On Thu, Nov 29, 2018 at 09:24:17AM -0800, Tomasz Figa wrote: > Whether the cache maintenance operation needs to actually do anything > or not is a function of `dev`. We can have some devices that are > coherent with CPU caches, and some that are not, on the same system. Yes, but that part is not decided by these low-level helpers but their callers in the DMA code (or maybe IOMMU code as well in the future). > There is also the use case of using CMA with device-specific pools of > memory reusable by the system when not used by the device and those > would have to somehow get the pool to allocate from, but I wonder if > struct device is the right way to pass such information. I'd see the > pool given explicitly like cma_alloc(struct cma_pool *, size, flags) > and perhaps a wrapper cma_alloc_default(size, flags) that is just a > simple macro calling cma_alloc(&cma_pool_default, size, flags). Yes, the cma APIs have quite a few warts that need addressing. I have a few of those things on my todo list, but help is always welcome.
