dri-devel February 2021

dri-devel@lists.freedesktop.org

314 participants
684 discussions

[PATCH 0/2] Fix purging buffers in the shmem helpers
by Neil Roberts 05 Mar '21

05 Mar '21

These two patches fix a problem with the madvise purging code for the shmem helpers where the mmaping for a purged buffer wouldn't get invalidated correctly. This presumably ends up as a security hole where the mapping can be accessed from user-space to read and write random pages from other buffers. This is currently affecting Panfrost. The second patch is a v2 from a patch that was sent standalone. There is a WIP IGT test for Panfrost which demonstrates the bug here: https://gitlab.freedesktop.org/nroberts/igt-gpu-tools/-/commits/panfrost-pu… Neil Roberts (2): drm/shmem-helper: Check for purged buffers in fault handler drm/shmem-helper: Don't remove the offset in vm_area_struct pgoff drivers/gpu/drm/drm_gem_shmem_helper.c | 25 ++++++++++++++++++------- 1 file changed, 18 insertions(+), 7 deletions(-) -- 2.29.2

2 5

[PATCH v3 0/8] Add support for SVM atomics in Nouveau
by Alistair Popple 04 Mar '21

04 Mar '21

This is the third version of a series to add support to Nouveau for atomic memory operations on OpenCL shared virtual memory (SVM) regions. This is achieved using the atomic PTE bits on the GPU to only permit atomic operations to system memory when a page is not mapped in userspace on the CPU. The previous version of this series used an unmap and pin page migration, however this resulted in problems with ZONE_MOVABLE and other issues so this series uses a different approach. Instead exclusive device access is implemented by adding a new swap entry type (SWAP_DEVICE_EXCLUSIVE) which is similar to a migration entry. The main difference is that on fault the original entry is immediately restored by the fault handler instead of waiting. Restoring the entry triggers calls to MMU notifers which allows a device driver to revoke the atomic access permission from the GPU prior to the CPU finalising the entry. Patches 1 & 2 are new for v3 and refactor the existing migration and device private entry functions. Patches 3 & 4 rework try_to_unmap_one() by splitting out unrelated functionality into separate functions - try_to_migrate_one() and try_to_munlock_one(). These should not change any functionality, but any help testing would be much appreciated as I have not been able to test every single usage of try_to_unmap_one(). Patch 5 contains the bulk of the implementation for device exclusive memory. Patch 6 contains some additions to the HMM selftests to ensure everything works as expected and has not changed significantly since v2. Patch 7 was posted previously and has not changed. Patch 8 was posted for v2 and has not changed significantly. This has been tested using the latest upstream Mesa userspace with a simple OpenCL test program which checks the results of atomic GPU operations on a SVM buffer whilst also writing to the same buffer from the CPU. v3: * Refactored some existing functionality. * Switched to using get_user_pages_remote() instead of open-coding it. * Moved code out of hmm. v2: * Changed implementation to use special swap entries instead of device private pages. Alistair Popple (8): mm: Remove special swap entry functions mm/swapops: Rework swap entry manipulation code mm/rmap: Split try_to_munlock from try_to_unmap mm/rmap: Split migration into its own function mm: Device exclusive memory access mm: Selftests for exclusive device memory nouveau/svm: Refactor nouveau_range_fault nouveau/svm: Implement atomic SVM access Documentation/vm/hmm.rst | 15 + arch/s390/mm/pgtable.c | 2 +- drivers/gpu/drm/nouveau/include/nvif/if000c.h | 1 + drivers/gpu/drm/nouveau/nouveau_svm.c | 120 +++- drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h | 1 + .../drm/nouveau/nvkm/subdev/mmu/vmmgp100.c | 6 + fs/proc/task_mmu.c | 23 +- include/linux/rmap.h | 8 +- include/linux/swap.h | 8 +- include/linux/swapops.h | 105 ++-- lib/test_hmm.c | 124 ++++ lib/test_hmm_uapi.h | 2 + mm/debug_vm_pgtable.c | 12 +- mm/hmm.c | 12 +- mm/huge_memory.c | 36 +- mm/hugetlb.c | 10 +- mm/memcontrol.c | 2 +- mm/memory.c | 128 +++- mm/migrate.c | 41 +- mm/mprotect.c | 18 +- mm/page_vma_mapped.c | 15 +- mm/rmap.c | 558 +++++++++++++++--- tools/testing/selftests/vm/hmm-tests.c | 219 +++++++ 23 files changed, 1207 insertions(+), 259 deletions(-) -- 2.20.1

4 25

[Bug 211649] New: "drm/amd/display: reuse current context instead of recreating one" cause hdmi hotplug blackscreen on amdgpu
by bugzilla-daemon＠bugzilla.kernel.org 04 Mar '21

04 Mar '21

https://bugzilla.kernel.org/show_bug.cgi?id=211649 Bug ID: 211649 Summary: "drm/amd/display: reuse current context instead of recreating one" cause hdmi hotplug blackscreen on amdgpu Product: Drivers Version: 2.5 Kernel Version: 5.11-rc7 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: youling257(a)gmail.com Regression: No 5.11-rc7 kernel,amdgpu 3400g, plug out hdmi, replug in hdmi, blackscreen. the bad commit is 8866a67 drm/amd/display: reuse current context instead of recreating one. -- 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 19

[Bug 211997] New: sys-kernel/gentoo-sources-5.11.2 crash upon disconnect HDMI monitor AMDGPU
by bugzilla-daemon＠bugzilla.kernel.org 04 Mar '21

04 Mar '21

https://bugzilla.kernel.org/show_bug.cgi?id=211997 Bug ID: 211997 Summary: sys-kernel/gentoo-sources-5.11.2 crash upon disconnect HDMI monitor AMDGPU Product: Drivers Version: 2.5 Kernel Version: 5.11.2-gentoo Hardware: x86-64 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: stefan(a)konink.de Regression: No Created attachment 295529 --> https://bugzilla.kernel.org/attachment.cgi?id=295529&action=edit dmesg When disconnecting a second monitor on a AMD Ryzen 5 2500U with Radeon Vega Mobile Gfx the system seems to hang. It is reachable over a wired connection, but the screen (other than a moving mouse) does not repond to any input. Connecting new devices such as an external mouse does not seem to influence the running session. Connected over netconsole, and later direct via the wired connection the error presented in dmesg: [drm:drm_atomic_helper_wait_for_flip_done [drm_kms_helper]] *ERROR* [CRTC:70:crtc-1] flip_done timed out [drm:drm_atomic_helper_wait_for_dependencies [drm_kms_helper]] *ERROR* [CRTC:70:crtc-1] flip_done timed out [drm:drm_atomic_helper_wait_for_dependencies [drm_kms_helper]] *ERROR* [PLANE:50:plane-2] flip_done timed out -- 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 5

[PATCH 0/2] drm/sun4i: fix DCLK and improve its handling
by Giulio Benetti 04 Mar '21

04 Mar '21

First patch is a tested by me fix, while the second need testing to understand if it works correctly with any sunxi SoC with DE peripheral. Already tested SoCs are: - A20 - A33 Need testing: - A10 - A10s - A13 Giulio Benetti (2): drm/sun4i: tcon: fix inverted DCLK polarity drm/sun4i: tcon: improve DCLK polarity handling drivers/gpu/drm/sun4i/sun4i_tcon.c | 20 +------------------- drivers/gpu/drm/sun4i/sun4i_tcon.h | 1 + 2 files changed, 2 insertions(+), 19 deletions(-) -- 2.25.1

6 27

[PATCH] dt-bindings: display: mediatek, dpi: Convert to use graph schema
by Rob Herring 03 Mar '21

03 Mar '21

Update the mediatek,dpi binding to use the graph schema. Missed this one from the mass conversion since it's not part of drm-misc. Cc: Chun-Kuang Hu <chunkuang.hu(a)kernel.org> Cc: Philipp Zabel <p.zabel(a)pengutronix.de> Cc: Matthias Brugger <matthias.bgg(a)gmail.com> Cc: CK Hu <ck.hu(a)mediatek.com> Cc: Jitao shi <jitao.shi(a)mediatek.com> Cc: dri-devel(a)lists.freedesktop.org Cc: linux-mediatek(a)lists.infradead.org Signed-off-by: Rob Herring <robh(a)kernel.org> --- .../bindings/display/mediatek/mediatek,dpi.yaml | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-) diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.yaml index 6cdb734c91a9..eb84b53cabb1 100644 --- a/Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.yaml +++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.yaml @@ -50,15 +50,10 @@ properties: - const: sleep port: - type: object + $ref: /schemas/graph.yaml#/properties/port description: - Output port node with endpoint definitions as described in - Documentation/devicetree/bindings/graph.txt. This port should be connected - to the input port of an attached HDMI or LVDS encoder chip. - - properties: - endpoint: - type: object + Output port node. This port should be connected to the input port of an + attached HDMI or LVDS encoder chip. required: - compatible -- 2.27.0

2 1

[PATCH -next v2] fbdev: atyfb: use LCD management functions for PPC_PMAC also
by Randy Dunlap 03 Mar '21

03 Mar '21

Include PPC_PMAC in the configs that use aty_ld_lcd() and aty_st_lcd() implementations so that the PM code may work correctly for PPC_PMAC. Suggested-by: Ville Syrjälä <ville.syrjala(a)linux.intel.com> Signed-off-by: Randy Dunlap <rdunlap(a)infradead.org> Cc: Daniel Vetter <daniel.vetter(a)ffwll.ch> Cc: Nick Desaulniers <ndesaulniers(a)google.com> Cc: linux-fbdev(a)vger.kernel.org Cc: dri-devel(a)lists.freedesktop.org Cc: Bartlomiej Zolnierkiewicz <b.zolnierkie(a)samsung.com> Cc: Sam Ravnborg <sam(a)ravnborg.org> Cc: David Airlie <airlied(a)linux.ie> Cc: Jani Nikula <jani.nikula(a)linux.intel.com> --- Daniel- We also need this patch: https://lore.kernel.org/dri-devel/20210224215528.822-1-rdunlap@infradead.or… to fix a kernel test robot build error. v2: send the correct version of this patch (thanks, Ville) drivers/video/fbdev/aty/atyfb_base.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) --- linux-next-20210219.orig/drivers/video/fbdev/aty/atyfb_base.c +++ linux-next-20210219/drivers/video/fbdev/aty/atyfb_base.c @@ -133,7 +133,7 @@ #define PRINTKE(fmt, args...) printk(KERN_ERR "atyfb: " fmt, ## args) #if defined(CONFIG_PMAC_BACKLIGHT) || defined(CONFIG_FB_ATY_GENERIC_LCD) || \ -defined(CONFIG_FB_ATY_BACKLIGHT) +defined(CONFIG_FB_ATY_BACKLIGHT) || defined (CONFIG_PPC_PMAC) static const u32 lt_lcd_regs[] = { CNFG_PANEL_LG, LCD_GEN_CNTL_LG, @@ -175,8 +175,8 @@ u32 aty_ld_lcd(int index, const struct a return aty_ld_le32(LCD_DATA, par); } } -#else /* defined(CONFIG_PMAC_BACKLIGHT) || defined(CONFIG_FB_ATY_BACKLIGHT) \ - defined(CONFIG_FB_ATY_GENERIC_LCD) */ +#else /* defined(CONFIG_PMAC_BACKLIGHT) || defined(CONFIG_FB_ATY_BACKLIGHT) || + defined(CONFIG_FB_ATY_GENERIC_LCD) || defined(CONFIG_PPC_PMAC) */ void aty_st_lcd(int index, u32 val, const struct atyfb_par *par) { } @@ -184,7 +184,8 @@ u32 aty_ld_lcd(int index, const struct a { return 0; } -#endif /* defined(CONFIG_PMAC_BACKLIGHT) || defined (CONFIG_FB_ATY_GENERIC_LCD) */ +#endif /* defined(CONFIG_PMAC_BACKLIGHT) || defined(CONFIG_FB_ATY_BACKLIGHT) || + defined (CONFIG_FB_ATY_GENERIC_LCD) || defined(CONFIG_PPC_PMAC) */ #ifdef CONFIG_FB_ATY_GENERIC_LCD /*

2 1

[Bug 211981] New: AMDGPU: without cable attached videocard shuts down and does wake up again, log prints "[drm] Cannot find any crtc or sizes"
by bugzilla-daemon＠bugzilla.kernel.org 03 Mar '21

03 Mar '21

https://bugzilla.kernel.org/show_bug.cgi?id=211981 Bug ID: 211981 Summary: AMDGPU: without cable attached videocard shuts down and does wake up again, log prints "[drm] Cannot find any crtc or sizes" Product: Drivers Version: 2.5 Kernel Version: 5.10.10 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: raulvior.bcn(a)gmail.com Regression: No Created attachment 295505 --> https://bugzilla.kernel.org/attachment.cgi?id=295505&action=edit Dmesg when no output is attached. Framebuffer not initalized. Linux boot shows the message in the title of this bug report if the videocard does not have any output attached. The computer boots up and when Linux starts loading, the card shuts down: fans stop spinning and the LED of the logo turns off. It's a Sapphire Radeon RX 580. If I attach the cable past this point the card does not turn on again. Pressing keys on the keyboard, moving the mouse or turning off and on the monitor does not awake the GPU. > amdgpu 0000:27:00.0: amdgpu: SE 4, SH per SE 1, CU per SH 9, active_cu_number > 36 > amdgpu 0000:27:00.0: [drm] Cannot find any crtc or sizes > [drm] Initialized amdgpu 3.40.0 20150101 for 0000:27:00.0 on minor 0 If there's an output device attached to the videocard, then it prints this text: > amdgpu 0000:27:00.0: amdgpu: SE 4, SH per SE 1, CU per SH 9, active_cu_number > 36 > [drm] fb mappable at 0xE0BEB000 > [drm] vram apper at 0xE0000000 > [drm] size 7299072 > [drm] fb depth is 24 > [drm] pitch is 6912 > fbcon: amdgpudrmfb (fb0) is primary device > EDAC amd64: F17h detected (node 0). > EDAC amd64: Node 0: DRAM ECC disabled. > Console: switching to colour frame buffer device 210x65 > amdgpu 0000:27:00.0: [drm] fb0: amdgpudrmfb frame buffer device > [drm] Initialized amdgpu 3.40.0 20150101 for 0000:27:00.0 on minor 0 -- 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 3

[PATCH 00/31] Introduce devm_pm_opp_* API
by Yangtao Li 03 Mar '21

03 Mar '21

Hi, This patchset add devm_pm_opp_set_clkname, devm_pm_opp_put_clkname, devm_pm_opp_set_regulators, devm_pm_opp_put_regulators, devm_pm_opp_set_supported_hw, devm_pm_opp_of_add_table and devm_pm_opp_register_notifier. Yangtao Li (31): opp: Add devres wrapper for dev_pm_opp_set_clkname and dev_pm_opp_put_clkname opp: Add devres wrapper for dev_pm_opp_set_regulators and dev_pm_opp_put_regulators opp: Add devres wrapper for dev_pm_opp_set_supported_hw opp: Add devres wrapper for dev_pm_opp_of_add_table opp: Add devres wrapper for dev_pm_opp_register_notifier serial: qcom_geni_serial: fix potential mem leak in qcom_geni_serial_probe() serial: qcom_geni_serial: convert to use devm_pm_opp_* API spi: spi-qcom-qspi: fix potential mem leak in spi_geni_probe() spi: spi-qcom-qspi: fix potential mem leak in spi_geni_probe() qcom-geni-se: remove opp_table mmc: sdhci-msm: fix potential mem leak in sdhci_msm_probe() mmc: sdhci-msm: convert to use devm_pm_opp_* API spi: spi-qcom-qspi: fix potential mem leak in qcom_qspi_probe() spi: spi-qcom-qspi: convert to use devm_pm_opp_* API drm/msm: fix potential mem leak drm/msm: convert to use devm_pm_opp_* API and remove dp_ctrl_put drm/lima: convert to use devm_pm_opp_* API drm/lima: remove unneeded devm_devfreq_remove_device() drm/panfrost: convert to use devm_pm_opp_* API media: venus: fix error check in core_get_v4() media: venus: convert to use devm_pm_opp_* API memory: samsung: exynos5422-dmc: fix return error in exynos5_init_freq_table memory: samsung: exynos5422-dmc: convert to use devm_pm_opp_* API memory: tegra20: convert to use devm_pm_opp_* API memory: tegra30: convert to use devm_pm_opp_* API PM / devfreq: tegra30: convert to use devm_pm_opp_* API PM / devfreq: rk3399_dmc: convert to use devm_pm_opp_* API PM / devfreq: imx8m-ddrc: convert to use devm_pm_opp_* API PM / devfreq: imx-bus: convert to use devm_pm_opp_* API PM / devfreq: exynos: convert to use devm_pm_opp_* API PM / devfreq: convert to devm_pm_opp_register_notifier and remove unused API drivers/devfreq/devfreq.c | 66 +------ drivers/devfreq/exynos-bus.c | 42 +---- drivers/devfreq/imx-bus.c | 14 +- drivers/devfreq/imx8m-ddrc.c | 15 +- drivers/devfreq/rk3399_dmc.c | 22 +-- drivers/devfreq/tegra30-devfreq.c | 21 +-- drivers/gpu/drm/lima/lima_devfreq.c | 45 +---- drivers/gpu/drm/lima/lima_devfreq.h | 2 - drivers/gpu/drm/msm/adreno/a5xx_gpu.c | 2 +- drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 2 +- drivers/gpu/drm/msm/adreno/adreno_gpu.c | 2 +- drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c | 31 ++-- drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h | 2 - drivers/gpu/drm/msm/dp/dp_ctrl.c | 29 +-- drivers/gpu/drm/msm/dp/dp_ctrl.h | 1 - drivers/gpu/drm/msm/dp/dp_display.c | 5 +- drivers/gpu/drm/msm/dsi/dsi_host.c | 23 ++- drivers/gpu/drm/panfrost/panfrost_devfreq.c | 34 +--- drivers/gpu/drm/panfrost/panfrost_devfreq.h | 1 - .../media/platform/qcom/venus/pm_helpers.c | 22 +-- drivers/memory/samsung/exynos5422-dmc.c | 13 +- drivers/memory/tegra/tegra20-emc.c | 29 +-- drivers/memory/tegra/tegra30-emc.c | 29 +-- drivers/mmc/host/sdhci-msm.c | 27 ++- drivers/opp/core.c | 173 ++++++++++++++++++ drivers/opp/of.c | 36 ++++ drivers/spi/spi-geni-qcom.c | 23 ++- drivers/spi/spi-qcom-qspi.c | 25 ++- drivers/tty/serial/qcom_geni_serial.c | 31 ++-- include/linux/devfreq.h | 23 --- include/linux/pm_opp.h | 38 ++++ include/linux/qcom-geni-se.h | 2 - 32 files changed, 402 insertions(+), 428 deletions(-) -- 2.25.1

7 51

[PATCH] drm: rcar-du: Use drmm_encoder_alloc() to manage encoder
by Kieran Bingham 02 Mar '21

02 Mar '21

The encoder allocation was converted to a DRM managed resource at the same time as the addition of a new helper drmm_encoder_alloc() which simplifies the same process. Convert the custom drm managed resource allocation of the encoder with the helper to simplify the implementation, and prevent hitting a WARN_ON() due to the handling the drm_encoder_init() call directly without registering a .destroy() function op. Fixes: f5f16725edbc ("drm: rcar-du: Use DRM-managed allocation for encoders") Reported-by: Geert Uytterhoeven <geert+renesas(a)glider.be> Signed-off-by: Kieran Bingham <kieran.bingham+renesas(a)ideasonboard.com> --- drivers/gpu/drm/rcar-du/rcar_du_encoder.c | 31 +++++------------------ 1 file changed, 6 insertions(+), 25 deletions(-) diff --git a/drivers/gpu/drm/rcar-du/rcar_du_encoder.c b/drivers/gpu/drm/rcar-du/rcar_du_encoder.c index ba8c6038cd63..ca3761772211 100644 --- a/drivers/gpu/drm/rcar-du/rcar_du_encoder.c +++ b/drivers/gpu/drm/rcar-du/rcar_du_encoder.c @@ -48,21 +48,12 @@ static unsigned int rcar_du_encoder_count_ports(struct device_node *node) static const struct drm_encoder_funcs rcar_du_encoder_funcs = { }; -static void rcar_du_encoder_release(struct drm_device *dev, void *res) -{ - struct rcar_du_encoder *renc = res; - - drm_encoder_cleanup(&renc->base); - kfree(renc); -} - int rcar_du_encoder_init(struct rcar_du_device *rcdu, enum rcar_du_output output, struct device_node *enc_node) { struct rcar_du_encoder *renc; struct drm_bridge *bridge; - int ret; /* * Locate the DRM bridge from the DT node. For the DPAD outputs, if the @@ -101,26 +92,16 @@ int rcar_du_encoder_init(struct rcar_du_device *rcdu, return -ENOLINK; } - renc = kzalloc(sizeof(*renc), GFP_KERNEL); - if (renc == NULL) - return -ENOMEM; - - renc->output = output; - dev_dbg(rcdu->dev, "initializing encoder %pOF for output %u\n", enc_node, output); - ret = drm_encoder_init(&rcdu->ddev, &renc->base, &rcar_du_encoder_funcs, - DRM_MODE_ENCODER_NONE, NULL); - if (ret < 0) { - kfree(renc); - return ret; - } + renc = drmm_encoder_alloc(&rcdu->ddev, struct rcar_du_encoder, base, + &rcar_du_encoder_funcs, DRM_MODE_ENCODER_NONE, + NULL); + if (!renc) + return -ENOMEM; - ret = drmm_add_action_or_reset(&rcdu->ddev, rcar_du_encoder_release, - renc); - if (ret) - return ret; + renc->output = output; /* * Attach the bridge to the encoder. The bridge will create the -- 2.25.1

3 6

← Newer
1
...
14
15
16
17
18
19
20
...
69
Older →

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

dri-devel February 2021