dri-devel December 2021

dri-devel@lists.freedesktop.org

310 participants
720 discussions

[PATCH 0/6] drm/meson: add support for AXG & MIPI-DSI
by Neil Armstrong 20 Jan '22

20 Jan '22

The Amlogic AXG SoC family has a downgraded VPU with the following changes : - Only a single OSD plane, no overlay video plane - The primary plane doesn't support HW scaling - The pixels are read directly from DDR without any Canvas module - Doesn't support HDMI or CVBS - Ouputs only with ENCL encoder to a DPI-to-DSI Synopsys DW-MIPI-DSI transceiver The Amlogic AXg SoCs embeds a Synopsys DW-MIPI-DSI transceiver (ver 1.21a), with a custom glue managing the IP resets, clock and data input similar to the DW-HDMI Glue on other Amlogic SoCs. This adds support for the Glue managing the transceiver, mimicing the init flow provided by Amlogic to setup the ENCl encoder, the glue, the transceiver, the digital D-PHY and the Analog PHY in the proper way. The DW-MIPI-DSI transceiver + D-PHY are directly clocked by the VCLK2 clock, which pixel clock is derived and feeds the ENCL encoder and the VIU pixel reader. An optional "MEAS" clock can be enabled to measure the delay between each vsync feeding the DW-MIPI-DSI transceiver. Neil Armstrong (6): dt-bindings: display: amlogic,meson-vpu: add bindings for VPU found in AXG SoCs dt-bindings: display: add Amlogic MIPI DSI Host Controller bindings drm/meson: add support for VPU found in AXG SoCs drm/meson: venc: add ENCL encoder setup for MIPI-DSI output drm/meson: remove useless recursive components matching drm/meson: add support for MIPI-DSI transceiver .../display/amlogic,meson-dw-mipi-dsi.yaml | 115 ++++ .../bindings/display/amlogic,meson-vpu.yaml | 36 +- drivers/gpu/drm/meson/Kconfig | 7 + drivers/gpu/drm/meson/Makefile | 1 + drivers/gpu/drm/meson/meson_crtc.c | 8 +- drivers/gpu/drm/meson/meson_drv.c | 163 ++--- drivers/gpu/drm/meson/meson_drv.h | 10 +- drivers/gpu/drm/meson/meson_dw_mipi_dsi.c | 562 ++++++++++++++++++ drivers/gpu/drm/meson/meson_plane.c | 74 ++- drivers/gpu/drm/meson/meson_registers.h | 1 + drivers/gpu/drm/meson/meson_venc.c | 230 ++++++- drivers/gpu/drm/meson/meson_venc.h | 6 + drivers/gpu/drm/meson/meson_viu.c | 50 +- drivers/gpu/drm/meson/meson_vpp.c | 6 +- drivers/gpu/drm/meson/meson_vpp.h | 2 + 15 files changed, 1176 insertions(+), 95 deletions(-) create mode 100644 Documentation/devicetree/bindings/display/amlogic,meson-dw-mipi-dsi.yaml create mode 100644 drivers/gpu/drm/meson/meson_dw_mipi_dsi.c -- 2.22.0

4 21

[PATCH] dma-buf: dma-heap: Add a size limitation for allocation
by guangming.cao＠mediatek.com 20 Jan '22

20 Jan '22

From: Guangming <Guangming.Cao(a)mediatek.com> Currently, there is no size check for allocation. If the alloc size is larger than DRAM, it will cause OOM issue. Besides, if it runs on a process that won't be killed by OOM flow, it will cause a kernel exception finally, and we couldn't find the correct memory usage by dma-buf dump api such as "dma_buf_debug_show" since the allocation is still on going and the corresponding dmabuf is not exported. However, it sounds not simple enough to adding a count to count how many pages has been allocated before allocating done. So adding a size limitation here to prevent this case. Signed-off-by: Guangming <Guangming.Cao(a)mediatek.com> Signed-off-by: jianjiao zeng <jianjiao.zeng(a)mediatek.com> --- drivers/dma-buf/dma-heap.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/dma-buf/dma-heap.c b/drivers/dma-buf/dma-heap.c index 56bf5ad01ad5..8b75998a106c 100644 --- a/drivers/dma-buf/dma-heap.c +++ b/drivers/dma-buf/dma-heap.c @@ -107,6 +107,9 @@ static long dma_heap_ioctl_allocate(struct file *file, void *data) if (heap_allocation->heap_flags & ~DMA_HEAP_VALID_HEAP_FLAGS) return -EINVAL; + if (heap_allocation->len / PAGE_SIZE > totalram_pages() / 2) + return -EINVAL; + fd = dma_heap_buffer_alloc(heap, heap_allocation->len, heap_allocation->fd_flags, heap_allocation->heap_flags); -- 2.17.1

6 23

linux-next: build warning after merge of the drm-misc tree
by Stephen Rothwell 20 Jan '22

20 Jan '22

Hi all, After merging the drm-misc tree, today's linux-next build (htmldocs) produced this warning: include/drm/drm_modeset_lock.h:74: warning: Function parameter or member 'stack_depot' not described in 'drm_modeset_acquire_ctx' Introduced by commit cd06ab2fd48f ("drm/locking: add backtrace for locking contended locks without backoff") -- Cheers, Stephen Rothwell

2 2

[Bug 213561] New: [bisected] AMD GPU can no longer idle state after commit 1c0b0efd148d5b24c4932ddb3fa03c8edd6097b3
by bugzilla-daemon＠bugzilla.kernel.org 20 Jan '22

20 Jan '22

https://bugzilla.kernel.org/show_bug.cgi?id=213561 Bug ID: 213561 Summary: [bisected] AMD GPU can no longer idle state after commit 1c0b0efd148d5b24c4932ddb3fa03c8edd6097b3 Product: Drivers Version: 2.5 Kernel Version: 5.13rc7 Hardware: All OS: Linux Tree: Mainline Status: NEW Severity: normal Priority: P1 Component: Video(DRI - non Intel) Assignee: drivers_video-dri(a)kernel-bugs.osdl.org Reporter: untaintableangel(a)hotmail.co.uk Regression: No Nature of the problem: RX 5700 is unable to enter low power state at idle (see below for usual behaviour) Sensors at idle prior to the commit: amdgpu-pci-0f00 Adapter: PCI adapter vddgfx: 775.00 mV fan1: 0 RPM (min = 0 RPM, max = 3200 RPM) edge: +48.0°C (crit = +100.0°C, hyst = -273.1°C) (emerg = +105.0°C) junction: +48.0°C (crit = +110.0°C, hyst = -273.1°C) (emerg = +115.0°C) mem: +52.0°C (crit = +105.0°C, hyst = -273.1°C) (emerg = +110.0°C) power1: 8.00 W (cap = 165.00 W) After the commit, the lowest is: amdgpu-pci-0f00 Adapter: PCI adapter vddgfx: 1.03 V fan1: 0 RPM (min = 0 RPM, max = 3200 RPM) edge: +54.0°C (crit = +100.0°C, hyst = -273.1°C) (emerg = +105.0°C) junction: +56.0°C (crit = +110.0°C, hyst = -273.1°C) (emerg = +115.0°C) mem: +52.0°C (crit = +105.0°C, hyst = -273.1°C) (emerg = +110.0°C) power1: 31.00 W (cap = 165.00 W) This problem wasn't present in rc6 but is present in 5.13rc7 and bisects to: 1c0b0efd148d5b24c4932ddb3fa03c8edd6097b3 is the first bad commit commit 1c0b0efd148d5b24c4932ddb3fa03c8edd6097b3 Author: Yifan Zhang <yifan1.zhang(a)amd.com> Date: Thu Jun 10 10:10:07 2021 +0800 drm/amdgpu/gfx10: enlarge CP_MEC_DOORBELL_RANGE_UPPER to cover full doorbell. If GC has entered CGPG, ringing doorbell > first page doesn't wakeup GC. Enlarge CP_MEC_DOORBELL_RANGE_UPPER to workaround this issue. Signed-off-by: Yifan Zhang <yifan1.zhang(a)amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling(a)amd.com> Reviewed-by: Alex Deucher <alexander.deucher(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> Cc: stable(a)vger.kernel.org The device is a Sapphire Pulse RX5700 and this problem is seen even with one monitor set at 60Hz. GPU: 0f:00.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc. [AMD/ATI] Navi 10 [Radeon RX 5600 OEM/5600 XT / 5700/5700 XT] [1002:731f] (rev c4) -- You may reply to this email to add a comment. You are receiving this mail because: You are watching the assignee of the bug.

1 20

[PATCH 1/3] dma-buf: make the flags can be configured during dma fence init
by Huang Rui 20 Jan '22

20 Jan '22

In some user scenarios, the get_timeline_name callback uses the flags to decide which way to return the timeline string name. Once the trace_dma_fence_init event is enabled, it will call get_timeline_name callback to dump the fence structure. However, at this moment, the flags are always 0, and it might trigger some issues in get_timeline_name callback implementation of different gpu driver. So make a member to initialize the flags in dma_fence_init(). Signed-off-by: Huang Rui <ray.huang(a)amd.com> --- drivers/dma-buf/dma-fence.c | 2 +- include/linux/dma-fence.h | 7 +++++++ 2 files changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/dma-buf/dma-fence.c b/drivers/dma-buf/dma-fence.c index 066400ed8841..3e0622bf385f 100644 --- a/drivers/dma-buf/dma-fence.c +++ b/drivers/dma-buf/dma-fence.c @@ -952,7 +952,7 @@ dma_fence_init(struct dma_fence *fence, const struct dma_fence_ops *ops, fence->lock = lock; fence->context = context; fence->seqno = seqno; - fence->flags = 0UL; + fence->flags = ops->init_flags; fence->error = 0; trace_dma_fence_init(fence); diff --git a/include/linux/dma-fence.h b/include/linux/dma-fence.h index 1ea691753bd3..f9270c5bc07a 100644 --- a/include/linux/dma-fence.h +++ b/include/linux/dma-fence.h @@ -131,6 +131,13 @@ struct dma_fence_ops { */ bool use_64bit_seqno; + /** + * @init_flags: + * + * The initial value of fence flags (A mask of DMA_FENCE_FLAG_* defined). + */ + unsigned long init_flags; + /** * @get_driver_name: * -- 2.25.1

3 11

[PATCH] drm/msm/dsi: Fix missing put_device() call in dsi_get_phy
by Miaoqian Lin 19 Jan '22

19 Jan '22

If of_find_device_by_node() succeeds, dsi_get_phy() doesn't a corresponding put_device(). Thus add put_device() to fix the exception handling. Fixes: ec31abf ("drm/msm/dsi: Separate PHY to another platform device") Signed-off-by: Miaoqian Lin <linmq006(a)gmail.com> --- drivers/gpu/drm/msm/dsi/dsi.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/msm/dsi/dsi.c b/drivers/gpu/drm/msm/dsi/dsi.c index 75ae3008b68f..35be526e907a 100644 --- a/drivers/gpu/drm/msm/dsi/dsi.c +++ b/drivers/gpu/drm/msm/dsi/dsi.c @@ -40,7 +40,12 @@ static int dsi_get_phy(struct msm_dsi *msm_dsi) of_node_put(phy_node); - if (!phy_pdev || !msm_dsi->phy) { + if (!phy_pdev) { + DRM_DEV_ERROR(&pdev->dev, "%s: phy driver is not ready\n", __func__); + return -EPROBE_DEFER; + } + if (!msm_dsi->phy) { + put_device(&phy_pdev->dev); DRM_DEV_ERROR(&pdev->dev, "%s: phy driver is not ready\n", __func__); return -EPROBE_DEFER; } -- 2.17.1

2 1

[PATCH -next] drm/msm: remove variable set but not used
by Yang Li 19 Jan '22

19 Jan '22

The code that uses variable mdss has been removed, So the declaration and assignment of the variable can be removed. Eliminate the following clang warning: drivers/gpu/drm/msm/msm_drv.c:513:19: warning: variable 'mdss' set but not used [-Wunused-but-set-variable] Reported-by: Abaci Robot <abaci(a)linux.alibaba.com> Fixes: 2027e5b3413d ("drm/msm: Initialize MDSS irq domain at probe time") Signed-off-by: Yang Li <yang.lee(a)linux.alibaba.com> --- drivers/gpu/drm/msm/msm_drv.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index ad35a5d94053..59e30192cdf6 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -510,7 +510,6 @@ static int msm_drm_init(struct device *dev, const struct drm_driver *drv) struct msm_drm_private *priv = dev_get_drvdata(dev); struct drm_device *ddev; struct msm_kms *kms; - struct msm_mdss *mdss; int ret, i; ddev = drm_dev_alloc(drv, dev); @@ -521,8 +520,6 @@ static int msm_drm_init(struct device *dev, const struct drm_driver *drv) ddev->dev_private = priv; priv->dev = ddev; - mdss = priv->mdss; - priv->wq = alloc_ordered_workqueue("msm", 0); priv->hangcheck_period = DRM_MSM_HANGCHECK_DEFAULT_PERIOD; -- 2.20.1.7.g153144c

2 1

[PATCH v2 0/2] drm/msm: rework clock handling
by Dmitry Baryshkov 19 Jan '22

19 Jan '22

msm_dss_clk_*() functions significantly duplicate clk_bulk_* family of functions. Drop custom code and use bulk clocks directly. This also removes dependency of DP driver on the DPU driver internals. Changes since v1: - Rebase on top of current tree to fix conflicts The following changes since commit e4840d537c2c6b1189d4de16ee0f4820e069dcea: drm/msm: Do hw_init() before capturing GPU state (2021-11-22 10:45:55 -0800) are available in the Git repository at: https://git.linaro.org/people/dmitry.baryshkov/kernel.git dpu-clocks-cleanup for you to fetch changes up to d6c3e05f71d0fe131f427c32553a01b6df0bec9d: drm/msm/dp: rewrite dss_module_power to use bulk clock functions (2021-11-25 12:20:56 +0300) ---------------------------------------------------------------- Dmitry Baryshkov (2): drm/msm/dpu: simplify clocks handling drm/msm/dp: rewrite dss_module_power to use bulk clock functions drivers/gpu/drm/msm/Makefile | 1 - drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c | 24 +--- drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.h | 6 +- drivers/gpu/drm/msm/disp/dpu1/dpu_io_util.c | 187 -------------------------- drivers/gpu/drm/msm/disp/dpu1/dpu_io_util.h | 40 ------ drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c | 46 ++----- drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h | 4 +- drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c | 26 ++-- drivers/gpu/drm/msm/dp/dp_ctrl.c | 19 ++- drivers/gpu/drm/msm/dp/dp_parser.c | 21 ++- drivers/gpu/drm/msm/dp/dp_parser.h | 17 ++- drivers/gpu/drm/msm/dp/dp_power.c | 81 ++++++----- drivers/gpu/drm/msm/msm_drv.c | 49 +++++++ drivers/gpu/drm/msm/msm_drv.h | 1 + 14 files changed, 164 insertions(+), 358 deletions(-) delete mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_io_util.c delete mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_io_util.h

4 9

[PATCH v5 0/7] MIPS: JZ4780 and CI20 HDMI
by H. Nikolaus Schaller 19 Jan '22

19 Jan '22

PATCH V5 2021-10-05 14:28:44: - dropped mode_fixup and timings support in dw-hdmi as it is no longer needed in this V5 (by hns(a)goldelico.com) - dropped "drm/ingenic: add some jz4780 specific features" (stimulated by paul(a)crapouillou.net) - fixed typo in commit subject: "synopsis" -> "synopsys" (by hns(a)goldelico.com) - swapped clocks in jz4780.dtsi to match synopsys,dw-hdmi.yaml (by hns(a)goldelico.com) - improved, simplified, fixed, dtbschecked ingenic-jz4780-hdmi.yaml and made dependent of bridge/synopsys,dw-hdmi.yaml (based on suggestions by maxime(a)cerno.tech) - fixed binding vs. driver&DTS use of hdmi-5v regulator (suggested by maxime(a)cerno.tech) - dropped "drm/bridge: synopsis: Fix to properly handle HPD" - was a no longer needed workaround for a previous version (suggested by maxime(a)cerno.tech) PATCH V4 2021-09-27 18:44:38: - fix setting output_port = 1 (issue found by paul(a)crapouillou.net) - ci20.dts: convert to use hdmi-connector (by hns(a)goldelico.com) - add a hdmi-regulator to control +5V power (by hns(a)goldelico.com) - added a fix to dw-hdmi to call drm_kms_helper_hotplug_event on plugin event detection (by hns(a)goldelico.com) - always allocate extended descriptor but initialize only for jz4780 (by hns(a)goldelico.com) - updated to work on top of "[PATCH v3 0/6] drm/ingenic: Various improvements v3" (by paul(a)crapouillou.net) - rebased to v5.13-rc3 PATCH V3 2021-08-08 07:10:50: This series adds HDMI support for JZ4780 and CI20 board (and fixes one IPU related issue in registration error path) - [patch 1/8] switched from mode_fixup to atomic_check (suggested by robert.foss(a)linaro.org) - the call to the dw-hdmi specialization is still called mode_fixup - [patch 3/8] diverse fixes for ingenic-drm-drv (suggested by paul(a)crapouillou.net) - factor out some non-HDMI features of the jz4780 into a separate patch - multiple fixes around max height - do not change regmap config but a copy on stack - define some constants - factor out fixing of drm_init error path for IPU into separate patch - use FIELD_PREP() - [patch 8/8] conversion to component framework dropped (suggested by Laurent.pinchart(a)ideasonboard.com and paul(a)crapouillou.net) PATCH V2 2021-08-05 16:08:05: This series adds HDMI support for JZ4780 and CI20 board V2: - code and commit messages revisited for checkpatch warnings - rebased on v5.14-rc4 - include (failed, hence RFC 8/8) attempt to convert to component framework (was suggested by Paul Cercueil <paul(a)crapouillou.net> a while ago) H. Nikolaus Schaller (1): MIPS: defconfig: CI20: configure for DRM_DW_HDMI_JZ4780 Paul Boddie (5): drm/ingenic: Fix drm_init error path if IPU was registered drm/ingenic: Add support for JZ4780 and HDMI output drm/ingenic: Add dw-hdmi driver for jz4780 MIPS: DTS: jz4780: Account for Synopsys HDMI driver and LCD controllers MIPS: DTS: CI20: Add DT nodes for HDMI setup Sam Ravnborg (1): dt-bindings: display: Add ingenic,jz4780-dw-hdmi DT Schema .../bindings/display/ingenic-jz4780-hdmi.yaml | 79 +++++++++++ arch/mips/boot/dts/ingenic/ci20.dts | 67 ++++++++++ arch/mips/boot/dts/ingenic/jz4780.dtsi | 45 +++++++ arch/mips/configs/ci20_defconfig | 6 + drivers/gpu/drm/ingenic/Kconfig | 9 ++ drivers/gpu/drm/ingenic/Makefile | 1 + drivers/gpu/drm/ingenic/ingenic-drm-drv.c | 96 +++++++++++++- drivers/gpu/drm/ingenic/ingenic-drm.h | 42 ++++++ drivers/gpu/drm/ingenic/ingenic-dw-hdmi.c | 125 ++++++++++++++++++ 9 files changed, 464 insertions(+), 6 deletions(-) create mode 100644 Documentation/devicetree/bindings/display/ingenic-jz4780-hdmi.yaml create mode 100644 drivers/gpu/drm/ingenic/ingenic-dw-hdmi.c -- 2.33.0

4 40

[PATCH v3 00/10] drm: Make drivers to honour the nomodeset parameter
by Javier Martinez Canillas 19 Jan '22

19 Jan '22

The nomodeset kernel command line parameter is used to prevent the KMS/DRM drivers to be registered/probed. But only a few drivers implement support for this and most DRM drivers just ignore it. This patch series is a v3 to make DRM drivers to honour nomodeset. It is posted as separate patches to make easier for drivers maintainers to ack or pick them independently at their own pace. The drm_module_{pci,platform}_driver() helper macros are added, which are just wrappers around module_{pci,platform}_driver() but adding a check for drm_firmware_drivers_only() and returning -ENODEV if that is true. PCI and platform DRM drivers are then modified in the following patches to make use of those macros. Only KMS drivers will be ported to use these new macros, and only for PCI and platform DRM drivers. A follow-up series might do the same for drivers that are rendering-only and for USB/SPI/I2C devices, but it will need more discussion to agree whether that's desirable or not. Not all drivers were posted in v3 to avoid flooding the list with too many patches. I'm only including the patches adding the macros and some patches as an example of their usage. I've built tested with 'make allmodconfig && make M=drivers/gpu/drm' but I don't have hardware to test the drivers, so review/testing is appreciated. Best regards, Javier Changes in v3: - Include Thomas Zimmermann's patches in the series and rebase on top. - Add collected Acked-by tags from v2. Changes in v2: - Add drm_module_{pci,platform}_driver() macros and put the check there (Thomas Zimmermann). - Use the drm_module_*_driver() macros if possible (Thomas Zimmermann). - Leave the DRM drivers that don't set the DRIVER_MODESET driver feature (Lucas Stach). - Leave USB/SPI/I2C drivers and only include PCI and platform ones (Noralf Trønnes). - Add collected Reviewed-by tags Javier Martinez Canillas (5): drm: Provide platform module-init macro drm/imx/dcss: Replace module initialization with DRM helpers drm/komeda: Replace module initialization with DRM helpers drm/arm/hdlcd: Replace module initialization with DRM helpers drm/malidp: Replace module initialization with DRM helpers Thomas Zimmermann (5): drm: Provide PCI module-init macros drm/ast: Replace module-init boiler-plate code with DRM helpers drm/bochs: Replace module-init boiler-plate code with DRM helpers drm/cirrus: Replace module-init boiler-plate code with DRM helpers drm/hisilicon/hibmc: Replace module initialization with DRM helpers Documentation/gpu/drm-internals.rst | 6 + .../gpu/drm/arm/display/komeda/komeda_drv.c | 3 +- drivers/gpu/drm/arm/hdlcd_drv.c | 3 +- drivers/gpu/drm/arm/malidp_drv.c | 3 +- drivers/gpu/drm/ast/ast_drv.c | 18 +-- .../gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c | 3 +- drivers/gpu/drm/imx/dcss/dcss-drv.c | 3 +- drivers/gpu/drm/tiny/bochs.c | 20 +-- drivers/gpu/drm/tiny/cirrus.c | 17 +-- include/drm/drm_module.h | 125 ++++++++++++++++++ 10 files changed, 147 insertions(+), 54 deletions(-) create mode 100644 include/drm/drm_module.h -- 2.33.1

3 15

← Newer
1
...
9
10
11
12
13
14
15
...
72
Older →

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

dri-devel December 2021