dri-devel June 2020

dri-devel@lists.freedesktop.org

315 participants
731 discussions

[PATCH v2 1/2] drm: vmwgfx: remove drm_driver::master_set() return typ
by Emil Velikov 15 Jun '20

15 Jun '20

The function always returns zero (success). Ideally we'll remove it all together - although that's requires a little more work. For now, we can drop the return type and simplify the drm core code surrounding it. v2: remove redundant assignment (Sam) Cc: David Airlie <airlied(a)linux.ie> Cc: Daniel Vetter <daniel(a)ffwll.ch> Cc: VMware Graphics <linux-graphics-maintainer(a)vmware.com> Cc: Roland Scheidegger <sroland(a)vmware.com> Signed-off-by: Emil Velikov <emil.l.velikov(a)gmail.com> Cc: Sam Ravnborg <sam(a)ravnborg.org> Reviewed-by: Sam Ravnborg <sam(a)ravnborg.org> --- VMWare team, I'm planning to push this via drm-misc. Review, ack and comments are appreciated. --- drivers/gpu/drm/drm_auth.c | 32 +++++++---------------------- drivers/gpu/drm/vmwgfx/vmwgfx_drv.c | 8 +++----- include/drm/drm_drv.h | 4 ++-- 3 files changed, 12 insertions(+), 32 deletions(-) diff --git a/drivers/gpu/drm/drm_auth.c b/drivers/gpu/drm/drm_auth.c index 74ce0c29c960..4c723e3a689c 100644 --- a/drivers/gpu/drm/drm_auth.c +++ b/drivers/gpu/drm/drm_auth.c @@ -122,27 +122,19 @@ struct drm_master *drm_master_create(struct drm_device *dev) return master; } -static int drm_set_master(struct drm_device *dev, struct drm_file *fpriv, - bool new_master) +static void drm_set_master(struct drm_device *dev, struct drm_file *fpriv, + bool new_master) { - int ret = 0; - dev->master = drm_master_get(fpriv->master); - if (dev->driver->master_set) { - ret = dev->driver->master_set(dev, fpriv, new_master); - if (unlikely(ret != 0)) { - drm_master_put(&dev->master); - } - } + if (dev->driver->master_set) + dev->driver->master_set(dev, fpriv, new_master); - fpriv->was_master = (ret == 0); - return ret; + fpriv->was_master = true; } static int drm_new_set_master(struct drm_device *dev, struct drm_file *fpriv) { struct drm_master *old_master; - int ret; lockdep_assert_held_once(&dev->master_mutex); @@ -157,22 +149,12 @@ static int drm_new_set_master(struct drm_device *dev, struct drm_file *fpriv) fpriv->is_master = 1; fpriv->authenticated = 1; - ret = drm_set_master(dev, fpriv, true); - if (ret) - goto out_err; + drm_set_master(dev, fpriv, true); if (old_master) drm_master_put(&old_master); return 0; - -out_err: - /* drop references and restore old master on failure */ - drm_master_put(&fpriv->master); - fpriv->master = old_master; - fpriv->is_master = 0; - - return ret; } /* @@ -265,7 +247,7 @@ int drm_setmaster_ioctl(struct drm_device *dev, void *data, goto out_unlock; } - ret = drm_set_master(dev, file_priv, false); + drm_set_master(dev, file_priv, false); out_unlock: mutex_unlock(&dev->master_mutex); return ret; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c index c2247a893ed4..470428387878 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c @@ -1129,9 +1129,9 @@ static long vmw_compat_ioctl(struct file *filp, unsigned int cmd, } #endif -static int vmw_master_set(struct drm_device *dev, - struct drm_file *file_priv, - bool from_open) +static void vmw_master_set(struct drm_device *dev, + struct drm_file *file_priv, + bool from_open) { /* * Inform a new master that the layout may have changed while @@ -1139,8 +1139,6 @@ static int vmw_master_set(struct drm_device *dev, */ if (!from_open) drm_sysfs_hotplug_event(dev); - - return 0; } static void vmw_master_drop(struct drm_device *dev, diff --git a/include/drm/drm_drv.h b/include/drm/drm_drv.h index bb924cddc09c..835c38a76ef6 100644 --- a/include/drm/drm_drv.h +++ b/include/drm/drm_drv.h @@ -311,8 +311,8 @@ struct drm_driver { * * Called whenever the minor master is set. Only used by vmwgfx. */ - int (*master_set)(struct drm_device *dev, struct drm_file *file_priv, - bool from_open); + void (*master_set)(struct drm_device *dev, struct drm_file *file_priv, + bool from_open); /** * @master_drop: * -- 2.25.1

3 7

[PATCH 1/3] drm/mm: remove unused rb_hole_size()
by Christian König 15 Jun '20

15 Jun '20

Just some code cleanup. Signed-off-by: Christian König <christian.koenig(a)amd.com> --- drivers/gpu/drm/drm_mm.c | 5 ----- 1 file changed, 5 deletions(-) diff --git a/drivers/gpu/drm/drm_mm.c b/drivers/gpu/drm/drm_mm.c index 82d2888eb7fe..425fcd3590e8 100644 --- a/drivers/gpu/drm/drm_mm.c +++ b/drivers/gpu/drm/drm_mm.c @@ -305,11 +305,6 @@ static inline struct drm_mm_node *rb_hole_addr_to_node(struct rb_node *rb) return rb_entry_safe(rb, struct drm_mm_node, rb_hole_addr); } -static inline u64 rb_hole_size(struct rb_node *rb) -{ - return rb_entry(rb, struct drm_mm_node, rb_hole_size)->hole_size; -} - static struct drm_mm_node *best_hole(struct drm_mm *mm, u64 size) { struct rb_node *rb = mm->holes_size.rb_root.rb_node; -- 2.17.1

2 5

[PATCH v16 0/1] mt8183 dpi support pin mode swap
by Jitao Shi 15 Jun '20

15 Jun '20

Changes since v15: - Fix YAML License to (GPL-2.0-only OR BSD-2-Clause). - "dt-bindings: display: mediatek: control dpi pins mode to avoid leakage" "drm/mediatek: set dpi pin mode to gpio low to avoid leakage current" applied v15. The links are https://patchwork.kernel.org/patch/11489545/ https://patchwork.kernel.org/patch/11489577/ Changes since v14: - add "Acked-by" and "Reviewed-by" - change port@0 to port in yaml Changes since v13: - move dpi dual edge patches to another series because it will have long time to implement the dual edge change base boris patches. https://patchwork.kernel.org/cover/11354279/ Changes since v12: - fix mediatek,dpi.yaml make_dt_binding_check errors. Change since v11: - fine tune mediatek,dpi.yaml. - add Acked-by: Rob Herring <robh(a)kernel.org>. Change since v10: - convert the Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.txt to yaml format. - read the pclk-sample in endpoint. Changes since v9: - rename pinctrl-names = "gpiomode", "dpimode" to "active", "idle". - fix some typo. Changes since v8: - drop pclk-sample redefine in mediatek,dpi.txt - only get the gpiomode and dpimode when dpi->pinctrl is successful. Changes since v7: - separate dt-bindings to independent patches. - move dpi dual edge to one patch. Changes since v6: - change dual_edge to pclk-sample - remove dpi_pin_mode_swap and Changes since v5: - fine tune the dt-bindings commit message. Changes since v4: - move pin mode control and dual edge control to deveice tree. - update dt-bindings document for pin mode swap and dual edge control. Changes since v3: - add dpi pin mode control when dpi on or off. - update dpi dual edge comment. Changes since v2: - update dt-bindings document for mt8183 dpi. - separate dual edge modfication as independent patch. Jitao Shi (1): dt-bindings: display: mediatek: convert the dpi bindings to yaml .../display/mediatek/mediatek,dpi.txt | 42 -------- .../display/mediatek/mediatek,dpi.yaml | 97 +++++++++++++++++++ 2 files changed, 97 insertions(+), 42 deletions(-) delete mode 100644 Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.txt create mode 100644 Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.yaml -- 2.25.1

2 2

Re: [PATCH v2] drm/etnaviv: fix ref count leak via pm_runtime_get_sync
by Markus Elfring 15 Jun '20

15 Jun '20

… > In case of failure, decrement the ref count before returning. Can it be nicer to use the term “reference count” here? Will the tag “Fixes” become helpful for the commit message? … > +++ b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c … > @@ -1326,6 +1331,7 @@ struct dma_fence *etnaviv_gpu_submit(struct etnaviv_gem_submit *submit) > ret = event_alloc(gpu, nr_events, event); > if (ret) { > DRM_ERROR("no free events\n"); > + pm_runtime_put_noidle(gpu->dev); > return NULL; > } I suggest to move a bit of exception handling code to the end of this function implementation so that it can be better reused after the addition of a jump target like “put_runtime”. Regards, Markus

1 0

[Bug 205497] New: Attempt to read amd gpu id causes a freeze
by bugzilla-daemon＠bugzilla.kernel.org 15 Jun '20

15 Jun '20

https://bugzilla.kernel.org/show_bug.cgi?id=205497 Bug ID: 205497 Summary: Attempt to read amd gpu id causes a freeze Product: Drivers Version: 2.5 Kernel Version: 5.3.9 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: luya(a)fedoraproject.org Regression: No Created attachment 285871 --> https://bugzilla.kernel.org/attachment.cgi?id=285871&action=edit Script from radeontop to read AMD gpu ids Running an utility named radeontop on an AMD APU causes a freeze while attempting to read amdgpu ids. Attached is the script. It will be nice to provide a better method to read AMD GPU cards. -- You are receiving this mail because: You are watching the assignee of the bug.

1 24

[PATCH v6 4/6] drm/msm: dsi: Use OPP API to set clk/perf state
by Rajendra Nayak 15 Jun '20

15 Jun '20

On SDM845 DSI needs to express a perforamnce state requirement on a power domain depending on the clock rates. Use OPP table from DT to register with OPP framework and use dev_pm_opp_set_rate() to set the clk/perf state. Signed-off-by: Rajendra Nayak <rnayak(a)codeaurora.org> Cc: Rob Clark <robdclark(a)gmail.com> Cc: Sean Paul <sean(a)poorly.run> Cc: dri-devel(a)lists.freedesktop.org --- No functional change in v6, rebased over 5.8-rc1 drivers/gpu/drm/msm/dsi/dsi.h | 2 ++ drivers/gpu/drm/msm/dsi/dsi_cfg.c | 4 +-- drivers/gpu/drm/msm/dsi/dsi_host.c | 58 ++++++++++++++++++++++++++++++++++++++ 3 files changed, 62 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/msm/dsi/dsi.h b/drivers/gpu/drm/msm/dsi/dsi.h index 4de771d..ba7583c 100644 --- a/drivers/gpu/drm/msm/dsi/dsi.h +++ b/drivers/gpu/drm/msm/dsi/dsi.h @@ -180,10 +180,12 @@ int msm_dsi_runtime_suspend(struct device *dev); int msm_dsi_runtime_resume(struct device *dev); int dsi_link_clk_set_rate_6g(struct msm_dsi_host *msm_host); int dsi_link_clk_set_rate_v2(struct msm_dsi_host *msm_host); +int dsi_link_clk_set_rate_6g_v2(struct msm_dsi_host *msm_host); int dsi_link_clk_enable_6g(struct msm_dsi_host *msm_host); int dsi_link_clk_enable_v2(struct msm_dsi_host *msm_host); void dsi_link_clk_disable_6g(struct msm_dsi_host *msm_host); void dsi_link_clk_disable_v2(struct msm_dsi_host *msm_host); +void dsi_link_clk_disable_6g_v2(struct msm_dsi_host *msm_host); int dsi_tx_buf_alloc_6g(struct msm_dsi_host *msm_host, int size); int dsi_tx_buf_alloc_v2(struct msm_dsi_host *msm_host, int size); void *dsi_tx_buf_get_6g(struct msm_dsi_host *msm_host); diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.c b/drivers/gpu/drm/msm/dsi/dsi_cfg.c index 813d69d..773c4fe 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_cfg.c +++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.c @@ -210,9 +210,9 @@ static const struct msm_dsi_host_cfg_ops msm_dsi_6g_host_ops = { }; static const struct msm_dsi_host_cfg_ops msm_dsi_6g_v2_host_ops = { - .link_clk_set_rate = dsi_link_clk_set_rate_6g, + .link_clk_set_rate = dsi_link_clk_set_rate_6g_v2, .link_clk_enable = dsi_link_clk_enable_6g, - .link_clk_disable = dsi_link_clk_disable_6g, + .link_clk_disable = dsi_link_clk_disable_6g_v2, .clk_init_ver = dsi_clk_init_6g_v2, .tx_buf_alloc = dsi_tx_buf_alloc_6g, .tx_buf_get = dsi_tx_buf_get_6g, diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c index 11ae5b8..890531c 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_host.c +++ b/drivers/gpu/drm/msm/dsi/dsi_host.c @@ -14,6 +14,7 @@ #include <linux/of_graph.h> #include <linux/of_irq.h> #include <linux/pinctrl/consumer.h> +#include <linux/pm_opp.h> #include <linux/regmap.h> #include <linux/regulator/consumer.h> #include <linux/spinlock.h> @@ -111,6 +112,9 @@ struct msm_dsi_host { struct clk *pixel_clk_src; struct clk *byte_intf_clk; + struct opp_table *opp_table; + bool has_opp_table; + u32 byte_clk_rate; u32 pixel_clk_rate; u32 esc_clk_rate; @@ -537,6 +541,38 @@ int dsi_link_clk_set_rate_6g(struct msm_dsi_host *msm_host) return 0; } +int dsi_link_clk_set_rate_6g_v2(struct msm_dsi_host *msm_host) +{ + int ret; + struct device *dev = &msm_host->pdev->dev; + + DBG("Set clk rates: pclk=%d, byteclk=%d", + msm_host->mode->clock, msm_host->byte_clk_rate); + + ret = dev_pm_opp_set_rate(dev, msm_host->byte_clk_rate); + if (ret) { + pr_err("%s: dev_pm_opp_set_rate failed %d\n", __func__, ret); + return ret; + } + + ret = clk_set_rate(msm_host->pixel_clk, msm_host->pixel_clk_rate); + if (ret) { + pr_err("%s: Failed to set rate pixel clk, %d\n", __func__, ret); + return ret; + } + + if (msm_host->byte_intf_clk) { + ret = clk_set_rate(msm_host->byte_intf_clk, + msm_host->byte_clk_rate / 2); + if (ret) { + pr_err("%s: Failed to set rate byte intf clk, %d\n", + __func__, ret); + return ret; + } + } + + return 0; +} int dsi_link_clk_enable_6g(struct msm_dsi_host *msm_host) { @@ -665,6 +701,13 @@ void dsi_link_clk_disable_6g(struct msm_dsi_host *msm_host) clk_disable_unprepare(msm_host->byte_clk); } +void dsi_link_clk_disable_6g_v2(struct msm_dsi_host *msm_host) +{ + /* Drop the performance state vote */ + dev_pm_opp_set_rate(&msm_host->pdev->dev, 0); + dsi_link_clk_disable_6g(msm_host); +} + void dsi_link_clk_disable_v2(struct msm_dsi_host *msm_host) { clk_disable_unprepare(msm_host->pixel_clk); @@ -1879,6 +1922,18 @@ int msm_dsi_host_init(struct msm_dsi *msm_dsi) goto fail; } + msm_host->opp_table = dev_pm_opp_set_clkname(&pdev->dev, "byte"); + if (IS_ERR(msm_host->opp_table)) + return PTR_ERR(msm_host->opp_table); + /* OPP table is optional */ + ret = dev_pm_opp_of_add_table(&pdev->dev); + if (!ret) { + msm_host->has_opp_table = true; + } else if (ret != -ENODEV) { + dev_err(&pdev->dev, "invalid OPP table in device tree\n"); + return ret; + } + init_completion(&msm_host->dma_comp); init_completion(&msm_host->video_comp); mutex_init(&msm_host->dev_mutex); @@ -1914,6 +1969,9 @@ void msm_dsi_host_destroy(struct mipi_dsi_host *host) mutex_destroy(&msm_host->cmd_mutex); mutex_destroy(&msm_host->dev_mutex); + if (msm_host->has_opp_table) + dev_pm_opp_of_remove_table(&msm_host->pdev->dev); + dev_pm_opp_put_clkname(msm_host->opp_table); pm_runtime_disable(&msm_host->pdev->dev); } -- QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by The Linux Foundation

1 0

[PATCH v6 3/6] drm/msm/dpu: Use OPP API to set clk/perf state
by Rajendra Nayak 15 Jun '20

15 Jun '20

On some qualcomm platforms DPU needs to express a performance state requirement on a power domain depending on the clock rates. Use OPP table from DT to register with OPP framework and use dev_pm_opp_set_rate() to set the clk/perf state. Signed-off-by: Rajendra Nayak <rnayak(a)codeaurora.org> Reviewed-by: Rob Clark <robdclark(a)chromium.org> Reviewed-by: Matthias Kaehlcke <mka(a)chromium.org> Cc: Rob Clark <robdclark(a)gmail.com> Cc: Sean Paul <sean(a)poorly.run> Cc: dri-devel(a)lists.freedesktop.org --- No functional change in v6, rebased over 5.8-rc1 drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c | 3 ++- drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c | 26 +++++++++++++++++++++++++- drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h | 4 ++++ 3 files changed, 31 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c index 7c230f7..b36919d 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c @@ -7,6 +7,7 @@ #include <linux/debugfs.h> #include <linux/errno.h> #include <linux/mutex.h> +#include <linux/pm_opp.h> #include <linux/sort.h> #include <linux/clk.h> #include <linux/bitmap.h> @@ -218,7 +219,7 @@ static int _dpu_core_perf_set_core_clk_rate(struct dpu_kms *kms, u64 rate) rate = core_clk->max_rate; core_clk->rate = rate; - return msm_dss_clk_set_rate(core_clk, 1); + return dev_pm_opp_set_rate(&kms->pdev->dev, core_clk->rate); } static u64 _dpu_core_perf_get_core_clk_rate(struct dpu_kms *kms) diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c index b8615d4..0bc8ec4 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c @@ -10,6 +10,7 @@ #include <linux/debugfs.h> #include <linux/dma-buf.h> #include <linux/of_irq.h> +#include <linux/pm_opp.h> #include <drm/drm_crtc.h> #include <drm/drm_file.h> @@ -1025,11 +1026,23 @@ static int dpu_bind(struct device *dev, struct device *master, void *data) if (!dpu_kms) return -ENOMEM; + dpu_kms->opp_table = dev_pm_opp_set_clkname(dev, "core"); + if (IS_ERR(dpu_kms->opp_table)) + return PTR_ERR(dpu_kms->opp_table); + /* OPP table is optional */ + ret = dev_pm_opp_of_add_table(dev); + if (!ret) { + dpu_kms->has_opp_table = true; + } else if (ret != -ENODEV) { + dev_err(dev, "invalid OPP table in device tree\n"); + return ret; + } + mp = &dpu_kms->mp; ret = msm_dss_parse_clock(pdev, mp); if (ret) { DPU_ERROR("failed to parse clocks, ret=%d\n", ret); - return ret; + goto err; } platform_set_drvdata(pdev, dpu_kms); @@ -1043,6 +1056,11 @@ static int dpu_bind(struct device *dev, struct device *master, void *data) priv->kms = &dpu_kms->base; return ret; +err: + if (dpu_kms->has_opp_table) + dev_pm_opp_of_remove_table(dev); + dev_pm_opp_put_clkname(dpu_kms->opp_table); + return ret; } static void dpu_unbind(struct device *dev, struct device *master, void *data) @@ -1057,6 +1075,10 @@ static void dpu_unbind(struct device *dev, struct device *master, void *data) if (dpu_kms->rpm_enabled) pm_runtime_disable(&pdev->dev); + + if (dpu_kms->has_opp_table) + dev_pm_opp_of_remove_table(dev); + dev_pm_opp_put_clkname(dpu_kms->opp_table); } static const struct component_ops dpu_ops = { @@ -1082,6 +1104,8 @@ static int __maybe_unused dpu_runtime_suspend(struct device *dev) struct dpu_kms *dpu_kms = platform_get_drvdata(pdev); struct dss_module_power *mp = &dpu_kms->mp; + /* Drop the performance state vote */ + dev_pm_opp_set_rate(dev, 0); rc = msm_dss_enable_clk(mp->clk_config, mp->num_clk, false); if (rc) DPU_ERROR("clock disable failed rc:%d\n", rc); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h index a3b122b..7400cd7 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h @@ -128,6 +128,10 @@ struct dpu_kms { struct platform_device *pdev; bool rpm_enabled; + + struct opp_table *opp_table; + bool has_opp_table; + struct dss_module_power mp; /* reference count bandwidth requests, so we know when we can -- QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by The Linux Foundation

1 0

[PATCH] drm/vc4: Fix an error code vc4_create_object()
by Dan Carpenter 15 Jun '20

15 Jun '20

The vc4_create_object() needs to return NULL on error, not error pointers. If it returns an error pointer then that will lead to an Oops in the callers. Fortunately, in current kernels small allocations always succed so this will never happen. Fixes: c826a6e10644 ("drm/vc4: Add a BO cache.") Signed-off-by: Dan Carpenter <dan.carpenter(a)oracle.com> --- drivers/gpu/drm/vc4/vc4_bo.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/vc4/vc4_bo.c b/drivers/gpu/drm/vc4/vc4_bo.c index 72d30d90b856c..0af246a5609ca 100644 --- a/drivers/gpu/drm/vc4/vc4_bo.c +++ b/drivers/gpu/drm/vc4/vc4_bo.c @@ -389,7 +389,7 @@ struct drm_gem_object *vc4_create_object(struct drm_device *dev, size_t size) bo = kzalloc(sizeof(*bo), GFP_KERNEL); if (!bo) - return ERR_PTR(-ENOMEM); + return NULL; bo->madv = VC4_MADV_WILLNEED; refcount_set(&bo->usecnt, 0); -- 2.26.2

1 0

[PATCH] drm/ttm: Fix dma_fence refcnt leak when adding move fence
by Xiyu Yang 15 Jun '20

15 Jun '20

ttm_bo_add_move_fence() invokes dma_fence_get(), which returns a reference of the specified dma_fence object to "fence" with increased refcnt. When ttm_bo_add_move_fence() returns, local variable "fence" becomes invalid, so the refcount should be decreased to keep refcount balanced. The reference counting issue happens in one exception handling path of ttm_bo_add_move_fence(). When no_wait_gpu flag is equals to true, the function forgets to decrease the refcnt increased by dma_fence_get(), causing a refcnt leak. Fix this issue by calling dma_fence_put() when no_wait_gpu flag is equals to true. Signed-off-by: Xiyu Yang <xiyuyang19(a)fudan.edu.cn> Signed-off-by: Xin Tan <tanxin.ctf(a)gmail.com> --- drivers/gpu/drm/ttm/ttm_bo.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c index f73b81c2576e..0f20e14a4cfd 100644 --- a/drivers/gpu/drm/ttm/ttm_bo.c +++ b/drivers/gpu/drm/ttm/ttm_bo.c @@ -883,8 +883,10 @@ static int ttm_bo_add_move_fence(struct ttm_buffer_object *bo, if (!fence) return 0; - if (no_wait_gpu) + if (no_wait_gpu) { + dma_fence_put(fence); return -EBUSY; + } dma_resv_add_shared_fence(bo->base.resv, fence); -- 2.7.4

2 1

[PATCH 0/9] shmem helper untangling
by Daniel Vetter 15 Jun '20

15 Jun '20

Hi all, I've started this a while ago, with the idea to move shmem helpers over to dma_resv_lock. Big prep work for that was to untangle the layering between functions called by drivers, and functions used to implement drm_gem_object_funcs. I didn't ever get to the locking part, but I think the cleanup here are worth it stand-alone still. Comments, review and testing very much welcome. Cheers, Daniel Daniel Vetter (9): drm/msm: Don't call dma_buf_vunmap without _vmap drm/gem: WARN if drm_gem_get_pages is called on a private obj drm/doc: Some polish for shmem helpers drm/virtio: Call the right shmem helpers drm/udl: Don't call get/put_pages on imported dma-buf drm/shmem-helpers: Don't call get/put_pages on imported dma-buf in vmap drm/shmem-helpers: Redirect mmap for imported dma-buf drm/shmem-helpers: Ensure get_pages is not called on imported dma-buf drm/shmem-helpers: Simplify dma-buf importing Documentation/gpu/drm-kms-helpers.rst | 12 --- Documentation/gpu/drm-mm.rst | 12 +++ drivers/gpu/drm/drm_gem.c | 8 ++ drivers/gpu/drm/drm_gem_shmem_helper.c | 128 ++++++++++++++---------- drivers/gpu/drm/msm/msm_gem.c | 3 +- drivers/gpu/drm/udl/udl_gem.c | 22 ++-- drivers/gpu/drm/virtio/virtgpu_object.c | 2 +- 7 files changed, 111 insertions(+), 76 deletions(-) -- 2.26.2

5 50

← Newer
1
...
49
50
51
52
53
54
55
...
74
Older →

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

dri-devel June 2020