Hi,
On 23/02/16 23:57, Laurent Pinchart wrote:
Hi Tomi,
Thank you for the patch.
Thanks for review, good points. I'll drop this patch from the series for now. The reason being, this issue is rather difficult to reproduce and test, so any changes I do to patch this will take time to verify.
I'll address the comments later.
On Friday 19 February 2016 11:47:40 Tomi Valkeinen wrote:
Errata i878 says that MPU should not be used to access RAM and DMM at the same time. As it's not possible to prevent MPU accessing RAM, we need to access DMM via a proxy.
This patch changes DMM driver to access DMM registers via sDMA. Instead of doing a normal readl/writel call to read/write a register, we use sDMA to copy 4 bytes from/to the DMM registers.
This patch provides only a partial workaround for i878, as not only DMM register reads/writes are affected, but also accesses to the DMM mapped buffers (framebuffers, usually).
Will this patch really improve the situation if the DMM mapping is accessed anyway ?
Depends. For TI environments it helps, as we don't draw with the CPU. If there's drawing done with CPU, the issue may still happen. I have never seen it happen in normal situations, though, but I did manage to trigger it with explicit stress testing.
Also, I don't have any good solution to the framebuffer part... Except doing a memcpy, ruining the performance.
Tomi