[Why]
hdmi_frl_status_polling_work() skipped any link whose connector_signal
was not SIGNAL_TYPE_HDMI_FRL. connector_signal is not reliably set to
SIGNAL_TYPE_HDMI_FRL while a link is actually running FRL, so links that
were operating in FRL mode were skipped and their status flags never got
polled, missing link-retrain events.
[How]
Use frl_link_settings.frl_link_rate to decide whether a link is running
FRL. A non-zero rate means FRL is active, so only links with a zero rate
are skipped. This ensures every link actually operating in FRL mode is
polled for status changes.
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Fangzhi Zuo <jerry.zuo@amd.com>
Signed-off-by: Roman Li <roman.li@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
VRR info packet previously only supported up to 10-bit refresh rate values
limiting the range of FreeSync minimum and maximum refresh rates that
could be encoded.
[How]
Expanded the bit masking in PB11/PB12 from 2 to 4 bits to capture bits
11:8 of the minimum and maximum FreeSync refresh rates, enabling the VRR
info packet to encode 12-bit refresh rate values.
Reviewed-by: Anthony Koo <anthony.koo@amd.com>
Signed-off-by: Harry VanZyllDeJong <hvanzyll@amd.com>
Signed-off-by: Roman Li <roman.li@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[why]
ensure dtbclk is enabled before hdmistreamclk_en
pmfw could stop dtbclk on idle.
driver needs to ensure dtbclk enabled is enabled before hdmistreamclk_en
also disable debounce timer on dcn42.
Reviewed-by: Chris Park <chris.park@amd.com>
Reviewed-by: Leo Chen <leo.chen@amd.com>
Signed-off-by: Charlene Liu <Charlene.Liu@amd.com>
Signed-off-by: Roman Li <roman.li@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
dm_dp_mst_get_modes() uses drm_connector->index to index the
per-connector HDCP arrays in struct hdcp_workqueue. Those arrays are
sized to AMDGPU_DM_MAX_DISPLAY_COUNT, which matches the DRM connector
index range (0..31). Add a defensive bounds check so that, should the
DRM connector index range ever grow beyond the array size, the access
is skipped instead of reading and writing out of bounds.
Assisted-by: Copilot:claude-opus-4.8
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Roman Li <roman.li@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
AMDGPU_DM_MAX_DISPLAY_INDEX is 31. It suggest a maximum number of
32 connectors. But the way it's used is like MAX_DISPLAY_COUNT.
Hence we're off by one with DRM core, which supports a max of 32
connectors.
Rename AMDGPU_DM_MAX_DISPLAY_INDEX to AMDGPU_DM_MAX_DISPLAY_COUNT
to match its actual use, and increase the size to 32 to match the
originally intended size.
Fixes: 82986fd631 ("drm/amd/display: save restore hdcp state when display is unplugged from mst hub")
Assisted-by: Copilot:claude-opus-4.8
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Roman Li <roman.li@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Set bits on dword 2 like GFX7-8 except load_global_uconfig
which doesn't exist on GFX6.
Emit VS_PARTIAL_FLUSH before VGT_FLUSH like GFX7-8.
For reference see old PAL which explains the bit fields in
this register and that load_global_uconfig doesn't exist on GFX6
and also see gfx_v7_ring_emit_cntxcntl() for the GFX7 code
which this commit follows.
Fixes: 2cd46ad223 ("drm/amdgpu: add graphic pipeline implementation for si v8")
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Check for mmhub0 rather than mmhub1. Looks like a copy
paste typo.
Fixes: d0c989a0aa ("drm/amd/amdgpu : Use the MES INV_TLBS API for tlb invalidation on gfx12_1")
Cc: Shaoyun Liu <shaoyun.liu@amd.com>
Reviewed-by: Shaoyun Liu <shaoyun.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
- Add hooks in various entry points to perform hw/sw init for DCN6 asic
- Add dependent changes needed to enable DCN6 asic
- Update the Makefiles so that DCN6 related newly added sources are compiled
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Roman Li <Roman.Li@amd.com>
Reviewed-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Allocate each ring buffer separately. A single allocation summing all
ring sizes can exceed the page allocator's MAX_ORDER limit and fail;
per-ring buffers stay small enough to satisfy. The existing allocation
style doesn't capture any ring data if the huge allocation fails.
Splitting into multiple allocations helps to capture as much data as
possible for the core dump.
A failed ring is left with a NULL buffer and skipped when formatting.
Fixes: eea85914d1 ("drm/amdgpu: save ring content before resetting the device")
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Assisted-by: Claude Code
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The number of rings with outstanding fences can be large, requiring a
bigger allocation. Such allocations don't need to be physically
contiguous, so use kvzalloc/kvcalloc which fall back to vmalloc when
contiguous memory isn't available. This also matches the existing
kvfree used to free these allocations.
Also guard the allocation with ring_count to avoid passing 0 size to
allocation routines.
Fixes: eea85914d1 ("drm/amdgpu: save ring content before resetting the device")
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
On GFX rings, amdgpu_cs_p2_ib() passed user-supplied ib_bytes through
to ib->length_dw without a limit, while ring_emit_ib() encodes length
into packet fields. Oversized values can corrupt adjacent control bits
and destabilize command submission.
Add a per-ring IB packet size limit helper and reject command
submissions exceeding the corresponding dword limit before IB
allocation. Use the documented 20-bit limit for GFX/compute/SDMA/VPE,
and apply the MM fallback limit for other ring types.
Signed-off-by: Candice Li <candice.li@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Creating a user queue can race with a GPU reset. While recovery holds
reset_domain->sem for write, MES is unresponsive, so the ADD_QUEUE from
amdgpu_userq_map_helper() times out (-110) and an otherwise valid queue
create fails:
amdgpu: MES(0) failed to respond to msg=ADD_QUEUE
[drm:mes_userq_map [amdgpu]] *ERROR* Failed to map queue in HW, err (-110)
amdgpu: [drm] *ERROR* ... Failed to map Queue
amdgpu: [drm] *ERROR* ... Failed to create usermode queue
Take reset_domain->sem for read around the map so it runs only once MES
is back up. This mirrors amdgpu_userq_cleanup() and honors the
userq_mutex -> reset_domain->sem order; the reset path never takes
userq_mutex, so there is no deadlock.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
mes_userq_reset() unmaps the queue via the low-level mes_userq_unmap()
(REMOVE_QUEUE) but does not update queue->state, so the queue still looks
MAPPED. The destroy path then issues a second, redundant REMOVE_QUEUE for
the already-removed queue; for gfx that unmap waits on an EOP that never
arrives, times out (-110) and escalates to a full GPU reset.
Mark the queue UNMAPPED on a successful reset-path unmap so destroy skips
the redundant REMOVE_QUEUE.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
If a priv/bad-op fault does not match a kernel queue slot, it belongs to a
MES-scheduled user queue. Extend the shared amdgpu_gfx_handle_priv_fault()
helper introduced by commit d8ab7636160e ("drm/amd/amdgpu: remove
duplicated code in gfx_v11 and gfx_v12") to recover it: gate on
adev->gfx.disable_uq, reset a compute user queue directly from its
doorbell, and for a gfx user queue (whose IV carries no doorbell) record
the HW slot and schedule the per-IP recovery worker.
v2:
- gate on adev->gfx.disable_uq instead of !adev->enable_mes (Alex)
- document why both the doorbell (compute) and HW-slot (gfx) reset
paths are needed (Alex)
v3:
- rebase amd-staging-drm-next. adapt to the
commit 9243cf4777 ("drm/amd/amdgpu: remove duplicated code in gfx_v11 and gfx_v12");
no functional change
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Suggested-by: Mario Sopena-Novales <Mario.Novales@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Mirror the gfx11 priv-fault user-queue recovery worker for GFX12, reading
the doorbell back from the HQD via soc24_grbm_select.
The shared amdgpu_gfx_handle_priv_fault() helper schedules this worker for
a gfx user-queue fault; wiring the helper up is done in a later patch.
v2:
- gate on adev->gfx.disable_uq instead of !adev->enable_mes (Alex)
- document why both the doorbell (compute) and HW-slot (gfx) reset
paths are needed (Alex)
v3:
- rebase amd-staging-drm-next. adapt to the commit 9243cf4777 ("drm/amd/amdgpu: remove duplicated code in gfx_v11 and gfx_v12"); no functional change
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Suggested-by: Mario Sopena-Novales <Mario.Novales@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
A gfx user-queue priv/bad-op fault is raised by the ME and carries only
the HW slot, not the faulting queue's doorbell. Add a per-IP worker that
drains adev->gfx.userq_priv_fault_slots, reads the doorbell back from each
HQD via soc21_grbm_select (regCP_RB_DOORBELL_CONTROL), looks up the user
queue and kicks its per-queue reset.
The shared amdgpu_gfx_handle_priv_fault() helper schedules this worker for
a gfx user-queue fault; wiring the helper up is done in a later patch.
v2:
- gate on adev->gfx.disable_uq instead of !adev->enable_mes (Alex)
- document why both the doorbell (compute) and HW-slot (gfx) reset
paths are needed (Alex)
v3:
- rebase amd-staging-drm-next. adapt to the commit 9243cf4777 ("drm/amd/amdgpu: remove duplicated code in gfx_v11 and gfx_v12"); no functional change
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Suggested-by: Mario Sopena-Novales <Mario.Novales@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
A gfx priv/bad-op fault IV carries only the HW slot (ring_id), not the
faulting user queue's doorbell. Add userq_priv_fault_slots (an atomic
bitmap of faulted slots, so concurrent faults are not dropped) and
userq_priv_fault_work to struct amdgpu_gfx; a worker drains the bitmap
and reads the doorbell back from each HQD to locate and reset the queue.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Move fs_reclaim_acquire() to before all lock acquisitions to eliminate
false positive circular locking dependency warning.
This is a 7.2-cycle regression fix suitable for stable backport.
v3: Address Mikhail Gavrilov technical review:
- Clarify that fs_reclaim_acquire/release pair only REGISTERS the
fs_reclaim lock class, does NOT create a static edge when called
with no locks held
- Explain that the actual fs_reclaim -> notifier_lock edge is
established at runtime during memory reclaim -> MMU notifier path
- Add Cc: Arunpravin PaneerSelvam
v2: Address Mikhail Gavrilov review feedback:
- Fix author name: Michael -> Mikhail Gavrilov in all trailers
- Add Fixes: tag to link regression to original commit
- Add Tested-by: Mikhail Gavrilov (tested on RX 7900 XTX)
Fixes: 1d0f5838b1 ("drm/amdgpu: Add lockdep annotations for lock ordering validation")
Reported-by: Mikhail Gavrilov <mikhail.v.gavrilov@gmail.com>
Analyzed-by: Mikhail Gavrilov <mikhail.v.gavrilov@gmail.com>
Test-case-by: Mikhail Gavrilov <mikhail.v.gavrilov@gmail.com>
Tested-by: Mikhail Gavrilov <mikhail.v.gavrilov@gmail.com>
Suggested-by: Christian König <christian.koenig@amd.com>
Tested-by: Vitaly Prosyak <vitaly.prosyak@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Arunpravin PaneerSelvam <Arunpravin.PaneerSelvam@amd.com>
Signed-off-by: Vitaly Prosyak <vitaly.prosyak@amd.com>
Acked-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Replace the simple exact-match loop in emit_clk_levels with a call
to smu_v14_0_0_find_clk_level() introduced in patch 1. The helper
already handles both exact and closest-match semantics.
Build a stack-local frequency table from the DPM levels (using
reverse index for SMU_MCLK since MemPstateTable stores levels
high-to-low), then call the helper once to find the active level.
The SMU reports time-filtered average frequencies that often do not
match any DPM table entry exactly. Without closest-match fallback,
MCLK, FCLK and other clocks show DPM levels but never display the
* marker, breaking userspace tools that rely on it to identify the
active frequency.
Signed-off-by: Priya Hosur <Priya.Hosur@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Add SMU_DCEFCLK case to smu_v14_0_1_get_dpm_freq_by_index and
smu_v14_0_0_get_dpm_freq_by_index using DcfClocks[] with
NumDcfClkLevelsEnabled bounds check. Add matching case in both
get_dpm_level_count functions.
Add SMU_DCEFCLK case in emit_clk_levels to list DCEF DPM levels.
No * marker is emitted since SmuMetrics_t has no DcfclkFrequency
field (same firmware limitation as Phoenix).
Without this, pp_dpm_dcefclk reports N/A on Strix Halo.
Signed-off-by: Priya Hosur <Priya.Hosur@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Add IP_VERSION(11,5,1) to pp_dpm_vclk and pp_dpm_dclk visibility
whitelists so these sysfs entries are exposed on Strix Halo (GC
11.5.1). Add IP_VERSION(11,5,1) to pp_dpm_vclk1 and pp_dpm_dclk1
whitelists with the existing num_vcn_inst >= 2 guard since Strix
Halo has two VCN instances.
Without this, amd-smi reports N/A for VCLK0, VCLK1, DCLK0 and
DCLK1 clocks.
Signed-off-by: Priya Hosur <Priya.Hosur@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
SmuMetrics_t has no DclkFrequency field but DCLK and VCLK have
separate DPM clock tables with different frequencies at each level.
Introduce smu_v14_0_0_find_clk_level(), a shared helper that
finds the closest DPM level for a given target frequency in a
frequency array.
For METRICS_AVERAGE_DCLK, use the helper to find the DPM level
whose VCLK frequency matches the reported VclkFrequency and
return the DCLK frequency at that same level index, since both
clocks share the same level count (VcnClkLevelsEnabled /
Vcn0ClkLevelsEnabled).
The original code returned 0 for METRICS_AVERAGE_DCLK, which broke
the active-level marker in pp_dpm_dclk entirely.
Signed-off-by: Priya Hosur <Priya.Hosur@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
GPU reset reinitializes PMFW and reloads the platform power table. The
saved user policy remains valid, but suspend-only restore does not run
because adev->in_suspend is clear.
Restore the active PPT policy from SMU late initialization while the
device is in reset recovery. Route each value through the common range
validation and ASIC setter before updating PMFW.
Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The existing user policy representation has three ambiguities:
- A numeric value cannot distinguish explicit zero from an unset policy.
- One value per controller cannot preserve independent AC and DC requests.
- Suspend-only restore misses runtime resume, GPU reset, and table reload.
Refactor policy storage and restore as follows:
- Store values and validity masks by power source and PPT controller.
- Save writes against the active source.
- Restore the active source after default SMU setup.
- Reapply the target policy after live AC/DC transitions.
- Use the target source default when no explicit request exists.
The late-init path now covers system resume, runtime resume, GPU reset,
and custom PPTable reload. Common code owns persistent policy; PMFW
continues to own effective current limits.
Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
SMU 13.0.0, SMU 13.0.7, and SMU 14.0.2 represent limits above the
message limit as a base value plus an overdrive percentage. GetPptLimit
returns only the base, which causes two incorrect results:
- hwmon reports the message limit instead of the effective limit.
- Lowering the cap can leave the previous OD percentage active.
Export the active overdrive table from PMFW and combine its PPT
percentage with the message result. Use the exported percentage to clear
OD state before programming a limit within the message range.
Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
amdgpu_userq_input_va_validate() converts expected_size to page count
with a plain right shift. For expected_size smaller than one GPU page,
the computed size becomes 0 and the range check is effectively bypassed.
This allows sub-page userq buffers (e.g. EOP/CSA) to pass validation
as long as the start VA is mapped, without verifying the full span.
Fix it by rejecting zero expected_size, checking overflow when
computing end address, and deriving page span from [start_addr,
end_addr] inclusive.
Signed-off-by: Candice Li <candice.li@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The scalar PPT model has several structural limitations:
- Controller identity, power source, capability, and PMFW state share
the same fields.
- Combined callbacks cannot address PPT0 and PPT1 independently.
- Cached current limits can become stale after PMFW reset.
Refactor the model around these rules:
- Index ranges by AC/DC source and PPT0/PPT1 controller.
- Store normal and overdrive capabilities separately.
- Pass the controller to get and set callbacks.
- Keep slow and fast names as semantic aliases.
Assign each value to a single owner:
- PPTable or platform initialization supplies constant capabilities.
- PMFW supplies runtime state.
- Common code selects the active source and overdrive range.
Van Gogh defaults are initialized once, while firmware without PPT query
support remains usable. Capable SMU 13.0.6 and SMU 15.0.8 platforms
expose PPT1 through the same interface. A writable controller must also
provide a setter callback.
Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
GetPptLimit reports mutable PMFW runtime state. Using that value for
static capabilities causes two externally visible problems:
- A userspace override changes the default reported through hwmon.
- A firmware reset changes the advertised minimum or maximum range.
Derive platform defaults and supported ranges from the driver PPTable
for Sienna Cichlid, Navi10, Arcturus, and Aldebaran. Reserve PMFW queries
for the effective current limit.
Runtime policy can no longer redefine immutable platform capabilities.
Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rename selected RAS core source and header filenames to drop
redundant ras_ prefixes in file names only.
Update related build object names, header include references, and
include guard macros in renamed headers.
Function/type prefixes and runtime behavior are unchanged.
No functional change intended.
Signed-off-by: YiPeng Chai <YiPeng.Chai@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When KFD_IOCTL_SVM_ATTR_NO_ACCESS is applied to a GPU that has an SVM
range mapped, unmap the range from that GPU. Once no GPU maps the range,
the MMU notifier can skip queue eviction on CPU page faults.
Replace the mapped_to_gpu boolean with bitmap_mapped to track which
GPUs currently have the range mapped. Set bits in svm_range_map_to_gpus()
and clear them in svm_range_unmap_from_gpus(). This is separate from
bitmap_access/bitmap_aip which track user-requested attributes and must
not be used to determine mapping state.
Add bitmap_needs_unmap to svm_range, set for each GPU given no-access.
Add svm_range_needs_unmap() to unmap the range from those GPUs when the
app sets the no-access attribute.
Bump the checkpoint timestamp on unmap so retry faults queued before the
no-access unmap are dropped instead of restoring the mapping.
v4:
- Rename and set prange->mapping_done to false if validate and map not
complete successfully (Felix)
v3:
- Correct error handling, support app retry update mapping (Felix)
v2:
- Add bitmap_mapped to not break get_attr (Felix)
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Extract svm_range_update_checkpoint_timestamp() from
svm_range_unmap_from_cpu(). The next patch calls it when the app sets
the no-access attribute.
Change checkpoint_ts in svm_range_list from uint64_t to atomic64_t so
svm_range_restore_pages() can read it from the page fault handler
without holding the svms lock.
No functional change, preparation for the next patch.
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Replace the legacy SMU table-copy metrics path with a DRAM-mapped,
accumulator-based approach for SMU v15.0.0, using a ping-pong buffer
to compute averaged metrics from deltas between consecutive samples.
- Add GetMetricsTableVersion, GetMetricsTableLogSample and
GetMetricsTableLogDramAddr messages and their MSG_MAP entries.
- Introduce SMU_15_0_0_MetricsInfo_t holding two MetricsTable_t
buffers, the mapped DRAM address, table size and pre-computed
avg_metric[] values.
- Resolve and ioremap_wc() the firmware log DRAM address once in
init_smc_tables(); iounmap() it in fini_smc_tables().
- Fetch samples into the inactive buffer, compute averaged clocks,
activity, power and temperature via wrapping_sub() accumulator
deltas, then swap the active index.
- Rename the old get_metrics_table() to get_gpu_metrics_table() for
the gpu_metrics v3.0 export path.
- Smartshift sensor reporting is dropped
Co-developed-by: Pratik Vishwakarma <Pratik.Vishwakarma@amd.com>
Signed-off-by: Pratik Vishwakarma <Pratik.Vishwakarma@amd.com>
Co-developed-by: Suresh Guttula <Suresh.Guttula@amd.com>
Signed-off-by: Suresh Guttula <Suresh.Guttula@amd.com>
Co-developed-by: Kanala Ramalingeswara Reddy <Kanala.RamalingeswaraReddy@amd.com>
Signed-off-by: Kanala Ramalingeswara Reddy <Kanala.RamalingeswaraReddy@amd.com>
Signed-off-by: Shubhankar Milind Sardeshpande <Shubhankar.MilindSardeshpande@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Acked-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rename the RAS core directory from rascore to core and update
build-path references accordingly.
The change is mechanical and done with git rename semantics so
history tracking is preserved.
No functional change intended.
Signed-off-by: YiPeng Chai <YiPeng.Chai@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Remove the obsolete gmc ras helpers including
amdgpu_gmc_ras_late_init and amdgpu_gmc_ras_fini
and their declarations, and call sites.
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Ce Sun <cesun102@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
TA_CNTL2.TRUNCATE_COORD_MODE selects whether texture coordinate
truncation is D3D9/GL/Vulkan conformant. gfx11 reads it and reports it to
userspace via AMDGPU_IDS_FLAGS_CONFORMANT_TRUNC_COORD, but gfx12 never
read it, so the flag was always reported as 0 and userspace fell back to
the non-conformant path.
Read it in gfx_v12_0_constants_init() like gfx11 does.
Fixes: 52cb80c12e ("drm/amdgpu: Add gfx v12_0 ip block support (v6)")
Signed-off-by: Qiang Yu <Qiang.Yu@amd.com>
Reviewed-by: Marek Olšák <maraeo@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Like jpeg_v5_0_0, in DPG mode the ring reset path only clears the
JPEG_PG_MODE bit and never resets a hung JRBC, so the post-reset ring test
times out and the driver falls back to a full MODE1 reset.
Temporarily force the static power-gating path during the reset so the
stop/start sequence power-cycles the JPEG block (JMI soft reset + power
off/on), matching the jpeg_v4_0 reset.
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Like jpeg_v5_0_0, in DPG mode the ring reset path only clears the
JPEG_PG_MODE bit and never resets a hung JRBC, so the post-reset ring test
times out and the driver falls back to a full MODE1 reset.
Temporarily force the static power-gating path during the reset so the
stop/start sequence power-cycles the JPEG block (JMI soft reset + power
off/on), matching the jpeg_v4_0 reset.
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
In DPG mode jpeg_v5_0_0_ring_reset() takes the DPG stop path, which only
clears the JPEG_PG_MODE bit and never resets the JRBC. A hung ring is not
recovered: the post-reset ring test times out and the driver falls back to
a full MODE1 reset.
Temporarily force the static power-gating path during the reset so the
stop/start sequence power-cycles the JPEG block (JMI soft reset + power
off/on), matching the jpeg_v4_0 reset which has no DPG path.
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The functions gfx_v11_0_handle_priv_fault and
gfx_v12_0_handle_priv_fault share the same logic for searching and
triggering a scheduler fault on a ring. This patch moves the shared
ring-searching logic to a common function, amdgpu_gfx_handle_priv_fault,
in amdgpu_gfx.c. The hardware-specific decoding of ring IDs remains in
the version-specific files to maintain proper architectural separation.
Signed-off-by: Ulisses Paixao <ulissespaixao@usp.br>
Co-developed-by: Felipe Sousa <felipesousa@usp.br>
Signed-off-by: Felipe Sousa <felipesousa@usp.br>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Enable IP block soft reset as a GPU recovery method for GFX7
graphics and compute rings.
This improves current user experience on all GFX7 chips:
* On Kaveri and Kabini there is currently no working
GPU recovery method so those chips currently require
the user to manually reset the computer when there
was a hang.
* On Hawaii and Bonaire, the current GPU recovery method
always clears the contents of VRAM, which means that
a buggy (hanging) app can crash the whole graphical
session, which is less than ideal.
Using GFX IP block soft reset means that we can now
have a working recovery on GFX7 APUs and we can also
move on from GFX hangs on dGPUs without crashing the
whole system.
Tested with the "hard_reset_cp_wait" test case from the
Hang Test Suite created by Natalie Vock and Konstantin Seurer.
This Vulkan testcase waits for an event that never occurs,
effectively a WAIT_REG_MEM packet that intentionally hangs.
IP block soft reset can resolve that hang and allow
the rest of the system to move on and keep functioning
without needing a full ASIC reset.
Tested on the following chips:
Bonaire (Radeon HD 7790)
Hawaii (Radeon R9 390X)
Kaveri (A10-7850K)
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Use basically the same implementation as GFX8,
except for the GFX7 specific MQD functions.
Reset every block using the GRBM, then proceed
to reset the GRBM and SEM blocks using the SRBM.
Remove the redundant gfx_v7_0_update_cg() function.
The soft reset now calls the clock and powergating
functions of the IP block instead.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
COND_EXEC tells the CP to discard the dwords following it
when its condition is zero (false).
This is useful for GPU recovery because it can help reduce
collateral damage during GFX IP block soft reset, meaning
that it reduces the likelyhood that we fail some jobs which
are not guilty of the hang as the IP block soft reset
mechanism clears the condition before doing the reset.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Clear the WPTR and RPTR at ring initialization.
Additionally clear the ring contents during reset.
After a reset, the ring contents could be "dirty"
and contain packets emitted before the reset.
and thus need to be cleared to prevent the command
processor from executing packets left over in the
ring from before the reset.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This packet is interpreted by the CE (constant engine).
The reason why this packet is emitted is basically to
make sure the CE can't start executing packets from the
next job submission until the current one is finished.
(Note that CE is not utilized by any maintained userspace
driver and is discontinued in new GPUs. It is now also
deprecated in the kernel.)
Implement the emit_switch_buffer() function instead of emitting
them duing emit_ib, emit_pipeline_sync and emit_vm_flush.
It isn't necessary to emit these in both emit_pipeline_sync()
and emit_vm_flush() because amdgpu_vm_flush() already calls
these when calling either of those functions.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
