dri-devel January 2022

dri-devel@lists.freedesktop.org

373 participants
842 discussions

[PATCH v2] Remove extra device acquisition method of i2c client in lt9611 driver
by Zhiming Liu 17 Jan '22

17 Jan '22

Signed-off-by: Zhiming Liu <lzmlzmhh(a)gmail.com> bridge : drm : Remove extra device acquisition method of i2c client in lt9611 driver. We have get the device of i2c client in probe function.So we should remove extra device acquisition method of i2c client. --- drivers/gpu/drm/bridge/lontium-lt9611.c | 4 ++-- drivers/gpu/drm/bridge/lontium-lt9611uxc.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/bridge/lontium-lt9611.c b/drivers/gpu/drm/bridge/lontium-lt9611.c index dafb1b47c15f..feb128a4557d 100644 --- a/drivers/gpu/drm/bridge/lontium-lt9611.c +++ b/drivers/gpu/drm/bridge/lontium-lt9611.c @@ -1090,7 +1090,7 @@ static int lt9611_probe(struct i2c_client *client, if (!lt9611) return -ENOMEM; - lt9611->dev = &client->dev; + lt9611->dev = dev; lt9611->client = client; lt9611->sleep = false; @@ -1100,7 +1100,7 @@ static int lt9611_probe(struct i2c_client *client, return PTR_ERR(lt9611->regmap); } - ret = lt9611_parse_dt(&client->dev, lt9611); + ret = lt9611_parse_dt(dev, lt9611); if (ret) { dev_err(dev, "failed to parse device tree\n"); return ret; diff --git a/drivers/gpu/drm/bridge/lontium-lt9611uxc.c b/drivers/gpu/drm/bridge/lontium-lt9611uxc.c index 33f9716da0ee..3d62e6bf6892 100644 --- a/drivers/gpu/drm/bridge/lontium-lt9611uxc.c +++ b/drivers/gpu/drm/bridge/lontium-lt9611uxc.c @@ -860,7 +860,7 @@ static int lt9611uxc_probe(struct i2c_client *client, if (!lt9611uxc) return -ENOMEM; - lt9611uxc->dev = &client->dev; + lt9611uxc->dev = dev; lt9611uxc->client = client; mutex_init(&lt9611uxc->ocm_lock); @@ -870,7 +870,7 @@ static int lt9611uxc_probe(struct i2c_client *client, return PTR_ERR(lt9611uxc->regmap); } - ret = lt9611uxc_parse_dt(&client->dev, lt9611uxc); + ret = lt9611uxc_parse_dt(dev, lt9611uxc); if (ret) { dev_err(dev, "failed to parse device tree\n"); return ret; -- 2.25.1

2 2

[PATCH v5] drm/bridge: anx7625: Return -EPROBE_DEFER if the dsi host was not found
by owen 17 Jan '22

17 Jan '22

From: owen <qwt9588(a)gmail.com> It will connect to the mipi dsi host and find the corresponding mipi dsi host node, but the node registered by the mipi dsi host has not been loaded yet. of_find_mipi_dsi_host_by_node() returns -EINVAL which causes the calling driver to fail. If the anx7625 driver is loaded afterwards the driver requesting the mipi dsi host will not notice this. Better approach is to return -EPROBE_DEFER in such case. Then when the anx7625 driver appears the driver requesting the mipi dsi host will be probed again. Signed-off-by: owen <qwt9588(a)gmail.com> --- drivers/gpu/drm/bridge/analogix/anx7625.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/bridge/analogix/anx7625.c b/drivers/gpu/drm/bridge/analogix/anx7625.c index 2346dbcc505f..297bbeb5aae0 100644 --- a/drivers/gpu/drm/bridge/analogix/anx7625.c +++ b/drivers/gpu/drm/bridge/analogix/anx7625.c @@ -1660,7 +1660,7 @@ static int anx7625_attach_dsi(struct anx7625_data *ctx) host = of_find_mipi_dsi_host_by_node(ctx->pdata.mipi_host_node); if (!host) { DRM_DEV_ERROR(dev, "fail to find dsi host.\n"); - return -EINVAL; + return -EPROBE_DEFER; } dsi = devm_mipi_dsi_device_register_full(dev, host, &info); -- 2.31.0

2 1

[PATCH] drm/i915/buddy: fixup potential uaf
by Matthew Auld 17 Jan '22

17 Jan '22

If we are unlucky and can't allocate enough memory when splitting blocks, where we temporarily end up with the given block and its buddy on the respective free list, then we need to ensure we delete both blocks, and no just the buddy, before potentially freeing them. Fixes: 14d1b9a6247c ("drm/i915: buddy allocator") Signed-off-by: Matthew Auld <matthew.auld(a)intel.com> Cc: Arunpravin <Arunpravin.PaneerSelvam(a)amd.com> Cc: Christian König <christian.koenig(a)amd.com> --- drivers/gpu/drm/i915/i915_buddy.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_buddy.c b/drivers/gpu/drm/i915/i915_buddy.c index 6e2ad68f8f3f..9ca81b095adb 100644 --- a/drivers/gpu/drm/i915/i915_buddy.c +++ b/drivers/gpu/drm/i915/i915_buddy.c @@ -293,8 +293,10 @@ i915_buddy_alloc(struct i915_buddy_mm *mm, unsigned int order) return block; out_free: - if (i != order) + if (i != order) { + list_del(&block->link); __i915_buddy_free(mm, block); + } return ERR_PTR(err); } @@ -401,8 +403,10 @@ int i915_buddy_alloc_range(struct i915_buddy_mm *mm, buddy = get_buddy(block); if (buddy && (i915_buddy_block_is_free(block) && - i915_buddy_block_is_free(buddy))) + i915_buddy_block_is_free(buddy))) { + list_del(&block->link); __i915_buddy_free(mm, block); + } err_free: i915_buddy_free_list(mm, &allocated); -- 2.31.1

1 0

dw_hdmi is showing wrong colour after commit 7cd70656d1285b79("drm/bridge: display-connector: implement bus fmts callbacks")
by Biju Das 17 Jan '22

17 Jan '22

Hi All, RZ/G2{H, M, N} SoC has dw_hdmi IP and it was working ok(colour) till the commit 7cd70656d1285b79("drm/bridge: display-connector: implement bus fmts callbacks"). After this patch, the screen becomes greenish(may be it is setting it into YUV format??). By checking the code, previously it used to call get_input_fmt callback and set colour as RGB24. After this commit, it calls get_output_fmt_callbck and returns 3 outputformats(YUV16, YUV24 and RGB24) And get_input_fmt callback, I see the outputformat as YUV16 instead of RGB24. Not sure, I am the only one seeing this issue with dw_HDMI driver. Regards, Biju

4 15

[PATCH] drm/panel-orientation-quirks: add Lenovo Ideapad Duet 3i
by Hendrik Borghorst 17 Jan '22

17 Jan '22

The Lenovo Ideapad Duet 3i has a 1200x1920 panel. Non-exact match is used because a non-LTE and LTE version exists with string "10IGL5" and "10IGL5-LTE". The patch is only tested on the non-LTE version. Signed-off-by: Hendrik Borghorst <hendrikborghorst(a)gmail.com> --- drivers/gpu/drm/drm_panel_orientation_quirks.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/gpu/drm/drm_panel_orientation_quirks.c b/drivers/gpu/drm/drm_panel_orientation_quirks.c index 042bb80383c9..a69caf423b48 100644 --- a/drivers/gpu/drm/drm_panel_orientation_quirks.c +++ b/drivers/gpu/drm/drm_panel_orientation_quirks.c @@ -262,6 +262,13 @@ static const struct dmi_system_id orientation_data[] = { DMI_EXACT_MATCH(DMI_PRODUCT_VERSION, "Lenovo ideapad D330-10IGM"), }, .driver_data = (void *)&lcd1200x1920_rightside_up, + }, { /* Lenovo IdeaPad Duet 3 10IGL5 */ + .matches = { + DMI_EXACT_MATCH(DMI_SYS_VENDOR, "LENOVO"), + /* Non exact match to match all versions */ + DMI_MATCH(DMI_PRODUCT_VERSION, "IdeaPad Duet 3 10IGL5"), + }, + .driver_data = (void *)&lcd1200x1920_rightside_up, }, { /* Lenovo Yoga Book X90F / X91F / X91L */ .matches = { /* Non exact match to match all versions */ -- 2.34.1

1 0

[PATCH] drm/panel: Add panel-edp: add lcd innolux,n140hca-eac
by Rex Nie 17 Jan '22

17 Jan '22

innolux,n140hca-eac is a eDP-based LCD panel. This panel has 1920x1080 resolution in 14-inch TFT panel. Signed-off-by: Rex Nie <rexnie3(a)gmail.com> --- .../display/panel/innolux,n140hca-eac.yaml | 43 +++++++++++++++++++ drivers/gpu/drm/panel/panel-edp.c | 26 +++++++++++ 2 files changed, 69 insertions(+) create mode 100644 Documentation/devicetree/bindings/display/panel/innolux,n140hca-eac.yaml diff --git a/Documentation/devicetree/bindings/display/panel/innolux,n140hca-eac.yaml b/Documentation/devicetree/bindings/display/panel/innolux,n140hca-eac.yaml new file mode 100644 index 000000000000..5493e383c97c --- /dev/null +++ b/Documentation/devicetree/bindings/display/panel/innolux,n140hca-eac.yaml @@ -0,0 +1,43 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/display/panel/innolux,n140hca-eac.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Innolux N140HCA-EAC 14 inch eDP 1080p display panel + +maintainers: + - Sandeep Panda <spanda(a)codeaurora.org> + - Douglas Anderson <dianders(a)chromium.org> + +allOf: + - $ref: panel-common.yaml# + +properties: + compatible: + const: innolux,n140hca-eac + + enable-gpios: true + power-supply: true + backlight: true + no-hpd: true + +required: + - compatible + - power-supply + +additionalProperties: false + +examples: + - | + #include <dt-bindings/gpio/gpio.h> + + panel_edp: panel-edp { + compatible = "innolux,n140hca-eac"; + enable-gpios = <&pio 96 GPIO_ACTIVE_HIGH>; + power-supply = <&pp3300_disp_x>; + backlight = <&backlight_lcd0>; + no-hpd; + }; + +... diff --git a/drivers/gpu/drm/panel/panel-edp.c b/drivers/gpu/drm/panel/panel-edp.c index 176ef0c3cc1d..3682986608dc 100644 --- a/drivers/gpu/drm/panel/panel-edp.c +++ b/drivers/gpu/drm/panel/panel-edp.c @@ -1303,6 +1303,29 @@ static const struct panel_desc innolux_n125hce_gn1 = { }, }; +static const struct display_timing innolux_n140hca_eac_timing = { + .pixelclock = { 72600000, 76420000, 80240000 }, + .hactive = { 1920, 1920, 1920 }, + .hfront_porch = { 80, 80, 80 }, + .hback_porch = { 190, 190, 190 }, + .hsync_len = { 60, 60, 60 }, + .vactive = { 1080, 1080, 1080 }, + .vfront_porch = { 6, 6, 6 }, + .vback_porch = { 38, 38, 38 }, + .vsync_len = { 8, 8, 8 }, + .flags = DISPLAY_FLAGS_VSYNC_LOW | DISPLAY_FLAGS_HSYNC_LOW, +}; + +static const struct panel_desc innolux_n140hca_eac = { + .timings = &innolux_n140hca_eac_timing, + .num_timings = 1, + .bpc = 6, + .size = { + .width = 309, + .height = 174, + }, +}; + static const struct drm_display_mode innolux_p120zdg_bf1_mode = { .clock = 206016, .hdisplay = 2160, @@ -1682,6 +1705,9 @@ static const struct of_device_id platform_of_match[] = { }, { .compatible = "innolux,n125hce-gn1", .data = &innolux_n125hce_gn1, + }, { + .compatible = "innolux,n140hca-eac", + .data = &innolux_n140hca_eac, }, { .compatible = "innolux,p120zdg-bf1", .data = &innolux_p120zdg_bf1, -- 2.25.1

1 0

[PATCH] vmwgfx: remove redundant ret variable
by cgel.zte＠gmail.com 17 Jan '22

17 Jan '22

From: Minghao Chi <chi.minghao(a)zte.com.cn> Return value directly instead of taking this in another redundant variable. Reported-by: Zeal Robot <zealci(a)zte.com.cn> Signed-off-by: Minghao Chi <chi.minghao(a)zte.com.cn> Signed-off-by: CGEL ZTE <cgel.zte(a)gmail.com> --- drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c | 13 +++---------- 1 file changed, 3 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index 44ca23b0ea4e..be5b1a9b48e1 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -1263,7 +1263,6 @@ static int vmw_cmd_dx_define_query(struct vmw_private *dev_priv, VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXDefineQuery); struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); struct vmw_resource *cotable_res; - int ret; if (!ctx_node) return -EINVAL; @@ -1275,9 +1274,8 @@ static int vmw_cmd_dx_define_query(struct vmw_private *dev_priv, return -EINVAL; cotable_res = vmw_context_cotable(ctx_node->ctx, SVGA_COTABLE_DXQUERY); - ret = vmw_cotable_notify(cotable_res, cmd->body.queryId); - return ret; + return vmw_cotable_notify(cotable_res, cmd->body.queryId); } /** @@ -2576,7 +2574,6 @@ static int vmw_cmd_dx_so_define(struct vmw_private *dev_priv, uint32 defined_id; } *cmd; enum vmw_so_type so_type; - int ret; if (!ctx_node) return -EINVAL; @@ -2586,9 +2583,8 @@ static int vmw_cmd_dx_so_define(struct vmw_private *dev_priv, if (IS_ERR(res)) return PTR_ERR(res); cmd = container_of(header, typeof(*cmd), header); - ret = vmw_cotable_notify(res, cmd->defined_id); - return ret; + return vmw_cotable_notify(res, cmd->defined_id); } /** @@ -2726,15 +2722,12 @@ static int vmw_cmd_dx_destroy_shader(struct vmw_private *dev_priv, struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXDestroyShader) = container_of(header, typeof(*cmd), header); - int ret; if (!ctx_node) return -EINVAL; - ret = vmw_shader_remove(sw_context->man, cmd->body.shaderId, 0, + return vmw_shader_remove(sw_context->man, cmd->body.shaderId, 0, &sw_context->staged_cmd_res); - - return ret; } /** -- 2.25.1

1 0

[PATCH v3 1/1] drm/bridge: anx7625: send DPCD command to downstream
by Xin Ji 17 Jan '22

17 Jan '22

Send DPCD command to downstream before anx7625 power down, let downstream monitor enter into standby mode. Signed-off-by: Xin Ji <xji(a)analogixsemi.com> --- drivers/gpu/drm/bridge/analogix/anx7625.c | 42 +++++++++++++++++++---- 1 file changed, 35 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/bridge/analogix/anx7625.c b/drivers/gpu/drm/bridge/analogix/anx7625.c index 33383f83255d..0b858c78abe8 100644 --- a/drivers/gpu/drm/bridge/analogix/anx7625.c +++ b/drivers/gpu/drm/bridge/analogix/anx7625.c @@ -129,6 +129,23 @@ static int anx7625_reg_write(struct anx7625_data *ctx, return ret; } +static int anx7625_reg_block_write(struct anx7625_data *ctx, + struct i2c_client *client, + u8 reg_addr, u8 len, u8 *buf) +{ + int ret; + struct device *dev = &client->dev; + + i2c_access_workaround(ctx, client); + + ret = i2c_smbus_write_i2c_block_data(client, reg_addr, len, buf); + if (ret < 0) + dev_err(dev, "write i2c block failed id=%x\n:%x", + client->addr, reg_addr); + + return ret; +} + static int anx7625_write_or(struct anx7625_data *ctx, struct i2c_client *client, u8 offset, u8 mask) @@ -214,8 +231,8 @@ static int wait_aux_op_finish(struct anx7625_data *ctx) return 0; } -static int anx7625_aux_dpcd_read(struct anx7625_data *ctx, - u32 address, u8 len, u8 *buf) +static int anx7625_aux_dpcd_trans(struct anx7625_data *ctx, u8 op, + u32 address, u8 len, u8 *buf) { struct device *dev = &ctx->client->dev; int ret; @@ -231,8 +248,7 @@ static int anx7625_aux_dpcd_read(struct anx7625_data *ctx, addrm = (address >> 8) & 0xFF; addrh = (address >> 16) & 0xFF; - cmd = DPCD_CMD(len, DPCD_READ); - cmd = ((len - 1) << 4) | 0x09; + cmd = DPCD_CMD(len, op); /* Set command and length */ ret = anx7625_reg_write(ctx, ctx->i2c.rx_p0_client, @@ -246,6 +262,9 @@ static int anx7625_aux_dpcd_read(struct anx7625_data *ctx, ret |= anx7625_reg_write(ctx, ctx->i2c.rx_p0_client, AP_AUX_ADDR_19_16, addrh); + if (op == DPCD_WRITE) + ret |= anx7625_reg_block_write(ctx, ctx->i2c.rx_p0_client, + AP_AUX_BUFF_START, len, buf); /* Enable aux access */ ret |= anx7625_write_or(ctx, ctx->i2c.rx_p0_client, AP_AUX_CTRL_STATUS, AP_AUX_CTRL_OP_EN); @@ -255,14 +274,17 @@ static int anx7625_aux_dpcd_read(struct anx7625_data *ctx, return -EIO; } - usleep_range(2000, 2100); - ret = wait_aux_op_finish(ctx); if (ret) { dev_err(dev, "aux IO error: wait aux op finish.\n"); return ret; } + /* Write done */ + if (op == DPCD_WRITE) + return 0; + + /* Read done, read out dpcd data */ ret = anx7625_reg_block_read(ctx, ctx->i2c.rx_p0_client, AP_AUX_BUFF_START, len, buf); if (ret < 0) { @@ -845,7 +867,7 @@ static int anx7625_hdcp_enable(struct anx7625_data *ctx) } /* Read downstream capability */ - anx7625_aux_dpcd_read(ctx, 0x68028, 1, &bcap); + anx7625_aux_dpcd_trans(ctx, DPCD_READ, 0x68028, 1, &bcap); if (!(bcap & 0x01)) { pr_warn("downstream not support HDCP 1.4, cap(%x).\n", bcap); return 0; @@ -918,6 +940,7 @@ static void anx7625_dp_stop(struct anx7625_data *ctx) { struct device *dev = &ctx->client->dev; int ret; + u8 data; DRM_DEV_DEBUG_DRIVER(dev, "stop dp output\n"); @@ -929,6 +952,11 @@ static void anx7625_dp_stop(struct anx7625_data *ctx) ret |= anx7625_write_and(ctx, ctx->i2c.tx_p2_client, 0x08, 0x7f); ret |= anx7625_video_mute_control(ctx, 1); + + dev_dbg(dev, "notify downstream enter into standby\n"); + /* Downstream monitor enter into standby mode */ + data = 2; + ret |= anx7625_aux_dpcd_trans(ctx, DPCD_WRITE, 0x000600, 1, &data); if (ret < 0) DRM_DEV_ERROR(dev, "IO error : mute video fail\n"); -- 2.25.1

2 1

[PATCH] drm/bridge: sii902x: add support for DRM_BRIDGE_ATTACH_NO_CONNECTOR
by Neil Armstrong 17 Jan '22

17 Jan '22

This adds support for DRM_BRIDGE_ATTACH_NO_CONNECTOR by adding the bridge get_edid() and detect() callbacks after refactoring the connector get_modes() and connector_detect() callbacks. In order to keep the bridge working, extra code in get_modes() has been moved to more logical places. Signed-off-by: Neil Armstrong <narmstrong(a)baylibre.com> --- drivers/gpu/drm/bridge/sii902x.c | 129 ++++++++++++++++++++++++------- 1 file changed, 99 insertions(+), 30 deletions(-) diff --git a/drivers/gpu/drm/bridge/sii902x.c b/drivers/gpu/drm/bridge/sii902x.c index 89558e581530..65549fbfdc87 100644 --- a/drivers/gpu/drm/bridge/sii902x.c +++ b/drivers/gpu/drm/bridge/sii902x.c @@ -166,10 +166,12 @@ struct sii902x { struct i2c_client *i2c; struct regmap *regmap; struct drm_bridge bridge; + struct drm_bridge *next_bridge; struct drm_connector connector; struct gpio_desc *reset_gpio; struct i2c_mux_core *i2cmux; struct regulator_bulk_data supplies[2]; + bool sink_is_hdmi; /* * Mutex protects audio and video functions from interfering * each other, by keeping their i2c command sequences atomic. @@ -245,10 +247,8 @@ static void sii902x_reset(struct sii902x *sii902x) gpiod_set_value(sii902x->reset_gpio, 0); } -static enum drm_connector_status -sii902x_connector_detect(struct drm_connector *connector, bool force) +static enum drm_connector_status sii902x_detect(struct sii902x *sii902x) { - struct sii902x *sii902x = connector_to_sii902x(connector); unsigned int status; mutex_lock(&sii902x->mutex); @@ -261,6 +261,14 @@ sii902x_connector_detect(struct drm_connector *connector, bool force) connector_status_connected : connector_status_disconnected; } +static enum drm_connector_status +sii902x_connector_detect(struct drm_connector *connector, bool force) +{ + struct sii902x *sii902x = connector_to_sii902x(connector); + + return sii902x_detect(sii902x); +} + static const struct drm_connector_funcs sii902x_connector_funcs = { .detect = sii902x_connector_detect, .fill_modes = drm_helper_probe_single_connector_modes, @@ -270,42 +278,40 @@ static const struct drm_connector_funcs sii902x_connector_funcs = { .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, }; -static int sii902x_get_modes(struct drm_connector *connector) +static struct edid *sii902x_get_edid(struct sii902x *sii902x, + struct drm_connector *connector) { - struct sii902x *sii902x = connector_to_sii902x(connector); - u32 bus_format = MEDIA_BUS_FMT_RGB888_1X24; - u8 output_mode = SII902X_SYS_CTRL_OUTPUT_DVI; struct edid *edid; - int num = 0, ret; mutex_lock(&sii902x->mutex); edid = drm_get_edid(connector, sii902x->i2cmux->adapter[0]); - drm_connector_update_edid_property(connector, edid); if (edid) { if (drm_detect_hdmi_monitor(edid)) - output_mode = SII902X_SYS_CTRL_OUTPUT_HDMI; - - num = drm_add_edid_modes(connector, edid); - kfree(edid); + sii902x->sink_is_hdmi = true; + else + sii902x->sink_is_hdmi = false; } - ret = drm_display_info_set_bus_formats(&connector->display_info, - &bus_format, 1); - if (ret) - goto error_out; + mutex_unlock(&sii902x->mutex); - ret = regmap_update_bits(sii902x->regmap, SII902X_SYS_CTRL_DATA, - SII902X_SYS_CTRL_OUTPUT_MODE, output_mode); - if (ret) - goto error_out; + return edid; +} - ret = num; +static int sii902x_get_modes(struct drm_connector *connector) +{ + struct sii902x *sii902x = connector_to_sii902x(connector); + struct edid *edid; + int num = 0; -error_out: - mutex_unlock(&sii902x->mutex); + edid = sii902x_get_edid(sii902x, connector); + drm_connector_update_edid_property(connector, edid); + if (edid) { + num = drm_add_edid_modes(connector, edid); + kfree(edid); + } - return ret; + return num; } static enum drm_mode_status sii902x_mode_valid(struct drm_connector *connector, @@ -354,12 +360,16 @@ static void sii902x_bridge_mode_set(struct drm_bridge *bridge, const struct drm_display_mode *adj) { struct sii902x *sii902x = bridge_to_sii902x(bridge); + u8 output_mode = SII902X_SYS_CTRL_OUTPUT_DVI; struct regmap *regmap = sii902x->regmap; u8 buf[HDMI_INFOFRAME_SIZE(AVI)]; struct hdmi_avi_infoframe frame; u16 pixel_clock_10kHz = adj->clock / 10; int ret; + if (sii902x->sink_is_hdmi) + output_mode = SII902X_SYS_CTRL_OUTPUT_HDMI; + buf[0] = pixel_clock_10kHz & 0xff; buf[1] = pixel_clock_10kHz >> 8; buf[2] = drm_mode_vrefresh(adj); @@ -375,6 +385,11 @@ static void sii902x_bridge_mode_set(struct drm_bridge *bridge, mutex_lock(&sii902x->mutex); + ret = regmap_update_bits(sii902x->regmap, SII902X_SYS_CTRL_DATA, + SII902X_SYS_CTRL_OUTPUT_MODE, output_mode); + if (ret) + goto out; + ret = regmap_bulk_write(regmap, SII902X_TPI_VIDEO_DATA, buf, 10); if (ret) goto out; @@ -405,13 +420,13 @@ static int sii902x_bridge_attach(struct drm_bridge *bridge, enum drm_bridge_attach_flags flags) { struct sii902x *sii902x = bridge_to_sii902x(bridge); + u32 bus_format = MEDIA_BUS_FMT_RGB888_1X24; struct drm_device *drm = bridge->dev; int ret; - if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) { - DRM_ERROR("Fix bridge driver to make connector optional!"); - return -EINVAL; - } + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) + return drm_bridge_attach(bridge->encoder, sii902x->next_bridge, + bridge, flags); drm_connector_helper_add(&sii902x->connector, &sii902x_connector_helper_funcs); @@ -433,16 +448,38 @@ static int sii902x_bridge_attach(struct drm_bridge *bridge, else sii902x->connector.polled = DRM_CONNECTOR_POLL_CONNECT; + ret = drm_display_info_set_bus_formats(&sii902x->connector.display_info, + &bus_format, 1); + if (ret) + return ret; + drm_connector_attach_encoder(&sii902x->connector, bridge->encoder); return 0; } +static enum drm_connector_status sii902x_bridge_detect(struct drm_bridge *bridge) +{ + struct sii902x *sii902x = bridge_to_sii902x(bridge); + + return sii902x_detect(sii902x); +} + +static struct edid *sii902x_bridge_get_edid(struct drm_bridge *bridge, + struct drm_connector *connector) +{ + struct sii902x *sii902x = bridge_to_sii902x(bridge); + + return sii902x_get_edid(sii902x, connector); +} + static const struct drm_bridge_funcs sii902x_bridge_funcs = { .attach = sii902x_bridge_attach, .mode_set = sii902x_bridge_mode_set, .disable = sii902x_bridge_disable, .enable = sii902x_bridge_enable, + .detect = sii902x_bridge_detect, + .get_edid = sii902x_bridge_get_edid, }; static int sii902x_mute(struct sii902x *sii902x, bool mute) @@ -829,8 +866,12 @@ static irqreturn_t sii902x_interrupt(int irq, void *data) mutex_unlock(&sii902x->mutex); - if ((status & SII902X_HOTPLUG_EVENT) && sii902x->bridge.dev) + if ((status & SII902X_HOTPLUG_EVENT) && sii902x->bridge.dev) { drm_helper_hpd_irq_event(sii902x->bridge.dev); + drm_bridge_hpd_notify(&sii902x->bridge, (status & SII902X_PLUGGED_STATUS) + ? connector_status_connected + : connector_status_disconnected); + } return IRQ_HANDLED; } @@ -1001,6 +1042,11 @@ static int sii902x_init(struct sii902x *sii902x) sii902x->bridge.funcs = &sii902x_bridge_funcs; sii902x->bridge.of_node = dev->of_node; sii902x->bridge.timings = &default_sii902x_timings; + sii902x->bridge.ops = DRM_BRIDGE_OP_DETECT | DRM_BRIDGE_OP_EDID; + + if (sii902x->i2c->irq > 0) + sii902x->bridge.ops |= DRM_BRIDGE_OP_HPD; + drm_bridge_add(&sii902x->bridge); sii902x_audio_codec_init(sii902x, dev); @@ -1022,6 +1068,7 @@ static int sii902x_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct device *dev = &client->dev; + struct device_node *endpoint; struct sii902x *sii902x; int ret; @@ -1049,6 +1096,28 @@ static int sii902x_probe(struct i2c_client *client, return PTR_ERR(sii902x->reset_gpio); } + endpoint = of_graph_get_endpoint_by_regs(dev->of_node, 1, -1); + if (endpoint) { + struct device_node *remote = of_graph_get_remote_port_parent(endpoint); + + of_node_put(endpoint); + if (!remote) { + dev_err(dev, "Endpoint in port@1 unconnected\n"); + return -ENODEV; + } + + if (!of_device_is_available(remote)) { + dev_err(dev, "port@1 remote device is disabled\n"); + of_node_put(remote); + return -ENODEV; + } + + sii902x->next_bridge = of_drm_find_bridge(remote); + of_node_put(remote); + if (!sii902x->next_bridge) + return -EPROBE_DEFER; + } + mutex_init(&sii902x->mutex); sii902x->supplies[0].supply = "iovcc"; -- 2.25.1

2 1

[PATCH 0/2] drm/panfrost: partial support of T628 GPUs
by asheplyakov＠basealt.ru 17 Jan '22

17 Jan '22

From: Alexey Sheplyakov <asheplyakov(a)basealt.ru> Hello, these patches address some problems which prevent panfrost from working with Mali T628 GPUs: - incomplete/incorrect mmu memory attributes - wrong jobs affinity on dual core group GPUs With these patches panfrost is able to drive mali T628 (r1p0) GPU on some armv8 SoCs (in particular BE-M1000). r0 GPUs are still not supported [yet] (tested with Exynos 5422). Alexey Sheplyakov (2): drm/panfrost: mmu: improved memory attributes drm/panfrost: adjusted job affinity for dual core group GPUs drivers/gpu/drm/panfrost/panfrost_device.h | 7 ++++ drivers/gpu/drm/panfrost/panfrost_gpu.c | 45 ++++++++++++++++++++++ drivers/gpu/drm/panfrost/panfrost_job.c | 14 ++++++- drivers/gpu/drm/panfrost/panfrost_mmu.c | 3 -- drivers/iommu/io-pgtable-arm.c | 16 ++++++-- 5 files changed, 76 insertions(+), 9 deletions(-) -- 2.30.2

5 15

← Newer
1
...
55
56
57
58
59
60
61
...
85
Older →

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

dri-devel January 2022