On Fri, Mar 10, 2017 at 7:40 AM, Daniel Vetter daniel@ffwll.ch wrote:
On Fri, Mar 10, 2017 at 10:31:13AM +0000, Brian Starkey wrote:
Hi,
On Thu, Mar 09, 2017 at 09:38:49AM -0800, Laura Abbott wrote:
On 03/09/2017 02:00 AM, Benjamin Gaignard wrote:
[snip]
For me those patches are going in the right direction.
I still have few questions:
- since alignment management has been remove from ion-core, should it
be also removed from ioctl structure ?
Yes, I think I'm going to go with the suggestion to fixup the ABI so we don't need the compat layer and as part of that I'm also dropping the align argument.
Is the only motivation for removing the alignment parameter that no-one got around to using it for something useful yet? The original comment was true - different devices do have different alignment requirements.
Better alignment can help SMMUs use larger blocks when mapping, reducing TLB pressure and the chance of a page table walk causing display underruns.
Extending ioctl uapi is easy, trying to get rid of bad uapi is much harder. Given that right now we don't have an ion allocator that does alignment I think removing it makes sense. And if we go with lots of heaps, we might as well have an ion heap per alignment that your hw needs, so there's different ways to implement this in the future.
slight correction: if you plan ahead (and do things like zero init if userspace passes in a smaller ioctl struct like drm_ioctl does), extending ioctl uapi is easy.. might be something worth fixing from the get-go..
BR, -R
At least from the unix device memory allocator pov it's probably simpler to encode stuff like this into the heap name, instead of having to pass heap + list of additional properties/constraints.
-Daniel
Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch _______________________________________________ dri-devel mailing list dri-devel@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/dri-devel