search for: nouveau_bo_fixup_align

...+++----------------- 1 file changed, 21 insertions(+), 19 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c index bb3734d..635a192 100644 --- a/drivers/gpu/drm/nouveau/nouveau_bo.c +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c @@ -160,24 +160,20 @@ nouveau_bo_fixup_align(struct nouveau_bo *nvbo, u32 flags, struct nouveau_drm *drm = nouveau_bdev(nvbo->bo.bdev); struct nouveau_device *device = nv_device(drm->device); - if (device->card_type < NV_50) { - if (nvbo->tile_mode) { - if (device->chipset >= 0x40) { - *align = 65536; - *si...

...> Date: Wed Aug 14 11:00:48 2019 +0200 > > drm/nouveau: Initialize GEM object before TTM object > > breaks nouveau userspace which tries to allocate GEM objects with a > non-page-aligned size. Previously nouveau_gem_new would just call > nouveau_bo_init which would call nouveau_bo_fixup_align before > initializing the GEM object. With this change, it is done after. What > do you think -- OK to just move that bit of logic into the new > nouveau_bo_alloc() (and make size/align be pointers so that they can > be fixed up?) Hi Ilia, sorry, got side-tracked earlier and forgot to...

On Wed, Aug 21, 2019 at 04:33:58PM +1000, Ben Skeggs wrote: > On Wed, 14 Aug 2019 at 20:14, Gerd Hoffmann <kraxel at redhat.com> wrote: > > > > Hi, > > > > > > Changing the order doesn't look hard. Patch attached (untested, have no > > > > test hardware). But maybe I missed some detail ... > > > > > > I came up with

...eau_bo.c | 9 ++++----- 1 files changed, 4 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c index e342a41..9fc4bd6 100644 --- a/drivers/gpu/drm/nouveau/nouveau_bo.c +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c @@ -65,8 +65,9 @@ nouveau_bo_fixup_align(struct drm_device *dev, /* * Some of the tile_flags have a periodic structure of N*4096 bytes, - * align to to that as well as the page size. Overallocate memory to - * avoid corruption of other buffer objects. + * align to to that as well as the page size. Align the size to the + * appr...

...u_bo.c | 10 +++++----- 1 files changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c index db0ed4c..028719f 100644 --- a/drivers/gpu/drm/nouveau/nouveau_bo.c +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c @@ -65,8 +65,10 @@ nouveau_bo_fixup_align(struct drm_device *dev, /* * Some of the tile_flags have a periodic structure of N*4096 bytes, - * align to to that as well as the page size. Overallocate memory to - * avoid corruption of other buffer objects. + * align to to that as well as the page size. Align the size to the + * appr...

From: Thierry Reding <treding at nvidia.com> Hi Ben, these are fixes for a couple of issues that I've been running into when testing on various Tegra boards. The first two patches fix up issues in the fix that I had sent out earlier to fix the regression introduced in drm-misc-next. The first one is critical because it avoids a BUG_ON as reported by Ilia, while the second is less

.../drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c > index f4a2eb9..c5fcbcc 100644 > --- a/drivers/gpu/drm/nouveau/nouveau_bo.c > +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c > @@ -231,6 +231,12 @@ nouveau_bo_new(struct drm_device *dev, int size, int align, > > nouveau_bo_fixup_align(nvbo, flags, &align, &size); > nvbo->bo.mem.num_pages = size >> PAGE_SHIFT; > + > + if (tile_flags & NOUVEAU_GEM_TILE_WCUS) > + nvbo->valid_caching = TTM_PL_FLAG_UNCACHED | TTM_PL_FLAG_WC; > + else > + nvbo-&g...

...| 1); phys += NV41_GART_PAGE; pte += 4; - cnt -= 1; } } } diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c index c0fde6b..16dce89 100644 --- a/drivers/gpu/drm/nouveau/nouveau_bo.c +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c @@ -178,7 +178,7 @@ nouveau_bo_fixup_align(struct nouveau_bo *nvbo, u32 flags, *size = roundup(*size, (1 << nvbo->page_shift)); *align = max((1 << nvbo->page_shift), *align); } - + *align = roundup(*align, PAGE_SIZE); *size = roundup(*size, PAGE_SIZE); } @@ -221,7 +221,7 @@ nouveau_bo_new(struct drm_device *d...

...hift = 12; > > if (drm->client.base.vm) { > > if (!(flags & TTM_PL_FLAG_TT) && size > 256 * 1024) > > - nvbo->page_shift = drm->client.base.vm->vmm->lpg_shift; > > + nvbo->page_shift = lpg_shift; > > } > > > > nouveau_bo_fixup_align(nvbo, flags, &align, &size); > Hm.. I don't know if it will be an issue here. But I'm concerned about the cases where nouveau_vm can end up unaligned. > This will not be an issue for the bar mappings, because they map the entire bo, and bo's are always page aligned. > S...

...ot use big pages at all. In effect, many buffers will be rejected for this reason. A behavior like "if the buffer size of more than 256KB, increase the size of the buffer to the next multiple of 128K and use big pages" would probably yield better results. > } > > nouveau_bo_fixup_align(nvbo, flags, &align, &size); > @@ -1641,6 +1646,10 @@ nouveau_bo_vma_add(struct nouveau_bo *nvbo, struct nvkm_vm *vm, > (nvbo->bo.mem.mem_type == TTM_PL_VRAM || > nvbo->page_shift != vma->vm->mmu->lpg_shift)) > nvkm_vm_map(...

..._bo_new(struct drm_device *dev, int size, int align, nvbo->page_shift = 12; if (drm->client.base.vm) { if (!(flags & TTM_PL_FLAG_TT) && size > 256 * 1024) - nvbo->page_shift = drm->client.base.vm->vmm->lpg_shift; + nvbo->page_shift = lpg_shift; } nouveau_bo_fixup_align(nvbo, flags, &align, &size); diff --git a/drivers/gpu/drm/nouveau/nouveau_sgdma.c b/drivers/gpu/drm/nouveau/nouveau_sgdma.c index ca5492a..494cf88 100644 --- a/drivers/gpu/drm/nouveau/nouveau_sgdma.c +++ b/drivers/gpu/drm/nouveau/nouveau_sgdma.c @@ -31,7 +31,7 @@ nv04_sgdma_bind(struct ttm_...

...uveau_drm.h | 8 ++++++ > 7 files changed, 95 insertions(+), 12 deletions(-) > > diff --git a/drm/nouveau/nouveau_bo.c b/drm/nouveau/nouveau_bo.c > index 6edcce1658b7..2a2ebbeb4fc0 100644 > --- a/drm/nouveau/nouveau_bo.c > +++ b/drm/nouveau/nouveau_bo.c > @@ -178,6 +178,24 @@ nouveau_bo_fixup_align(struct nouveau_bo *nvbo, u32 flags, > *size = roundup(*size, PAGE_SIZE); > } > > +void > +nouveau_bo_update_tiling(struct nouveau_drm *drm, struct nouveau_bo *nvbo, > + struct nvkm_mem *mem) > +{ > + switch (drm->device.info.family) { > + case NV_DEVICE_INFO_V0...

..., int align, > nvbo->page_shift = 12; > if (drm->client.base.vm) { > if (!(flags & TTM_PL_FLAG_TT) && size > 256 * 1024) > - nvbo->page_shift = drm->client.base.vm->vmm->lpg_shift; > + nvbo->page_shift = lpg_shift; > } > > nouveau_bo_fixup_align(nvbo, flags, &align, &size); Hm.. I don't know if it will be an issue here. But I'm concerned about the cases where nouveau_vm can end up unaligned. This will not be an issue for the bar mappings, because they map the entire bo, and bo's are always page aligned. See nouveau_bo_f...

...gt; } > } > } Ack > diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c > index c0fde6b..16dce89 100644 > --- a/drivers/gpu/drm/nouveau/nouveau_bo.c > +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c > @@ -178,7 +178,7 @@ nouveau_bo_fixup_align(struct nouveau_bo *nvbo, u32 flags, > *size = roundup(*size, (1 << nvbo->page_shift)); > *align = max((1 << nvbo->page_shift), *align); > } > - > + *align = roundup(*align, PAGE_SIZE); > *size = roundup(*size...

...ding <treding at nvidia.com> Date: Wed Aug 14 11:00:48 2019 +0200 drm/nouveau: Initialize GEM object before TTM object breaks nouveau userspace which tries to allocate GEM objects with a non-page-aligned size. Previously nouveau_gem_new would just call nouveau_bo_init which would call nouveau_bo_fixup_align before initializing the GEM object. With this change, it is done after. What do you think -- OK to just move that bit of logic into the new nouveau_bo_alloc() (and make size/align be pointers so that they can be fixed up?) Cheers, -ilia

...etion(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c index f4a2eb9..c5fcbcc 100644 --- a/drivers/gpu/drm/nouveau/nouveau_bo.c +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c @@ -231,6 +231,12 @@ nouveau_bo_new(struct drm_device *dev, int size, int align, nouveau_bo_fixup_align(nvbo, flags, &align, &size); nvbo->bo.mem.num_pages = size >> PAGE_SHIFT; + + if (tile_flags & NOUVEAU_GEM_TILE_WCUS) + nvbo->valid_caching = TTM_PL_FLAG_UNCACHED | TTM_PL_FLAG_WC; + else + nvbo->valid_caching = TTM_PL_MASK_CACHING; + nouveau_bo_placement_set(nvbo,...

...if (!(flags & TTM_PL_FLAG_TT) && size > 256 * 1024) >> > - nvbo->page_shift = drm->client.base.vm->vmm->lpg_shift; >> > + nvbo->page_shift = lpg_shift; >> > } >> > >> > nouveau_bo_fixup_align(nvbo, flags, &align, &size); >> Hm.. I don't know if it will be an issue here. But I'm concerned about the cases where nouveau_vm can end up unaligned. >> This will not be an issue for the bar mappings, because they map the entire bo, and bo's are always page aligned...

...++++++---------------- 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 +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c @@ -176,8 +176,8 @@ nouveau_bo_fixup_align(struct nouveau_bo *nvbo, u32 flags, *size = roundup(*size, 32 * nvbo->tile_mode); } } else { - *size = roundup(*size, (1 << nvbo->page_shift)); - *align = max((1 << nvbo->page_shift), *align); + *align = 1 << nvbo->page_shift; + *size = roundup(*size, *al...

Second version of this patchset addressing Ben's comments and fixing a few extra things. This patchset proposes to introduce a "staging" module option to dynamically enable features (mostly ioctls) that are merged but may be refined before they are declared "stable". The second patch illustrates the use of this staging option with the SET_TILING ioctl, which can be used to

...32c 100644 --- a/drivers/gpu/drm/nouveau/nouveau_bo.c +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c @@ -143,14 +143,6 @@ nouveau_bo_del_ttm(struct ttm_buffer_object *bo) kfree(nvbo); } -static inline u64 -roundup_64(u64 x, u32 y) -{ - x += y - 1; - do_div(x, y); - return x * y; -} - static void nouveau_bo_fixup_align(struct nouveau_bo *nvbo, u32 flags, int *align, u64 *size) diff --git a/fs/xfs/xfs_linux.h b/fs/xfs/xfs_linux.h index edbd5a210df2..13de9d49bd52 100644 --- a/fs/xfs/xfs_linux.h +++ b/fs/xfs/xfs_linux.h @@ -207,13 +207,6 @@ static inline xfs_dev_t linux_to_xfs_dev_t(dev_t dev) #define xfs...