aboutsummaryrefslogtreecommitdiff
path: root/drivers/gpu
AgeCommit message (Collapse)AuthorFilesLines
2023-08-10drm/panfrost: Sync IRQ by job's timeout handlerDmitry Osipenko1-0/+16
Panfrost IRQ handler may stuck for a long time, for example this happens when there is a bad HDMI connection and HDMI handler takes a long time to finish processing, holding Panfrost. Make Panfrost's job timeout handler to sync IRQ before checking fence signal status in order to prevent spurious job timeouts due to a slow IRQ processing. Reviewed-by: Steven Price <[email protected]> Reviewed-by: Boris Brezillon <[email protected]> Reviewed-by: AngeloGioacchino Del Regno <[email protected]> Tested-by: AngeloGioacchino Del Regno <[email protected]> # MediaTek MT8192 and MT8195 Chromebooks Signed-off-by: Dmitry Osipenko <[email protected]> Signed-off-by: Boris Brezillon <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2023-08-10drm/exec: add test case for using a drm_exec multiple timesChristian König1-0/+21
Not really a common use case, but let's make sure that we don't accidentially break that somehow. CC: Boris Brezillon <[email protected]> Signed-off-by: Christian König <[email protected]> Reviewed-by: Boris Brezillon <[email protected]> Signed-off-by: Boris Brezillon <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2023-08-10drm/shmem-helper: Reset vma->vm_ops before calling dma_buf_mmap()Boris Brezillon1-0/+6
The dma-buf backend is supposed to provide its own vm_ops, but some implementation just have nothing special to do and leave vm_ops untouched, probably expecting this field to be zero initialized (this is the case with the system_heap implementation for instance). Let's reset vma->vm_ops to NULL to keep things working with these implementations. Fixes: 26d3ac3cb04d ("drm/shmem-helpers: Redirect mmap for imported dma-buf") Cc: <[email protected]> Cc: Daniel Vetter <[email protected]> Reported-by: Roman Stratiienko <[email protected]> Signed-off-by: Boris Brezillon <[email protected]> Tested-by: Roman Stratiienko <[email protected]> Reviewed-by: Thomas Zimmermann <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2023-08-09drm/mediatek: dp: Don't register HPD interrupt handler for eDP caseAngeloGioacchino Del Regno1-27/+46
The interrupt handler for HPD is useful only if a display is actually supposed to be hotpluggable, as that manages the machinery to perform cable (un)plug detection, debouncing and setup for re-training. Since eDP panels are not supposed to be hotpluggable we can avoid using the HPD interrupts altogether and rely on HPD polling only for the suspend/resume case, saving us some spinlocking action and the overhead of interrupts firing at every suspend/resume cycle, achieving a faster (even if just slightly) display resume. Signed-off-by: AngeloGioacchino Del Regno <[email protected]> Tested-by: Chen-Yu Tsai <[email protected]> Reviewed-by: Alexandre Mergnat <[email protected]> Reviewed-by: CK Hu <[email protected]> Link: https://patchwork.kernel.org/project/dri-devel/patch/[email protected]/ Signed-off-by: Chun-Kuang Hu <[email protected]>
2023-08-09drm/mediatek: dp: Add .wait_hpd_asserted() for AUX busAngeloGioacchino Del Regno1-7/+38
In order to support usecases in which the panel regulator can be switched on and off to save power, and usecases in which the panel regulator is off at boot, add a .wait_hpd_asserted() callback for the AUX bus: this will make sure to wait until the panel is fully ready after power-on before trying to communicate with it. Also, parse the eDP display capabilities in that callback, so that we can also avoid using the .get_edid() callback from this bridge. Since at this point the hpd machinery is performed in the new hpd callback and the detection and edid reading are done outside of this driver, assign the DRM_BRIDGE_OP_{DETECT, EDID, HPD} ops and register the bridge unconditionally at probe time only if we are probing full DisplayPort and not eDP while, for the latter, we register the bridge in the .done_probing() callback and only if the panel was found and triggered HPD. Signed-off-by: AngeloGioacchino Del Regno <[email protected]> Tested-by: Chen-Yu Tsai <[email protected]> Reviewed-by: Alexandre Mergnat <[email protected]> Reviewed-by: CK Hu <[email protected]> Link: https://patchwork.kernel.org/project/dri-devel/patch/[email protected]/ Signed-off-by: Chun-Kuang Hu <[email protected]>
2023-08-09drm/mediatek: dp: Add support for embedded DisplayPort aux-busAngeloGioacchino Del Regno2-14/+79
For the eDP case we can support using aux-bus on MediaTek DP: this gives us the possibility to declare our panel as generic "panel-edp" which will automatically configure the timings and available modes via the EDID that we read from it. To do this, move the panel parsing at the end of the probe function so that the hardware is initialized beforehand and also initialize the DPTX AUX block and power both on as, when we populate the aux-bus, the panel driver will trigger an EDID read to perform panel detection. Last but not least, since now the AUX transfers can happen in the separated aux-bus, it was necessary to add an exclusion for the cable_plugged_in check in `mtk_dp_aux_transfer()` and the easiest way to do this is to simply ignore checking that when the bridge type is eDP. Signed-off-by: AngeloGioacchino Del Regno <[email protected]> Tested-by: Chen-Yu Tsai <[email protected]> Reviewed-by: Alexandre Mergnat <[email protected]> Reviewed-by: CK Hu <[email protected]> Link: https://patchwork.kernel.org/project/dri-devel/patch/[email protected]/ Signed-off-by: Chun-Kuang Hu <[email protected]>
2023-08-09drm/mediatek: dp: Move PHY registration to new functionAngeloGioacchino Del Regno1-17/+26
In preparation for adding support for eDP, move the PHY registration code to a new mtk_dp_register_phy() function for better readability. This commit brings no functional changes. Signed-off-by: AngeloGioacchino Del Regno <[email protected]> Tested-by: Chen-Yu Tsai <[email protected]> Reviewed-by: Alexandre Mergnat <[email protected]> Reviewed-by: CK Hu <[email protected]> Link: https://patchwork.kernel.org/project/dri-devel/patch/[email protected]/ Signed-off-by: Chun-Kuang Hu <[email protected]>
2023-08-09drm/mediatek: dp: Avoid mutex locks if audio is not supported/enabledAngeloGioacchino Del Regno1-2/+5
If a controller (usually, eDP!) does not support audio, or audio is not enabled because the endpoint has no audio support, it's useless to lock a mutex only to unlock it right after because there's no .plugged_cb(). Check if the audio is supported and enabled before locking the mutex in mtk_dp_update_plugged_status(): if not, we simply return immediately. While at it, since the update_plugged_status_lock mutex would not be used if the controller doesn't support audio at all, initialize it only if `audio_supported` is true. Signed-off-by: AngeloGioacchino Del Regno <[email protected]> Tested-by: Chen-Yu Tsai <[email protected]> Reviewed-by: Alexandre Mergnat <[email protected]> Reviewed-by: CK Hu <[email protected]> Link: https://patchwork.kernel.org/project/dri-devel/patch/[email protected]/ Signed-off-by: Chun-Kuang Hu <[email protected]>
2023-08-09drm/mediatek: dp: Enable event interrupt only when bridge attachedAngeloGioacchino Del Regno1-5/+10
It is useless and error-prone to enable the DisplayPort event interrupt before finishing to probe and install the driver, as the DP training cannot happen before the entire pipeline is correctly set up, as the interrupt handler also requires the full hardware to be initialized by mtk_dp_bridge_attach(). Anyway, depending in which state the controller is left from the bootloader, this may cause an interrupt storm and consequently hang the kernel during boot, so, avoid enabling the interrupt until we reach a clean state by adding the IRQ_NOAUTOEN flag before requesting it at probe time and manage the enablement of the ISR in the .attach() and .detach() handlers for the DP bridge. Signed-off-by: AngeloGioacchino Del Regno <[email protected]> Tested-by: Chen-Yu Tsai <[email protected]> Reviewed-by: Alexandre Mergnat <[email protected]> Reviewed-by: CK Hu <[email protected]> Link: https://patchwork.kernel.org/project/dri-devel/patch/[email protected]/ Signed-off-by: Chun-Kuang Hu <[email protected]>
2023-08-09drm/mediatek: dp: Move AUX_P0 setting to mtk_dp_initialize_aux_settings()AngeloGioacchino Del Regno1-4/+5
Move the register write to MTK_DP_AUX_P0_3690 to set the AUX reply mode to function mtk_dp_initialize_aux_settings(), as this is effectively part of the DPTX AUX setup sequence. Signed-off-by: AngeloGioacchino Del Regno <[email protected]> Tested-by: Chen-Yu Tsai <[email protected]> Reviewed-by: CK Hu <[email protected]> Reviewed-by: Alexandre Mergnat <[email protected]> Link: https://patchwork.kernel.org/project/dri-devel/patch/[email protected]/ Signed-off-by: Chun-Kuang Hu <[email protected]>
2023-08-09drm/mediatek: dp: Use devm variant of drm_bridge_add()AngeloGioacchino Del Regno1-2/+3
In preparation for adding support for aux-bus, which will add a code path that may fail after the drm_bridge_add() call, change that to devm_drm_bridge_add() to simplify failure paths later. Signed-off-by: AngeloGioacchino Del Regno <[email protected]> Tested-by: Chen-Yu Tsai <[email protected]> Reviewed-by: CK Hu <[email protected]> Reviewed-by: Alexandre Mergnat <[email protected]> Link: https://patchwork.kernel.org/project/dri-devel/patch/[email protected]/ Signed-off-by: Chun-Kuang Hu <[email protected]>
2023-08-09drm/mediatek: dp: Change logging to dev for mtk_dp_aux_transfer()AngeloGioacchino Del Regno1-3/+3
Change logging from drm_{err,info}() to dev_{err,info}() in functions mtk_dp_aux_transfer() and mtk_dp_aux_do_transfer(): this will be essential to avoid getting NULL pointer kernel panics if any kind of error happens during AUX transfers happening before the bridge is attached. This may potentially start happening in a later commit implementing aux-bus support, as AUX transfers will be triggered from the panel driver (for EDID) before the mtk-dp bridge gets attached, and it's done in preparation for the same. Signed-off-by: AngeloGioacchino Del Regno <[email protected]> Tested-by: Chen-Yu Tsai <[email protected]> Reviewed-by: CK Hu <[email protected]> Reviewed-by: Alexandre Mergnat <[email protected]> Link: https://patchwork.kernel.org/project/dri-devel/patch/[email protected]/ Signed-off-by: Chun-Kuang Hu <[email protected]>
2023-08-09drm/mediatek: dp: Move AUX and panel poweron/off sequence to functionAngeloGioacchino Del Regno1-46/+30
Everytime we run bridge detection and/or EDID read we run a poweron and poweroff sequence for both the AUX and the panel; moreover, this is also done when enabling the bridge in the .atomic_enable() callback. Move this power on/off sequence to a new mtk_dp_aux_panel_poweron() function as to commonize it. Note that, before this commit, in mtk_dp_bridge_atomic_enable() only the AUX was getting powered on but the panel was left powered off if the DP cable wasn't plugged in while now we unconditionally send a D0 request and this is done for two reasons: - First, whether this request fails or not, it takes the same time and anyway the DP hardware won't produce any error (or, if it does, it's ignorable because it won't block further commands) - Second, training the link between a sleeping/standby/unpowered display makes little sense. Signed-off-by: AngeloGioacchino Del Regno <[email protected]> Tested-by: Chen-Yu Tsai <[email protected]> Reviewed-by: CK Hu <[email protected]> Reviewed-by: Alexandre Mergnat <[email protected]> Link: https://patchwork.kernel.org/project/dri-devel/patch/[email protected]/ Signed-off-by: Chun-Kuang Hu <[email protected]>
2023-08-09drm/mediatek: dp: Add missing error checks in mtk_dp_parse_capabilitiesAngeloGioacchino Del Regno1-5/+10
If reading the RX capabilities fails the training pattern will be set wrongly: add error checking for drm_dp_read_dpcd_caps() and return if anything went wrong with it. While at it, also add a less critical error check when writing to clear the ESI0 IRQ vector. Fixes: f70ac097a2cf ("drm/mediatek: Add MT8195 Embedded DisplayPort driver") Signed-off-by: AngeloGioacchino Del Regno <[email protected]> Tested-by: Chen-Yu Tsai <[email protected]> Reviewed-by: Alexandre Mergnat <[email protected]> Reviewed-by: CK Hu <[email protected]> Link: https://patchwork.kernel.org/project/dri-devel/patch/[email protected]/ Signed-off-by: Chun-Kuang Hu <[email protected]>
2023-08-09drm/mediatek: mtk_dpi: Compress struct of_device_id entriesAngeloGioacchino Del Regno1-22/+8
Reduce line count by compressing the entries of struct of_device_id; while at it, also add the usual /* sentinel */ comment to the last entry. This commit brings no functional changes. Signed-off-by: AngeloGioacchino Del Regno <[email protected]> Reviewed-by: Fei Shao <[email protected]> Reviewed-by: CK Hu <[email protected]> Link: https://patchwork.kernel.org/project/dri-devel/patch/[email protected]/ Signed-off-by: Chun-Kuang Hu <[email protected]>
2023-08-09drm/mediatek: mtk_dpi: Use devm_platform_ioremap_resource()AngeloGioacchino Del Regno1-3/+1
Instead of the open-coded platform_get_resource, devm_ioremap_resource switch to devm_platform_ioremap_resource(), also dropping the useless struct resource pointer, which becomes unused. Signed-off-by: AngeloGioacchino Del Regno <[email protected]> Reviewed-by: Fei Shao <[email protected]> Reviewed-by: CK Hu <[email protected]> Link: https://patchwork.kernel.org/project/dri-devel/patch/[email protected]/ Signed-off-by: Chun-Kuang Hu <[email protected]>
2023-08-09drm/mediatek: mtk_dpi: Switch to .remove_new() void callbackAngeloGioacchino Del Regno1-4/+2
The .remove() callback cannot fail: switch to .remove_new() and change mtk_dpi_remove() to void. Signed-off-by: AngeloGioacchino Del Regno <[email protected]> Reviewed-by: Fei Shao <[email protected]> Reviewed-by: CK Hu <[email protected]> Link: https://patchwork.kernel.org/project/dri-devel/patch/[email protected]/ Signed-off-by: Chun-Kuang Hu <[email protected]>
2023-08-09drm/mediatek: mtk_dpi: Switch to devm_drm_of_get_bridge()AngeloGioacchino Del Regno1-4/+4
Function drm_of_find_panel_or_bridge() is marked as deprecated: since the usage of that in this driver exactly corresponds to the new function devm_drm_of_get_bridge(), switch to it. Signed-off-by: AngeloGioacchino Del Regno <[email protected]> Reviewed-by: Fei Shao <[email protected]> Reviewed-by: CK Hu <[email protected]> Link: https://patchwork.kernel.org/project/dri-devel/patch/[email protected]/ Signed-off-by: Chun-Kuang Hu <[email protected]>
2023-08-09drm/mediatek: mtk_dpi: Simplify with dev_err_probe()AngeloGioacchino Del Regno1-30/+14
Use dev_err_probe() across the entire probe function of this driver to shrink the size. Signed-off-by: AngeloGioacchino Del Regno <[email protected]> Reviewed-by: Fei Shao <[email protected]> Reviewed-by: CK Hu <[email protected]> Link: https://patchwork.kernel.org/project/dri-devel/patch/[email protected]/ Signed-off-by: Chun-Kuang Hu <[email protected]>
2023-08-09drm/mediatek: mtk_dpi: Simplify with devm_drm_bridge_add()AngeloGioacchino Del Regno1-5/+3
Change drm_bridge_add() to its devm variant to slightly simplify the probe function. Signed-off-by: AngeloGioacchino Del Regno <[email protected]> Reviewed-by: Fei Shao <[email protected]> Reviewed-by: CK Hu <[email protected]> Link: https://patchwork.kernel.org/project/dri-devel/patch/[email protected]/ Signed-off-by: Chun-Kuang Hu <[email protected]>
2023-08-09drm/amdgpu: Use local64_try_cmpxchg in amdgpu_perf_readUros Bizjak1-3/+2
Use local64_try_cmpxchg instead of local64_cmpxchg (*ptr, old, new) == old in amdgpu_perf_read. x86 CMPXCHG instruction returns success in ZF flag, so this change saves a compare after cmpxchg (and related move instruction in front of cmpxchg). Also, try_cmpxchg implicitly assigns old *ptr value to "old" when cmpxchg fails. There is no need to re-read the value in the loop. No functional change intended. Cc: Alex Deucher <[email protected]> Cc: "Christian König" <[email protected]> Cc: "Pan, Xinhui" <[email protected]> Cc: David Airlie <[email protected]> Cc: Daniel Vetter <[email protected]> Signed-off-by: Uros Bizjak <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/ttm/tests: Add tests for ttm_poolKarolina Stolarek4-0/+459
Add KUnit tests that exercise page allocation using page pools and freeing pages, either by returning them to the pool or freeing them. Add a basic test for ttm_pool cleanup. Introduce helpers to create a dummy ttm_buffer_object. Signed-off-by: Karolina Stolarek <[email protected]> Reviewed-by: Christian König <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/427ea373357d0b6cb376c9d7ebc33c930bf1d28a.1691487006.git.karolina.stolarek@intel.com Signed-off-by: Christian König <[email protected]>
2023-08-09drm/ttm/tests: Add tests for ttm_deviceKarolina Stolarek1-0/+158
Test initialization and cleanup of the ttm_device struct, including some error paths. Verify the creation of page pools if use_dma_alloc param is true. Signed-off-by: Karolina Stolarek <[email protected]> Reviewed-by: Christian König <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/2abb0e53e7d56b0a24d0255f9075e2123b991278.1691487006.git.karolina.stolarek@intel.com Signed-off-by: Christian König <[email protected]>
2023-08-09drm/ttm: Introduce KUnit testKarolina Stolarek7-0/+212
Add the initial version of unit tests for ttm_device struct, together with helper functions. Signed-off-by: Karolina Stolarek <[email protected]> Reviewed-by: Christian König <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/3d1cc45c8a0cf536b92a850e0025f6c555de0169.1691487006.git.karolina.stolarek@intel.com Signed-off-by: Christian König <[email protected]>
2023-08-09drm/amdkfd: disable IOMMUv2 support for RavenAlex Deucher1-7/+0
Use the dGPU path instead. There were a lot of platform issues with IOMMU in general on these chips due to windows not enabling IOMMU at the time. The dGPU path has been used for a long time with newer APUs and works fine. This also paves the way to simplify the driver significantly. Reviewed-by: Felix Kuehling <[email protected]> Acked-by: Christian König <[email protected]> Tested-by: Mike Lothian <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdkfd: disable IOMMUv2 support for KV/CZAlex Deucher2-13/+1
Use the dGPU path instead. There were a lot of platform issues with IOMMU in general on these chips due to windows not enabling IOMMU at the time. The dGPU path has been used for a long time with newer APUs and works fine. This also paves the way to simplify the driver significantly. v2: use the dGPU queue manager functions Reviewed-by: Felix Kuehling <[email protected]> Acked-by: Christian König <[email protected]> Tested-by: Mike Lothian <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdkfd: ignore crat by defaultAlex Deucher1-4/+0
We are dropping the IOMMUv2 path, so no need to enable this. It's often buggy on consumer platforms anyway. Reviewed-by: Felix Kuehling <[email protected]> Acked-by: Christian König <[email protected]> Tested-by: Mike Lothian <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdgpu/gfx11: only enable CP GFX shadowing on SR-IOVAlex Deucher1-2/+6
This is only required for SR-IOV world switches, but it adds additional latency leading to reduced performance in some benchmarks. Disable for now on bare metal. Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amd/display: Fix a regression on Polaris cardsMario Limonciello1-1/+1
DCE products don't define a `remove_stream_from_ctx` like DCN ones do. This means that when compute_mst_dsc_configs_for_state() is called it always returns -EINVAL which causes MST to fail to setup. Cc: [email protected] # 6.4.y Cc: Harry Wentland <[email protected]> Reported-by: [email protected] Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/2671 Fixes: efa4c4df864e ("drm/amd/display: call remove_stream_from_ctx from res_pool funcs") Signed-off-by: Mario Limonciello <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdgpu: fix possible UAF in amdgpu_cs_pass1()Alex Deucher1-1/+1
Since the gang_size check is outside of chunk parsing loop, we need to reset i before we free the chunk data. Suggested by Ye Zhang (@VAR10CK) of Baidu Security. Reviewed-by: Guchun Chen <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected]
2023-08-09drm/amd/pm: Fix SMU v13.0.6 energy reportingLijo Lazar1-3/+2
Energy counter should be reported in units of 15.259 uJ. Don't apply any conversion. Signed-off-by: Lijo Lazar <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amd/display: check attr flag before set cursor degamma on DCN3+Melissa Wen1-2/+5
Don't set predefined degamma curve to cursor plane if the cursor attribute flag is not set. Applying a degamma curve to the cursor by default breaks userspace expectation. Checking the flag before performing any color transformation prevents too dark cursor gamma in DCN3+ on many Linux desktop environment (KDE Plasma, GNOME, wlroots-based, etc.) as reported at: - https://gitlab.freedesktop.org/drm/amd/-/issues/1513 This is the same approach followed by DCN2 drivers where the issue is not present. Fixes: 03f54d7d3448 ("drm/amd/display: Add DCN3 DPP") Link: https://gitlab.freedesktop.org/drm/amd/-/issues/1513 Signed-off-by: Melissa Wen <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Tested-by: Alex Hung <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected]
2023-08-09drm/amd/pm: disable the SMU13 OD feature support temporarilyEvan Quan2-6/+23
The existing OD interface cannot support the growing demand for more OD features. We are in the transition to a new OD mechanism. So, disable the SMU13 OD feature support temporarily. And this should be reverted when the new OD mechanism online. Signed-off-by: Evan Quan <[email protected]> Reviewed-by: Guchun Chen <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amd/pm: correct the pcie width for smu 13.0.0Kenneth Feng1-2/+1
correct the pcie width value in pp_dpm_pcie for smu 13.0.0 Signed-off-by: Kenneth Feng <[email protected]> Reviewed-by: Harish Kasiviswanathan <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected]
2023-08-09drm/amd/display: Don't show stack trace for missing eDPMario Limonciello1-1/+2
Some systems are only connected by HDMI or DP, so warning related to missing eDP is unnecessary. Downgrade to debug instead. Cc: Hamza Mahfooz <[email protected]> Fixes: 6d9b6dceaa51 ("drm/amd/display: only warn once in dce110_edp_wait_for_hpd_ready()") Reported-by: [email protected] Signed-off-by: Mario Limonciello <[email protected]> Reviewed-by: Hamza Mahfooz <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdgpu: Match against exact bootloader statusLijo Lazar1-7/+8
On PSP v13.x ASICs, boot loader will set only the MSB to 1 and clear the least significant bits for any command submission. Hence match against the exact register value, otherwise a register value of all 0xFFs also could falsely indicate that boot loader is ready. Also, from PSP v13.0.6 and newer, bits[7:0] will be used to indicate command error status. Signed-off-by: Lijo Lazar <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amd/pm: skip the RLC stop when S0i3 suspend for SMU v13.0.4/11Tim Huang1-2/+2
For SMU v13.0.4/11, driver does not need to stop RLC for S0i3, the firmwares will handle that properly. Signed-off-by: Tim Huang <[email protected]> Reviewed-by: Mario Limonciello <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amd: Disable S/G for APUs when 64GB or more host memoryMario Limonciello3-3/+29
Users report a white flickering screen on multiple systems that is tied to having 64GB or more memory. When S/G is enabled pages will get pinned to both VRAM carve out and system RAM leading to this. Until it can be fixed properly, disable S/G when 64GB of memory or more is detected. This will force pages to be pinned into VRAM. This should fix white screen flickers but if VRAM pressure is encountered may lead to black screens. It's a trade-off for now. Fixes: 81d0bcf99009 ("drm/amdgpu: make display pinning more flexible (v2)") Cc: Hamza Mahfooz <[email protected]> Cc: Roman Li <[email protected]> Cc: <[email protected]> # 6.1.y: bf0207e172703 ("drm/amdgpu: add S/G display parameter") Cc: <[email protected]> # 6.4.y Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2735 Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2354 Signed-off-by: Mario Limonciello <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdgpu: Add pci usage to nbio v7.9Asad Kamal3-1/+72
Add implementation to get pcie usage for nbio v7.9. Signed-off-by: Asad Kamal <[email protected]> Reviewed-by: Lijo Lazar <[email protected]> Reviewed-by: Shiwu Zhang <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdgpu: Add pcie usage callback to nbioAsad Kamal2-0/+11
Add a callback in nbio to get pcie usage Signed-off-by: Asad Kamal <[email protected]> Reviewed-by: Lijo Lazar <[email protected]> Reviewed-by: Shiwu Zhang <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdgpu: Extend poison mode check to SDMA/VCN/JPEGCandice Li1-1/+4
Treat SDMA/VCN/JPEG as RAS capable IP blocks in poison mode. Signed-off-by: Candice Li <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/buddy: Fix drm buddy info output formatMa Jun1-4/+4
[1] Change pages to blocks to avoid confusion. [2] Fix output format to align the output info. Signed-off-by: Ma Jun <[email protected]> Reviewed-by: Arunpravin Paneer Selvam <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdgpu/irq: Move irq resume to the beginningEmily Deng1-1/+1
Need to move irq resume to the beginning of reset sriov, or if one interrupt occurs before irq resume, then the irq won't work anymore. Signed-off-by: Emily Deng <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdgpu: add RAS fatal error handler for NBIO v7.9Tao Zhou3-0/+193
Register RAS fatal error interrupt and add handler. v2: only register NBIO RAS for dGPU platform. change nbio_v7_9_set_ras_controller_irq_state and nbio_v7_9_set_ras_err_event_athub_irq_state to dummy functions. Signed-off-by: Tao Zhou <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdgpu: Fix identation issues in 'kgd_gfx_v9_program_trap_handler_settings'Srinivasan Shanmugam1-2/+2
Fixes the following: ERROR: code indent should use tabs where possible WARNING: please, no spaces at the start of a line Cc: Guchun Chen <[email protected]> Cc: Christian König <[email protected]> Cc: Alex Deucher <[email protected]> Cc: "Pan, Xinhui" <[email protected]> Signed-off-by: Srinivasan Shanmugam <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdgpu/gfx11: only enable CP GFX shadowing on SR-IOVAlex Deucher1-2/+6
This is only required for SR-IOV world switches, but it adds additional latency leading to reduced performance in some benchmarks. Disable for now on bare metal. Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amd/display: Get replay info from VSDBBhawanpreet Lakha3-0/+64
We need to make sure that the panel supports replay. This info is inside the amd vsdb (vendor specific data block). Create a function to parse the block and read the replay_mode bit. Signed-off-by: Bhawanpreet Lakha <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdgpu: Add FRU sysfs nodes only if neededLijo Lazar3-68/+94
Create sysfs nodes for FRU data only if FRU data is available. Move the logic to FRU specific file. Signed-off-by: Lijo Lazar <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amd/pm: Clean up errors in smu73_discrete.hRan Sun1-44/+29
Fix the following errors reported by checkpatch: ERROR: open brace '{' following struct go on the same line ERROR: trailing whitespace ERROR: space prohibited before open square bracket '[' Signed-off-by: Ran Sun <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdgpu: Clean up errors in vcn_v3_0.cRan Sun1-6/+5
Fix the following errors reported by checkpatch: ERROR: space required before the open brace '{' ERROR: "foo * bar" should be "foo *bar" ERROR: space required before the open parenthesis '(' ERROR: that open brace { should be on the previous line Signed-off-by: Ran Sun <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.