search for: nvif_notifi

From: Jason Gunthorpe <jgg at mellanox.com> Remove the hmm_mirror object and use the mmu_range_notifier API instead for the range, and use the normal mmu_notifier API for the general invalidation callback. While here re-organize the pagefault path so the locking pattern is clear. nouveau is the only driver that uses a temporary range object and instead forwards nearly every invalidation

On Wed, Aug 01, 2018 at 05:14:58PM -0400, Lyude Paul wrote: > We can't and don't need to try resuming the device from our hotplug > handlers, but hotplug events are generally something we'd like to keep > the device awake for whenever possible. So, grab a PM ref safely in our > hotplug handlers using pm_runtime_get_noresume() and mark the device as > busy once we're

From: Jason Gunthorpe <jgg at mellanox.com> There is no reason to get the invalidate_range_start() callback via an indirection through hmm_mirror, just register a normal notifier directly. Cc: Ben Skeggs <bskeggs at redhat.com> Cc: dri-devel at lists.freedesktop.org Cc: nouveau at lists.freedesktop.org Cc: Ralph Campbell <rcampbell at nvidia.com> Signed-off-by: Jason Gunthorpe

On Wed, Aug 01, 2018 at 05:14:57PM -0400, Lyude Paul wrote: > When we disable hotplugging on the GPU, we need to be able to > synchronize with each connector's hotplug interrupt handler before the > interrupt is finally disabled. This can be a problem however, since > nouveau_connector_detect() currently grabs a runtime power reference > when handling connector probing. This

This is the latest version of https://patchwork.freedesktop.org/series/46815/ With a bunch of fixes to the new fb_helper to prevent it from breaking module loading/unloading with nouveau. Also; lots of documentation fixes and one fix in response to a kbuild bot. Lyude Paul (8): drm/nouveau: Fix bogus drm_kms_helper_poll_enable() placement drm/nouveau: Enable polling even if we have runtime

Hi Jérôme, Ben and Jason, below is a series against the hmm tree which fixes up the mmap_sem locking in nouveau and while at it also removes leftover legacy HMM APIs only used by nouveau.

Update nouveau to only use the mmu interval notifiers. Signed-off-by: Ralph Campbell <rcampbell at nvidia.com> --- drivers/gpu/drm/nouveau/nouveau_svm.c | 313 +++++++++++++++++--------- 1 file changed, 201 insertions(+), 112 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_svm.c b/drivers/gpu/drm/nouveau/nouveau_svm.c index df9bf1fd1bc0..0343e48d41d7 100644 ---

We can't and don't need to try resuming the device from our hotplug handlers, but hotplug events are generally something we'd like to keep the device awake for whenever possible. So, grab a PM ref safely in our hotplug handlers using pm_runtime_get_noresume() and mark the device as busy once we're finished. Signed-off-by: Lyude Paul <lyude at redhat.com> Cc: stable at

We can't and don't need to try resuming the device from our hotplug handlers, but hotplug events are generally something we'd like to keep the device awake for whenever possible. So, grab a PM ref safely in our hotplug handlers using pm_runtime_get_noresume() and mark the device as busy once we're finished. Signed-off-by: Lyude Paul <lyude at redhat.com> Cc: stable at

Like other cases, it should use rcu protected 'chan' rather than 'fence->channel' in nouveau_fence_wait_uevent_handler. Fixes: 0ec5f02f0e2c ("drm/nouveau: prevent stale fence->channel pointers, and protect with rcu") Signed-off-by: YueHaibing <yuehaibing at huawei.com> --- drivers/gpu/drm/nouveau/nouveau_fence.c | 2 +- 1 file changed, 1 insertion(+), 1

When we disable hotplugging on the GPU, we need to be able to synchronize with each connector's hotplug interrupt handler before the interrupt is finally disabled. This can be a problem however, since nouveau_connector_detect() currently grabs a runtime power reference when handling connector probing. This will deadlock the runtime suspend handler like so: [ 861.480896] INFO: task

In order for suspend/resume reprobing to work, we need to be able to perform sideband communications during suspend/resume, along with runtime PM suspend/resume. In order to do so, we also need to make sure that nouveau doesn't bother grabbing a runtime PM reference to do so, since otherwise we'll start deadlocking runtime PM again. Note that we weren't able to do this before, because

In order for suspend/resume reprobing to work, we need to be able to perform sideband communications during suspend/resume, along with runtime PM suspend/resume. In order to do so, we also need to make sure that nouveau doesn't bother grabbing a runtime PM reference to do so, since otherwise we'll start deadlocking runtime PM again. Note that we weren't able to do this before, because

In order for suspend/resume reprobing to work, we need to be able to perform sideband communications during suspend/resume, along with runtime PM suspend/resume. In order to do so, we also need to make sure that nouveau doesn't bother grabbing a runtime PM reference to do so, since otherwise we'll start deadlocking runtime PM again. Note that we weren't able to do this before, because

From: Lyude Paul <lyude at redhat.com> [ Upstream commit 09e530657e1c982d3dbc5e4302bf9207950c3d0a ] In order for suspend/resume reprobing to work, we need to be able to perform sideband communications during suspend/resume, along with runtime PM suspend/resume. In order to do so, we also need to make sure that nouveau doesn't bother grabbing a runtime PM reference to do so, since

The SVM page fault handler groups faults into a range of contiguous virtual addresses and requests hmm_range_fault() to populate and return the page frame number of system memory mapped by the CPU. In preparation for supporting large pages to be mapped by the GPU, process faults one page at a time. In addition, use the hmm_range default_flags to fix a corner case where the input hmm_pfns array is

The SVM page fault handler groups faults into a range of contiguous virtual addresses and requests hmm_range_fault() to populate and return the page frame number of system memory mapped by the CPU. In preparation for supporting large pages to be mapped by the GPU, process faults one page at a time. In addition, use the hmm_range default_flags to fix a corner case where the input hmm_pfns array is

These two functions are marked as a legacy APIs to get rid of, but seem to suit the current nouveau flow. Move it to the only user in preparation for fixing a locking bug involving caller and callee. All comments referring to the old API have been removed as this now is a driver private helper. Signed-off-by: Christoph Hellwig <hch at lst.de> --- drivers/gpu/drm/nouveau/nouveau_svm.c | 45

hmm_vma_fault is marked as a legacy API to get rid of, but quite suites the current nouvea flow. Move it to the only user in preparation for fixing a locking bug involving caller and callee. Signed-off-by: Christoph Hellwig <hch at lst.de> --- drivers/gpu/drm/nouveau/nouveau_svm.c | 54 ++++++++++++++++++++++++++- include/linux/hmm.h | 54 --------------------------- 2

Hi Jérôme, Ben and Jason, below is a series against the hmm tree which fixes up the mmap_sem locking in nouveau and while at it also removes leftover legacy HMM APIs only used by nouveau. Changes since v1: - don't return the valid state from hmm_range_unregister - additional nouveau cleanups