search for: io_reserve_lru

...ed... In a "manager" kind of way, despite being neither a manager nor an insider or active contributor. ;-) On 24/01/2020 14:30, Christian K?nig wrote: > From: Christian K?nig <ckoenig.leichtzumerken at gmail.com> > > While working on TTM cleanups I've found that the io_reserve_lru used by > Nouveau is actually not working at all. What made you conclude this mechanism doesn't work? Does this imply that Nouveau is broken? If so: Does the migration of this code from TTM to Nouveau mean that Nouveau remains broken, but it's now simply no longer your problem? And w...

From: Christian K?nig <ckoenig.leichtzumerken at gmail.com> While working on TTM cleanups I've found that the io_reserve_lru used by Nouveau is actually not working at all. In general we should remove driver specific handling from the memory management, so this patch moves the io_reserve_lru handling into Nouveau instead. The patch should be functional correct, but is only compile tested! v2: don't call ttm_bo_unm...

While working on TTM cleanups I've found that the io_reserve_lru used by Nouveau is actually not working at all. In general we should remove driver specific handling from the memory management, so this patch moves the io_reserve_lru handling into Nouveau instead. v2: don't call ttm_bo_unmap_virtual in nouveau_ttm_io_mem_reserve v3: rebased and use both bas...

On Mon, Sep 30, 2019 at 03:12:53PM +0200, Christian König wrote: > While working on TTM cleanups I've found that the io_reserve_lru used by > Nouveau is actually not working at all. > > In general we should remove driver specific handling from the memory > management, so this patch moves the io_reserve_lru handling into Nouveau > instead. > > The patch should be functional correct, but is only compile tes...

On Sat, 25 Jan 2020 at 00:30, Christian K?nig <ckoenig.leichtzumerken at gmail.com> wrote: > > From: Christian K?nig <ckoenig.leichtzumerken at gmail.com> > > While working on TTM cleanups I've found that the io_reserve_lru used by > Nouveau is actually not working at all. > > In general we should remove driver specific handling from the memory > management, so this patch moves the io_reserve_lru handling into Nouveau > instead. > > The patch should be functional correct, but is only compile teste...

Hi guys, while working through more old TTM functionality I stumbled over the io_reserve_lru. Basic idea is that when this flag is set the driver->io_mem_reserve() callback can return -EAGAIN resulting in unmapping of other BOs. But Nouveau doesn't seem to return -EAGAIN in the call path of io_mem_reserve anywhere. So is this unused or am I missing something? Regards, Christia...

While working on TTM cleanups I've found that the io_reserve_lru used by Nouveau is actually not working at all. In general we should remove driver specific handling from the memory management, so this patch moves the io_reserve_lru handling into Nouveau instead. The patch should be functional correct, but is only compile tested! Signed-off-by: Christian Köni...

Hi guys, I've already send this out in September last year, but only got a response from Daniel. Could you guys please test this and tell me what you think about it? Basically I'm trying to remove all driver specific features from TTM which don't need to be inside the framework. Thanks, Christian.

Just a gentle ping on this. Already got the Acked-by from Daniel, but I need some of the nouveau guys to test this since I can only compile test it. Regards, Christian.

Hi guys, so I got some hardware and tested this and after hammering out tons of typos it now seems to work fine. Could you give it more testing? Thanks in advance, Christian

...reserved_count = 0; mem.bus.base = 0; mem.bus.offset = 0; mem.bus.addr = NULL; @@ -1238,7 +1227,6 @@ int ttm_bo_init_reserved(struct ttm_bo_device *bdev, INIT_LIST_HEAD(&bo->lru); INIT_LIST_HEAD(&bo->ddestroy); INIT_LIST_HEAD(&bo->swap); - INIT_LIST_HEAD(&bo->io_reserve_lru); bo->bdev = bdev; bo->type = type; bo->num_pages = num_pages; @@ -1247,8 +1235,6 @@ int ttm_bo_init_reserved(struct ttm_bo_device *bdev, bo->mem.num_pages = bo->num_pages; bo->mem.mm_node = NULL; bo->mem.page_alignment = page_alignment; - bo->mem.bus.io_reserved...

Hello Christian K?nig, The patch 141b15e59175: "drm/nouveau: move io_reserve_lru handling into the driver v5" from Aug 21, 2020, leads to the following static checker warning: drivers/gpu/drm/nouveau/nouveau_ttm.c:148 nouveau_ttm_fault() warn: inconsistent returns '*bo->base.resv'. drivers/gpu/drm/nouveau/nouveau_ttm.c 126 static vm_fault_t nouveau_ttm_f...

Implementing those is completely unecessary. Signed-off-by: Christian K?nig <christian.koenig at amd.com> --- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 5 ----- drivers/gpu/drm/drm_gem_vram_helper.c | 5 ----- drivers/gpu/drm/qxl/qxl_ttm.c | 6 ------ drivers/gpu/drm/radeon/radeon_ttm.c | 5 ----- drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c | 11 -----------

We are trying to remove the io_lru handling and depend on zero init base, offset and addr here. v2: init addr as well Signed-off-by: Christian K?nig <christian.koenig at amd.com> --- drivers/gpu/drm/ttm/ttm_bo.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c index e3931e515906..772c640a6046 100644 ---

On Tue, 24 Sep 2019 at 22:19, Christian König <ckoenig.leichtzumerken at gmail.com> wrote: > > Hi guys, > > while working through more old TTM functionality I stumbled over the > io_reserve_lru. > > Basic idea is that when this flag is set the driver->io_mem_reserve() > callback can return -EAGAIN resulting in unmapping of other BOs. > > But Nouveau doesn't seem to return -EAGAIN in the call path of > io_mem_reserve anywhere. I believe this is a bug in Nouveau. W...

...9 at 5:44 PM Ben Skeggs <skeggsb at gmail.com> wrote: > > On Tue, 24 Sep 2019 at 22:19, Christian König > <ckoenig.leichtzumerken at gmail.com> wrote: > > > > Hi guys, > > > > while working through more old TTM functionality I stumbled over the > > io_reserve_lru. > > > > Basic idea is that when this flag is set the driver->io_mem_reserve() > > callback can return -EAGAIN resulting in unmapping of other BOs. > > > > But Nouveau doesn't seem to return -EAGAIN in the call path of > > io_mem_reserve anywhere. > I b...

Hi guys, I already tried this a few month ago, but since I don't have NVidia hardware its rather hard to test for me (need to get some ordered). Dave brought up the topic that we should probably try to move the handling into Nouveau once more, so I tried to fix the problem Ben reported and rebased on top of current drm-misc-next. Dave can you test this? At least in theory the approach

...struct vm_area_struct *vma = vmf->vma; >> + struct ttm_buffer_object *bo = vma->vm_private_data; >> + pgprot_t prot; >> + vm_fault_t ret; >> + >> + ret = ttm_bo_vm_reserve(bo, vmf); >> + if (ret) >> + return ret; >> + >> + nouveau_bo_del_io_reserve_lru(bo); > Isn't this opening up a can of worms (in theory) where a lot of concurrent > faults will all fail because they're all removed themselves from the lru, > so can't see anyone else to throw out? > > Same problem really that ttm had (well still has I think for !amdgpu)...

...rm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c index 3d2617dd63e3..59f1b979b0da 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_util.c +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c @@ -510,7 +510,7 @@ static int ttm_buffer_object_transfer(struct ttm_buffer_object *bo, INIT_LIST_HEAD(&fbo->base.io_reserve_lru); mutex_init(&fbo->base.wu_mutex); fbo->base.moving = NULL; - drm_vma_node_reset(&fbo->base.vma_node); + drm_vma_node_reset(&fbo->base.base.vma_node); atomic_set(&fbo->base.cpu_writers, 0); kref_init(&fbo->base.list_kref); diff --git a/drivers/gpu/drm...

...rm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c index 05fbcaf6a3f2..f5009c1b6a9c 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_util.c +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c @@ -510,7 +510,7 @@ static int ttm_buffer_object_transfer(struct ttm_buffer_object *bo, INIT_LIST_HEAD(&fbo->base.io_reserve_lru); mutex_init(&fbo->base.wu_mutex); fbo->base.moving = NULL; - drm_vma_node_reset(&fbo->base.vma_node); + drm_vma_node_reset(&fbo->base.base.vma_node); atomic_set(&fbo->base.cpu_writers, 0); kref_init(&fbo->base.list_kref); diff --git a/drivers/gpu/drm...