search for: gk20a_instobj_bar2_iommu

...y) +{ + 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(&iobj->base.memory); + + gk20a_instobj_release_iommu(&iobj->base.memory); +...

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