similar to: [PATCH] drm/nouveau: do not map evicted bo's in nouveau_bo_vma_add

From: Maarten Lankhorst <maarten.lankhorst at canonical.com> Signed-off-by: Maarten Lankhorst <maarten.lankhorst at canonical.com> --- drivers/gpu/drm/nouveau/nouveau_bo.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c index 9ecb874..bb3734d 100644 ---

On Wed, Sep 04, 2013 at 08:59:13AM +0200, Maarten Lankhorst wrote: > > When looking into this bug I noticed that nouveau_bo_vma_add needs to have a check for nvbo->page_shift == vma->vm->vmm->spg_shift, > and only if the check is true it should map the page in TTM_PL_TT. Patch below. > Should probably also be cc'd to stable. > How about this patch? Is it ready to

Op 04-09-13 05:41, Ben Skeggs schreef: > On Thu, Aug 22, 2013 at 5:12 PM, Maarten Lankhorst > <maarten.lankhorst at canonical.com> wrote: >> Op 22-08-13 02:10, Ilia Mirkin schreef: >>> The code expects non-VRAM mem nodes to have a pages list. If that's not >>> set, it will do a null deref down the line. Warn on that condition and >>> return an

move_notify is called by ttm after the the object is idle and about to be destroyed. Clean up the vm list properly in that case. This is not a problem right now, since the list should already be empty, but if it wasn't empty, vm_put was not called which leads to random corruption later. With this fix, nouveau_gem_object_close can be safely changed to a noop, forcing the vm bindings to be

This flag allows userspace to give the kernel a hint that it should use a non-snooped resource. To guarantee coherency at all times mappings into userspace are done write combined, so userspace should avoid reading back from those resources. Signed-off-by: Lucas Stach <dev at lynxeye.de> --- On x86 an optimized userspace can save up on snoop traffic in the system, on ARM the benefits are

On Wed, Aug 28, 2013 at 10:00 AM, Lucas Stach <dev at lynxeye.de> wrote: > This flag allows userspace to give the kernel a hint that it should use > a non-snooped resource. To guarantee coherency at all times mappings > into userspace are done write combined, so userspace should avoid > reading back from those resources. Do any other combinations of cached/uncached and

On Thu, Aug 22, 2013 at 5:12 PM, Maarten Lankhorst <maarten.lankhorst at canonical.com> wrote: > Op 22-08-13 02:10, Ilia Mirkin schreef: >> The code expects non-VRAM mem nodes to have a pages list. If that's not >> set, it will do a null deref down the line. Warn on that condition and >> return an error. >> >> See

Op 11-08-13 10:04, Benjamin Herrenschmidt schreef: > On Sun, 2013-08-11 at 17:06 +1000, Benjamin Herrenschmidt wrote: > >> I think I found at least two cases where "12" was used where it should >> have been PAGE_SHIFT (basically ttm_mem_reg->num_pages). This >> is only the tip of the iceberg, so this isn't a formal patch submission, >> but I would

From: Maarten Lankhorst <maarten.lankhorst at canonical.com> Map the GART to the bar and use that mapping, to hide all the tiling details from users. Signed-off-by: Maarten Lankhorst <maarten.lankhorst at canonical.com> --- drivers/gpu/drm/nouveau/core/subdev/bar/nv50.c | 5 ++++- drivers/gpu/drm/nouveau/core/subdev/bar/nvc0.c | 5 ++++- drivers/gpu/drm/nouveau/nouveau_bo.c

On Sun, Aug 11, 2013 at 7:35 PM, Benjamin Herrenschmidt <benh at kernel.crashing.org> wrote: > On Sun, 2013-08-11 at 11:02 +0200, Maarten Lankhorst wrote: > >> > diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv40.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv40.c >> > index 5c7433d..c314a5f 100644 >> > ---

nv50_bo_move_m2mf might copy to tiled gart, in which case linear copy is not appropriate. --- drivers/gpu/drm/nouveau/nouveau_bo.c | 42 ++++++++++++++++++++---------------- 1 file changed, 24 insertions(+), 18 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c index 88f0c45..0daf3f0 100644 --- a/drivers/gpu/drm/nouveau/nouveau_bo.c +++

Hi folks ! So I've been trying to figure out what it would take to make nouveau work properly on architectures where PAGE_SIZE isn't 4k such as most ppc64's. An initial patch from Dave fixed a bogon in nv41.c nv41_vm_map_sg() which was trying to handle the case at that low level, but this isn't enough, and after a bit of digging, I also think that's not the right approach:

Op 11-08-13 07:36, Benjamin Herrenschmidt schreef: > On Sun, 2013-08-11 at 10:41 +1000, Benjamin Herrenschmidt wrote: >> Now, to do that, I need a better understanding of the various things >> in there since I'm not familiar with nouveau at all. What I think I've >> figured out is with a few questions, it would be awesome if you could >> answer them so I can have

On Sun, 2013-08-11 at 17:06 +1000, Benjamin Herrenschmidt wrote: > I think I found at least two cases where "12" was used where it should > have been PAGE_SHIFT (basically ttm_mem_reg->num_pages). This > is only the tip of the iceberg, so this isn't a formal patch submission, > but I would appreciate your thought as to whether the below is correct > (and thus

On Sun, 2013-08-11 at 11:02 +0200, Maarten Lankhorst wrote: > > diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv40.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv40.c > > index 5c7433d..c314a5f 100644 > > --- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv40.c > > +++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv40.c > > @@ -190,8 +190,8 @@

On Fri, Nov 29, 2013 at 4:01 PM, Benjamin Herrenschmidt <benh at kernel.crashing.org> wrote: > On Thu, 2013-08-29 at 16:49 +1000, Ben Skeggs wrote: > >> > Additionally the current code is broken in that the upper layer in >> > vm/base.c doesn't increment "pte" by the right amount. >> > >> > Now, if those two assertions can be made

On Thu, 2013-08-29 at 16:49 +1000, Ben Skeggs wrote: > > Additionally the current code is broken in that the upper layer in > > vm/base.c doesn't increment "pte" by the right amount. > > > > Now, if those two assertions can be made always true: > > > > - Those two functions (map_sg and map_sg_table) never deal with the > > "big"

On Sun, 2013-08-11 at 10:41 +1000, Benjamin Herrenschmidt wrote: > Now, to do that, I need a better understanding of the various things > in there since I'm not familiar with nouveau at all. What I think I've > figured out is with a few questions, it would be awesome if you could > answer them so I can have a shot at fixing it all :-) Ok, a few more questions :-) - in struct

From: Maarten Lankhorst <maarten.lankhorst at canonical.com> Commit de7b7d59d54852c introduced tiled GART, but a linear copy is still performed. This may result in errors on eviction, fix it by checking tiling from memtype. Signed-off-by: Maarten Lankhorst <maarten.lankhorst at canonical.com> Cc: stable at vger.kernel.org #3.10+ --- drivers/gpu/drm/nouveau/nouveau_bo.c | 33

Hi everyone, I have been studing the VM code in order to come with a better implementation of my fix for large pages usage on GK20A (https://lkml.org/lkml/2014/6/3/375 ). Following some older threads about this code (http://marc.info/?l=dri-devel&m=137618326022935 ) also left me kind of puzzled, and I wonder if we could not simplify and improve things there. It seems like arbitrary