Messages in this thread | | | Date | Fri, 11 Oct 2019 02:26:22 -0700 | From | Christoph Hellwig <> | Subject | Re: [PATCH 1/2] habanalabs: support vmalloc memory mapping |
| |
On Fri, Oct 11, 2019 at 12:19:36PM +0300, Oded Gabbay wrote: > We first allocate, using vmalloc_user, a certain memory block that > will be used by the ASIC and the user (ASIC is producer, user is > consumer). > After we use vmalloc_user, we map the *kernel* pointer we got from the > vmalloc_user() to the ASIC MMU. We reuse our driver's generic code > path to map host memory to ASIC MMU and that's why we need the patch > above. The user does NOT send us the pointer. He doesn't have this > pointer. It is internal to the kernel driver. To do this reuse, we > added a call to the is_vmalloc_addr(), so the function will know if it > is called to work on user pointers, or on vmalloc *kernel* pointers.
But the function can't decided that. As I said before you can't just take a value that possibly contains user pointers and call is_vmalloc_addr on it, as kernel and user address can overlap on various architectures.
You need to restructure your code to keep the kernel and user pointer code paths entirely separate.
| |