Thank you for your contribution! Sashiko AI review found 2 potential issues to consider: - [High] drm/msm/dpu: KMS init failure on SC8180X in _dpu_kms_initialize_displayport() - [Medium] drm/msm/dp: devres leak of msm_dp_panel in msm_dp_mst_attach_encoder() --
commit 866340b38e0036903bdb9a49a2dd242e4d969ea0 Author: Yongxing Mou <[email protected]> drm/msm/dpu: add per-stream MST encoders This commit initializes one DRM encoder per DisplayPort MST stream with MST-specific helper callbacks and attaches each encoder and stream panel to the DP MST state. > diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c > b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c > index 3c3203ca32c76..92691d39dcb24 100644 > --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c > +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c > @@ -688,6 +688,21 @@ static int _dpu_kms_initialize_displayport(struct > drm_device *dev, > return rc; > } > } > + > + for (int stream_id = 0; stream_cnt > 1 && stream_id < > stream_cnt; stream_id++) { > + info.stream_id = stream_id; > + encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_DPMST, > &info); > + if (IS_ERR(encoder)) { > + DPU_ERROR("encoder init failed for dp mst > display\n"); > + return PTR_ERR(encoder); > + } > + > + rc = msm_dp_mst_attach_encoder(priv->kms->dp[i], > stream_id, encoder); > + if (rc) { > + DPU_ERROR("dp_mst attach_encoder failed, rc = > %d\n", rc); > + return rc; > + } > + } > } > > return 0; [Severity: High] Does this break DPU KMS initialization on SC8180X? On SC8180X, arch/arm64/boot/dts/qcom/sc8180x.dtsi defines both stream_pixel and stream_1_pixel on mdss_dp0 and mdss_dp1, so msm_dp_get_mst_max_stream() returns stream_cnt = 2. However, sc8180x_intf[] only defines one INTF_DP entry for MSM_DP_CONTROLLER_0 and one for MSM_DP_CONTROLLER_1, while the shared MST interface has a dummy controller_id: drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h: /* INTF_3 is for MST, wired to INTF_DP 0 and 1, use dummy index until this is supported */ { .name = "intf_3", .id = INTF_3, .base = 0x6b800, .len = 0x280, .type = INTF_DP, .controller_id = 999, When stream_id == 1, dpu_encoder_init() -> dpu_encoder_setup_display() -> dpu_encoder_get_intf() cannot find a second matching INTF_DP entry and returns NULL, causing dpu_encoder_init() to return ERR_PTR(-ENOMEM) and aborting _dpu_kms_initialize_displayport() and all display initialization. > diff --git a/drivers/gpu/drm/msm/dp/dp_mst_drm.c > b/drivers/gpu/drm/msm/dp/dp_mst_drm.c > index a2900ff3b0ab3..ef5fb70863b81 100644 > --- a/drivers/gpu/drm/msm/dp/dp_mst_drm.c > +++ b/drivers/gpu/drm/msm/dp/dp_mst_drm.c > @@ -147,6 +147,26 @@ static int msm_dp_mst_encoder_stream_id(struct > msm_dp_mst *mst, > return -1; > } > > +int msm_dp_mst_attach_encoder(struct msm_dp *dp_display, unsigned int > stream_id, > + struct drm_encoder *encoder) > +{ > + struct msm_dp_mst *mst = dp_display->msm_dp_mst; > + struct msm_dp_panel *dp_panel; > + > + dp_panel = msm_dp_display_get_panel(dp_display, stream_id); > + if (!dp_panel) { > + drm_err(dp_display->drm_dev, > + "[MST] failed to allocate panel for stream %d\n", > stream_id); > + return -ENOMEM; > + } > + > + mst->mst_encoders[stream_id].enc = encoder; > + mst->mst_encoders[stream_id].stream_id = stream_id; > + mst->mst_encoders[stream_id].dp_panel = dp_panel; > + > + return 0; > +} [Severity: Medium] Does calling msm_dp_display_get_panel() here leak struct msm_dp_panel_private across DRM component master unbind/rebind cycles or KMS init failure? msm_dp_mst_attach_encoder() runs during DRM master bind in _dpu_kms_initialize_displayport(), but msm_dp_display_get_panel() -> msm_dp_panel_get() allocates the panel with devm_kzalloc() on the persistent subordinate DP platform device (&dp->msm_dp_display.pdev->dev). When the DRM master unbinds or msm_drm_init() fails later, dpu_kms_destroy() -> msm_dp_mst_unregister() -> msm_dp_mst_mgr_destroy() frees mst and drops mst->mst_encoders[stream_id].dp_panel, leaving the devres allocation behind on the DP device and allocating a new panel on the next bind. -- Sashiko AI review ยท https://sashiko.dev/#/patchset/[email protected]?part=16
