in more
recent GuC versions the RC6 exit sequence now finishes with status
READY as well. So the test can be simplified.
Also, add an enum giving all the current status codes that GuC loading
can report as a reference without having to pull and search through
the GuC source files.
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220107000622.292081-4-John.C.Harrison@Intel.com
Update to the latest GuC release.
The latest GuC firmware introduces a number of interface changes:
GuC may return NO_RESPONSE_RETRY message for requests sent over CTB.
Add support for this reply and try resending the request again as a
new CTB message.
A KLV (key-length-value) mechanism is now used for passing
configuration data such as CTB management.
With the new KLV scheme, the old CTB management actions are no longer
used and are removed.
Register capture on hang is now supported by GuC. Full i915 support
for this will be added by a later patch. A minimum support of
providing capture memory and register lists is required though, so add
that in.
The device id of the current platform needs to be provided at init time.
The 'poll CS' w/a (Wa_22012773006) was blanket enabled by previous
versions of GuC. It must now be explicitly requested by the KMD. So,
add in the code to turn it on when relevant.
The GuC log entry format has changed. This requires adding a new field
to the log header structure to mark the wrap point at the end of the
buffer (as the buffer size is no longer a multiple of the log entry
size).
New CTB notification messages are now sent for some things that were
previously only sent via MMIO notifications.
Of these, the crash dump notification was not really being handled by
i915. It called the log flush code but that only flushed the regular
debug log and then only if relay logging was enabled. So just report
an error message instead.
The 'exception' notification was just being ignored completely. So add
an error message for that as well.
Note that in either the crash dump or the exception case, the GuC is
basically dead. The KMD will detect this via the heartbeat and trigger
both an error log (which will include the crash dump as part of the
GuC log) and a GT reset. So no other processing is really required.
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220107000622.292081-3-John.C.Harrison@Intel.com
There is a known (but exceedingly unlikely) race condition where the
asynchronous frequency management code could reduce the GT clock while
a GuC reload is in progress (during a full GT reset). A fix is in
progress but there are complex locking issues to be resolved. In the
meantime bump the timeout to 200ms. Even at slowest clock, this
should be sufficient. And in the working case, a larger timeout makes
no difference.
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220107000622.292081-2-John.C.Harrison@Intel.com
Actual hardware state of CRTC is controlled by the member 'active' in
struct drm_crtc_state instead of the member 'enable', according to the
kernel doc of the member 'enable'. In fact, the drm client modeset
and atomic helpers are using the member 'active' to do the control.
Referencing the member 'enable' of new_crtc_state, the function
crtc_needs_disable() may fail to reflect if CRTC needs disable in
self refresh mode, e.g., when the framebuffer emulation will be blanked
through the client modeset helper with the next commit, the member
'enable' of new_crtc_state is still true while the member 'active' is
false, hence the relevant potential encoder and bridges won't be disabled.
So, let's check new_crtc_state->active to determine if CRTC needs disable
in self refresh mode instead of new_crtc_state->enable.
Fixes: 1452c25b0e ("drm: Add helpers to kick off self refresh mode in drivers")
Cc: Sean Paul <seanpaul@chromium.org>
Cc: Rob Clark <robdclark@chromium.org>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Maxime Ripard <mripard@kernel.org>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: David Airlie <airlied@linux.ie>
Cc: Daniel Vetter <daniel@ffwll.ch>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Liu Ying <victor.liu@nxp.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211230040626.646807-1-victor.liu@nxp.com
Implement async (non-blocking) unbinding by not syncing the vma before
calling unbind on the vma_resource.
Add the resulting unbind fence to the object's dma_resv from where it is
picked up by the ttm migration code.
Ideally these unbind fences should be coalesced with the migration blit
fence to avoid stalling the migration blit waiting for unbind, as they
can certainly go on in parallel, but since we don't yet have a
reasonable data structure to use to coalesce fences and attach the
resulting fence to a timeline, we defer that for now.
Note that with async unbinding, even while the unbind waits for the
preceding bind to complete before unbinding, the vma itself might have been
destroyed in the process, clearing the vma pages. Therefore we can
only allow async unbinding if we have a refcounted sg-list and keep a
refcount on that for the vma resource pages to stay intact until
binding occurs. If this condition is not met, a request for an async
unbind is diverted to a sync unbind.
v2:
- Use a separate kmem_cache for vma resources for now to isolate their
memory allocation and aid debugging.
- Move the check for vm closed to the actual unbinding thread. Regardless
of whether the vm is closed, we need the unbind fence to properly wait
for capture.
- Clear vma_res::vm on unbind and update its documentation.
v4:
- Take cache coloring into account when searching for vma resources
pending unbind. (Matthew Auld)
v5:
- Fix timeout and error check in i915_vma_resource_bind_dep_await().
- Avoid taking a reference on the object for async binding if
async unbind capable.
- Fix braces around a single-line if statement.
v6:
- Fix up the cache coloring adjustment. (Kernel test robot <lkp@intel.com>)
- Don't allow async unbinding if the vma_res pages are not the same as
the object pages. (Matthew Auld)
v7:
- s/unsigned long/u64/ in a number of places (Matthew Auld)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220110172219.107131-5-thomas.hellstrom@linux.intel.com
