dri-devel August 2019

dri-devel@lists.freedesktop.org

318 participants
2338 discussions

[Bug 204345] New: Unigine Superposition benchmark notoriously unstable since 5.1.18 (and possibly before)
by bugzilla-daemon＠bugzilla.kernel.org 21 Aug '19

21 Aug '19

https://bugzilla.kernel.org/show_bug.cgi?id=204345 Bug ID: 204345 Summary: Unigine Superposition benchmark notoriously unstable since 5.1.18 (and possibly before) Product: Drivers Version: 2.5 Kernel Version: 5.1.18 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: edisonalvaringo(a)bol.com.br Regression: No At first glance I though it was an Overclocking problem because I oc my hardware just slightly. But no even with no OCing, zero, nada... since 5.1.18 Unigine benchmark crashes. It is really unstable 90% of the times it will crash when it's about to display the final results, on 5.1.20 is even worse and it gets crazy because fps drop really hard. If you are really luck it will finish. I have tried 5.1.18/19 and 20. 5.0.20 doesn't not have this bug. I will try to investigate further to at least know which is the precise point release when this bug was introduced. -- You are receiving this mail because: You are watching the assignee of the bug.

1 11

[Bug 110659] pageflipping seems to cause jittering on mouse input when running Hitman 2 in Wine/DXVK with amdgpu.dc=1
by bugzilla-daemon＠freedesktop.org 21 Aug '19

21 Aug '19

https://bugs.freedesktop.org/show_bug.cgi?id=110659 --- Comment #38 from tempel.julian(a)gmail.com --- Huh, that seems suspicious. I'm not aware of any such a tool which would be active for me all the time. I had redshift installed, but it wasn't running in background, so not surprisingly uninstalling it and restarting the system didn't help. There is colord running, but killing its process didn't help either. So I'm a bit clueless how this comes. If I knew how, I'd try to turn it off. Could you try manually disabling KWin compositing via Shift + Alt + F12 before starting Hitman 2? I just tricked myself when trying out Xfce instead of Plasma, but the reason why Xfce wasn't showing the issue was that it didn't turn off compositing in fullscreen (despite of an option telling otherwise). After manually turning off compositing, the issue was the same as on Plasma without compositor. -- You are receiving this mail because: You are the assignee for the bug.

1 0

[PATCH] dma-fence: Serialise signal enabling (dma_fence_enable_sw_signaling)
by Chris Wilson 21 Aug '19

21 Aug '19

Make dma_fence_enable_sw_signaling() behave like its dma_fence_add_callback() and dma_fence_default_wait() counterparts and perform the test to enable signaling under the fence->lock, along with the action to do so. This ensure that should an implementation be trying to flush the cb_list (by signaling) on retirement before freeing the fence, it can do so in a race-free manner. See also 0fc89b6802ba ("dma-fence: Simply wrap dma_fence_signal_locked with dma_fence_signal"). Signed-off-by: Chris Wilson <chris(a)chris-wilson.co.uk> --- drivers/dma-buf/dma-fence.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/drivers/dma-buf/dma-fence.c b/drivers/dma-buf/dma-fence.c index 2c136aee3e79..587727089134 100644 --- a/drivers/dma-buf/dma-fence.c +++ b/drivers/dma-buf/dma-fence.c @@ -285,19 +285,18 @@ void dma_fence_enable_sw_signaling(struct dma_fence *fence) { unsigned long flags; + if (test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &fence->flags)) + return; + + spin_lock_irqsave(fence->lock, flags); if (!test_and_set_bit(DMA_FENCE_FLAG_ENABLE_SIGNAL_BIT, &fence->flags) && - !test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &fence->flags) && fence->ops->enable_signaling) { trace_dma_fence_enable_signal(fence); - - spin_lock_irqsave(fence->lock, flags); - if (!fence->ops->enable_signaling(fence)) dma_fence_signal_locked(fence); - - spin_unlock_irqrestore(fence->lock, flags); } + spin_unlock_irqrestore(fence->lock, flags); } EXPORT_SYMBOL(dma_fence_enable_sw_signaling); -- 2.23.0

1 0

Recall: [PATCH v2 11/14] drm/amd/display: Validate DSC caps on MST endpoints
by Liu, Wenjing 21 Aug '19

21 Aug '19

Liu, Wenjing would like to recall the message, "[PATCH v2 11/14] drm/amd/display: Validate DSC caps on MST endpoints".

1 0

[Bug 110659] pageflipping seems to cause jittering on mouse input when running Hitman 2 in Wine/DXVK with amdgpu.dc=1
by bugzilla-daemon＠freedesktop.org 21 Aug '19

21 Aug '19

https://bugs.freedesktop.org/show_bug.cgi?id=110659 --- Comment #37 from Nicholas Kazlauskas <nicholas.kazlauskas(a)amd.com> --- Are you running a color management tool in the background? The difference between my setup and yours is that there isn't anything locking the connectors hundreds of times per second and performing full updates. This is confirmed in your log by lines like the following: [ 2788.165907] [drm:drm_atomic_add_affected_connectors [drm]] Adding all current connectors for [CRTC:47:crtc-0] to 00000000acb155e9 -- You are receiving this mail because: You are the assignee for the bug.

1 0

[PATCH v3 hmm 00/11] Add mmu_notifier_get/put for managing mmu notifier registrations
by Jason Gunthorpe 21 Aug '19

21 Aug '19

From: Jason Gunthorpe <jgg(a)mellanox.com> This series introduces a new registration flow for mmu_notifiers based on the idea that the user would like to get a single refcounted piece of memory for a mm, keyed to its use. For instance many users of mmu_notifiers use an interval tree or similar to dispatch notifications to some object. There are many objects but only one notifier subscription per mm holding the tree. Of the 12 places that call mmu_notifier_register: - 7 are maintaining some kind of obvious mapping of mm_struct to mmu_notifier registration, ie in some linked list or hash table. Of the 7 this series converts 4 (gru, hmm, RDMA, radeon) - 3 (hfi1, gntdev, vhost) are registering multiple notifiers, but each one immediately does some VA range filtering, ie with an interval tree. These would be better with a global subsystem-wide range filter and could convert to this API. - 2 (kvm, amd_iommu) are deliberately using a single mm at a time, and really can't use this API. One of the intel-svm's modes is also in this list The 3/7 unconverted drivers are: - intel-svm This driver tracks mm's in a global linked list 'global_svm_list' and would benefit from this API. Its flow is a bit complex, since it also wants a set of non-shared notifiers. - i915_gem_usrptr This driver tracks mm's in a per-device hash table (dev_priv->mm_structs), but only has an optional use of mmu_notifiers. Since it still seems to need the hash table it is difficult to convert. - amdkfd/kfd_process This driver is using a global SRCU hash table to track mm's The control flow here is very complicated and the driver is relying on this hash table to be fast on the ioctl syscall path. It would definitely benefit, but only if the ioctl path didn't need to do the search so often. This series is already entangled with patches in the hmm & RDMA tree and will require some git topic branches for the RDMA ODP stuff. I intend for it to go through the hmm tree. There is a git version here: https://github.com/jgunthorpe/linux/commits/mmu_notifier Which has the required pre-patches for the RDMA ODP conversion that are still being reviewed. Jason Gunthorpe (11): mm/mmu_notifiers: hoist do_mmu_notifier_register down_write to the caller mm/mmu_notifiers: do not speculatively allocate a mmu_notifier_mm mm/mmu_notifiers: add a get/put scheme for the registration misc/sgi-gru: use mmu_notifier_get/put for struct gru_mm_struct hmm: use mmu_notifier_get/put for 'struct hmm' RDMA/odp: use mmu_notifier_get/put for 'struct ib_ucontext_per_mm' RDMA/odp: remove ib_ucontext from ib_umem drm/radeon: use mmu_notifier_get/put for struct radeon_mn drm/amdkfd: fix a use after free race with mmu_notifer unregister drm/amdkfd: use mmu_notifier_put mm/mmu_notifiers: remove unregister_no_release drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 1 + drivers/gpu/drm/amd/amdkfd/kfd_priv.h | 3 - drivers/gpu/drm/amd/amdkfd/kfd_process.c | 88 ++++----- drivers/gpu/drm/nouveau/nouveau_drm.c | 3 + drivers/gpu/drm/radeon/radeon.h | 3 - drivers/gpu/drm/radeon/radeon_device.c | 2 - drivers/gpu/drm/radeon/radeon_drv.c | 2 + drivers/gpu/drm/radeon/radeon_mn.c | 157 ++++------------ drivers/infiniband/core/umem.c | 4 +- drivers/infiniband/core/umem_odp.c | 183 ++++++------------ drivers/infiniband/core/uverbs_cmd.c | 3 - drivers/infiniband/core/uverbs_main.c | 1 + drivers/infiniband/hw/mlx5/main.c | 5 - drivers/misc/sgi-gru/grufile.c | 1 + drivers/misc/sgi-gru/grutables.h | 2 - drivers/misc/sgi-gru/grutlbpurge.c | 84 +++------ include/linux/hmm.h | 12 +- include/linux/mm_types.h | 6 - include/linux/mmu_notifier.h | 40 +++- include/rdma/ib_umem.h | 2 +- include/rdma/ib_umem_odp.h | 10 +- include/rdma/ib_verbs.h | 3 - kernel/fork.c | 1 - mm/hmm.c | 121 +++--------- mm/mmu_notifier.c | 230 +++++++++++++++++------ 25 files changed, 408 insertions(+), 559 deletions(-) -- 2.22.0

5 25

[PATCH v5 0/7] qcom: add OCMEM support
by Brian Masney 21 Aug '19

21 Aug '19

This patch series adds support for Qualcomm's On Chip MEMory (OCMEM) that is needed in order to support some a3xx and a4xx-based GPUs upstream. This is based on Rob Clark's patch series that he submitted in October 2015 and I am resubmitting updated patches with his permission. See the individual patches for the changelog. This was tested with the GPU on a LG Nexus 5 (hammerhead) phone and this will work on other msm8974-based systems. For a summary of what currently works upstream on the Nexus 5, see my status page at https://masneyb.github.io/nexus-5-upstream/. Brian Masney (5): dt-bindings: soc: qcom: add On Chip MEMory (OCMEM) bindings dt-bindings: display: msm: gmu: add optional ocmem property soc: qcom: add OCMEM driver drm/msm/gpu: add ocmem init/cleanup functions ARM: qcom_defconfig: add ocmem support Rob Clark (2): firmware: qcom: scm: add OCMEM lock/unlock interface firmware: qcom: scm: add support to restore secure config to qcm_scm-32 .../devicetree/bindings/display/msm/gmu.txt | 50 ++ .../devicetree/bindings/sram/qcom,ocmem.yaml | 96 ++++ arch/arm/configs/qcom_defconfig | 1 + drivers/firmware/qcom_scm-32.c | 52 ++- drivers/firmware/qcom_scm-64.c | 12 + drivers/firmware/qcom_scm.c | 53 +++ drivers/firmware/qcom_scm.h | 9 + drivers/gpu/drm/msm/Kconfig | 1 + drivers/gpu/drm/msm/adreno/a3xx_gpu.c | 28 +- drivers/gpu/drm/msm/adreno/a3xx_gpu.h | 3 +- drivers/gpu/drm/msm/adreno/a4xx_gpu.c | 25 +- drivers/gpu/drm/msm/adreno/a4xx_gpu.h | 3 +- drivers/gpu/drm/msm/adreno/adreno_gpu.c | 40 ++ drivers/gpu/drm/msm/adreno/adreno_gpu.h | 10 + drivers/soc/qcom/Kconfig | 10 + drivers/soc/qcom/Makefile | 1 + drivers/soc/qcom/ocmem.c | 433 ++++++++++++++++++ include/linux/qcom_scm.h | 26 ++ include/soc/qcom/ocmem.h | 62 +++ 19 files changed, 870 insertions(+), 45 deletions(-) create mode 100644 Documentation/devicetree/bindings/sram/qcom,ocmem.yaml create mode 100644 drivers/soc/qcom/ocmem.c create mode 100644 include/soc/qcom/ocmem.h -- 2.21.0

2 9

[Bug 110659] pageflipping seems to cause jittering on mouse input when running Hitman 2 in Wine/DXVK with amdgpu.dc=1
by bugzilla-daemon＠freedesktop.org 21 Aug '19

21 Aug '19

https://bugs.freedesktop.org/show_bug.cgi?id=110659 --- Comment #36 from tempel.julian(a)gmail.com --- The log now starts at [ 2788.164016], I hope nothing important is cut out. Else I'd have to recheck my log size limits. -- You are receiving this mail because: You are the assignee for the bug.

1 0

[Bug 110659] pageflipping seems to cause jittering on mouse input when running Hitman 2 in Wine/DXVK with amdgpu.dc=1
by bugzilla-daemon＠freedesktop.org 21 Aug '19

21 Aug '19

https://bugs.freedesktop.org/show_bug.cgi?id=110659 --- Comment #35 from tempel.julian(a)gmail.com --- Created attachment 145119 --> https://bugs.freedesktop.org/attachment.cgi?id=145119&action=edit debug dmesg.log -- You are receiving this mail because: You are the assignee for the bug.

1 0

[PATCH] drm/ttm: properly unmap dma page upon failure
by Yu Zhao 21 Aug '19

21 Aug '19

dma_unmap_page() must be called with exactly the same dma address and size returned by dma_map_page(). Otherwise, the function may fail. This is at least the case for debug_dma_unmap_page() and AMD iommu unmap_page callback. Signed-off-by: Yu Zhao <yuzhao(a)google.com> --- drivers/gpu/drm/ttm/ttm_page_alloc.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc.c b/drivers/gpu/drm/ttm/ttm_page_alloc.c index f841accc2c00..c1636595f63c 100644 --- a/drivers/gpu/drm/ttm/ttm_page_alloc.c +++ b/drivers/gpu/drm/ttm/ttm_page_alloc.c @@ -1118,10 +1118,19 @@ int ttm_populate_and_map_pages(struct device *dev, struct ttm_dma_tt *tt, 0, num_pages * PAGE_SIZE, DMA_BIDIRECTIONAL); if (dma_mapping_error(dev, tt->dma_address[i])) { + num_pages = 1; while (i--) { + if (i && tt->ttm.pages[i] - + tt->ttm.pages[i - 1] == 1) { + tt->dma_address[i] = 0; + num_pages++; + continue; + } dma_unmap_page(dev, tt->dma_address[i], - PAGE_SIZE, DMA_BIDIRECTIONAL); + num_pages * PAGE_SIZE, + DMA_BIDIRECTIONAL); tt->dma_address[i] = 0; + num_pages = 1; } ttm_pool_unpopulate(&tt->ttm); return -EFAULT; -- 2.21.0.392.gf8f6787159e-goog

1 2

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

dri-devel August 2019