dri-devel October 2017

dri-devel@lists.freedesktop.org

228 participants
1625 discussions

[Bug 29725] New: EXT_framebuffer_object extension with free radeon driver
by bugzilla-daemon＠freedesktop.org 08 Oct '17

08 Oct '17

https://bugs.freedesktop.org/show_bug.cgi?id=29725 Summary: EXT_framebuffer_object extension with free radeon driver Product: Mesa Version: unspecified Platform: Other OS/Version: All Status: NEW Severity: normal Priority: medium Component: Drivers/DRI/Radeon AssignedTo: dri-devel(a)lists.freedesktop.org ReportedBy: bioinfornatics(a)gmail.com hi, i am fedora packager and i package this game: http://os.wildfiregames.com/ i see for radeon driver a crash because with EXT_framebuffer_object extension with free radeon driver. beta rpm can be found here: http://koji.fedoraproject.org/koji/taskinfo?taskID=2415654 rpm -qa | grep mesa mesa-dri-drivers-7.8.1-8.fc13.x86_64 mesa-libGL-devel-7.8.1-8.fc13.x86_64 mesa-libGL-7.8.1-8.fc13.x86_64 mesa-debuginfo-7.8.1-8.fc13.x86_64 mesa-libGLU-devel-7.8.1-8.fc13.x86_64 mesa-libGLU-7.8.1-8.fc13.x86_64 -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug.

1 9

[PATCH] drm: vc4: Fix race during binding
by Stefan Wahren 08 Oct '17

08 Oct '17

This fixes the race between vc4_overflow_mem_work and the init of the job lock. Otherwise we could trigger a NULL pointer dereference during VC4 binding. Link: https://github.com/anholt/linux/issues/114 Signed-off-by: Stefan Wahren <stefan.wahren(a)i2se.com> Fixes: d5b1a78a772f ("drm/vc4: Add support for drawing 3D frames.") --- drivers/gpu/drm/vc4/vc4_gem.c | 1 - drivers/gpu/drm/vc4/vc4_irq.c | 1 + 2 files changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/vc4/vc4_gem.c b/drivers/gpu/drm/vc4/vc4_gem.c index d0c6bfb..47d964f 100644 --- a/drivers/gpu/drm/vc4/vc4_gem.c +++ b/drivers/gpu/drm/vc4/vc4_gem.c @@ -1088,7 +1088,6 @@ vc4_gem_init(struct drm_device *dev) INIT_LIST_HEAD(&vc4->render_job_list); INIT_LIST_HEAD(&vc4->job_done_list); INIT_LIST_HEAD(&vc4->seqno_cb_list); - spin_lock_init(&vc4->job_lock); INIT_WORK(&vc4->hangcheck.reset_work, vc4_reset_work); setup_timer(&vc4->hangcheck.timer, diff --git a/drivers/gpu/drm/vc4/vc4_irq.c b/drivers/gpu/drm/vc4/vc4_irq.c index 7d7af3a..d508d13 100644 --- a/drivers/gpu/drm/vc4/vc4_irq.c +++ b/drivers/gpu/drm/vc4/vc4_irq.c @@ -195,6 +195,7 @@ vc4_irq_preinstall(struct drm_device *dev) struct vc4_dev *vc4 = to_vc4_dev(dev); init_waitqueue_head(&vc4->job_wait_queue); + spin_lock_init(&vc4->job_lock); INIT_WORK(&vc4->overflow_mem_work, vc4_overflow_mem_work); /* Clear any pending interrupts someone might have left around -- 2.7.4

2 4

[PATCH v3] drm/gem-fb-helper: Improve documentation
by Noralf Trønnes 08 Oct '17

08 Oct '17

Make the docs read a little better. Cc: Laurent Pinchart <laurent.pinchart(a)ideasonboard.com> Cc: Daniel Vetter <daniel.vetter(a)ffwll.ch> Signed-off-by: Noralf Trønnes <noralf(a)tronnes.org> --- Changes since version 2: - Capitalized start of sentences/descriptions - Made sure all functions have Returns: - Added some info about how things tie together drivers/gpu/drm/drm_gem_framebuffer_helper.c | 82 ++++++++++++++++++---------- 1 file changed, 52 insertions(+), 30 deletions(-) diff --git a/drivers/gpu/drm/drm_gem_framebuffer_helper.c b/drivers/gpu/drm/drm_gem_framebuffer_helper.c index fc7e995..aa8cb9b 100644 --- a/drivers/gpu/drm/drm_gem_framebuffer_helper.c +++ b/drivers/gpu/drm/drm_gem_framebuffer_helper.c @@ -27,19 +27,24 @@ * DOC: overview * * This library provides helpers for drivers that don't subclass - * &drm_framebuffer and and use &drm_gem_object for their backing storage. + * &drm_framebuffer and use &drm_gem_object for their backing storage. * * Drivers without additional needs to validate framebuffers can simply use - * drm_gem_fb_create() and everything is wired up automatically. But all - * parts can be used individually. + * drm_gem_fb_create() and everything is wired up automatically. Other drivers + * can use all parts independently. */ /** - * drm_gem_fb_get_obj() - Get GEM object for framebuffer - * @fb: The framebuffer - * @plane: Which plane + * drm_gem_fb_get_obj() - Get GEM object backing the framebuffer + * @fb: Framebuffer + * @plane: Plane index * - * Returns the GEM object for given framebuffer. + * No additional reference is taken beyond the one that the &drm_frambuffer + * already holds. + * + * Returns: + * Pointer to &drm_gem_object for the given framebuffer and plane index or NULL + * if it does not exist. */ struct drm_gem_object *drm_gem_fb_get_obj(struct drm_framebuffer *fb, unsigned int plane) @@ -82,7 +87,7 @@ drm_gem_fb_alloc(struct drm_device *dev, /** * drm_gem_fb_destroy - Free GEM backed framebuffer - * @fb: DRM framebuffer + * @fb: Framebuffer * * Frees a GEM backed framebuffer with its backing buffer(s) and the structure * itself. Drivers can use this as their &drm_framebuffer_funcs->destroy @@ -102,12 +107,13 @@ EXPORT_SYMBOL(drm_gem_fb_destroy); /** * drm_gem_fb_create_handle - Create handle for GEM backed framebuffer - * @fb: DRM framebuffer - * @file: drm file - * @handle: handle created + * @fb: Framebuffer + * @file: DRM file to register the handle for + * @handle: Pointer to return the created handle * + * This function creates a handle for the GEM object backing the framebuffer. * Drivers can use this as their &drm_framebuffer_funcs->create_handle - * callback. + * callback. The GETFB IOCTL calls into this callback. * * Returns: * 0 on success or a negative error code on failure. @@ -120,18 +126,21 @@ int drm_gem_fb_create_handle(struct drm_framebuffer *fb, struct drm_file *file, EXPORT_SYMBOL(drm_gem_fb_create_handle); /** - * drm_gem_fb_create_with_funcs() - helper function for the + * drm_gem_fb_create_with_funcs() - Helper function for the * &drm_mode_config_funcs.fb_create * callback * @dev: DRM device - * @file: drm file for the ioctl call - * @mode_cmd: metadata from the userspace fb creation request + * @file: DRM file that holds the GEM handle(s) backing the framebuffer + * @mode_cmd: Metadata from the userspace framebuffer creation request * @funcs: vtable to be used for the new framebuffer object * * This can be used to set &drm_framebuffer_funcs for drivers that need the * &drm_framebuffer_funcs.dirty callback. Use drm_gem_fb_create() if you don't * need to change &drm_framebuffer_funcs. * The function does buffer size validation. + * + * Returns: + * Pointer to a &drm_framebuffer on success or an error pointer on failure. */ struct drm_framebuffer * drm_gem_fb_create_with_funcs(struct drm_device *dev, struct drm_file *file, @@ -192,15 +201,26 @@ static const struct drm_framebuffer_funcs drm_gem_fb_funcs = { }; /** - * drm_gem_fb_create() - &drm_mode_config_funcs.fb_create callback function + * drm_gem_fb_create() - Helper function for the + * &drm_mode_config_funcs.fb_create callback * @dev: DRM device - * @file: drm file for the ioctl call - * @mode_cmd: metadata from the userspace fb creation request + * @file: DRM file that holds the GEM handle(s) backing the framebuffer + * @mode_cmd: Metadata from the userspace framebuffer creation request + * + * This function creates a new framebuffer object described by + * &drm_mode_fb_cmd2. This description includes handles for the buffer(s) + * backing the framebuffer. * * If your hardware has special alignment or pitch requirements these should be * checked before calling this function. The function does buffer size * validation. Use drm_gem_fb_create_with_funcs() if you need to set * &drm_framebuffer_funcs.dirty. + * + * Drivers can use this as their &drm_mode_config_funcs.fb_create callback. + * The ADDFB2 IOCTL calls into this callback. + * + * Returns: + * Pointer to a &drm_framebuffer on success or an error pointer on failure. */ struct drm_framebuffer * drm_gem_fb_create(struct drm_device *dev, struct drm_file *file, @@ -212,15 +232,15 @@ drm_gem_fb_create(struct drm_device *dev, struct drm_file *file, EXPORT_SYMBOL_GPL(drm_gem_fb_create); /** - * drm_gem_fb_prepare_fb() - Prepare gem framebuffer - * @plane: Which plane - * @state: Plane state attach fence to + * drm_gem_fb_prepare_fb() - Prepare a GEM backed framebuffer + * @plane: Plane + * @state: Plane state the fence will be attached to * - * This can be used as the &drm_plane_helper_funcs.prepare_fb hook. - * - * This function checks if the plane FB has an dma-buf attached, extracts - * the exclusive fence and attaches it to plane state for the atomic helper - * to wait on. + * This function prepares a GEM backed framebuffer for scanout by checking if + * the plane framebuffer has a DMA-BUF attached. If it does, it extracts the + * exclusive fence and attaches it to the plane state for the atomic helper to + * wait on. This function can be used as the &drm_plane_helper_funcs.prepare_fb + * callback. * * There is no need for &drm_plane_helper_funcs.cleanup_fb hook for simple * gem based framebuffer drivers which have their buffers always pinned in @@ -246,17 +266,19 @@ int drm_gem_fb_prepare_fb(struct drm_plane *plane, EXPORT_SYMBOL_GPL(drm_gem_fb_prepare_fb); /** - * drm_gem_fbdev_fb_create - Create a drm_framebuffer for fbdev emulation + * drm_gem_fbdev_fb_create - Create a GEM backed &drm_framebuffer for fbdev + * emulation * @dev: DRM device * @sizes: fbdev size description - * @pitch_align: optional pitch alignment + * @pitch_align: Optional pitch alignment * @obj: GEM object backing the framebuffer * @funcs: vtable to be used for the new framebuffer object * - * This function creates a framebuffer for use with fbdev emulation. + * This function creates a framebuffer from a &drm_fb_helper_surface_size + * description for use in the &drm_fb_helper_funcs.fb_probe callback. * * Returns: - * Pointer to a drm_framebuffer on success or an error pointer on failure. + * Pointer to a &drm_framebuffer on success or an error pointer on failure. */ struct drm_framebuffer * drm_gem_fbdev_fb_create(struct drm_device *dev, -- 2.7.4

2 2

[PATCH v2] drm/i915: Replace *_reference/unreference() or *_ref/unref with _get/put()
by Harsha Sharma 08 Oct '17

08 Oct '17

Replace instances of drm_framebuffer_reference/unreference() with *_get/put() suffixes and drm_dev_unref with *_put() suffix because get/put is shorter and consistent with the kernel use of *_get/put suffixes. Done with following coccinelle semantic patch @@ expression ex; @@ ( -drm_framebuffer_unreference(ex); +drm_framebuffer_put(ex); | -drm_dev_unref(ex); +drm_dev_put(ex); | -drm_framebuffer_reference(ex); +drm_framebuffer_get(ex); ) Signed-off-by: Harsha Sharma <harshasharmaiitr(a)gmail.com> --- Changes in v2: -Added coccinelle patch in log message -cc to all driver-specific mailing lists drivers/gpu/drm/i915/i915_pci.c | 2 +- drivers/gpu/drm/i915/intel_display.c | 10 +++++----- drivers/gpu/drm/i915/intel_fbdev.c | 4 ++-- drivers/gpu/drm/i915/selftests/i915_gem_dmabuf.c | 2 +- drivers/gpu/drm/i915/selftests/i915_gem_evict.c | 2 +- drivers/gpu/drm/i915/selftests/i915_gem_gtt.c | 2 +- drivers/gpu/drm/i915/selftests/i915_gem_object.c | 2 +- drivers/gpu/drm/i915/selftests/i915_gem_request.c | 2 +- drivers/gpu/drm/i915/selftests/i915_vma.c | 2 +- drivers/gpu/drm/i915/selftests/intel_breadcrumbs.c | 2 +- 10 files changed, 15 insertions(+), 15 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c index 09d97e0..2f106cc 100644 --- a/drivers/gpu/drm/i915/i915_pci.c +++ b/drivers/gpu/drm/i915/i915_pci.c @@ -510,7 +510,7 @@ static void i915_pci_remove(struct pci_dev *pdev) struct drm_device *dev = pci_get_drvdata(pdev); i915_driver_unload(dev); - drm_dev_unref(dev); + drm_dev_put(dev); } static int i915_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index f172755..92f8304 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -2856,7 +2856,7 @@ static int skl_format_to_fourcc(int format, bool rgb_order, bool alpha) if (intel_plane_ggtt_offset(state) == plane_config->base) { fb = c->primary->fb; - drm_framebuffer_reference(fb); + drm_framebuffer_get(fb); goto valid_fb; } } @@ -2887,7 +2887,7 @@ static int skl_format_to_fourcc(int format, bool rgb_order, bool alpha) intel_crtc->pipe, PTR_ERR(intel_state->vma)); intel_state->vma = NULL; - drm_framebuffer_unreference(fb); + drm_framebuffer_put(fb); return; } @@ -2908,7 +2908,7 @@ static int skl_format_to_fourcc(int format, bool rgb_order, bool alpha) if (i915_gem_object_is_tiled(obj)) dev_priv->preserve_bios_swizzle = true; - drm_framebuffer_reference(fb); + drm_framebuffer_get(fb); primary->fb = primary->state->fb = fb; primary->crtc = primary->state->crtc = &intel_crtc->base; @@ -9847,7 +9847,7 @@ struct drm_framebuffer * if (obj->base.size < mode->vdisplay * fb->pitches[0]) return NULL; - drm_framebuffer_reference(fb); + drm_framebuffer_get(fb); return fb; #else return NULL; @@ -10028,7 +10028,7 @@ int intel_get_load_detect_pipe(struct drm_connector *connector, if (ret) goto fail; - drm_framebuffer_unreference(fb); + drm_framebuffer_put(fb); ret = drm_atomic_set_mode_for_crtc(&crtc_state->base, mode); if (ret) diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c index 262e75c..1ff7149 100644 --- a/drivers/gpu/drm/i915/intel_fbdev.c +++ b/drivers/gpu/drm/i915/intel_fbdev.c @@ -189,7 +189,7 @@ static int intelfb_create(struct drm_fb_helper *helper, " releasing it\n", intel_fb->base.width, intel_fb->base.height, sizes->fb_width, sizes->fb_height); - drm_framebuffer_unreference(&intel_fb->base); + drm_framebuffer_put(&intel_fb->base); intel_fb = ifbdev->fb = NULL; } if (!intel_fb || WARN_ON(!intel_fb->obj)) { @@ -624,7 +624,7 @@ static bool intel_fbdev_init_bios(struct drm_device *dev, ifbdev->preferred_bpp = fb->base.format->cpp[0] * 8; ifbdev->fb = fb; - drm_framebuffer_reference(&ifbdev->fb->base); + drm_framebuffer_put(&ifbdev->fb->base); /* Final pass to check if any active pipes don't have fbs */ for_each_crtc(dev, crtc) { diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_dmabuf.c b/drivers/gpu/drm/i915/selftests/i915_gem_dmabuf.c index 89dc25a..a7055b1 100644 --- a/drivers/gpu/drm/i915/selftests/i915_gem_dmabuf.c +++ b/drivers/gpu/drm/i915/selftests/i915_gem_dmabuf.c @@ -389,7 +389,7 @@ int i915_gem_dmabuf_mock_selftests(void) err = i915_subtests(tests, i915); - drm_dev_unref(&i915->drm); + drm_dev_put(&i915->drm); return err; } diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_evict.c b/drivers/gpu/drm/i915/selftests/i915_gem_evict.c index 5ea3732..75764fd 100644 --- a/drivers/gpu/drm/i915/selftests/i915_gem_evict.c +++ b/drivers/gpu/drm/i915/selftests/i915_gem_evict.c @@ -345,6 +345,6 @@ int i915_gem_evict_mock_selftests(void) err = i915_subtests(tests, i915); mutex_unlock(&i915->drm.struct_mutex); - drm_dev_unref(&i915->drm); + drm_dev_put(&i915->drm); return err; } diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c index 6b132ca..bc3c4a5 100644 --- a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c @@ -1540,7 +1540,7 @@ int i915_gem_gtt_mock_selftests(void) err = i915_subtests(tests, i915); mutex_unlock(&i915->drm.struct_mutex); - drm_dev_unref(&i915->drm); + drm_dev_put(&i915->drm); return err; } diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_object.c b/drivers/gpu/drm/i915/selftests/i915_gem_object.c index 8f011c4..7bfd990 100644 --- a/drivers/gpu/drm/i915/selftests/i915_gem_object.c +++ b/drivers/gpu/drm/i915/selftests/i915_gem_object.c @@ -586,7 +586,7 @@ int i915_gem_object_mock_selftests(void) err = i915_subtests(tests, i915); - drm_dev_unref(&i915->drm); + drm_dev_put(&i915->drm); return err; } diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_request.c b/drivers/gpu/drm/i915/selftests/i915_gem_request.c index 6664cb2..8985578 100644 --- a/drivers/gpu/drm/i915/selftests/i915_gem_request.c +++ b/drivers/gpu/drm/i915/selftests/i915_gem_request.c @@ -260,7 +260,7 @@ int i915_gem_request_mock_selftests(void) return -ENOMEM; err = i915_subtests(tests, i915); - drm_dev_unref(&i915->drm); + drm_dev_put(&i915->drm); return err; } diff --git a/drivers/gpu/drm/i915/selftests/i915_vma.c b/drivers/gpu/drm/i915/selftests/i915_vma.c index 2e86ec1..d8cc0a6 100644 --- a/drivers/gpu/drm/i915/selftests/i915_vma.c +++ b/drivers/gpu/drm/i915/selftests/i915_vma.c @@ -734,7 +734,7 @@ int i915_vma_mock_selftests(void) err = i915_subtests(tests, i915); mutex_unlock(&i915->drm.struct_mutex); - drm_dev_unref(&i915->drm); + drm_dev_put(&i915->drm); return err; } diff --git a/drivers/gpu/drm/i915/selftests/intel_breadcrumbs.c b/drivers/gpu/drm/i915/selftests/intel_breadcrumbs.c index 828904b..3c8bf44 100644 --- a/drivers/gpu/drm/i915/selftests/intel_breadcrumbs.c +++ b/drivers/gpu/drm/i915/selftests/intel_breadcrumbs.c @@ -475,7 +475,7 @@ int intel_breadcrumbs_mock_selftests(void) return -ENOMEM; err = i915_subtests(tests, i915->engine[RCS]); - drm_dev_unref(&i915->drm); + drm_dev_put(&i915->drm); return err; } -- 1.9.1

1 1

[BUG] gma500: Possible sleep-in-atomic bugs in gma_resume_pci
by Jia-Ju Bai 08 Oct '17

08 Oct '17

The driver may sleep under a spinlock, and the function call paths are: gma_power_begin (acquire the spinlock) gma_resume_pci pci_set_power_state __pci_start_power_transition (drivers/pci/pci.c) msleep --> may sleep gma_power_begin (acquire the spinlock) gma_resume_pci pci_enable_device pci_enable_device_flags (drivers/pci/pci.c) do_pci_enable_device pci_set_power_state __pci_start_power_transition msleep --> may sleep A possible fix is to replace msleep with mdelay in __pci_start_power_transition in drivers/pci/pci.c. These bugs are found by my static analysis tool and my code review. Thanks, Jia-Ju Bai

1 0

[Bug 102431] Talos Principle renders many small black patches with OpenGL on Vega
by bugzilla-daemon＠freedesktop.org 07 Oct '17

07 Oct '17

https://bugs.freedesktop.org/show_bug.cgi?id=102431 Bug ID: 102431 Summary: Talos Principle renders many small black patches with OpenGL on Vega Product: Mesa Version: git Hardware: All OS: All Status: NEW Severity: major Priority: medium Component: Drivers/Gallium/radeonsi Assignee: dri-devel(a)lists.freedesktop.org Reporter: vedran(a)miletic.net QA Contact: dri-devel(a)lists.freedesktop.org Created attachment 133814 --> https://bugs.freedesktop.org/attachment.cgi?id=133814&action=edit Screenshot Kernel amd-staging-drm-next, LLVM git, Mesa git, XFX RX Vega 64. -- You are receiving this mail because: You are the assignee for the bug.

1 1

[drm-intel:for-linux-next 18/27] drivers/gpu/drm/i915/i915_gem_gtt.c:1087: warning: 'iter' is used uninitialized in this function
by kbuild test robot 07 Oct '17

07 Oct '17

tree: git://anongit.freedesktop.org/drm-intel for-linux-next head: a883241c3922000b21b58b5740c55badfe09940f commit: 0a03852e049af91da9ae70326c44bb5d9b0d377a [18/27] drm/i915: support 2M pages for the 48b PPGTT config: x86_64-randconfig-a0-10072329 (attached as .config) compiler: gcc-4.4 (Debian 4.4.7-8) 4.4.7 reproduce: git checkout 0a03852e049af91da9ae70326c44bb5d9b0d377a # save the attached .config to linux build tree make ARCH=x86_64 All warnings (new ones prefixed by >>): drivers/gpu/drm/i915/i915_gem_gtt.c: In function 'gen8_ppgtt_insert_3lvl': drivers/gpu/drm/i915/i915_gem_gtt.c:1007: warning: 'iter.sg' is used uninitialized in this function drivers/gpu/drm/i915/i915_gem_gtt.c:1008: warning: 'iter.dma' is used uninitialized in this function drivers/gpu/drm/i915/i915_gem_gtt.c: In function 'gen8_ppgtt_insert_4lvl': >> drivers/gpu/drm/i915/i915_gem_gtt.c:1087: warning: 'iter' is used uninitialized in this function drivers/gpu/drm/i915/i915_gem_gtt.c: In function 'intel_ppat_get': drivers/gpu/drm/i915/i915_gem_gtt.c:2969: warning: 'entry' may be used uninitialized in this function vim +/iter +1087 drivers/gpu/drm/i915/i915_gem_gtt.c 9df15b499 Ben Widawsky 2013-11-02 998 894ccebee Chris Wilson 2017-02-15 999 static void gen8_ppgtt_insert_3lvl(struct i915_address_space *vm, 4a234c5fa Matthew Auld 2017-06-22 1000 struct i915_vma *vma, f9b5b782c Michel Thierry 2015-07-30 1001 enum i915_cache_level cache_level, f9b5b782c Michel Thierry 2015-07-30 1002 u32 unused) f9b5b782c Michel Thierry 2015-07-30 1003 { 17369ba08 Chuanxiao Dong 2017-07-07 1004 struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm); 894ccebee Chris Wilson 2017-02-15 1005 struct sgt_dma iter = { 4a234c5fa Matthew Auld 2017-06-22 1006 .sg = vma->pages->sgl, 894ccebee Chris Wilson 2017-02-15 @1007 .dma = sg_dma_address(iter.sg), 894ccebee Chris Wilson 2017-02-15 1008 .max = iter.dma + iter.sg->length, 894ccebee Chris Wilson 2017-02-15 1009 }; 4a234c5fa Matthew Auld 2017-06-22 1010 struct gen8_insert_pte idx = gen8_insert_pte(vma->node.start); de5ba8eb9 Michel Thierry 2015-08-03 1011 9e89f9ee3 Chris Wilson 2017-02-25 1012 gen8_ppgtt_insert_pte_entries(ppgtt, &ppgtt->pdp, &iter, &idx, 9e89f9ee3 Chris Wilson 2017-02-25 1013 cache_level); de5ba8eb9 Michel Thierry 2015-08-03 1014 } 894ccebee Chris Wilson 2017-02-15 1015 0a03852e0 Matthew Auld 2017-10-06 1016 static void gen8_ppgtt_insert_huge_entries(struct i915_vma *vma, 0a03852e0 Matthew Auld 2017-10-06 1017 struct i915_page_directory_pointer **pdps, 0a03852e0 Matthew Auld 2017-10-06 1018 struct sgt_dma *iter, 0a03852e0 Matthew Auld 2017-10-06 1019 enum i915_cache_level cache_level) 0a03852e0 Matthew Auld 2017-10-06 1020 { 0a03852e0 Matthew Auld 2017-10-06 1021 const gen8_pte_t pte_encode = gen8_pte_encode(0, cache_level); 0a03852e0 Matthew Auld 2017-10-06 1022 u64 start = vma->node.start; 0a03852e0 Matthew Auld 2017-10-06 1023 dma_addr_t rem = iter->sg->length; 0a03852e0 Matthew Auld 2017-10-06 1024 0a03852e0 Matthew Auld 2017-10-06 1025 do { 0a03852e0 Matthew Auld 2017-10-06 1026 struct gen8_insert_pte idx = gen8_insert_pte(start); 0a03852e0 Matthew Auld 2017-10-06 1027 struct i915_page_directory_pointer *pdp = pdps[idx.pml4e]; 0a03852e0 Matthew Auld 2017-10-06 1028 struct i915_page_directory *pd = pdp->page_directory[idx.pdpe]; 0a03852e0 Matthew Auld 2017-10-06 1029 unsigned int page_size; 0a03852e0 Matthew Auld 2017-10-06 1030 gen8_pte_t encode = pte_encode; 0a03852e0 Matthew Auld 2017-10-06 1031 gen8_pte_t *vaddr; 0a03852e0 Matthew Auld 2017-10-06 1032 u16 index, max; 0a03852e0 Matthew Auld 2017-10-06 1033 0a03852e0 Matthew Auld 2017-10-06 1034 if (vma->page_sizes.sg & I915_GTT_PAGE_SIZE_2M && 0a03852e0 Matthew Auld 2017-10-06 1035 IS_ALIGNED(iter->dma, I915_GTT_PAGE_SIZE_2M) && 0a03852e0 Matthew Auld 2017-10-06 1036 rem >= I915_GTT_PAGE_SIZE_2M && !idx.pte) { 0a03852e0 Matthew Auld 2017-10-06 1037 index = idx.pde; 0a03852e0 Matthew Auld 2017-10-06 1038 max = I915_PDES; 0a03852e0 Matthew Auld 2017-10-06 1039 page_size = I915_GTT_PAGE_SIZE_2M; 0a03852e0 Matthew Auld 2017-10-06 1040 0a03852e0 Matthew Auld 2017-10-06 1041 encode |= GEN8_PDE_PS_2M; 0a03852e0 Matthew Auld 2017-10-06 1042 0a03852e0 Matthew Auld 2017-10-06 1043 vaddr = kmap_atomic_px(pd); 0a03852e0 Matthew Auld 2017-10-06 1044 } else { 0a03852e0 Matthew Auld 2017-10-06 1045 struct i915_page_table *pt = pd->page_table[idx.pde]; 0a03852e0 Matthew Auld 2017-10-06 1046 0a03852e0 Matthew Auld 2017-10-06 1047 index = idx.pte; 0a03852e0 Matthew Auld 2017-10-06 1048 max = GEN8_PTES; 0a03852e0 Matthew Auld 2017-10-06 1049 page_size = I915_GTT_PAGE_SIZE; 0a03852e0 Matthew Auld 2017-10-06 1050 0a03852e0 Matthew Auld 2017-10-06 1051 vaddr = kmap_atomic_px(pt); 0a03852e0 Matthew Auld 2017-10-06 1052 } 0a03852e0 Matthew Auld 2017-10-06 1053 0a03852e0 Matthew Auld 2017-10-06 1054 do { 0a03852e0 Matthew Auld 2017-10-06 1055 GEM_BUG_ON(iter->sg->length < page_size); 0a03852e0 Matthew Auld 2017-10-06 1056 vaddr[index++] = encode | iter->dma; 0a03852e0 Matthew Auld 2017-10-06 1057 0a03852e0 Matthew Auld 2017-10-06 1058 start += page_size; 0a03852e0 Matthew Auld 2017-10-06 1059 iter->dma += page_size; 0a03852e0 Matthew Auld 2017-10-06 1060 rem -= page_size; 0a03852e0 Matthew Auld 2017-10-06 1061 if (iter->dma >= iter->max) { 0a03852e0 Matthew Auld 2017-10-06 1062 iter->sg = __sg_next(iter->sg); 0a03852e0 Matthew Auld 2017-10-06 1063 if (!iter->sg) 0a03852e0 Matthew Auld 2017-10-06 1064 break; 0a03852e0 Matthew Auld 2017-10-06 1065 0a03852e0 Matthew Auld 2017-10-06 1066 rem = iter->sg->length; 0a03852e0 Matthew Auld 2017-10-06 1067 iter->dma = sg_dma_address(iter->sg); 0a03852e0 Matthew Auld 2017-10-06 1068 iter->max = iter->dma + rem; 0a03852e0 Matthew Auld 2017-10-06 1069 0a03852e0 Matthew Auld 2017-10-06 1070 if (unlikely(!IS_ALIGNED(iter->dma, page_size))) 0a03852e0 Matthew Auld 2017-10-06 1071 break; 0a03852e0 Matthew Auld 2017-10-06 1072 } 0a03852e0 Matthew Auld 2017-10-06 1073 } while (rem >= page_size && index < max); 0a03852e0 Matthew Auld 2017-10-06 1074 0a03852e0 Matthew Auld 2017-10-06 1075 kunmap_atomic(vaddr); 0a03852e0 Matthew Auld 2017-10-06 1076 } while (iter->sg); 0a03852e0 Matthew Auld 2017-10-06 1077 } 0a03852e0 Matthew Auld 2017-10-06 1078 894ccebee Chris Wilson 2017-02-15 1079 static void gen8_ppgtt_insert_4lvl(struct i915_address_space *vm, 4a234c5fa Matthew Auld 2017-06-22 1080 struct i915_vma *vma, 894ccebee Chris Wilson 2017-02-15 1081 enum i915_cache_level cache_level, 894ccebee Chris Wilson 2017-02-15 1082 u32 unused) 894ccebee Chris Wilson 2017-02-15 1083 { 894ccebee Chris Wilson 2017-02-15 1084 struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm); 894ccebee Chris Wilson 2017-02-15 1085 struct sgt_dma iter = { 4a234c5fa Matthew Auld 2017-06-22 1086 .sg = vma->pages->sgl, 894ccebee Chris Wilson 2017-02-15 @1087 .dma = sg_dma_address(iter.sg), 894ccebee Chris Wilson 2017-02-15 1088 .max = iter.dma + iter.sg->length, 894ccebee Chris Wilson 2017-02-15 1089 }; 894ccebee Chris Wilson 2017-02-15 1090 struct i915_page_directory_pointer **pdps = ppgtt->pml4.pdps; 0a03852e0 Matthew Auld 2017-10-06 1091 0a03852e0 Matthew Auld 2017-10-06 1092 if (vma->page_sizes.sg > I915_GTT_PAGE_SIZE) { 0a03852e0 Matthew Auld 2017-10-06 1093 gen8_ppgtt_insert_huge_entries(vma, pdps, &iter, cache_level); 0a03852e0 Matthew Auld 2017-10-06 1094 } else { 4a234c5fa Matthew Auld 2017-06-22 1095 struct gen8_insert_pte idx = gen8_insert_pte(vma->node.start); 894ccebee Chris Wilson 2017-02-15 1096 0a03852e0 Matthew Auld 2017-10-06 1097 while (gen8_ppgtt_insert_pte_entries(ppgtt, pdps[idx.pml4e++], 0a03852e0 Matthew Auld 2017-10-06 1098 &iter, &idx, cache_level)) 9e89f9ee3 Chris Wilson 2017-02-25 1099 GEM_BUG_ON(idx.pml4e >= GEN8_PML4ES_PER_PML4); f9b5b782c Michel Thierry 2015-07-30 1100 } 0a03852e0 Matthew Auld 2017-10-06 1101 } f9b5b782c Michel Thierry 2015-07-30 1102 :::::: The code at line 1087 was first introduced by commit :::::: 894ccebee2b0e606ba9638d20dd87b33568482d7 drm/i915: Micro-optimise gen8_ppgtt_insert_entries() :::::: TO: Chris Wilson <chris(a)chris-wilson.co.uk> :::::: CC: Chris Wilson <chris(a)chris-wilson.co.uk> --- 0-DAY kernel test infrastructure Open Source Technology Center https://lists.01.org/pipermail/kbuild-all Intel Corporation

1 0

[Bug 99851] [drm:.r600_ring_test [radeon]] *ERROR* radeon: ring 0 test failed (scratch(0x8504)=0xCAFEDEAD)
by bugzilla-daemon＠freedesktop.org 07 Oct '17

07 Oct '17

https://bugs.freedesktop.org/show_bug.cgi?id=99851 --- Comment #55 from erhard_f(a)mailbox.org --- Joshuas comment #54 made me curious if it's a card specific issue. So I swapped some cards around my boxes. I run both machines on Kernel 4.13.5. on my PowerMac G5 11,2: Radeon HD 5450: hits the bug Radeon HD 6570: hits the bug Radeon X600: WORKS on Athlon X4 845: Radeon HD 5450: WORKS Radeon HD 6570: WORKS Radeon X600: WORKS The r380 Radeon X600 passes the ring test whereas the r600 cards fail: # dmesg | grep -i drm [ 9.523830] [drm] radeon kernel modesetting enabled. [ 9.525292] [drm] initializing kernel modesetting (RV380 0x1002:0x5B62 0x1002:0x0F02 0x00). [ 9.595072] [drm] GPU not posted. posting now... [ 9.740409] [drm] Generation 2 PCI interface, using max accessible memory [ 9.740442] [drm] Detected VRAM RAM=128M, BAR=128M [ 9.740445] [drm] RAM width 64bits DDR [ 9.740785] [drm] radeon: 128M of VRAM memory ready [ 9.740789] [drm] radeon: 512M of GTT memory ready. [ 9.740810] [drm] GART: num cpu pages 131072, num gpu pages 131072 [ 9.742591] [drm] radeon: 1 quad pipes, 1 Z pipes initialized [ 9.763583] [drm] PCIE GART of 512M enabled (table at 0x0000000088040000). [ 9.763705] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013). [ 9.763708] [drm] Driver supports precise vblank timestamp query. [ 9.763818] [drm] radeon: irq initialized. [ 9.763825] [drm] Loading R300 Microcode [ 10.233486] [drm] radeon: ring at 0x0000000068001000 [ 10.233545] [drm] ring test succeeded in 0 usecs [ 10.234665] [drm] ib test succeeded in 0 usecs [ 10.235279] [drm] Radeon Display Connectors [ 10.235284] [drm] Connector 0: [ 10.235287] [drm] DVI-I-1 [ 10.235289] [drm] HPD1 [ 10.235293] [drm] DDC: 0x60 0x60 0x60 0x60 0x60 0x60 0x60 0x60 [ 10.235296] [drm] Encoders: [ 10.235299] [drm] CRT1: INTERNAL_DAC1 [ 10.235302] [drm] DFP1: INTERNAL_TMDS1 [ 10.308511] [drm] fb mappable at 0x880C0000 [ 10.308521] [drm] vram apper at 0x88000000 [ 10.308525] [drm] size 8294400 [ 10.308527] [drm] fb depth is 24 [ 10.308530] [drm] pitch is 7680 [ 10.372575] radeon 0001:06:00.0: fb0: radeondrmfb frame buffer device [ 10.372649] [drm] Initialized radeon 2.50.0 20080528 for 0001:06:00.0 on minor 0 So I conclued this issue is produced by some interaction of the r600 DRM with PowerPC-specific or Mac-PCIe-specific things? -- You are receiving this mail because: You are the assignee for the bug.

1 0

[radeon-alex:drm-next-4.15-dc 1065/1129] drivers/gpu/drm/amd/amdgpu/../display/dc/basics/logger.c:73:23-27: WARNING: casting value returned by memory allocation function to (char *) is useless.
by kbuild test robot 06 Oct '17

06 Oct '17

tree: git://people.freedesktop.org/~agd5f/linux.git drm-next-4.15-dc head: 8ee5702afdd48b5864c46418ad310d6a23c8e9ab commit: 2004f45ef83f07f43f5da6ede780b08068c7583d [1065/1129] drm/amd/display: Use kernel alloc/free coccinelle warnings: (new ones prefixed by >>) >> drivers/gpu/drm/amd/amdgpu/../display/dc/basics/logger.c:73:23-27: WARNING: casting value returned by memory allocation function to (char *) is useless. Please review and possibly fold the followup patch. --- 0-DAY kernel test infrastructure Open Source Technology Center https://lists.01.org/pipermail/kbuild-all Intel Corporation

1 1

[PATCH v3 0/3] Split default display handling out from VGA arbiter
by Daniel Axtens 06 Oct '17

06 Oct '17

This patch set: - splits the default display handling out from VGA arbiter, into its own file and behind its own Kconfig option (and gives the functions better names). - adds extra detection of default devices. To be nominated, the vga arbiter and platform hooks must not have nominated a default. A card will then only be nominated if it has a driver attached and has IO or memory decoding enabled. - adds relevant documentation. The practical impact of this is improved X autoconfiguration on some arm64 systems. Changes in v3: - Add documentation - thanks Daniel Vetter for pointing it out. - Clarify explanations. Thanks to everyone for continuing to bear with my incomplete understanding of PCI and provide some clarity. - Split refactoring and adding functionality. Changes in v2: https://www.spinics.net/lists/linux-pci/msg64007.html Drop all the powerpc patches. [explanation snipped] v1: https://www.spinics.net/lists/linux-pci/msg63581.html Regards, Daniel Daniel Axtens (3): drm: split default display handler out of VGA arbiter drm: add fallback default device detection drm: documentation for default display device Documentation/gpu/default_display.rst | 93 +++++++++++++++++++ Documentation/gpu/index.rst | 1 + arch/ia64/pci/fixup.c | 6 +- arch/powerpc/kernel/pci-common.c | 6 +- arch/x86/pci/fixup.c | 6 +- arch/x86/video/fbdev.c | 4 +- drivers/gpu/vga/Kconfig | 12 +++ drivers/gpu/vga/Makefile | 1 + drivers/gpu/vga/default_display.c | 163 ++++++++++++++++++++++++++++++++++ drivers/gpu/vga/vga_switcheroo.c | 8 +- drivers/gpu/vga/vgaarb.c | 61 +++---------- drivers/pci/pci-sysfs.c | 4 +- include/linux/default_display.h | 44 +++++++++ include/linux/vgaarb.h | 15 ---- 14 files changed, 344 insertions(+), 80 deletions(-) create mode 100644 Documentation/gpu/default_display.rst create mode 100644 drivers/gpu/vga/default_display.c create mode 100644 include/linux/default_display.h -- 2.11.0

5 12

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

dri-devel October 2017