aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2021-10-15drm/i915/guc: Assign contexts in parent-child relationship consecutive guc_idsMatthew Brost2-24/+86
Assign contexts in parent-child relationship consecutive guc_ids. This is accomplished by partitioning guc_id space between ones that need to be consecutive (1/16 available guc_ids) and ones that do not (15/16 of available guc_ids). The consecutive search is implemented via the bitmap API. This is a precursor to the full GuC multi-lrc implementation but aligns to how GuC mutli-lrc interface is defined - guc_ids must be consecutive when using the GuC multi-lrc interface. v2: (Daniel Vetter) - Explicitly state why we assign consecutive guc_ids v3: (John Harrison) - Bring back in spin lock Signed-off-by: Matthew Brost <[email protected]> Reviewed-by: John Harrison <[email protected]> Signed-off-by: John Harrison <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-10-15drm/i915/guc: Ensure GuC schedule operations do not operate on child contextsMatthew Brost1-1/+12
In GuC parent-child contexts the parent context controls the scheduling, ensure only the parent does the scheduling operations. Signed-off-by: Matthew Brost <[email protected]> Reviewed-by: John Harrison <[email protected]> Signed-off-by: John Harrison <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-10-15drm/i915/guc: Add multi-lrc context registrationMatthew Brost6-3/+134
Add multi-lrc context registration H2G. In addition a workqueue and process descriptor are setup during multi-lrc context registration as these data structures are needed for multi-lrc submission. v2: (John Harrison) - Move GuC specific fields into sub-struct - Clean up WQ defines - Add comment explaining math to derive WQ / PD address v3: (John Harrison) - Add PARENT_SCRATCH_SIZE define - Update comment explaining multi-lrc register v4: (John Harrison) - Move PARENT_SCRATCH_SIZE to common file Signed-off-by: Matthew Brost <[email protected]> Reviewed-by: John Harrison <[email protected]> Signed-off-by: John Harrison <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-10-15drm/i915/guc: Introduce context parent-child relationshipMatthew Brost3-0/+91
Introduce context parent-child relationship. Once this relationship is created all pinning / unpinning operations are directed to the parent context. The parent context is responsible for pinning all of its children and itself. This is a precursor to the full GuC multi-lrc implementation but aligns to how GuC mutli-lrc interface is defined - a single H2G is used register / deregister all of the contexts simultaneously. Subsequent patches in the series will implement the pinning / unpinning operations for parent / child contexts. v2: (Daniel Vetter) - Add kernel doc, add wrapper to access parent to ensure safety v3: (John Harrison) - Fix comment explaing GEM_BUG_ON in to_parent() - Make variable names generic (non-GuC specific) v4: (John Harrison) - s/its'/its/g Signed-off-by: Matthew Brost <[email protected]> Reviewed-by: John Harrison <[email protected]> Signed-off-by: John Harrison <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-10-15drm/i915: Expose logical engine instance to userMatthew Brost2-1/+9
Expose logical engine instance to user via query engine info IOCTL. This is required for split-frame workloads as these needs to be placed on engines in a logically contiguous order. The logical mapping can change based on fusing. Rather than having user have knowledge of the fusing we simply just expose the logical mapping with the existing query engine info IOCTL. IGT: https://patchwork.freedesktop.org/patch/445637/?series=92854&rev=1 media UMD: https://github.com/intel/media-driver/pull/1252 v2: (Daniel Vetter) - Add IGT link, placeholder for media UMD Cc: Tvrtko Ursulin <[email protected]> Signed-off-by: Matthew Brost <[email protected]> Reviewed-by: John Harrison <[email protected]> Signed-off-by: John Harrison <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-10-15drm/i915: Add logical engine mappingMatthew Brost5-29/+62
Add logical engine mapping. This is required for split-frame, as workloads need to be placed on engines in a logically contiguous manner. v2: (Daniel Vetter) - Add kernel doc for new fields v3: (Tvrtko) - Update comment for new logical_mask field v4: (John Harrison) - Update comment for new logical_mask field Signed-off-by: Matthew Brost <[email protected]> Reviewed-by: John Harrison <[email protected]> Signed-off-by: John Harrison <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-10-15drm/i915/guc: Don't call switch_to_kernel_context with GuC submissionMatthew Brost1-0/+13
Calling switch_to_kernel_context isn't needed if the engine PM reference is taken while all user contexts are pinned as if don't have PM ref that guarantees that all user contexts scheduling is disabled. By not calling switch_to_kernel_context we save on issuing a request to the engine. v2: (Daniel Vetter) - Add FIXME comment about pushing switch_to_kernel_context to backend v3: (John Harrison) - Update commit message - Fix workding comment Signed-off-by: Matthew Brost <[email protected]> Reviewed-by: Daniel Vetter <[email protected]> Reviewed-by: John Harrison <[email protected]> Signed-off-by: John Harrison <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-10-15drm/i915/guc: Take engine PM when a context is pinned with GuC submissionMatthew Brost5-3/+89
Taking a PM reference to prevent intel_gt_wait_for_idle from short circuiting while any user context has scheduling enabled. Returning GT idle when it is not can cause all sorts of issues throughout the stack. v2: (Daniel Vetter) - Add might_lock annotations to pin / unpin function v3: (CI) - Drop intel_engine_pm_might_put from unpin path as an async put is used v4: (John Harrison) - Make intel_engine_pm_might_get/put work with GuC virtual engines - Update commit message v5: - Update commit message again Signed-off-by: Matthew Brost <[email protected]> Reviewed-by: John Harrison <[email protected]> Signed-off-by: John Harrison <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-10-15drm/i915/guc: Take GT PM ref when deregistering contextMatthew Brost6-54/+121
Taking a PM reference to prevent intel_gt_wait_for_idle from short circuiting while a deregister context H2G is in flight. To do this must issue the deregister H2G from a worker as context can be destroyed from an atomic context and taking GT PM ref blows up. Previously we took a runtime PM from this atomic context which worked but will stop working once runtime pm autosuspend in enabled. So this patch is two fold, stop intel_gt_wait_for_idle from short circuting and fix runtime pm autosuspend. v2: (John Harrison) - Split structure changes out in different patch (Tvrtko) - Don't drop lock in deregister_destroyed_contexts v3: (John Harrison) - Flush destroyed contexts before destroying context reg pool Signed-off-by: Matthew Brost <[email protected]> Reviewed-by: John Harrison <[email protected]> Signed-off-by: John Harrison <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-10-15drm/i915/guc: Move GuC guc_id allocation under submission state sub-structMatthew Brost3-34/+46
Move guc_id allocation under submission state sub-struct as a future patch will reuse the spin lock as a global submission state lock. Moving this into sub-struct makes ownership of fields / lock clear. v2: (Docs) - Add comment for submission_state sub-structure v3: (John Harrison) - Fixup a few comments v4: (John Harrison) - Fix typo Signed-off-by: Matthew Brost <[email protected]> Reviewed-by: John Harrison <[email protected]> Signed-off-by: John Harrison <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-10-14drm/i915/gt: move remaining debugfs interfaces into gtAndi Shyti5-36/+111
The following interfaces: i915_wedged i915_forcewake_user are dependent on gt values. Put them inside gt/ and drop the "i915_" prefix name. This would be the new structure: dri/0/gt | +-- forcewake_user | \-- reset For backwards compatibility with existing igt (and the slight semantic difference between operating on the i915 abi entry points and the deep gt info): dri/0 | +-- i915_wedged | \-- i915_forcewake_user remain at the top level. Signed-off-by: Andi Shyti <[email protected]> Cc: Tvrtko Ursulin <[email protected]> Cc: Chris Wilson <[email protected]> Reviewed-by: Lucas De Marchi <[email protected]> Signed-off-by: Lucas De Marchi <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-10-14drm/i915: fix blank screen booting crashesHugh Dickins1-2/+2
5.15-rc1 crashes with blank screen when booting up on two ThinkPads using i915. Bisections converge convincingly, but arrive at different and surprising "culprits", none of them the actual culprit. netconsole (with init_netconsole() hacked to call i915_init() when logging has started, instead of by module_init()) tells the story: kernel BUG at drivers/gpu/drm/i915/i915_sw_fence.c:245! with RSI: ffffffff814d408b pointing to sw_fence_dummy_notify(). I've been building with CONFIG_CC_OPTIMIZE_FOR_SIZE=y, and that function needs to be 4-byte aligned. v2: (Jani Nikula) - Change BUG_ON to WARN_ON v3: (Jani / Tvrtko) - Short circuit __i915_sw_fence_init on WARN_ON v4: (Lucas) - Break WARN_ON changes out in a different patch Fixes: 62eaf0ae217d ("drm/i915/guc: Support request cancellation") Signed-off-by: Hugh Dickins <[email protected]> Signed-off-by: Matthew Brost <[email protected]> Reviewed-by: Matthew Brost <[email protected]> Signed-off-by: John Harrison <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-10-14drm/i915/uapi: Add comment clarifying purpose of I915_TILING_* valuesMatt Roper1-0/+6
The I915_TILING_* values in our uapi header are intended solely for use with the old get_tiling/set_tiling ioctls that operate on hardware de-tiling fences; all other uapi communication about tiling types is done via framebuffer modifiers rather than with these old values. On newer Intel platforms detiling fences no longer exist so the old get_tiling/set_tiling ioctls are no longer usable and will always return -EOPNOTSUPP. This means there's no reason to add new tiling types (such as the Tile4 format introduced by Xe_HP) to the uapi header here. Any kernel-internal code that needs to represent tiling format should either rely on framebuffer modifiers (as the display code does) or use some kind of non-uapi enum (as the GEM blt selftest now does). References: https://patchwork.freedesktop.org/patch/456656/?series=95308 Cc: Ville Syrjälä <[email protected]> Signed-off-by: Matt Roper <[email protected]> Reviewed-by: Caz Yokoyama <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-10-14drm/i915: Fix oops on platforms w/o hpd supportVille Syrjälä1-1/+1
We don't have hpd support on i8xx/i915 which means hotplug_funcs==NULL. Let's not oops when loading the driver on one those machines. v2: Drop the redundant function pointer check (Jani) Cc: Dave Airlie <[email protected]> Cc: Jani Nikula <[email protected]> Fixes: cd030c7c11a4 ("drm/i915: constify hotplug function vtable.") Signed-off-by: Ville Syrjälä <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] Reviewed-by: Jani Nikula <[email protected]>
2021-10-14drm/i915: Remove memory frequency calculationJosé Roberto de Souza2-36/+2
This memory frequency calculated is only used to check if it is zero, what is not useful as it will never actually be zero. Also the calculation is wrong, we should be checking other bit to select the appropriate frequency multiplier while this code is stuck with a fixed multiplier. So here dropping it as whole. v2: - Also remove memory frequency calculation for gen9 LP platforms Cc: Yakui Zhao <[email protected]> Cc: Matt Roper <[email protected]> Fixes: 5d0c938ec9cc ("drm/i915/gen11+: Only load DRAM information from pcode") Signed-off-by: José Roberto de Souza <[email protected]> Reviewed-by: Matt Roper <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-10-14drm/i915: Add all per-lane register definitions for icl combo phyVille Syrjälä4-24/+22
Add the FOO_LN() register macros for all the icl combo phy registers. Also get rid of the semi-pointless FOO_LN0() variants and just use the parametrized version. Signed-off-by: Ville Syrjälä <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] Reviewed-by: Jani Nikula <[email protected]>
2021-10-14drm/i915: Extract icl_combo_phy_loadgen_select()Ville Syrjälä1-9/+14
Pull the convoluted loadgen calculation into a small helper. Signed-off-by: Ville Syrjälä <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] Reviewed-by: Jani Nikula <[email protected]>
2021-10-14drm/i915: Remove dead DKL_TX_LOADGEN_SHARING_PMD_DISABLE stuffVille Syrjälä2-9/+0
DKL_TX_LOADGEN_SHARING_PMD_DISABLE doesn't even seem to exist, also the spec says to skip all loadgen stuff. The code was dead anyway since it wasn't actually writing the value anywhere. Signed-off-by: Ville Syrjälä <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] Reviewed-by: Jani Nikula <[email protected]>
2021-10-14drm/i915: Use standard form terminating condition for lane for loopsVille Syrjälä1-2/+2
Use <4 instead of <=3 as the terminating condition for the loops over the 4 lanes. Signed-off-by: Ville Syrjälä <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] Reviewed-by: Jani Nikula <[email protected]>
2021-10-14drm/i915: Shrink {icl_mg,tgl_dkl}_phy_ddi_buf_transVille Syrjälä1-6/+6
All the values we have in {icl_mg,tgl_dkl}_phy_ddi_buf_trans fit into u8. Shrink the types accordingly. Signed-off-by: Ville Syrjälä <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] Reviewed-by: Jani Nikula <[email protected]>
2021-10-14drm/i915: Remove pointless extra namespace from dkl/snps buf trans structsVille Syrjälä3-12/+12
The struct itself already has sufficient namespace. No need to duplicate it in the members. Signed-off-by: Ville Syrjälä <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] Reviewed-by: Jani Nikula <[email protected]>
2021-10-14drm/i915: rename intel_sideband.[ch] to intel_sbi.[ch]Jani Nikula5-29/+13
Now that intel_sideband.[ch] has been decluttered, it's pure lpt/wpt iosf sideband. Let's call it intel_sbi, following the function naming. Cc: Ville Syrjälä <[email protected]> Reviewed-by: Ville Syrjälä <[email protected]> Reviewed-by: Lucas De Marchi <[email protected]> Signed-off-by: Jani Nikula <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/183423ff23b2d259e4a197e74daf6bcd750bfe14.1634207064.git.jani.nikula@intel.com
2021-10-14drm/i915: split out intel_pcode.[ch] to separate fileJani Nikula17-255/+277
The snb+ pcode mailbox code is not sideband, so split it out to a separate file. As can be seen from the #include changes, very few places use both sideband and pcode. Code movement only. Cc: Ville Syrjälä <[email protected]> Reviewed-by: Ville Syrjälä <[email protected]> Reviewed-by: Lucas De Marchi <[email protected]> Signed-off-by: Jani Nikula <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/185deb18eb739e5ae019e27834b9997dcc1347bc.1634207064.git.jani.nikula@intel.com
2021-10-14drm/i915/bios: gracefully disable dual eDP for nowJani Nikula1-0/+47
For the time being, neither the power sequencer nor the backlight code properly support two eDP panels simultaneously. While the software states will be independent, the same sets of registers will be used for both eDP panels, clobbering the hardware state and leading to errors. Gracefully disable dual eDP until proper support has been added. Cc: José Roberto de Souza <[email protected]> Cc: Uma Shankar <[email protected]> Cc: Ville Syrjälä <[email protected]> Cc: Swati Sharma <[email protected]> Reviewed-by: José Roberto de Souza <[email protected]> Acked-by: Ville Syrjälä <[email protected]> Tested-by: Swati Sharma <[email protected]> Signed-off-by: Jani Nikula <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-10-14drm/i915: split out vlv sideband to a separate fileJani Nikula22-382/+405
The VLV/CHV sideband code is pretty distinct from the rest of the sideband code. Split it out to new vlv_sideband.[ch]. Pure code movement with relevant #include changes, and a tiny checkpatch fix on top. Cc: Lucas De Marchi <[email protected]> Cc: Ville Syrjälä <[email protected]> Reviewed-by: Hans de Goede <[email protected]> Acked-by: Lucas De Marchi <[email protected]> Signed-off-by: Jani Nikula <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/755ebbbaf01fc6d306b763b6ef60f45e671ba290.1634119597.git.jani.nikula@intel.com
2021-10-13drm/amdgpu/psp: add some missing cases to ↵Alex Deucher1-1/+2
psp_check_pmfw_centralized_cstate_management Missed a few asics. v2: update comment Fixes: 82d05736c47b19 ("drm/amdgpu/amdgpu_psp: convert to IP version checking") Reviewed-by: Evan Quan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13drm/amdgpu/swsmu: fix is_support_sw_smu() for VEGA20Alex Deucher1-0/+4
VEGA20 is 11.0.2, but it's handled by powerplay, not swsmu. Fixes: a8967967f6a554 ("drm/amdgpu/amdgpu_smu: convert to IP version checking") Reviewed-by: Evan Quan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13drm/amdkfd: unregistered svm range not overlap with TTM rangePhilip Yang3-7/+94
When creating unregistered new svm range to recover retry fault, avoid new svm range to overlap with ranges or userptr ranges managed by TTM, otherwise svm migration will trigger TTM or userptr eviction, to evict user queues unexpectedly. Change helper amdgpu_ttm_tt_affect_userptr to return userptr which is inside the range. Add helper svm_range_check_vm_userptr to scan all userptr of the vm, and return overlap userptr bo start, last. Signed-off-by: Philip Yang <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13drm/amdgpu: fix out of bounds writeThelford Williams1-1/+1
Size can be any value and is user controlled resulting in overwriting the 40 byte array wr_buf with an arbitrary length of data from buf. Signed-off-by: Thelford Williams <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13drm/amdgpu/smu11: fix firmware version check for vangoghAlex Deucher1-1/+1
Was missed in the conversion to IP version checking. Fixes: af3b89d3a639d5 ("drm/amdgpu/smu11.0: convert to IP version checking") Reviewed-by: Evan Quan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13MAINTAINERS: Add Siqueira for AMD DCHarry Wentland1-0/+1
He's been helping maintain it for quite a while now. Make it official. Signed-off-by: Harry Wentland <[email protected]> Acked-by: Christian König <[email protected]> Acked-by: Alex Deucher <[email protected]> Acked-by: Rodrigo Siqueira <[email protected]> Signed-off-by: Rodrigo Siqueira <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13drm/amdkfd: fix resume error when iommu disabled in PicassoYifan Zhang1-0/+1
When IOMMU disabled in sbios and kfd in iommuv2 path, IOMMU resume failure blocks system resume. Don't allow kfd to use iommu v2 when iommu is disabled. Reported-by: youling <[email protected]> Tested-by: youling <[email protected]> Signed-off-by: Yifan Zhang <[email protected]> Reviewed-by: James Zhu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13drm/amdkfd: fix boot failure when iommu is disabled in Picasso.Yifan Zhang2-4/+3
When IOMMU disabled in sbios and kfd in iommuv2 path, iommuv2 init will fail. But this failure should not block amdgpu driver init. Reported-by: youling <[email protected]> Tested-by: youling <[email protected]> Signed-off-by: Yifan Zhang <[email protected]> Reviewed-by: James Zhu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13drm/amdkfd: handle svm partial migration cpages 0Philip Yang1-34/+52
migrate_vma_setup may return cpages 0, means 0 page can be migrated, treat this as error case to skip the rest of vma migration steps. Change svm_migrate_vma_to_vram and svm_migrate_vma_to_ram to return the number of pages migrated successfully or error code. The caller add up all the successful migration pages and update prange->actual_loc only if the total migrated pages is not 0. This also removes the warning message "VRAM BO missing during validation" if migration cpages is 0. Signed-off-by: Philip Yang <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13drm/amdkfd: ratelimited svm debug messagesPhilip Yang2-21/+30
No function change, use pr_debug_ratelimited to avoid per page debug message overflowing dmesg buf and console log. use dev_err to show error message from unexpected situation, to provide clue to help debug without enabling dynamic debug log. Define dev_fmt to output function name in error message. Signed-off-by: Philip Yang <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13drm/amd/pm: Fix incorrect power limit readback in smu11 if POWER_SOURCE_DCDarren Powell1-4/+5
when smu->adev->pm.ac_power == 0, message parameter with bit 16 set is saved to smu->current_power_limit. Fixes: 0cb4c62125a9 ("drm/amd/pm: correct power limit setting for SMU V11)" Signed-off-by: Darren Powell <[email protected]> Reviewed-by: Lijo Lazar <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13amdgpu/pm: (v2) add limit_type to (pptable_funcs)->set_power_limit signatureDarren Powell8-19/+39
v2 add check for SMU_DEFAULT_PPT_LIMIT v1 modify (pptable_funcs)->set_power_limit signature modify smu11 set_power_limit signature (arcturus, navi10, sienna_cichlid) modify smu13 set_power_limit signature (aldabaran) modify vangogh_set_power_limit signature (vangogh) === Test === sudo bash AMDGPU_PCI_ADDR=`lspci -nn | grep "VGA\|Display" | cut -d " " -f 1` AMDGPU_HWMON=`ls -la /sys/class/hwmon | grep $AMDGPU_PCI_ADDR | awk '{print $9}'` HWMON_DIR=/sys/class/hwmon/${AMDGPU_HWMON} LOGFILE=pp_show_power_cap.log cp $LOGFILE{,.old} lspci -nn | grep "VGA\|Display" > $LOGFILE FILES=" power1_cap power2_cap" for f in $FILES do if test -f "$HWMON_DIR/$f"; then echo === $f === >> $LOGFILE cat $HWMON_DIR/$f >> $LOGFILE RESTORE_VALUE=`cat $HWMON_DIR/$f` 2>&1 >> $LOGFILE echo RESTORE_VALUE $RESTORE_VALUE >> $LOGFILE echo 120000000 > $HWMON_DIR/$f sleep 3 cat $HWMON_DIR/$f >> $LOGFILE echo $RESTORE_VALUE > $HWMON_DIR/$f sleep 3 cat $HWMON_DIR/$f >> $LOGFILE else echo === $f === >> $LOGFILE echo File Not Found >> $LOGFILE fi done cat $LOGFILE Signed-off-by: Darren Powell <[email protected]> Reviewed-by: Lijo Lazar <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13drm/amdgpu: Fix RAS page retirement with mode2 reset on AldebaranMukul Joshi1-12/+21
During mode2 reset, the GPU is temporarily removed from the mgpu_info list. As a result, page retirement fails because it cannot find the GPU in the GPU list. To fix this, create our own list of GPUs that support MCE notifier based page retirement and use that list to check if the UMC error occurred on a GPU that supports MCE notifier based page retirement. Signed-off-by: Mukul Joshi <[email protected]> Reviewed-by: Tao Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13drm/amdgpu: Enable RAS error injection after mode2 reset on AldebaranMukul Joshi1-0/+2
Add the missing call to re-enable RAS error injections on the Aldebaran mode2 reset code path. Signed-off-by: Mukul Joshi <[email protected]> Reviewed-by: Tao Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13drm/amd/display: fix null pointer deref when plugging in displayAurabindo Pillai1-1/+2
[Why&How] When system boots in headless mode, connecting a 4k display creates a null pointer dereference due to hubp for a certain plane being null. Add a condition to check for null hubp before dereferencing it. Signed-off-by: Aurabindo Pillai <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13drm/amd/display: Fix surface optimization regression on CarrizoNicholas Kazlauskas1-2/+13
[Why] DCE legacy optimization path isn't well tested under new DC optimization flow which can result in underflow occuring when initializing X11 on Carrizo. [How] Retain the legacy optimization flow for DCE and keep the new one for DCN to satisfy optimizations being correctly applied for ASIC that can support it. Fixes: 34316c1e561db0 ("drm/amd/display: Optimize bandwidth on following fast update") Reported-by: Tom St Denis <[email protected]> Tested-by: Tom St Denis <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Nicholas Kazlauskas <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13drm/amd/display: Enable PSR by default on newer DCNNicholas Kazlauskas2-3/+19
[Why] For optimal power savings on panels that can support it. This was previously left disabled by default because of issues with compositors that do not pageflip and scan out directly to the frontbuffer. For these compositors we now have detection methods that wait for x number of pageflips after a full update - triggered by a buffer or format change typically. This may introduce bugs or new cases not tested by users so this is only currently targeting newer DCN. [How] Add code in DM to set PSR state by default for newer DCN while falling back to the feature mask for older. Add a global debug flag that can be set to disable it for either. Signed-off-by: Nicholas Kazlauskas <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13drm/amdgpu: enable display for cyan skillfishLang Yu1-2/+1
Display support for cyan skillfish is ready now. Enable it! Signed-off-by: Lang Yu <[email protected]> Reviewed-by: Huang Rui <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13amd/display: check cursor plane matches underlying planeSimon Ser1-15/+34
The current logic checks whether the cursor plane blending properties match the primary plane's. However that's wrong, because the cursor is painted on all planes underneath. If the cursor is over the primary plane and the overlay plane, it's painted on both pipes. Iterate over the CRTC planes and check their scaling match the cursor's. v4: fix typo in commit message (Harry) Signed-off-by: Simon Ser <[email protected]> Cc: Alex Deucher <[email protected]> Cc: Harry Wentland <[email protected]> Cc: Nicholas Kazlauskas <[email protected]> Cc: Bas Nieuwenhuizen <[email protected]> Cc: Rodrigo Siqueira <[email protected]> Cc: Sean Paul <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13amd/amdkfd: remove svms declaration to avoid werrorAlex Sierra1-6/+5
svm_range_list svms declaration removed to avoid werror when CONFIG_HSA_AMD_SVM is not enabled. Signed-off-by: Alex Sierra <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13drm/amdkfd: fix KFDSVMRangeTest.PartialUnmapSysMemTest failsYifan Zhang1-0/+5
[ RUN ] KFDSVMRangeTest.PartialUnmapSysMemTest /home/yifan/brahma/libhsakmt/tests/kfdtest/src/KFDTestUtil.cpp:245: Failure Value of: (hsaKmtAllocMemory(m_Node, m_Size, m_Flags, &m_pBuf)) Actual: 1 Expected: HSAKMT_STATUS_SUCCESS Which is: 0 /home/yifan/brahma/libhsakmt/tests/kfdtest/src/KFDTestUtil.cpp:248: Failure Value of: (hsaKmtMapMemoryToGPUNodes(m_pBuf, m_Size, __null, mapFlags, 1, &m_Node)) Actual: 1 Expected: HSAKMT_STATUS_SUCCESS Which is: 0 /home/yifan/brahma/libhsakmt/tests/kfdtest/src/KFDTestUtil.cpp:306: Failure Expected: ((void *)__null) != (ptr), actual: NULL vs NULL Segmentation fault (core dumped) [ ] Profile: Full Test [ ] HW capabilities: 0x9 kernel log: [ 102.029150] ret_from_fork+0x22/0x30 [ 102.029158] ---[ end trace 15c34e782714f9a3 ]--- [ 3613.603598] amdgpu: Address: 0x7f7149ccc000 already allocated by SVM [ 3613.610620] show_signal_msg: 27 callbacks suppressed These is race with deferred actions from previous memory map changes (e.g. munmap).Flush pending deffered work to avoid such case. Signed-off-by: Yifan Zhang <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13drm/amdkfd: export svm_range_list_lock_and_flush_workYifan Zhang2-1/+2
export svm_range_list_lock_and_flush_work to make other kfd parts be able to sync svm_range_list. Signed-off-by: Yifan Zhang <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13drm/amdkfd: avoid conflicting address mappingsAlex Sierra2-17/+75
[Why] Avoid conflict with address ranges mapped by SVM mechanism that try to be allocated again through ioctl_alloc in the same process. And viceversa. [How] For ioctl_alloc_memory_of_gpu allocations Check if the address range passed into ioctl memory alloc does not exist already in the kfd_process svms->objects interval tree. For SVM allocations Look for the address range into the interval tree VA from the VM inside of each pdds used in a kfd_process. Signed-off-by: Alex Sierra <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-13drm/amdgpu/nbio2.3: don't use GPU_HDP_FLUSH bit 12Alex Deucher3-1/+36
It's used internally by firmware. Using it in the driver could conflict with firmware. v2: squash in fix for navi1x (Alex) Reviewed-by: James Zhu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-12drm/i915: Stop using I915_TILING_* in client blit selftestMatt Roper1-11/+18
The I915_TILING_* definitions in the uapi header are intended solely for tiling modes that are visible to the old de-tiling fence ioctls. Since modern hardware does not support de-tiling fences, we should not add new definitions for new tiling types going forward. However we do want the client blit selftest to eventually cover other new tiling modes (such as Tile4), so switch it to using its own enum of tiling modes. Cc: Ville Syrjälä <[email protected]> Cc: Stanislav Lisovskiy <[email protected]> Signed-off-by: Matt Roper <[email protected]> Reviewed-by: Ville Syrjälä <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Powered by cgit, Gruvboxed by Blaster4385.