search for: gk20a_instobj_release_dma

When using the DMA-API for instmem, we may obtain a write-combined mapping. For such cases, add a write barrier in gk20a_instobj_release_dma() to make sure that all writes have reached memory at this time. Signed-off-by: Alexandre Courbot <acourbot at nvidia.com> --- drm/nouveau/nvkm/subdev/instmem/gk20a.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drm/nouveau/nvkm/subdev/instmem/gk20a.c b/drm/nouveau/nvkm/subdev/inst...

...; "this is not going to end well...\n"); goto out; @@ -232,15 +218,25 @@ out: node->use_cpt++; spin_unlock_irqrestore(&imem->lock, flags); - return node->vaddr; + return node->base.vaddr; } static void -gk20a_instobj_release(struct nvkm_memory *memory) +gk20a_instobj_release_dma(struct nvkm_memory *memory) { struct gk20a_instobj *node = gk20a_instobj(memory); struct gk20a_instmem *imem = node->imem; struct nvkm_ltc *ltc = imem->base.subdev.device->ltc; + + nvkm_ltc_invalidate(ltc); +} + +static void +gk20a_instobj_release_iommu(struct nvkm_memory *memory) +...

...t; goto out; > @@ -232,15 +218,25 @@ out: > node->use_cpt++; > spin_unlock_irqrestore(&imem->lock, flags); > > - return node->vaddr; > + return node->base.vaddr; > } > > static void > -gk20a_instobj_release(struct nvkm_memory *memory) > +gk20a_instobj_release_dma(struct nvkm_memory *memory) > { > struct gk20a_instobj *node = gk20a_instobj(memory); > struct gk20a_instmem *imem = node->imem; > struct nvkm_ltc *ltc = imem->base.subdev.device->ltc; > + > + nvkm_ltc_invalidate(ltc); > +} > + > +static void > +gk20a_i...

...m_ltc_invalidate(ltc); } +static u64 +gk20a_instobj_bar2_dma(struct nvkm_memory *memory) +{ + struct gk20a_instobj_dma *iobj = gk20a_instobj_dma(memory); + u64 addr = ~0ULL; + + if (gk20a_instobj_acquire_dma(&iobj->base.memory)) + addr = gk20a_instobj_addr(&iobj->base.memory); + + gk20a_instobj_release_dma(&iobj->base.memory); + + return addr; +} + +static u64 +gk20a_instobj_bar2_iommu(struct nvkm_memory *memory) +{ + struct gk20a_instobj_iommu *iobj = gk20a_instobj_iommu(memory); + u64 addr = ~0ULL; + + if (gk20a_instobj_acquire_iommu(&iobj->base.memory)) + addr = gk20a_instobj_addr(&...

From: Thierry Reding <treding at nvidia.com> Hi Ben, I messed up the ordering of patches in my tree a bit, so these two fixes got separated from the others. I don't consider these particularily urgent because the crash that the first one fixes only happens on gp10b which we don't enable by default yet and the second patch fixes a crash that only happens on module unload (or driver